@inproceedings{zawad_hybrid_2023,
	address = {Singapore},
	series = {Lecture {Notes} in {Networks} and {Systems}},
	title = {A {Hybrid} {Approach} for {Stress} {Prediction} from {Heart} {Rate} {Variability}},
	isbn = {978-981-19519-1-6},
	doi = {10.1007/978-981-19-5191-6_10},
	abstract = {Stress is a condition that causes a specific physiologicsal response. Heart rate variability (HRV) is a critical aspect in identifying stress. It is crucial for those who want to keep track of their wellness. Currently, numerous research is being conducted on stress prediction from HRV. The existing works in this field cover different data sets to identify stress, where significantly few models can predict stress with high accuracy. This work combines two well-known stress prediction data sets comprising HRV features named WESAD and SWELL-KW to compare twelve classical machine learning models and hybrid models. Finally, it proposes a hybrid stress prediction model that combines Artificial Neural Network (ANN) with Naive Bayes (NB). The proposed model performed auspiciously, having an accuracy of 95.75\% within only 0.80 s. A stress prediction framework is also suggested based on the findings.},
	language = {en},
	booktitle = {Frontiers of {ICT} in {Healthcare}},
	publisher = {Springer Nature},
	author = {Zawad, Md. Rahat Shahriar and Rony, Chowdhury Saleh Ahmed and Haque, Md. Yeaminul and Banna, Md. Hasan Al and Mahmud, Mufti and Kaiser, M. Shamim},
	editor = {Mandal, Jyotsna Kumar and De, Debashis},
	year = {2023},
	keywords = {HRV, Hybrid method, Machine learning, Stress},
	pages = {111--121},
}

Stress is a condition that causes a specific physiologicsal response. Heart rate variability (HRV) is a critical aspect in identifying stress. It is crucial for those who want to keep track of their wellness. Currently, numerous research is being conducted on stress prediction from HRV. The existing works in this field cover different data sets to identify stress, where significantly few models can predict stress with high accuracy. This work combines two well-known stress prediction data sets comprising HRV features named WESAD and SWELL-KW to compare twelve classical machine learning models and hybrid models. Finally, it proposes a hybrid stress prediction model that combines Artificial Neural Network (ANN) with Naive Bayes (NB). The proposed model performed auspiciously, having an accuracy of 95.75% within only 0.80 s. A stress prediction framework is also suggested based on the findings.

@article{ghosh_attention-based_2023,
	title = {An attention-based hybrid architecture with explainability for depressive social media text detection in {Bangla}},
	volume = {213},
	issn = {0957-4174},
	url = {https://www.sciencedirect.com/science/article/pii/S0957417422020255},
	doi = {10.1016/j.eswa.2022.119007},
	abstract = {Mental health has become a major concern in recent years. Social media have been increasingly used as platforms to gain insight into a person’s mental health condition by analysing the posts and comments, which are textual in nature. By analysing these texts, depressive posts can be detected. To facilitate this process, this work presents an attention-based bidirectional Long Short-Term Memory (LSTM)- Convolutional Neural Network (CNN) based model to detect depressive Bangla social media texts, which is lighter and more robust than the conventional models and provides better performance. A dataset containing such Bangla texts was also developed in this work to mitigate the scarcity. Different preprocessing stages were followed, and three embeddings were used in this task. Thanks to the attention mechanism, the proposed model achieved an accuracy of 94.3\% with 92.63\% of sensitivity and 95.12\% of specificity. When tested on other languages, such as English, the proposed model performed remarkably. The robustness and explainability of the proposed model were also discussed in this paper. Additionally, when compared with classical machine learning models, ensemble approaches, transformers, other similar models, and existing architectures, the proposed model outperformed them.},
	language = {en},
	urldate = {2022-11-20},
	journal = {Expert Systems with Applications},
	author = {Ghosh, Tapotosh and Banna, Md. Hasan Al and Nahian, Md. Jaber Al and Uddin, Mohammed Nasir and Kaiser, M. Shamim and Mahmud, Mufti},
	month = mar,
	year = {2023},
	keywords = {Attention, Depression, Mental health, Social media, Suicide},
	pages = {119007},
}

Mental health has become a major concern in recent years. Social media have been increasingly used as platforms to gain insight into a person’s mental health condition by analysing the posts and comments, which are textual in nature. By analysing these texts, depressive posts can be detected. To facilitate this process, this work presents an attention-based bidirectional Long Short-Term Memory (LSTM)- Convolutional Neural Network (CNN) based model to detect depressive Bangla social media texts, which is lighter and more robust than the conventional models and provides better performance. A dataset containing such Bangla texts was also developed in this work to mitigate the scarcity. Different preprocessing stages were followed, and three embeddings were used in this task. Thanks to the attention mechanism, the proposed model achieved an accuracy of 94.3% with 92.63% of sensitivity and 95.12% of specificity. When tested on other languages, such as English, the proposed model performed remarkably. The robustness and explainability of the proposed model were also discussed in this paper. Additionally, when compared with classical machine learning models, ensemble approaches, transformers, other similar models, and existing architectures, the proposed model outperformed them.

@inproceedings{raisa_cyber-physical_2022,
	address = {Singapore},
	series = {Lecture {Notes} in {Networks} and {Systems}},
	title = {A {Cyber}-{Physical} {Fusion} {System} for {Stress} {Detection} {Using} {Multimodal} and {Social} {Media} {Data}},
	isbn = {978-981-19244-5-3},
	doi = {10.1007/978-981-19-2445-3_43},
	abstract = {Stress is identified as one of the most common human responses to physical, mental or emotional pressure. Long-term stress can cause cardiovascular diseases, depression, anxiety and even death. Stress can be recognized by observing physiological activity data and social media posts of individuals. This explorative study is performed to find the effect of fusion of physiological measurement with social media textual posts for classifying stress. The proposed model implements Heart Rate Variability (HRV) datasets as physiological stress datasets and social media post dataset as textual dataset. At first the datasets were individually implemented with different machine learning models to find the best fit model. It is shown that Random Forest showed the best classification result with an accuracy of 99.85\% for the HRV data and the Logistic Regression model performed best for the social media data with an accuracy of 96.4\%. The two models are combined using fuzzy fusion technique with an accuracy of 98\%. To our knowledge, the fuzzy fusion technique for combining physiological and textual data is a novel approach for stress detection with significant applicability.},
	language = {en},
	booktitle = {Proceedings of {International} {Conference} on {Fourth} {Industrial} {Revolution} and {Beyond} 2021},
	publisher = {Springer Nature},
	author = {Raisa, Jasiya Fairiz and Jahan, Sobhana and Kaiser, M. Shamim},
	editor = {Hossain, Sazzad and Hossain, Md. Shahadat and Kaiser, M. Shamim and Majumder, Satya Prasad and Ray, Kanad},
	year = {2022},
	keywords = {Artificial intelligence, Fusion, Heart rate variability, Natural language processing, Stress detection},
	pages = {615--627},
}

Stress is identified as one of the most common human responses to physical, mental or emotional pressure. Long-term stress can cause cardiovascular diseases, depression, anxiety and even death. Stress can be recognized by observing physiological activity data and social media posts of individuals. This explorative study is performed to find the effect of fusion of physiological measurement with social media textual posts for classifying stress. The proposed model implements Heart Rate Variability (HRV) datasets as physiological stress datasets and social media post dataset as textual dataset. At first the datasets were individually implemented with different machine learning models to find the best fit model. It is shown that Random Forest showed the best classification result with an accuracy of 99.85% for the HRV data and the Logistic Regression model performed best for the social media data with an accuracy of 96.4%. The two models are combined using fuzzy fusion technique with an accuracy of 98%. To our knowledge, the fuzzy fusion technique for combining physiological and textual data is a novel approach for stress detection with significant applicability.

@incollection{ahmed_computational_2022,
	address = {Singapore},
	series = {Brain {Informatics} and {Health}},
	title = {Computational {Intelligence} in {Detection} and {Support} of {Autism} {Spectrum} {Disorder}},
	isbn = {978-981-19527-2-2},
	url = {https://doi.org/10.1007/978-981-19-5272-2_9},
	abstract = {Autism Spectrum Disorder (ASD) refers to a spectrum of conditions characterised mainly by impairments in social interaction, speech and nonverbal communication, and restricted—repetitive behaviour. The lack of physical testing, done primarily via behaviour analysis, makes ASD diagnosis more difficult. The emergence of Computational Intelligence techniques has resulted in the development of a variety of fast and early ASD diagnosis methods based on multiple input modalities. The premise of computational intelligence (CI) and its efficiency in detecting and monitoring ASD has been examined in this chapter, which has recently advanced. Two types of studies have been discussed in this article. Several aspects of ASD screening, including questionnaires, eye scan paths, movement tracking, behavioural analysis from video, brain scans, and more, have been discussed using machine learning and deep learning. Secondly, ASD detection and monitoring applications have been studied extensively in the past year, with significant advances. Finally, a discussion has been made on the challenges faced in ASD detection and management with future research scopes.},
	language = {en},
	urldate = {2022-11-20},
	booktitle = {Artificial {Intelligence} in {Healthcare}: {Recent} {Applications} and {Developments}},
	publisher = {Springer Nature},
	author = {Ahmed, Sabbir and Nur, Silvia Binte and Farhad Hossain, Md. and Kaiser, M. Shamim and Mahmud, Mufti and Chen, Tianhua},
	editor = {Chen, Tianhua and Carter, Jenny and Mahmud, Mufti and Khuman, Arjab Singh},
	year = {2022},
	doi = {10.1007/978-981-19-5272-2_9},
	pages = {179--197},
}

Autism Spectrum Disorder (ASD) refers to a spectrum of conditions characterised mainly by impairments in social interaction, speech and nonverbal communication, and restricted—repetitive behaviour. The lack of physical testing, done primarily via behaviour analysis, makes ASD diagnosis more difficult. The emergence of Computational Intelligence techniques has resulted in the development of a variety of fast and early ASD diagnosis methods based on multiple input modalities. The premise of computational intelligence (CI) and its efficiency in detecting and monitoring ASD has been examined in this chapter, which has recently advanced. Two types of studies have been discussed in this article. Several aspects of ASD screening, including questionnaires, eye scan paths, movement tracking, behavioural analysis from video, brain scans, and more, have been discussed using machine learning and deep learning. Secondly, ASD detection and monitoring applications have been studied extensively in the past year, with significant advances. Finally, a discussion has been made on the challenges faced in ASD detection and management with future research scopes.

@incollection{rahat_shahriar_zawad_computational_2022,
	address = {Singapore},
	series = {Brain {Informatics} and {Health}},
	title = {Computational {Intelligence} in {Depression} {Detection}},
	isbn = {978-981-19527-2-2},
	url = {https://doi.org/10.1007/978-981-19-5272-2_7},
	abstract = {According to the World Health Organisation, depression is the prime contributor to mental disability worldwide. Depression is a severe threat to people’s public and private lives because it causes catastrophic alterations in feelings and emotions. The recent rise in mental health issues and major depressive disorder has spurred many depression detection studies. Computational intelligence-based depression detection has piqued the scientific community’s interest due to its increased efficiency and low mistake rate. This work presented a systematic review of recent works on computational intelligence-based depression detection based on their detection models, preprocessing, and data types. Discussing the findings, frameworks for social media, smartphone data, image/video and biosignal based depression detection were suggested. Finally, challenges and future research scopes in depression detection using computational intelligence have also been discussed.},
	language = {en},
	urldate = {2022-11-20},
	booktitle = {Artificial {Intelligence} in {Healthcare}: {Recent} {Applications} and {Developments}},
	publisher = {Springer Nature},
	author = {Rahat Shahriar Zawad, Md. and Yeaminul Haque, Md. and Kaiser, M. Shamim and Mahmud, Mufti and Chen, Tianhua},
	editor = {Chen, Tianhua and Carter, Jenny and Mahmud, Mufti and Khuman, Arjab Singh},
	year = {2022},
	doi = {10.1007/978-981-19-5272-2_7},
	pages = {145--163},
}

According to the World Health Organisation, depression is the prime contributor to mental disability worldwide. Depression is a severe threat to people’s public and private lives because it causes catastrophic alterations in feelings and emotions. The recent rise in mental health issues and major depressive disorder has spurred many depression detection studies. Computational intelligence-based depression detection has piqued the scientific community’s interest due to its increased efficiency and low mistake rate. This work presented a systematic review of recent works on computational intelligence-based depression detection based on their detection models, preprocessing, and data types. Discussing the findings, frameworks for social media, smartphone data, image/video and biosignal based depression detection were suggested. Finally, challenges and future research scopes in depression detection using computational intelligence have also been discussed.

@misc{jahan_explainable_2022,
	title = {Explainable {AI}-based {Alzheimer}’s {Prediction} and {Management} {Using} {Multimodal} {Data}},
	url = {https://www.preprints.org/manuscript/202203.0214/v1},
	doi = {10.20944/preprints202203.0214.v1},
	abstract = {According to the World Health Organization (WHO), dementia is the seventh leading reason of death among all illnesses and one of the leading causes of disability among the world’s elderly people. Day by day the number of Alzheimer’s patients is raising. Considering the increasing rate and the dangers, Alzheimer’s disease should be diagnosed carefully. Machine learning is a potential technique for Alzheimer’s diagnosis but general users do not trust machine learning models due to the black-box nature. Even, some of those models do not provide the best performance because of using only neuroimaging data. To solve these issues, this paper proposes a novel explainable Alzheimer’s disease prediction model using a multimodal dataset. This approach performs a data-level fusion using clinical data, Freesurfer MRI segmentation data, and psychological data. For Alzheimer’s disease vs cognitively normal prediction, the random forest classifier provides 100\% accuracy. Furthermore, Alzheimer’s disease and non-Alzheimer’s dementia should be classified properly because their symptoms are similar. To the best of our knowledge, we are the first to present a three-class classification on Alzheimer’s disease vs cognitively normal vs non-Alzheimer’s dementia and achieved 99.86\% accuracy using an ensemble model. Besides, a novel Alzheimer’s patient management architecture is also proposed in this work..},
	language = {en},
	urldate = {2022-11-20},
	publisher = {Preprints},
	author = {Jahan, Sobhana and Taher, Kazi Abu and Kaiser, M. Shamim and Mahmud, Mufti and Rahman, Md Sazzadur and Hosen, A. S. M. Sanwar and Ra, In-Ho},
	month = mar,
	year = {2022},
	keywords = {Data-level fusion, Dementia, Machine learning},
}

According to the World Health Organization (WHO), dementia is the seventh leading reason of death among all illnesses and one of the leading causes of disability among the world’s elderly people. Day by day the number of Alzheimer’s patients is raising. Considering the increasing rate and the dangers, Alzheimer’s disease should be diagnosed carefully. Machine learning is a potential technique for Alzheimer’s diagnosis but general users do not trust machine learning models due to the black-box nature. Even, some of those models do not provide the best performance because of using only neuroimaging data. To solve these issues, this paper proposes a novel explainable Alzheimer’s disease prediction model using a multimodal dataset. This approach performs a data-level fusion using clinical data, Freesurfer MRI segmentation data, and psychological data. For Alzheimer’s disease vs cognitively normal prediction, the random forest classifier provides 100% accuracy. Furthermore, Alzheimer’s disease and non-Alzheimer’s dementia should be classified properly because their symptoms are similar. To the best of our knowledge, we are the first to present a three-class classification on Alzheimer’s disease vs cognitively normal vs non-Alzheimer’s dementia and achieved 99.86% accuracy using an ensemble model. Besides, a novel Alzheimer’s patient management architecture is also proposed in this work..

@article{fabietti_abot_2022,
	title = {{ABOT}: an open-source online benchmarking tool for machine learning-based artefact detection and removal methods from neuronal signals},
	volume = {9},
	issn = {2198-4026},
	shorttitle = {{ABOT}},
	url = {https://doi.org/10.1186/s40708-022-00167-3},
	doi = {10.1186/s40708-022-00167-3},
	abstract = {Brain signals are recorded using different techniques to aid an accurate understanding of brain function and to treat its disorders. Untargeted internal and external sources contaminate the acquired signals during the recording process. Often termed as artefacts, these contaminations cause serious hindrances in decoding the recorded signals; hence, they must be removed to facilitate unbiased decision-making for a given investigation. Due to the complex and elusive manifestation of artefacts in neuronal signals, computational techniques serve as powerful tools for their detection and removal. Machine learning (ML) based methods have been successfully applied in this task. Due to ML’s popularity, many articles are published every year, making it challenging to find, compare and select the most appropriate method for a given experiment. To this end, this paper presents ABOT (Artefact removal Benchmarking Online Tool) as an online benchmarking tool which allows users to compare existing ML-driven artefact detection and removal methods from the literature. The characteristics and related information about the existing methods have been compiled as a knowledgebase (KB) and presented through a user-friendly interface with interactive plots and tables for users to search it using several criteria. Key characteristics extracted from over 120 articles from the literature have been used in the KB to help compare the specific ML models. To comply with the FAIR (Findable, Accessible, Interoperable and Reusable) principle, the source code and documentation of the toolbox have been made available via an open-access repository.},
	number = {1},
	urldate = {2022-11-20},
	journal = {Brain Informatics},
	author = {Fabietti, Marcos and Mahmud, Mufti and Lotfi, Ahmad and Kaiser, M. Shamim},
	month = sep,
	year = {2022},
	keywords = {Computational neuroscience, Electrocorticogram, Electroencephalogram, Local field potentials, Magnetoencephalogram, Neuronal spikes},
	pages = {19},
}

Brain signals are recorded using different techniques to aid an accurate understanding of brain function and to treat its disorders. Untargeted internal and external sources contaminate the acquired signals during the recording process. Often termed as artefacts, these contaminations cause serious hindrances in decoding the recorded signals; hence, they must be removed to facilitate unbiased decision-making for a given investigation. Due to the complex and elusive manifestation of artefacts in neuronal signals, computational techniques serve as powerful tools for their detection and removal. Machine learning (ML) based methods have been successfully applied in this task. Due to ML’s popularity, many articles are published every year, making it challenging to find, compare and select the most appropriate method for a given experiment. To this end, this paper presents ABOT (Artefact removal Benchmarking Online Tool) as an online benchmarking tool which allows users to compare existing ML-driven artefact detection and removal methods from the literature. The characteristics and related information about the existing methods have been compiled as a knowledgebase (KB) and presented through a user-friendly interface with interactive plots and tables for users to search it using several criteria. Key characteristics extracted from over 120 articles from the literature have been used in the KB to help compare the specific ML models. To comply with the FAIR (Findable, Accessible, Interoperable and Reusable) principle, the source code and documentation of the toolbox have been made available via an open-access repository.

@book{kaiser_rhythms_2022,
	address = {Singapore},
	series = {Studies in {Rhythm} {Engineering}},
	title = {Rhythms in {Healthcare}},
	isbn = {978-981-19418-8-7 978-981-19418-9-4},
	url = {https://link.springer.com/10.1007/978-981-19-4189-4},
	language = {en},
	urldate = {2022-11-20},
	publisher = {Springer Nature},
	editor = {Kaiser, M. Shamim and Mahmud, Mufti and Al Mamun, Shamim},
	year = {2022},
	doi = {10.1007/978-981-19-4189-4},
	keywords = {Biological Data, Biosignals, Computational Intelligence, Explainable AI, Health Informatics, Machine Learning, Time Series Analysis},
}

@article{sazzad_rosenet_2022,
	title = {{RoseNet}: {Rose} leave dataset for the development of an automation system to recognize the diseases of rose},
	volume = {44},
	issn = {2352-3409},
	shorttitle = {{RoseNet}},
	url = {https://www.data-in-brief.com/article/S2352-3409(22)00691-6/fulltext},
	doi = {10.1016/j.dib.2022.108497},
	language = {English},
	urldate = {2022-11-20},
	journal = {Data in Brief},
	author = {Sazzad, Sadia and Rajbongshi, Aditya and Shakil, Rashiduzzaman and Akter, Bonna and Kaiser, M. Shamim},
	month = oct,
	year = {2022},
	pmid = {35966946},
	note = {Publisher: Elsevier},
	keywords = {Dataset, Feature ranking, Machine learning, Rose leave},
}

@article{chakraborty_guest_2022,
	title = {Guest editorial: intelligent ubiquitous computing and advanced learning systems for biomedical engineering},
	volume = {2022},
	issn = {2051-3305},
	shorttitle = {Guest editorial},
	url = {https://onlinelibrary.wiley.com/doi/abs/10.1049/tje2.12206},
	doi = {10.1049/tje2.12206},
	abstract = {This special issue editorial introduces the latest development in emerging technologies of biomedical engineering, including big medical data, artificial intelligence, cloud/fog computing, federated learning, ubiquitous computing and communication, internet of things, wireless technologies, and security and privacy. In this special issue, nine manuscripts are published related to advanced learning and computing systems for biomedical engineering.},
	language = {en},
	number = {11},
	urldate = {2022-11-20},
	journal = {The Journal of Engineering},
	author = {Chakraborty, Chinmay and Khosravi, Mohammad and Garg, Lalit and Kaiser, M. Shamim and Li, Xingwang and Song, Houbing},
	year = {2022},
	note = {\_eprint: https://onlinelibrary.wiley.com/doi/pdf/10.1049/tje2.12206},
	pages = {1037--1040},
}

This special issue editorial introduces the latest development in emerging technologies of biomedical engineering, including big medical data, artificial intelligence, cloud/fog computing, federated learning, ubiquitous computing and communication, internet of things, wireless technologies, and security and privacy. In this special issue, nine manuscripts are published related to advanced learning and computing systems for biomedical engineering.

@inproceedings{chowdhury_low-cost_2022,
	address = {Singapore},
	series = {Lecture {Notes} in {Networks} and {Systems}},
	title = {Low-{Cost} {Stand}-{Alone} {Smart} {Irrigation} {System}: {A} {Case} {Study}},
	isbn = {9789811675973},
	shorttitle = {Low-{Cost} {Stand}-{Alone} {Smart} {Irrigation} {System}},
	doi = {10.1007/978-981-16-7597-3_28},
	abstract = {Bangladesh is heavily dependent on agriculture for its crop production, food supply, and crop rotation. About 50\% of the population in Bangladesh is working in the agriculture sector; agriculture occupies 70\% of the country’s territory. To ensure a bountiful harvest, a soil condition suitable for cultivation and the judicious use of irrigation is essential. A fuzzy neural network-controlled irrigation controller system was developed using the research presented here. The system comprises a feedback Fuzzy Neural Network (FNN) controller that keeps track of important system measurements using sensors. The controller bases its findings on crop production, which guides it in determining when it is appropriate to irrigate. MATLAB may assign triangular and trapezoidal membership functions to every input variable. This inference engine uses Max-Min methods, which serve to derive the optimum answer for every case. Also, water consumption is lessened, and freshwater supplies are thereby protected. the system is created and tested for plant growth that reduces water usage by about 50–60\% and reduces energy generating costs by the same amount. Improved irrigation management can be achieved when FNN is combined with data logging. By implementing this strategy, the overall energy use, water demand, total energy use, battery, and power control unit expenses can be reduced.},
	language = {en},
	booktitle = {Proceedings of the {Third} {International} {Conference} on {Trends} in {Computational} and {Cognitive} {Engineering}},
	publisher = {Springer Nature},
	author = {Chowdhury, Farzana Haque and Raisa, Roksana Akter and Azad, Md. Sharif Uddin and Kaiser, M. Shamim and Mahmud, Mufti},
	editor = {Kaiser, M. Shamim and Ray, Kanad and Bandyopadhyay, Anirban and Jacob, Kavikumar and Long, Kek Sie},
	year = {2022},
	keywords = {Fuzzy Logic, Irrigation, Micro-controller, Sensors, Smart pump},
	pages = {349--356},
}

Bangladesh is heavily dependent on agriculture for its crop production, food supply, and crop rotation. About 50% of the population in Bangladesh is working in the agriculture sector; agriculture occupies 70% of the country’s territory. To ensure a bountiful harvest, a soil condition suitable for cultivation and the judicious use of irrigation is essential. A fuzzy neural network-controlled irrigation controller system was developed using the research presented here. The system comprises a feedback Fuzzy Neural Network (FNN) controller that keeps track of important system measurements using sensors. The controller bases its findings on crop production, which guides it in determining when it is appropriate to irrigate. MATLAB may assign triangular and trapezoidal membership functions to every input variable. This inference engine uses Max-Min methods, which serve to derive the optimum answer for every case. Also, water consumption is lessened, and freshwater supplies are thereby protected. the system is created and tested for plant growth that reduces water usage by about 50–60% and reduces energy generating costs by the same amount. Improved irrigation management can be achieved when FNN is combined with data logging. By implementing this strategy, the overall energy use, water demand, total energy use, battery, and power control unit expenses can be reduced.

@inproceedings{mahbub_deep_2022,
	address = {Singapore},
	series = {Lecture {Notes} in {Networks} and {Systems}},
	title = {Deep {Neural} {Networks} for {Brain} {Tumor} {Detection} from {MRI} {Images}},
	isbn = {9789811675973},
	doi = {10.1007/978-981-16-7597-3_39},
	abstract = {Brain tumor patients have a significant mortality rate. If the tumors are misdiagnosed, it may result in ineffective medical treatment and reduce their life chances. As the risk of brain tumors increases with age and the world’s population ages, there is an urgent need to develop low-cost, easy-to-use early detection technologies. MRI scans are commonly used to visualize a patient’s brain. Artificial intelligence (AI), deep learning (DL), and its sub-domain have recently reduced the need for human judgment in detecting disorders. DL models are increasingly being employed in traditional supervised learning algorithms due to their inherent advantages of automatically obtaining the required features from images. The detection of brain tumors is one of the most challenging tasks in biomedical imaging. This study is intended to propose a deep neural network (DNN) based solution with a limited number of epochs and parameters. The experiment was conducted on two different datasets, and the proposed DNN obtained 99.22\% accuracy, 98.94\% sensitivity, 99.53\% specificity, 99.57\% precision, and 99.26\% F1-Score for Dataset (D1) and 99.43\% accuracy, 98.86\% sensitivity, 100.0\% specificity, 100.0\% precision, and 99.43\% F1-Score for dataset (D2). The results are comparable with the current state-of-the-art.},
	language = {en},
	booktitle = {Proceedings of the {Third} {International} {Conference} on {Trends} in {Computational} and {Cognitive} {Engineering}},
	publisher = {Springer Nature},
	author = {Mahbub, Md. Kawsher and Biswas, Milon and Miah, Md. Abdul Mozid and Kaiser, M. Shamim},
	editor = {Kaiser, M. Shamim and Ray, Kanad and Bandyopadhyay, Anirban and Jacob, Kavikumar and Long, Kek Sie},
	year = {2022},
	keywords = {Brain tumor, Deep learning, MRI},
	pages = {473--485},
}

Brain tumor patients have a significant mortality rate. If the tumors are misdiagnosed, it may result in ineffective medical treatment and reduce their life chances. As the risk of brain tumors increases with age and the world’s population ages, there is an urgent need to develop low-cost, easy-to-use early detection technologies. MRI scans are commonly used to visualize a patient’s brain. Artificial intelligence (AI), deep learning (DL), and its sub-domain have recently reduced the need for human judgment in detecting disorders. DL models are increasingly being employed in traditional supervised learning algorithms due to their inherent advantages of automatically obtaining the required features from images. The detection of brain tumors is one of the most challenging tasks in biomedical imaging. This study is intended to propose a deep neural network (DNN) based solution with a limited number of epochs and parameters. The experiment was conducted on two different datasets, and the proposed DNN obtained 99.22% accuracy, 98.94% sensitivity, 99.53% specificity, 99.57% precision, and 99.26% F1-Score for Dataset (D1) and 99.43% accuracy, 98.86% sensitivity, 100.0% specificity, 100.0% precision, and 99.43% F1-Score for dataset (D2). The results are comparable with the current state-of-the-art.

@inproceedings{kipli_development_2022,
	address = {Singapore},
	series = {Lecture {Notes} in {Networks} and {Systems}},
	title = {Development of {Mobile} {Application} for {Detection} and {Grading} of {Diabetic} {Retinopathy}},
	isbn = {9789811688263},
	doi = {10.1007/978-981-16-8826-3_29},
	abstract = {The key to preventing blindness caused by diabetic retinopathy (DR) is regular screening and early recognition during its early stages. Currently, DR grading is done manually by ophthalmologists and trained graders where the process is time-consuming. Therefore, this paper aims to develop a mobile app that can provide DR detection and grading without a professional or doctor. The patients will be referred to ophthalmologists if further evaluations are required. This research builds an image classification within a mobile application by using deep learning techniques which utilized the Google AI technologies: Google TensorFlow and Google Cloud Platform (Cloud AutoML and Cloud storage). Image classification is performed in two layers which involve DR detection and grading. A total of 12,062 fundus images are chosen from the dataset collected and undergo image preprocessing. The preprocessed images are used to train the model in TensorFlow and Cloud AutoML, respectively. The model will be implemented into the mobile application after being trained with high accuracy. The final test accuracy for the MobileNet pretrained model is 82.9\%, while averaging precision for the model of Cloud AutoML is 75\%. Further research is required to improve the stability of this algorithm and mobile app for real clinical environment settings.},
	language = {en},
	booktitle = {Proceedings of {Trends} in {Electronics} and {Health} {Informatics}},
	publisher = {Springer},
	author = {Kipli, Kuryati and Hui, Lee Yee and Tajudin, Nurul Mirza Afiqah and Sapawi, Rohana and Sahari, Siti Kudnie and Mat, Dayang Azra Awang and Jalil, M. A. and Ray, Kanad and Shamim Kaiser, M. and Mahmud, Mufti},
	editor = {Kaiser, M. Shamim and Bandyopadhyay, Anirban and Ray, Kanad and Singh, Raghvendra and Nagar, Vishal},
	year = {2022},
	keywords = {Cloud AutoML, Diabetic retinopathy, Google TensorFlow, Google cloud platform, Image classification, Mobile application},
	pages = {339--349},
}

The key to preventing blindness caused by diabetic retinopathy (DR) is regular screening and early recognition during its early stages. Currently, DR grading is done manually by ophthalmologists and trained graders where the process is time-consuming. Therefore, this paper aims to develop a mobile app that can provide DR detection and grading without a professional or doctor. The patients will be referred to ophthalmologists if further evaluations are required. This research builds an image classification within a mobile application by using deep learning techniques which utilized the Google AI technologies: Google TensorFlow and Google Cloud Platform (Cloud AutoML and Cloud storage). Image classification is performed in two layers which involve DR detection and grading. A total of 12,062 fundus images are chosen from the dataset collected and undergo image preprocessing. The preprocessed images are used to train the model in TensorFlow and Cloud AutoML, respectively. The model will be implemented into the mobile application after being trained with high accuracy. The final test accuracy for the MobileNet pretrained model is 82.9%, while averaging precision for the model of Cloud AutoML is 75%. Further research is required to improve the stability of this algorithm and mobile app for real clinical environment settings.

@inproceedings{hasan_implementation_2022,
	address = {Singapore},
	series = {Lecture {Notes} in {Networks} and {Systems}},
	title = {Implementation of {Real}-{Time} {Automated} {Attendance} {System} {Using} {Deep} {Learning}},
	isbn = {9789811675973},
	doi = {10.1007/978-981-16-7597-3_10},
	abstract = {In comparison to general manual operations, contemporary technology always saves time and is often more hassle-free when it comes to verifying human authenticity using their biometrical components. However, despite the fact that face recognition technology has been used in a variety of sectors such as human identification systems, this work is the first to describe how the Face Recognition Technique can be integrated with a deep learning approach. Advanced deep learning techniques can make the attendance system completely automated, highly secure, easier to use, and faster to implement than older systems. Nowadays, the Attendance System is becoming increasingly automated, resulting in time-saving, effective, and beneficial solutions that reduce the burden on administration and organizations. In this paper, we suggest an automatic attendance mechanism that is based on Deep Convolutional Neural Networks (DCNN). SeetaFace, a deep convolutional neural network-based face detection system, is employed in this research effort to detect faces in real-time video capture. This implementation is a VIPLFaceNet implementation, to be more specific. AlexNet, which is also a DCNN, is used for image categorization. The experimental results bring four short similarity situations of the classroom such as absence, delayed appearances, early leave, and unauthorized entry during class or session along with the name, student id, and section and passes this information to the attendance sheet which will evaluate the students/persons in the classroom. This methodology saves time when compared to the traditional method of attendance marking, as well as allows organizations to conduct stress-free observations of students and staff.},
	language = {en},
	booktitle = {Proceedings of the {Third} {International} {Conference} on {Trends} in {Computational} and {Cognitive} {Engineering}},
	publisher = {Springer Nature},
	author = {Hasan, Hafiz Mahdi and Rahman, Md. Mahfujur and Khan, Md. Al-Amin and Meghla, Tamara Islam and Al Mamun, Shamim and Kaiser, M. Shamim},
	editor = {Kaiser, M. Shamim and Ray, Kanad and Bandyopadhyay, Anirban and Jacob, Kavikumar and Long, Kek Sie},
	year = {2022},
	keywords = {AlexNet, Bio-metric identification, DCNN, DLA, Open CV, SeetaFace, VIPLFaceNet},
	pages = {121--132},
}

In comparison to general manual operations, contemporary technology always saves time and is often more hassle-free when it comes to verifying human authenticity using their biometrical components. However, despite the fact that face recognition technology has been used in a variety of sectors such as human identification systems, this work is the first to describe how the Face Recognition Technique can be integrated with a deep learning approach. Advanced deep learning techniques can make the attendance system completely automated, highly secure, easier to use, and faster to implement than older systems. Nowadays, the Attendance System is becoming increasingly automated, resulting in time-saving, effective, and beneficial solutions that reduce the burden on administration and organizations. In this paper, we suggest an automatic attendance mechanism that is based on Deep Convolutional Neural Networks (DCNN). SeetaFace, a deep convolutional neural network-based face detection system, is employed in this research effort to detect faces in real-time video capture. This implementation is a VIPLFaceNet implementation, to be more specific. AlexNet, which is also a DCNN, is used for image categorization. The experimental results bring four short similarity situations of the classroom such as absence, delayed appearances, early leave, and unauthorized entry during class or session along with the name, student id, and section and passes this information to the attendance sheet which will evaluate the students/persons in the classroom. This methodology saves time when compared to the traditional method of attendance marking, as well as allows organizations to conduct stress-free observations of students and staff.

@inproceedings{sharmin_interplanetary_2022,
	address = {Singapore},
	series = {Lecture {Notes} on {Data} {Engineering} and {Communications} {Technologies}},
	title = {{InterPlanetary} {File} {System}-{Based} {Decentralized} and {Secured} {Electronic} {Health} {Record} {System} {Using} {Lightweight} {Algorithm}},
	isbn = {9789811666360},
	doi = {10.1007/978-981-16-6636-0_52},
	abstract = {The electronic health record (EHR) system is a cloud-based patient health record in digital format that often includes contact information about the patient, test reports, medical history, and current and previous prescriptions. However, data breaches in cloud-based EHRs pose significant privacy and security concerns for a variety of health care organizations. Cryptographic techniques currently in use are inadequate to secure EHR data in the cloud from data breaches. Blockchain technology is a new technology that can be used to address security and privacy problems with EHR data on the blockchain in a decentralized manner. We have created a stable decentralized medical blockchain in this paper to address privacy and security concerns when sharing patient data on health care between medical organizations. The health care data is encrypted using Advanced Encryption Standard-based lightweight authenticated encryption algorithm before being uploaded to a cloud-based blockchain and Solidity smart code built on Ethereum to restrict access to EHR data in the cloud. We have used an InterPlanetary file system to store data because it is distributed and ensures record immutability. The medical blockchain also ensures that patient EHR data is interoperable, traceable, and anonymous across organizations. The stable cloud-based blockchain of medical records visualizes patient care data in a distributed and immutable environment with enhanced protection.},
	language = {en},
	booktitle = {Proceedings of the {International} {Conference} on {Big} {Data}, {IoT}, and {Machine} {Learning}},
	publisher = {Springer},
	author = {Sharmin, Sanjida and Sarker, Iqbal H. and Shamim Kaiser, M. and Arefin, Mohammad Shamsul},
	editor = {Arefin, Mohammad Shamsul and Kaiser, M. Shamim and Bandyopadhyay, Anirban and Ahad, Md. Atiqur Rahman and Ray, Kanad},
	year = {2022},
	keywords = {Blockchain, Ethereum, IPFS, Smart contract},
	pages = {691--702},
}

The electronic health record (EHR) system is a cloud-based patient health record in digital format that often includes contact information about the patient, test reports, medical history, and current and previous prescriptions. However, data breaches in cloud-based EHRs pose significant privacy and security concerns for a variety of health care organizations. Cryptographic techniques currently in use are inadequate to secure EHR data in the cloud from data breaches. Blockchain technology is a new technology that can be used to address security and privacy problems with EHR data on the blockchain in a decentralized manner. We have created a stable decentralized medical blockchain in this paper to address privacy and security concerns when sharing patient data on health care between medical organizations. The health care data is encrypted using Advanced Encryption Standard-based lightweight authenticated encryption algorithm before being uploaded to a cloud-based blockchain and Solidity smart code built on Ethereum to restrict access to EHR data in the cloud. We have used an InterPlanetary file system to store data because it is distributed and ensures record immutability. The medical blockchain also ensures that patient EHR data is interoperable, traceable, and anonymous across organizations. The stable cloud-based blockchain of medical records visualizes patient care data in a distributed and immutable environment with enhanced protection.

@inproceedings{arafath_developing_2022,
	address = {Singapore},
	series = {Lecture {Notes} on {Data} {Engineering} and {Communications} {Technologies}},
	title = {Developing a {Framework} for {Credit} {Card} {Fraud} {Detection}},
	isbn = {9789811666360},
	doi = {10.1007/978-981-16-6636-0_48},
	abstract = {Credit card is one of the most popular online or manual payment methods, and credit card fraud is increasing that can cause enormous financial damage. Protective action must therefore be taken to stop the credit card. Several new technologies can be utilized to detect frauduleous transactions based on artificial intelligence, data mining, and machine learning. Several new techniques may be used to identify artificial intelligence, data mining, machine learning, sequence alignment, genetic programming, etc. This article provides a new paradigm for credit card fraud detection based on the characteristics of previous user credit card transactions. Detecting fraudulent purchases from past credit card transactions is a difficult challenge, as it depends on a number of factors, like timing, amount, etc. The data on credit card transactions is rising at a huge rate every day. This constant influx of new data is also challenging to handle and to construct new models to determine if a transaction is fraudulent or not. To do this, we use the PCA data type of user transformation since user data supply sensitive information about users and user transactions. We utilize a tweaked fraud detection model utilizing a RandomizedSearchCV hyperparameter tweaking for detecting fraudulent transactions. Our mechanism provided can determine whether a transaction is fraudulent based on the data patterns of prior transactions of the user.},
	language = {en},
	booktitle = {Proceedings of the {International} {Conference} on {Big} {Data}, {IoT}, and {Machine} {Learning}},
	publisher = {Springer},
	author = {Arafath, Yeasin and Roy, Animesh Chandra and Shamim Kaiser, M. and Arefin, Mohammad Shamsul},
	editor = {Arefin, Mohammad Shamsul and Kaiser, M. Shamim and Bandyopadhyay, Anirban and Ahad, Md. Atiqur Rahman and Ray, Kanad},
	year = {2022},
	keywords = {Decision tree classifier, GaussianNB, K-neighbors classifier, LinearSVC, Logistic regression, Random forest classifier},
	pages = {637--651},
}

Credit card is one of the most popular online or manual payment methods, and credit card fraud is increasing that can cause enormous financial damage. Protective action must therefore be taken to stop the credit card. Several new technologies can be utilized to detect frauduleous transactions based on artificial intelligence, data mining, and machine learning. Several new techniques may be used to identify artificial intelligence, data mining, machine learning, sequence alignment, genetic programming, etc. This article provides a new paradigm for credit card fraud detection based on the characteristics of previous user credit card transactions. Detecting fraudulent purchases from past credit card transactions is a difficult challenge, as it depends on a number of factors, like timing, amount, etc. The data on credit card transactions is rising at a huge rate every day. This constant influx of new data is also challenging to handle and to construct new models to determine if a transaction is fraudulent or not. To do this, we use the PCA data type of user transformation since user data supply sensitive information about users and user transactions. We utilize a tweaked fraud detection model utilizing a RandomizedSearchCV hyperparameter tweaking for detecting fraudulent transactions. Our mechanism provided can determine whether a transaction is fraudulent based on the data patterns of prior transactions of the user.

@inproceedings{salleh_classification_2022,
	address = {Singapore},
	series = {Lecture {Notes} in {Networks} and {Systems}},
	title = {Classification of {ECG} {Ventricular} {Beats} {Assisted} by {Gaussian} {Parameters}’ {Dictionary}},
	isbn = {9789811675973},
	doi = {10.1007/978-981-16-7597-3_44},
	abstract = {Automatic processing and diagnosis of electrocardiogram (ECG) signals remain a very challenging problem, especially with the growth of advanced monitoring technologies. A particular task in ECG processing that has received tremendous attention is to detect and identify pathological heartbeats, e.g., those caused by premature ventricular contraction (PVC). This paper aims to build on the existing methods of heartbeat classification and introduce a new approach to detect ventricular beats using a dictionary of Gaussian-based parameters that model ECG signals. The proposed approach relies on new techniques to segment the stream of ECG signals and automatically cluster the beats for each patient. Two benchmark datasets have been used to evaluate the classification performance, namely, the QTDB and MIT-BIH Arrhythmia databases, based on a single lead short ECG segment. Using the QTDB database, the method achieved the average accuracies of 99.3\% ± 0.7 and 99.4\% ± 0.6\% for lead-1 and lead-2, respectively. On the other hand, identifying ventricular beats in the MIT-BIH Arrhythmia dataset resulted in a sensitivity of 82.8\%, a positive predictivity of 62.0\%, and F1 score of 70.9\%. For non-ventricular beats, the method achieved a sensitivity of 96.0\%, a positive predictivity of 98.6\%, and F1 score of 97.3\%. The proposed technique represents an improvement in the field of ventricular beat classification compared with the conventional methods.},
	language = {en},
	booktitle = {Proceedings of the {Third} {International} {Conference} on {Trends} in {Computational} and {Cognitive} {Engineering}},
	publisher = {Springer Nature},
	author = {Salleh, Sh Hussain and Noman, Fuad and Hussain, Hadri and Ting, Chee-Ming and Hamid, Syed Rasul bin G. Syed and Sh-Hussain, Hadrina and Jalil, M. A. and Zubaidi, A. L. Ahmad and Rizvi, Syed Zuhaib Haider and Kipli, Kuryati and Jacob, Kavikumar and Ray, Kanad and Kaiser, M. Shamim and Mahmud, Mufti and Ali, Jalil},
	editor = {Kaiser, M. Shamim and Ray, Kanad and Bandyopadhyay, Anirban and Jacob, Kavikumar and Long, Kek Sie},
	year = {2022},
	keywords = {Classification, ECG, Gaussian kernels, Segmentation, Template extraction},
	pages = {533--548},
}

Automatic processing and diagnosis of electrocardiogram (ECG) signals remain a very challenging problem, especially with the growth of advanced monitoring technologies. A particular task in ECG processing that has received tremendous attention is to detect and identify pathological heartbeats, e.g., those caused by premature ventricular contraction (PVC). This paper aims to build on the existing methods of heartbeat classification and introduce a new approach to detect ventricular beats using a dictionary of Gaussian-based parameters that model ECG signals. The proposed approach relies on new techniques to segment the stream of ECG signals and automatically cluster the beats for each patient. Two benchmark datasets have been used to evaluate the classification performance, namely, the QTDB and MIT-BIH Arrhythmia databases, based on a single lead short ECG segment. Using the QTDB database, the method achieved the average accuracies of 99.3% ± 0.7 and 99.4% ± 0.6% for lead-1 and lead-2, respectively. On the other hand, identifying ventricular beats in the MIT-BIH Arrhythmia dataset resulted in a sensitivity of 82.8%, a positive predictivity of 62.0%, and F1 score of 70.9%. For non-ventricular beats, the method achieved a sensitivity of 96.0%, a positive predictivity of 98.6%, and F1 score of 97.3%. The proposed technique represents an improvement in the field of ventricular beat classification compared with the conventional methods.

@inproceedings{ritu_facial_2022,
	address = {Singapore},
	series = {Lecture {Notes} in {Networks} and {Systems}},
	title = {Facial {Detection} for {Neonatal} {Infant} {Pain} {Using} {Facial} {Geometry} {Features} and {LBP}},
	isbn = {9789811675973},
	doi = {10.1007/978-981-16-7597-3_42},
	abstract = {Neonatal pain assessment is essential for infants concerning their health issues. There have been several studies to assess the pain of infants using image processing in the field of computer vision. In this paper, we propose a different approach to detect pain in infants that outperforms previous research in this field. We merged a face area-based feature collection method with a local binary pattern (LBP). Moreover, three different machine learning algorithms have been executed to find the best parameter to get a decent accuracy on the iCOPE dataset. The proposed method uses the SVM classifier to achieve 86\% of testing accuracy compared to other methods.},
	language = {en},
	booktitle = {Proceedings of the {Third} {International} {Conference} on {Trends} in {Computational} and {Cognitive} {Engineering}},
	publisher = {Springer Nature},
	author = {Ritu, Jarin Tasnim and Shakil, Md. Shahadat Hossen and Hasan, Md. Nahian Imtiaz and Al Mamun, Shamim and Kaiser, M. Shamim and Mahmud, Mufti},
	editor = {Kaiser, M. Shamim and Ray, Kanad and Bandyopadhyay, Anirban and Jacob, Kavikumar and Long, Kek Sie},
	year = {2022},
	keywords = {Facialgeometry, LBP, Neonates, Pain detection, iCOPE},
	pages = {509--518},
}

Neonatal pain assessment is essential for infants concerning their health issues. There have been several studies to assess the pain of infants using image processing in the field of computer vision. In this paper, we propose a different approach to detect pain in infants that outperforms previous research in this field. We merged a face area-based feature collection method with a local binary pattern (LBP). Moreover, three different machine learning algorithms have been executed to find the best parameter to get a decent accuracy on the iCOPE dataset. The proposed method uses the SVM classifier to achieve 86% of testing accuracy compared to other methods.

@inproceedings{garhwal_drop-shaped_2022,
	address = {Singapore},
	series = {Lecture {Notes} in {Networks} and {Systems}},
	title = {Drop-{Shaped} {Fractal} {Patch} {Antenna} for {THz} {Applications}},
	isbn = {9789811675973},
	doi = {10.1007/978-981-16-7597-3_33},
	abstract = {In this paper, a drop-shaped fractal patch antenna is designed and simulated using Polyamide substrate. The designed antenna is simulated for 4.35–4.42 THz. The designed antenna resonates at 4.4 THz frequency. The maximum gain of 9.34 dBi is achieved. The designed antenna has applications in THz for communication, sensing, and 4.2, 4.3, and 4.4 THz frequency is used for quantum cascade laser. The proposed antenna is designed using CST software.},
	language = {en},
	booktitle = {Proceedings of the {Third} {International} {Conference} on {Trends} in {Computational} and {Cognitive} {Engineering}},
	publisher = {Springer Nature},
	author = {Garhwal, Anita and Jalil, Muhammad Arif and Mahmud, Mufti and Kaiser, M. Shamim and Ray, Kanad and Yupapin, Preecha and Prabpal, P. and Rizvi, Syed Zuhaib Haider and Jacob, Kavikumar and Bandyopadhyay, Anirban and Ali, Jalil},
	editor = {Kaiser, M. Shamim and Ray, Kanad and Bandyopadhyay, Anirban and Jacob, Kavikumar and Long, Kek Sie},
	year = {2022},
	keywords = {Drop shaped, Fractal, Patch, Quantum cascade laser, THz},
	pages = {405--410},
}

In this paper, a drop-shaped fractal patch antenna is designed and simulated using Polyamide substrate. The designed antenna is simulated for 4.35–4.42 THz. The designed antenna resonates at 4.4 THz frequency. The maximum gain of 9.34 dBi is achieved. The designed antenna has applications in THz for communication, sensing, and 4.2, 4.3, and 4.4 THz frequency is used for quantum cascade laser. The proposed antenna is designed using CST software.

@inproceedings{kipli_gsr_2022,
	address = {Singapore},
	series = {Lecture {Notes} in {Networks} and {Systems}},
	title = {{GSR} {Signals} {Features} {Extraction} for {Emotion} {Recognition}},
	isbn = {9789811688263},
	doi = {10.1007/978-981-16-8826-3_28},
	abstract = {Over the years, the recognition of emotion has become more efficient, diverse, and easily accessible. In general, emotion recognition is conducted in four main steps which are signal acquisition, preprocessing, feature extraction, and classification. Galvanic skin response (GSR) is the autonomic activation of sweat glands in the skin when an individual gets triggered through emotional stimulation. The paper provides an overview of emotion recognition, GSR signals, and how GSR signals are analyzed for emotion recognition. The focus of this research is on the performance of feature extraction of GSR signals. Therefore, related sources were identified using combinations of keywords and terms such as feature extraction, emotion recognition, and galvanic skin response. Existing emotion recognition methods were investigated which focused more on the different feature extraction methods. Research conducted has shown that feature extraction method in time–frequency domain has the best accuracy rate overall compared to time domain and frequency domain. Current GSR-based technology also has the potential to be improved more toward the implementation of a more efficient and reliable emotion recognition system.},
	language = {en},
	booktitle = {Proceedings of {Trends} in {Electronics} and {Health} {Informatics}},
	publisher = {Springer},
	author = {Kipli, Kuryati and Latip, Aisya Amelia Abdul and Lias, Kasumawati and Bateni, Norazlina and Yusoff, Salmah Mohamad and Tajudin, Nurul Mirza Afiqah and Jalil, M. A. and Ray, Kanad and Shamim Kaiser, M. and Mahmud, Mufti},
	editor = {Kaiser, M. Shamim and Bandyopadhyay, Anirban and Ray, Kanad and Singh, Raghvendra and Nagar, Vishal},
	year = {2022},
	keywords = {Emotion recognition, Feature extraction, Galvanic skin response},
	pages = {329--338},
}

Over the years, the recognition of emotion has become more efficient, diverse, and easily accessible. In general, emotion recognition is conducted in four main steps which are signal acquisition, preprocessing, feature extraction, and classification. Galvanic skin response (GSR) is the autonomic activation of sweat glands in the skin when an individual gets triggered through emotional stimulation. The paper provides an overview of emotion recognition, GSR signals, and how GSR signals are analyzed for emotion recognition. The focus of this research is on the performance of feature extraction of GSR signals. Therefore, related sources were identified using combinations of keywords and terms such as feature extraction, emotion recognition, and galvanic skin response. Existing emotion recognition methods were investigated which focused more on the different feature extraction methods. Research conducted has shown that feature extraction method in time–frequency domain has the best accuracy rate overall compared to time domain and frequency domain. Current GSR-based technology also has the potential to be improved more toward the implementation of a more efficient and reliable emotion recognition system.

@inproceedings{ahmed_toward_2022,
	address = {Singapore},
	series = {Lecture {Notes} in {Networks} and {Systems}},
	title = {Toward {Machine} {Learning}-{Based} {Psychological} {Assessment} of {Autism} {Spectrum} {Disorders} in {School} and {Community}},
	isbn = {9789811688263},
	doi = {10.1007/978-981-16-8826-3_13},
	abstract = {Ahmed, SabbirHossain, Md. FarhadNur, Silvia BinteShamim Kaiser, M.Mahmud, MuftiThe sensory processing system of the human body is capable of collecting, developing, and integrating information through sensory organs. Sensory impairment has been discovered in children with autism spectrum disorder (ASD). People with ASD are susceptible to hyper/hypo-sensitivity that might cause changes in information management, affect cognitive impairment, and social reactions to everyday events. This article proposed a questionnaire based on ASD symptoms found in previous studies with 82 questions. Following that, a dataset is created by conducting a survey using the questionnaire. Several machine learning models that can identify ASD and its types are also compared. Among the machine learning models, the artificial neural network achieved an accuracy of 89.8\%. Implicit measurements and ecologically sound settings have shown excellent precision in predicting outcomes and the correct classification of populations into categories.},
	language = {en},
	booktitle = {Proceedings of {Trends} in {Electronics} and {Health} {Informatics}},
	publisher = {Springer},
	author = {Ahmed, Sabbir and Hossain, Md. Farhad and Nur, Silvia Binte and Shamim Kaiser, M. and Mahmud, Mufti},
	editor = {Kaiser, M. Shamim and Bandyopadhyay, Anirban and Ray, Kanad and Singh, Raghvendra and Nagar, Vishal},
	year = {2022},
	keywords = {Artificial neural network (ANN), Autism spectrum disorder (ASD), Questionnaire, Random forest (RF), Support vector machine (SVM), k-nearest neighbors (KNN)},
	pages = {139--149},
}

Ahmed, SabbirHossain, Md. FarhadNur, Silvia BinteShamim Kaiser, M.Mahmud, MuftiThe sensory processing system of the human body is capable of collecting, developing, and integrating information through sensory organs. Sensory impairment has been discovered in children with autism spectrum disorder (ASD). People with ASD are susceptible to hyper/hypo-sensitivity that might cause changes in information management, affect cognitive impairment, and social reactions to everyday events. This article proposed a questionnaire based on ASD symptoms found in previous studies with 82 questions. Following that, a dataset is created by conducting a survey using the questionnaire. Several machine learning models that can identify ASD and its types are also compared. Among the machine learning models, the artificial neural network achieved an accuracy of 89.8%. Implicit measurements and ecologically sound settings have shown excellent precision in predicting outcomes and the correct classification of populations into categories.

@inproceedings{abdullah-al-mahmod_toward_2022,
	address = {Singapore},
	series = {Lecture {Notes} in {Networks} and {Systems}},
	title = {Toward {Deep} {Learning}-{Based} {Automated} {Speed} and {Line} {Change} {Detection} {System} in {Perspective} of {Bangladesh}},
	isbn = {9789811688263},
	doi = {10.1007/978-981-16-8826-3_30},
	abstract = {Abdullah-Al-MahmodUsmani, Sabbir AhmedSalam, Mohammad AbdusFoyjul Haque Somrat, Md.Shamim Kaiser, M.In the smart city, major crossing and most part of the road will be under the CCTV surveillance system. This influenced the community to investigate a vision-based speed and line change detection system for traffic management in the city, ensuring both road safety and efficient road design. In this paper, we proposed a deep learning model for detecting vehicle type, speed and abrupt line change using the CCTV footage in real-time. The faster region-based convolutional neural network (fr-CNN) model is chosen in this scenario, which demonstrates amazing performance in object detection. The model is trained and validated using data acquired from a self-created traffic dataset from Dhaka. According to the results of the performance evaluation, the suggested fr-CNN model for moving vehicle status detection system outperforms the mobile-net single-shot multibox detection technique in terms of overall performance.},
	language = {en},
	booktitle = {Proceedings of {Trends} in {Electronics} and {Health} {Informatics}},
	publisher = {Springer},
	author = {{Abdullah-Al-Mahmod} and Usmani, Sabbir Ahmed and Salam, Mohammad Abdus and Foyjul Haque Somrat, Md. and Shamim Kaiser, M.},
	editor = {Kaiser, M. Shamim and Bandyopadhyay, Anirban and Ray, Kanad and Singh, Raghvendra and Nagar, Vishal},
	year = {2022},
	keywords = {CNN, Object detection, Open CV, Road-traffic, Single-shot multibox detection, Tensor flow},
	pages = {351--361},
}

Abdullah-Al-MahmodUsmani, Sabbir AhmedSalam, Mohammad AbdusFoyjul Haque Somrat, Md.Shamim Kaiser, M.In the smart city, major crossing and most part of the road will be under the CCTV surveillance system. This influenced the community to investigate a vision-based speed and line change detection system for traffic management in the city, ensuring both road safety and efficient road design. In this paper, we proposed a deep learning model for detecting vehicle type, speed and abrupt line change using the CCTV footage in real-time. The faster region-based convolutional neural network (fr-CNN) model is chosen in this scenario, which demonstrates amazing performance in object detection. The model is trained and validated using data acquired from a self-created traffic dataset from Dhaka. According to the results of the performance evaluation, the suggested fr-CNN model for moving vehicle status detection system outperforms the mobile-net single-shot multibox detection technique in terms of overall performance.

@inproceedings{kabir_performance_2022,
	address = {Singapore},
	series = {Lecture {Notes} in {Networks} and {Systems}},
	title = {Performance {Analysis} of {MC}-{CDMA}-{Based} {Cognitive} {Radio} {Network} {Under} {Rayleigh} {Fading} {Channel}},
	isbn = {9789811688263},
	doi = {10.1007/978-981-16-8826-3_56},
	abstract = {Higher rate of mobile data traffic demand is increased with the advent of the Internet of things (IoT) and advanced network services (ANS) operators that have begun to develop fifth generation (5G) cellular networks in order to overcome the limitations of the current fourth generation (4G) cellular network. In order to solve the bandwidth scarcity and effective allocation of spectrum resources and also provide higher demand of bandwidth, a multi-carrier code division multiple access (MC-CDMA)-based cognitive radio network (CRN) is proposed and the performance of this system is investigated in this research. MC-CDMA-based CRN improves the channel capacity of the cognitive cooperative network (CCN). Moreover, CCN enhances the spectrum utilization efficiency. Signal to noise plus interference ratio (SNIR) and the bit error rate (BER) are explored, as well as analytical derivations are investigated for performance analysis of our proposed model under Rayleigh fading channels. The comparison between our proposed model and conventional decode and forward (DAF) relaying is also included in the research and MC-CDMA-based cooperative relaying system with multiple receiving antenna schemes to show that the recommended approach is effective. The simulation as well as the numerical results are presented to demonstrate that the suggested cooperative relaying spectrum sharing technique is efficient.},
	language = {en},
	booktitle = {Proceedings of {Trends} in {Electronics} and {Health} {Informatics}},
	publisher = {Springer},
	author = {Kabir, Md. Alomgir and Kaiser, M. Shamim},
	editor = {Kaiser, M. Shamim and Bandyopadhyay, Anirban and Ray, Kanad and Singh, Raghvendra and Nagar, Vishal},
	year = {2022},
	keywords = {BER, CRN, MC-CDMA, Outage capacity, Outage probability, SNIR},
	pages = {651--662},
}

Higher rate of mobile data traffic demand is increased with the advent of the Internet of things (IoT) and advanced network services (ANS) operators that have begun to develop fifth generation (5G) cellular networks in order to overcome the limitations of the current fourth generation (4G) cellular network. In order to solve the bandwidth scarcity and effective allocation of spectrum resources and also provide higher demand of bandwidth, a multi-carrier code division multiple access (MC-CDMA)-based cognitive radio network (CRN) is proposed and the performance of this system is investigated in this research. MC-CDMA-based CRN improves the channel capacity of the cognitive cooperative network (CCN). Moreover, CCN enhances the spectrum utilization efficiency. Signal to noise plus interference ratio (SNIR) and the bit error rate (BER) are explored, as well as analytical derivations are investigated for performance analysis of our proposed model under Rayleigh fading channels. The comparison between our proposed model and conventional decode and forward (DAF) relaying is also included in the research and MC-CDMA-based cooperative relaying system with multiple receiving antenna schemes to show that the recommended approach is effective. The simulation as well as the numerical results are presented to demonstrate that the suggested cooperative relaying spectrum sharing technique is efficient.

@incollection{kumar_chapter_2022,
	title = {Chapter 12 - {Prediction} of energy generation target of hydropower plants using artificial neural networks},
	isbn = {978-0-323-91228-0},
	url = {https://www.sciencedirect.com/science/article/pii/B9780323912280000057},
	abstract = {Hydropower is a renewable, reliable, and highly predictable source of energy. It has been used for centuries. The tariff of energy generation is divided into two parts: fixed charges and variable charges. Fixed charges are based on the availability of machinery (i.e., plant availability factor) and variable charges are based on the actual energy generation. The energy generation targets are decided by the local regulatory authorities for individual power plants. In this chapter, a scientific approach has been proposed to predict the energy generation target of individual power plants by using artificial neural networks (ANN). The yearly energy generation data of 12 hydropower plants, which are owned by UJVN Ltd., were selected. Past energy generation data from the financial year of 2011–12 to 2019–20 were utilized for the prediction. The prediction of yearly energy generation targets of individual power plants with a correction coefficient higher than 0.99 has been achieved.},
	language = {en},
	urldate = {2022-05-21},
	booktitle = {Sustainable {Developments} by {Artificial} {Intelligence} and {Machine} {Learning} for {Renewable} {Energies}},
	publisher = {Elsevier},
	author = {Kumar, Krishna and Saini, Gaurav and Kumar, Narendra and Kaiser, M. Shamim and Kannan, Ramani and Shah, Rachna},
	editor = {Kumar, Krishna and Rao, Ram Shringar and Kaiwartya, Omprakash and Kaiser, M. Shamim and Padmanaban, Sanjeevikumar},
	month = jan,
	year = {2022},
	doi = {10.1016/B978-0-323-91228-0.00005-7},
	keywords = {Artificial neural networks, Hydropower, Modeling, Renewable energy},
	pages = {309--320},
}

Hydropower is a renewable, reliable, and highly predictable source of energy. It has been used for centuries. The tariff of energy generation is divided into two parts: fixed charges and variable charges. Fixed charges are based on the availability of machinery (i.e., plant availability factor) and variable charges are based on the actual energy generation. The energy generation targets are decided by the local regulatory authorities for individual power plants. In this chapter, a scientific approach has been proposed to predict the energy generation target of individual power plants by using artificial neural networks (ANN). The yearly energy generation data of 12 hydropower plants, which are owned by UJVN Ltd., were selected. Past energy generation data from the financial year of 2011–12 to 2019–20 were utilized for the prediction. The prediction of yearly energy generation targets of individual power plants with a correction coefficient higher than 0.99 has been achieved.

@incollection{islam_chapter_2022,
	title = {Chapter 3 - {IoET}-{SG}: {Integrating} internet of energy things with smart grid},
	isbn = {978-0-323-91228-0},
	shorttitle = {Chapter 3 - {IoET}-{SG}},
	url = {https://www.sciencedirect.com/science/article/pii/B9780323912280000136},
	abstract = {The term smart grid refers to an electrical grid system that incorporates a range of operating and energy initiatives including advanced metering, intelligent distribution boards and circuit breakers, load control switches and smart devices, and renewable energy resources. The incorporation of the internet of things (IoT) into the smart grid, called the internet of energy things-smart grid (IoET-SG), has many benefits, such as a decrease in costs or saving time through smart grid devices. This system can include a wide range of home appliances, gadgets, clothes, intelligent trackers, smart meters, and small-scale vehicles. These devices would record usage data in real-time from all those objects and enable a two-way exchange of information which allows optimization of their usage. This study focuses mainly on IoET-SG research, advantages, and future challenges, along with effective solutions to these challenges.},
	language = {en},
	urldate = {2022-05-21},
	booktitle = {Sustainable {Developments} by {Artificial} {Intelligence} and {Machine} {Learning} for {Renewable} {Energies}},
	publisher = {Elsevier},
	author = {Islam, M. Shahidul and Islam, Md. Mehedi and Ahmed, Sabbir and Rahman, Md. Sazzadur and Kumar, Krishna and Kaiser, M. Shamim},
	editor = {Kumar, Krishna and Rao, Ram Shringar and Kaiwartya, Omprakash and Kaiser, M. Shamim and Padmanaban, Sanjeevikumar},
	month = jan,
	year = {2022},
	doi = {10.1016/B978-0-323-91228-0.00013-6},
	keywords = {Big data, Energy utilization, Internet of things, Machine learning, Smart grid},
	pages = {49--61},
}

The term smart grid refers to an electrical grid system that incorporates a range of operating and energy initiatives including advanced metering, intelligent distribution boards and circuit breakers, load control switches and smart devices, and renewable energy resources. The incorporation of the internet of things (IoT) into the smart grid, called the internet of energy things-smart grid (IoET-SG), has many benefits, such as a decrease in costs or saving time through smart grid devices. This system can include a wide range of home appliances, gadgets, clothes, intelligent trackers, smart meters, and small-scale vehicles. These devices would record usage data in real-time from all those objects and enable a two-way exchange of information which allows optimization of their usage. This study focuses mainly on IoET-SG research, advantages, and future challenges, along with effective solutions to these challenges.

@article{jahan_explainable_2022-1,
	title = {Explainable {AI}-based {Alzheimer}’s {Prediction} and {Management} {Using} {Multimodal} {Data}},
	url = {https://www.preprints.org/manuscript/202203.0214/v1},
	doi = {10.20944/preprints202203.0214.v1},
	abstract = {According to the World Health Organization (WHO), dementia is the seventh leading reason of death among all illnesses and one of the leading causes of disability among the world’s elderly people. Day by day the number of Alzheimer’s patients is raising. Considering the increasing rate and the dangers, Alzheimer’s disease should be diagnosed carefully. Machine learning is a potential technique for Alzheimer’s diagnosis but general users do not trust machine learning models due to the black-box nature. Even, some of those models do not provide the best performance because of using only neuroimaging data. To solve these issues, this paper proposes a novel explainable Alzheimer’s disease prediction model using a multimodal dataset. This approach performs a data-level fusion using clinical data, Freesurfer MRI segmentation data, and psychological data. For Alzheimer’s disease vs cognitively normal prediction, the random forest classifier provides 100\% accuracy. Furthermore, Alzheimer’s disease and non-Alzheimer’s dementia should be classified properly because their symptoms are similar. To the best of our knowledge, we are the first to present a three-class classification on Alzheimer’s disease vs cognitively normal vs non-Alzheimer’s dementia and achieved 99.86\% accuracy using an ensemble model. Besides, a novel Alzheimer’s patient management architecture is also proposed in this work..},
	language = {en},
	urldate = {2022-05-21},
	author = {Jahan, Sobhana and Taher, Kazi Abu and Kaiser, M. Shamim and Mahmud, Mufti and Rahman, Md Sazzadur and Hosen, A. S. M. Sanwar and Ra, In-Ho},
	month = mar,
	year = {2022},
	note = {Publisher: Preprints},
}

According to the World Health Organization (WHO), dementia is the seventh leading reason of death among all illnesses and one of the leading causes of disability among the world’s elderly people. Day by day the number of Alzheimer’s patients is raising. Considering the increasing rate and the dangers, Alzheimer’s disease should be diagnosed carefully. Machine learning is a potential technique for Alzheimer’s diagnosis but general users do not trust machine learning models due to the black-box nature. Even, some of those models do not provide the best performance because of using only neuroimaging data. To solve these issues, this paper proposes a novel explainable Alzheimer’s disease prediction model using a multimodal dataset. This approach performs a data-level fusion using clinical data, Freesurfer MRI segmentation data, and psychological data. For Alzheimer’s disease vs cognitively normal prediction, the random forest classifier provides 100% accuracy. Furthermore, Alzheimer’s disease and non-Alzheimer’s dementia should be classified properly because their symptoms are similar. To the best of our knowledge, we are the first to present a three-class classification on Alzheimer’s disease vs cognitively normal vs non-Alzheimer’s dementia and achieved 99.86% accuracy using an ensemble model. Besides, a novel Alzheimer’s patient management architecture is also proposed in this work..

@inproceedings{biswas_survey_2022,
	address = {Singapore},
	series = {Lecture {Notes} in {Networks} and {Systems}},
	title = {A {Survey} on {Predicting} {Player}’s {Performance} and {Team} {Recommendation} in {Game} of {Cricket} {Using} {Machine} {Learning}},
	isbn = {9789811607394},
	doi = {10.1007/978-981-16-0739-4_22},
	abstract = {Machine learning in sports analytics is a hot field in computer science. Using machine learning algorithms, we can predict the outcome of a game or performance of teams or individual players and building new strategies for upcoming competitions. Cricket is one of the foremost popular games in the world. Choosing the right player is one of the most challenging work for all kinds of sport and no exception in cricket. In the field of machine learning, several algorithms are used for prediction and classifications. Machine learning algorithms like linear regression, support vector machine, random forest, and naive Bayes with linear and polynomial kernel showed good results to predict the runs scored by a batsman and runs given by a bowler. In this work, we explored the techniques that have been applied to solve the challenges in cricket.},
	language = {en},
	booktitle = {Information and {Communication} {Technology} for {Competitive} {Strategies} ({ICTCS} 2020)},
	publisher = {Springer},
	author = {Biswas, Milon and Niamat Ullah Akhund, Tajim Md. and Mahbub, Md. Kawsher and Saiful Islam, Sikder Md. and Sorna, Sadia and Shamim Kaiser, M.},
	editor = {Joshi, Amit and Mahmud, Mufti and Ragel, Roshan G. and Thakur, Nileshsingh V.},
	year = {2022},
	keywords = {Cricket, Machine learning, Performance, Players, Survey, Team recommendation},
	pages = {223--230},
}

Machine learning in sports analytics is a hot field in computer science. Using machine learning algorithms, we can predict the outcome of a game or performance of teams or individual players and building new strategies for upcoming competitions. Cricket is one of the foremost popular games in the world. Choosing the right player is one of the most challenging work for all kinds of sport and no exception in cricket. In the field of machine learning, several algorithms are used for prediction and classifications. Machine learning algorithms like linear regression, support vector machine, random forest, and naive Bayes with linear and polynomial kernel showed good results to predict the runs scored by a batsman and runs given by a bowler. In this work, we explored the techniques that have been applied to solve the challenges in cricket.

@article{banna_attention-based_2021,
	title = {Attention-{Based} {Bi}-{Directional} {Long}-{Short} {Term} {Memory} {Network} for {Earthquake} {Prediction}},
	volume = {9},
	issn = {2169-3536},
	doi = {10.1109/ACCESS.2021.3071400},
	abstract = {An earthquake is a tremor felt on the surface of the earth created by the movement of the major pieces of its outer shell. Till now, many attempts have been made to forecast earthquakes, which saw some success, but these attempted models are specific to a region. In this paper, an earthquake occurrence and location prediction model is proposed. After reviewing the literature, long short-term memory (LSTM) is found to be a good option for building the model because of its memory-keeping ability. Using the Keras tuner, the best model was selected from candidate models, which are composed of combinations of various LSTM architectures and dense layers. This selected model used seismic indicators from the earthquake catalog of Bangladesh as features to predict earthquakes of the following month. Attention mechanism was added to the LSTM architecture to improve the model’s earthquake occurrence prediction accuracy, which was 74.67\%. Additionally, a regression model was built using LSTM and dense layers to predict the earthquake epicenter as a distance from a predefined location, which provided a root mean square error of 1.25.},
	journal = {IEEE Access},
	author = {Banna, Md. Hasan Al and Ghosh, Tapotosh and Nahian, Md. Jaber Al and Taher, Kazi Abu and Kaiser, M. Shamim and Mahmud, Mufti and Hossain, Mohammad Shahadat and Andersson, Karl},
	year = {2021},
	note = {Conference Name: IEEE Access},
	keywords = {Attention, Computer architecture, Earthquakes, LSTM, Load modeling, Neural networks, Predictive models, Recurrent neural networks, Support vector machines, earthquake, location, occurrence},
	pages = {56589--56603},
}

An earthquake is a tremor felt on the surface of the earth created by the movement of the major pieces of its outer shell. Till now, many attempts have been made to forecast earthquakes, which saw some success, but these attempted models are specific to a region. In this paper, an earthquake occurrence and location prediction model is proposed. After reviewing the literature, long short-term memory (LSTM) is found to be a good option for building the model because of its memory-keeping ability. Using the Keras tuner, the best model was selected from candidate models, which are composed of combinations of various LSTM architectures and dense layers. This selected model used seismic indicators from the earthquake catalog of Bangladesh as features to predict earthquakes of the following month. Attention mechanism was added to the LSTM architecture to improve the model’s earthquake occurrence prediction accuracy, which was 74.67%. Additionally, a regression model was built using LSTM and dense layers to predict the earthquake epicenter as a distance from a predefined location, which provided a root mean square error of 1.25.

@inproceedings{kaiser_6g_2021,
	address = {Singapore},
	series = {Advances in {Intelligent} {Systems} and {Computing}},
	title = {{6G} {Access} {Network} for {Intelligent} {Internet} of {Healthcare} {Things}: {Opportunity}, {Challenges}, and {Research} {Directions}},
	isbn = {978-981-334-673-4},
	shorttitle = {{6G} {Access} {Network} for {Intelligent} {Internet} of {Healthcare} {Things}},
	doi = {10.1007/978-981-33-4673-4_25},
	abstract = {The Internet of Healthcare Things (IoHT) demands massive and smart connectivity, huge bandwidth, lower latency with ultra-high data rate and better quality of healthcare experience. Unlike the 5G wireless network, the upcoming 6G communication system is expected to provide Intelligent IoHT (IIoHT) services everywhere at any time to improve the quality of life of the human being. In this paper, we present the framework of 6G cellular networks, its aggregation with multidimensional communication techniques such as optical wireless communication network, cell-free communication system, backhaul network, and quantum communication, as well as distributed security paradigm in the context of IIoHT. Such low latency and ultra-high-speed communication network will provide a new paradigm for connecting homes to hospitals, healthcare people, medical devices, hospital infrastructure, etc. Also, the requirements of 6G wireless networking, other key techniques, challenges and research direction of deploying IIoHT are outlined in the article.},
	language = {en},
	booktitle = {Proceedings of {International} {Conference} on {Trends} in {Computational} and {Cognitive} {Engineering}},
	publisher = {Springer},
	author = {Kaiser, M. Shamim and Zenia, Nusrat and Tabassum, Fariha and Mamun, Shamim Al and Rahman, M. Arifur and Islam, Md. Shahidul and Mahmud, Mufti},
	editor = {Kaiser, M. Shamim and Bandyopadhyay, Anirban and Mahmud, Mufti and Ray, Kanad},
	year = {2021},
	keywords = {Distributed security, Internet of everything (IoE), Machine learning, Massive MIMO, holographic beamforming},
	pages = {317--328},
}

The Internet of Healthcare Things (IoHT) demands massive and smart connectivity, huge bandwidth, lower latency with ultra-high data rate and better quality of healthcare experience. Unlike the 5G wireless network, the upcoming 6G communication system is expected to provide Intelligent IoHT (IIoHT) services everywhere at any time to improve the quality of life of the human being. In this paper, we present the framework of 6G cellular networks, its aggregation with multidimensional communication techniques such as optical wireless communication network, cell-free communication system, backhaul network, and quantum communication, as well as distributed security paradigm in the context of IIoHT. Such low latency and ultra-high-speed communication network will provide a new paradigm for connecting homes to hospitals, healthcare people, medical devices, hospital infrastructure, etc. Also, the requirements of 6G wireless networking, other key techniques, challenges and research direction of deploying IIoHT are outlined in the article.

@inproceedings{al_mamun_combined_2021,
	address = {Singapore},
	series = {Advances in {Intelligent} {Systems} and {Computing}},
	title = {A {Combined} {Framework} of {InterPlanetary} {File} {System} and {Blockchain} to {Securely} {Manage} {Electronic} {Medical} {Records}},
	isbn = {978-981-334-673-4},
	doi = {10.1007/978-981-33-4673-4_40},
	abstract = {Blockchain has become a popular research area since its introduction, as the benefit has been seen in various industries. It could greatly benefit the healthcare sector as it offers anonymity, immutability, and decentralization of data. Electronic Medical Record (EMR) systems face crucial problems regarding data security, accessibility, and management. A great deal of security threats relating to patient privacy involves unauthorized access to medical records, misuse of patient’s disease reports, and so on. To address these issues, we have proposed a blockchain combined with the InterPlanetary File System solution framework for EMR in the healthcare industry. The aim is to implement the blockchain for EMR and provide access rules for various users of it. The proposed framework, while protecting patient privacy, allows convenient access by approved authorities such as healthcare providers to medical data.},
	language = {en},
	booktitle = {Proceedings of {International} {Conference} on {Trends} in {Computational} and {Cognitive} {Engineering}},
	publisher = {Springer},
	author = {Al Mamun, Abdullah and Faruk Jahangir, Md. Umor and Azam, Sami and Kaiser, M. Shamim and Karim, Asif},
	editor = {Kaiser, M. Shamim and Bandyopadhyay, Anirban and Mahmud, Mufti and Ray, Kanad},
	year = {2021},
	keywords = {Blockchain, Data Security, EMR, Flask, IPFS, Postman, Python, SHA256},
	pages = {501--511},
}

Blockchain has become a popular research area since its introduction, as the benefit has been seen in various industries. It could greatly benefit the healthcare sector as it offers anonymity, immutability, and decentralization of data. Electronic Medical Record (EMR) systems face crucial problems regarding data security, accessibility, and management. A great deal of security threats relating to patient privacy involves unauthorized access to medical records, misuse of patient’s disease reports, and so on. To address these issues, we have proposed a blockchain combined with the InterPlanetary File System solution framework for EMR in the healthcare industry. The aim is to implement the blockchain for EMR and provide access rules for various users of it. The proposed framework, while protecting patient privacy, allows convenient access by approved authorities such as healthcare providers to medical data.

@inproceedings{arifeen_blockchain-based_2021,
	address = {Singapore},
	series = {Advances in {Intelligent} {Systems} and {Computing}},
	title = {A {Blockchain}-{Based} {Scheme} for {Sybil} {Attack} {Detection} in {Underwater} {Wireless} {Sensor} {Networks}},
	isbn = {978-981-334-673-4},
	doi = {10.1007/978-981-33-4673-4_37},
	abstract = {Underwater wireless sensor network (UWSN) has become popular because of its diverse applications and massive improvements in sensing technologies in recent years. However, this sensor network is vulnerable to various types of cyber-attacks due to its inherent characteristics. Among many cyber-attacks, the Sybil attack is one of the fatal attacks and damages the network severely. In this work, we have proposed a blockchain-based Sybil attack detection scheme in UWSN. We have also integrated one of our previous trust model with the blockchain-based method to make it resilient against the attacks detection. We have conducted an experimented in Flask and discussed the implementation with code details.},
	language = {en},
	booktitle = {Proceedings of {International} {Conference} on {Trends} in {Computational} and {Cognitive} {Engineering}},
	publisher = {Springer},
	author = {Arifeen, Md. Murshedul and Al Mamun, Abdullah and Ahmed, Tanvir and Kaiser, M. Shamim and Mahmud, Mufti},
	editor = {Kaiser, M. Shamim and Bandyopadhyay, Anirban and Mahmud, Mufti and Ray, Kanad},
	year = {2021},
	keywords = {Blockchain, Cluster, Flask, Postman, Sybil attack, UWSN, Wireless sensor netwrok},
	pages = {467--476},
}

Underwater wireless sensor network (UWSN) has become popular because of its diverse applications and massive improvements in sensing technologies in recent years. However, this sensor network is vulnerable to various types of cyber-attacks due to its inherent characteristics. Among many cyber-attacks, the Sybil attack is one of the fatal attacks and damages the network severely. In this work, we have proposed a blockchain-based Sybil attack detection scheme in UWSN. We have also integrated one of our previous trust model with the blockchain-based method to make it resilient against the attacks detection. We have conducted an experimented in Flask and discussed the implementation with code details.

@inproceedings{tahura_anomaly_2021,
	address = {Singapore},
	series = {Advances in {Intelligent} {Systems} and {Computing}},
	title = {Anomaly {Detection} in {Electroencephalography} {Signal} {Using} {Deep} {Learning} {Model}},
	isbn = {978-981-334-673-4},
	doi = {10.1007/978-981-33-4673-4_18},
	abstract = {Biosignals such as Electroencephalogram (EEG), Electrocardiogram (ECG), Electromyogram (EMG) represent the electrical activities of various parts of human body. Various low cost non-invasive bio-sensors measures bio-signals and assist medical practitioner to monitor physiological conditions of a human health and identify associated risk. The volume bio-signals is a big data and can not be analyzed and identify anomaly manually, therefore intelligent algorithms have been proposed to detect personalized anomaly in real time data. This paper presents a review on Deep Learning (DL) based anomaly detection techniques in EEG. The convolutional neural network, recurrent neural network and autoencoder based DL algorithms are considered. Here EEG signal acquisition, feature extracting techniques and key-anomaly features and corresponding performance of the various techniques found in the literature are also discussed. The challenges and open research questions are outlined at the end of the article.},
	language = {en},
	booktitle = {Proceedings of {International} {Conference} on {Trends} in {Computational} and {Cognitive} {Engineering}},
	publisher = {Springer},
	author = {Tahura, Sharaban and Hasnat Samiul, S. M. and Shamim Kaiser, M. and Mahmud, Mufti},
	editor = {Kaiser, M. Shamim and Bandyopadhyay, Anirban and Mahmud, Mufti and Ray, Kanad},
	year = {2021},
	keywords = {Autoencoder, Convolutional neural network, Machine learning, Prediction, Recurrent neural network},
	pages = {205--217},
}

Biosignals such as Electroencephalogram (EEG), Electrocardiogram (ECG), Electromyogram (EMG) represent the electrical activities of various parts of human body. Various low cost non-invasive bio-sensors measures bio-signals and assist medical practitioner to monitor physiological conditions of a human health and identify associated risk. The volume bio-signals is a big data and can not be analyzed and identify anomaly manually, therefore intelligent algorithms have been proposed to detect personalized anomaly in real time data. This paper presents a review on Deep Learning (DL) based anomaly detection techniques in EEG. The convolutional neural network, recurrent neural network and autoencoder based DL algorithms are considered. Here EEG signal acquisition, feature extracting techniques and key-anomaly features and corresponding performance of the various techniques found in the literature are also discussed. The challenges and open research questions are outlined at the end of the article.

@inproceedings{satu_application_2021,
	address = {Singapore},
	series = {Advances in {Intelligent} {Systems} and {Computing}},
	title = {Application of {Feature} {Engineering} with {Classification} {Techniques} to {Enhance} {Corporate} {Tax} {Default} {Detection} {Performance}},
	isbn = {978-981-334-673-4},
	doi = {10.1007/978-981-33-4673-4_5},
	abstract = {The objective of this work is to propose a methodology that is helpful in analyzing tax data and predict significant features that cause tax defaulting. In this work, we gathered a Finnish tax default data of different firms and then split it according to primary and transformed feature sets. Different feature selection techniques were used to explore significant feature sets. After that, we applied various classification techniques into primary and transformed data sets and analyzed experimental outcomes. Besides, almost all classification techniques are represented the highest results for correlation-based feature selection subset evaluation, information gain feature selection and gain ratio attribute evaluation techniques. But, information gain feature selection is found as the most reliable feature selection method in this work. This analysis can be useful as a complementary tool to assess tax default factors in corporate sectors.},
	language = {en},
	booktitle = {Proceedings of {International} {Conference} on {Trends} in {Computational} and {Cognitive} {Engineering}},
	publisher = {Springer},
	author = {Satu, Md. Shahriare and Zoynul Abedin, Mohammad and Khanom, Shoma and Ouenniche, Jamal and Shamim Kaiser, M.},
	editor = {Kaiser, M. Shamim and Bandyopadhyay, Anirban and Mahmud, Mufti and Ray, Kanad},
	year = {2021},
	keywords = {Classification techniques, Feature selection, Tax default detection},
	pages = {53--63},
}

The objective of this work is to propose a methodology that is helpful in analyzing tax data and predict significant features that cause tax defaulting. In this work, we gathered a Finnish tax default data of different firms and then split it according to primary and transformed feature sets. Different feature selection techniques were used to explore significant feature sets. After that, we applied various classification techniques into primary and transformed data sets and analyzed experimental outcomes. Besides, almost all classification techniques are represented the highest results for correlation-based feature selection subset evaluation, information gain feature selection and gain ratio attribute evaluation techniques. But, information gain feature selection is found as the most reliable feature selection method in this work. This analysis can be useful as a complementary tool to assess tax default factors in corporate sectors.

@inproceedings{rahman_cascade_2021,
	address = {Singapore},
	series = {Advances in {Intelligent} {Systems} and {Computing}},
	title = {Cascade {Classification} of {Face} {Liveliness} {Detection} {Using} {Heart} {Beat} {Measurement}},
	isbn = {978-981-334-673-4},
	doi = {10.1007/978-981-33-4673-4_47},
	abstract = {Face detection and recognition is a prevalent concept in security and access control area which is commonly used in surveillance cameras at public places, attendance etc. But often this type of system can be circumvented by holding a photo or running a video of authorized person to the camera. Therefore, liveliness concept comes up with a solution to detect the person is real or spoofed. In this paper, We proposed a cascade classifier based model for detecting liveliness using deep-learning and Heart-beat measurement. Moreover, we have evaluated our model accuracy with our own dataset of real and fake videos and photos. By using our proposed model of face liveliness detection model, FPR and FNR have declined 16\% and 5.22\% respectively. In addition, we have also compared proposed system with other state-of-art methods. And here proposed study has achieved an accuracy of 99.46\%.},
	language = {en},
	booktitle = {Proceedings of {International} {Conference} on {Trends} in {Computational} and {Cognitive} {Engineering}},
	publisher = {Springer},
	author = {Rahman, Md. Mahfujur and Mamun, Shamim Al and Kaiser, M. Shamim and Islam, Md. Shahidul and Rahman, Md. Arifur},
	editor = {Kaiser, M. Shamim and Bandyopadhyay, Anirban and Mahmud, Mufti and Ray, Kanad},
	year = {2021},
	keywords = {CNN, Deep Learning, Face Detection, Face Liveliness, FaceNet, Features, Heart Beat, PCA},
	pages = {581--590},
}

Face detection and recognition is a prevalent concept in security and access control area which is commonly used in surveillance cameras at public places, attendance etc. But often this type of system can be circumvented by holding a photo or running a video of authorized person to the camera. Therefore, liveliness concept comes up with a solution to detect the person is real or spoofed. In this paper, We proposed a cascade classifier based model for detecting liveliness using deep-learning and Heart-beat measurement. Moreover, we have evaluated our model accuracy with our own dataset of real and fake videos and photos. By using our proposed model of face liveliness detection model, FPR and FNR have declined 16% and 5.22% respectively. In addition, we have also compared proposed system with other state-of-art methods. And here proposed study has achieved an accuracy of 99.46%.

@inproceedings{sharma_comparative_2021,
	address = {Singapore},
	series = {Advances in {Intelligent} {Systems} and {Computing}},
	title = {Comparative {Analysis} of {Different} {Classifiers} on {EEG} {Signals} for {Predicting} {Epileptic} {Seizure}},
	isbn = {978-981-334-673-4},
	doi = {10.1007/978-981-33-4673-4_17},
	abstract = {Epilepsy is a neurological disease that’s characterized by perennial seizures. In this neurological condition the transient electrical phenomenon within the brain occurs that produces an amendment in sensation, awareness, and behavior of an individuals that leads to risk. To understand the brain behavior Electroencephalogram (EEG) signals are used in six different sub-bands viz. Alpha (αα{\textbackslash}alpha ), Beta (ββ{\textbackslash}beta ), Gamma1 (γγ{\textbackslash}gamma 1), Gamma2 (γγ{\textbackslash}gamma 2), Theta (θθ{\textbackslash}theta ) and Delta (δδ{\textbackslash}delta ). The Brainstorm software is used for visualizing, analyzing and filtration of EEG signals in each sub-band. This paper deals with the extraction of the various features in each sub-bands and different Machine Learning classifiers were used on these extracted features for comparative analysis in terms of Accuracy, prediction Speed and training time in MatLab. The various statistical and spectral methods are applied on EEG signals to obtained the distinct features in each sub-band. After compared these classifiers on the performance parameters.we have 8 best classifier trained Models that were utilized in checking effectiveness to clearly distinguish between Epileptic and Normal cases.},
	language = {en},
	booktitle = {Proceedings of {International} {Conference} on {Trends} in {Computational} and {Cognitive} {Engineering}},
	publisher = {Springer},
	author = {Sharma, M. K. and Ray, K. and Yupapin, P. and Kaiser, M. S. and Ong, C. T. and Ali, J.},
	editor = {Kaiser, M. Shamim and Bandyopadhyay, Anirban and Mahmud, Mufti and Ray, Kanad},
	year = {2021},
	keywords = {Electroencephalogram (EEG), Epilepsy, Interictal and ictal, Preictal, Seizure, Spectral analysis},
	pages = {193--204},
}

Epilepsy is a neurological disease that’s characterized by perennial seizures. In this neurological condition the transient electrical phenomenon within the brain occurs that produces an amendment in sensation, awareness, and behavior of an individuals that leads to risk. To understand the brain behavior Electroencephalogram (EEG) signals are used in six different sub-bands viz. Alpha (αα\alpha ), Beta (ββ\beta ), Gamma1 (γγ\gamma 1), Gamma2 (γγ\gamma 2), Theta (θθ\theta ) and Delta (δδ\delta ). The Brainstorm software is used for visualizing, analyzing and filtration of EEG signals in each sub-band. This paper deals with the extraction of the various features in each sub-bands and different Machine Learning classifiers were used on these extracted features for comparative analysis in terms of Accuracy, prediction Speed and training time in MatLab. The various statistical and spectral methods are applied on EEG signals to obtained the distinct features in each sub-band. After compared these classifiers on the performance parameters.we have 8 best classifier trained Models that were utilized in checking effectiveness to clearly distinguish between Epileptic and Normal cases.

@article{arifeen_blockchain-enable_2020,
	title = {Blockchain-enable {Contact} {Tracing} for {Preserving} {User} {Privacy} {During} {COVID}-19 {Outbreak}},
	url = {https://www.preprints.org/manuscript/202007.0502/v1},
	doi = {10.20944/preprints202007.0502.v1},
	abstract = {Contact tracing has become an indispensable tool of various extensive measures to control the spread of COVID-19 pandemic due to novel coronavirus. This essential tool helps to identify, isolate and quarantine the contacted persons of a COVID-19 patient. However, the existing contact tracing applications developed by various countries, health organizations to trace down the contacts after identifying a COVID-19 patient suffers from several security and privacy concerns. In this work, we have identified those security and privacy issues of several leading contact tracing applications and proposed a blockchain-based framework to overcome the major security and privacy challenges imposed by the applications. We have discussed the security and privacy measures that are achieved by the proposed framework to show the effectiveness against the security and privacy issues raised by the existing mobile contact tracing applications.},
	language = {en},
	urldate = {2021-02-21},
	author = {Arifeen, Md Murshedul and Mamun, Abdullah Al and Kaiser, M. Shamim and Mahmud, Mufti},
	month = jul,
	year = {2020},
	note = {Publisher: Preprints},
}

Contact tracing has become an indispensable tool of various extensive measures to control the spread of COVID-19 pandemic due to novel coronavirus. This essential tool helps to identify, isolate and quarantine the contacted persons of a COVID-19 patient. However, the existing contact tracing applications developed by various countries, health organizations to trace down the contacts after identifying a COVID-19 patient suffers from several security and privacy concerns. In this work, we have identified those security and privacy issues of several leading contact tracing applications and proposed a blockchain-based framework to overcome the major security and privacy challenges imposed by the applications. We have discussed the security and privacy measures that are achieved by the proposed framework to show the effectiveness against the security and privacy issues raised by the existing mobile contact tracing applications.

@inproceedings{ruiz_3d_2020,
	address = {Cham},
	series = {Lecture {Notes} in {Computer} {Science}},
	title = {{3D} {DenseNet} {Ensemble} in 4-{Way} {Classification} of {Alzheimer}’s {Disease}},
	isbn = {978-3-030-59277-6},
	doi = {10.1007/978-3-030-59277-6_8},
	abstract = {One of the major causes of death in developing nations is the Alzheimer’s Disease (AD). For the treatment of this illness, is crucial to early diagnose mild cognitive impairment (MCI) and AD, with the help of feature extraction from magnetic resonance images (MRI). This paper proposes a 4-way classification of 3D MRI images using an ensemble implementation of 3D Densely Connected Convolutional Networks (3D DenseNets) models. The research makes use of dense connections that improve the movement of data within the model, due to having each layer linked with all the subsequent layers in a block. Afterwards, a probability-based fusion method is employed to merge the probabilistic output of each unique individual classifier model. Available through the ADNI dataset, preprocessed 3D MR images from four subject groups (i.e., AD, healthy control, early MCI, and late MCI) were acquired to perform experiments. In the tests, the proposed approach yields better results than other state-of-the-art methods dealing with 3D MR images.},
	language = {en},
	booktitle = {Brain {Informatics}},
	publisher = {Springer International Publishing},
	author = {Ruiz, Juan and Mahmud, Mufti and Modasshir, Md and Shamim Kaiser, M. and Alzheimer’s Disease Neuroimaging Initiative, for the},
	editor = {Mahmud, Mufti and Vassanelli, Stefano and Kaiser, M. Shamim and Zhong, Ning},
	year = {2020},
	keywords = {Convolutional neural network, Deep learning, Machine learning, Magnetic resonance imaging, Neuroimaging},
	pages = {85--96},
}

One of the major causes of death in developing nations is the Alzheimer’s Disease (AD). For the treatment of this illness, is crucial to early diagnose mild cognitive impairment (MCI) and AD, with the help of feature extraction from magnetic resonance images (MRI). This paper proposes a 4-way classification of 3D MRI images using an ensemble implementation of 3D Densely Connected Convolutional Networks (3D DenseNets) models. The research makes use of dense connections that improve the movement of data within the model, due to having each layer linked with all the subsequent layers in a block. Afterwards, a probability-based fusion method is employed to merge the probabilistic output of each unique individual classifier model. Available through the ADNI dataset, preprocessed 3D MR images from four subject groups (i.e., AD, healthy control, early MCI, and late MCI) were acquired to perform experiments. In the tests, the proposed approach yields better results than other state-of-the-art methods dealing with 3D MR images.

@inproceedings{farhin_attack_2020,
	title = {Attack {Detection} in {Internet} of {Things} using {Software} {Defined} {Network} and {Fuzzy} {Neural} {Network}},
	doi = {10.1109/ICIEVicIVPR48672.2020.9306666},
	abstract = {Internet of Things (IoT) is a dynamic and distributed wide network system that can integrate a gigantic number of pervasive sensors (i.e., physical objects), wireless nodes, and ubiquitous computing systems. These sensors can collect tons of raw data, send them to the internet at an unprecedented rate, and convert them to actionable insights using computing systems. These sensing nodes or physical objects are vulnerable and have upraised cybersecurity threats. In this work, we proposed the attack detection model for IoT using Software-defined network (SDN). The SDN controller can analyze the traffic flow, detect the anomaly, and block incoming traffic as well as the source nodes. In the SDN, a Fuzzy neural network (FNN) based attack detection system is considered which can detect attacks such as man-in-the-middle, distributed denial of service, side-channel, and malicious code. The FNN is trained and tested using NSL-KDD datasets. The evaluated performance exhibits that the FNN based attack detection system can detect the mentioned attack with an accuracy of 83\%.},
	booktitle = {2020 {Joint} 9th {International} {Conference} on {Informatics}, {Electronics} {Vision} ({ICIEV}) and 2020 4th {International} {Conference} on {Imaging}, {Vision} {Pattern} {Recognition} ({icIVPR})},
	author = {Farhin, F. and Sultana, I. and Islam, N. and Kaiser, M. S. and Rahman, M. S. and Mahmud, M.},
	month = aug,
	year = {2020},
	keywords = {Attack Detection, FNN, FNN based attack detection system, Feature extraction, Fuzzy control, Fuzzy logic, Fuzzy neural networks, Internet of Things, Internet of things, IoT, NSL-KDD Dataset, NSL-KDD datasets, SDN, SDN controller, Security, Servers, computer crime, computer network security, distributed denial of service, dynamic distributed wide network system, fuzzy neural nets, fuzzy neural network based attack detection system, malicious code, man-in-the-middle, pervasive sensors, physical objects, side-channel, software defined network, software defined networking, source nodes, telecommunication traffic, ubiquitous computing systems, wireless nodes},
	pages = {1--6},
}

Internet of Things (IoT) is a dynamic and distributed wide network system that can integrate a gigantic number of pervasive sensors (i.e., physical objects), wireless nodes, and ubiquitous computing systems. These sensors can collect tons of raw data, send them to the internet at an unprecedented rate, and convert them to actionable insights using computing systems. These sensing nodes or physical objects are vulnerable and have upraised cybersecurity threats. In this work, we proposed the attack detection model for IoT using Software-defined network (SDN). The SDN controller can analyze the traffic flow, detect the anomaly, and block incoming traffic as well as the source nodes. In the SDN, a Fuzzy neural network (FNN) based attack detection system is considered which can detect attacks such as man-in-the-middle, distributed denial of service, side-channel, and malicious code. The FNN is trained and tested using NSL-KDD datasets. The evaluated performance exhibits that the FNN based attack detection system can detect the mentioned attack with an accuracy of 83%.

@inproceedings{al_banna_monitoring_2020,
	address = {Cham},
	series = {Lecture {Notes} in {Computer} {Science}},
	title = {A {Monitoring} {System} for {Patients} of {Autism} {Spectrum} {Disorder} {Using} {Artificial} {Intelligence}},
	isbn = {978-3-030-59277-6},
	doi = {10.1007/978-3-030-59277-6_23},
	abstract = {When the world is suffering from the deadliest consequences of COVID-19, people with autism find themselves in the worst possible situation. The patients of autism lack social skills, and in many cases, show repetitive behavior. Many of them need outside support throughout their life. During the COVID-19 pandemic, as many of the places are in lockdown conditions, it is very tough for them to find help from their doctors and therapists. Suddenly, the caregivers and parents of the ASD patients find themselves in a strange situation. Therefore, we are proposing an artificial intelligence-based system that uses sensor data to monitor the patient’s condition, and based on the emotion and facial expression of the patient, adjusts the learning method through exciting games and tasks. Whenever something goes wrong with the patient’s behavior, the caregivers and the parents are alerted about it. We then presented how this AI-based system can help them during COVID-19 pandemic. This system can help the parents to adjust to the new situation and continue the mental growth of the patients.},
	language = {en},
	booktitle = {Brain {Informatics}},
	publisher = {Springer International Publishing},
	author = {Al Banna, Md. Hasan and Ghosh, Tapotosh and Taher, Kazi Abu and Kaiser, M. Shamim and Mahmud, Mufti},
	editor = {Mahmud, Mufti and Vassanelli, Stefano and Kaiser, M. Shamim and Zhong, Ning},
	year = {2020},
	keywords = {Artificial intelligence, Autism, COVID-19, Emotion Detection},
	pages = {251--262},
}

When the world is suffering from the deadliest consequences of COVID-19, people with autism find themselves in the worst possible situation. The patients of autism lack social skills, and in many cases, show repetitive behavior. Many of them need outside support throughout their life. During the COVID-19 pandemic, as many of the places are in lockdown conditions, it is very tough for them to find help from their doctors and therapists. Suddenly, the caregivers and parents of the ASD patients find themselves in a strange situation. Therefore, we are proposing an artificial intelligence-based system that uses sensor data to monitor the patient’s condition, and based on the emotion and facial expression of the patient, adjusts the learning method through exciting games and tasks. Whenever something goes wrong with the patient’s behavior, the caregivers and the parents are alerted about it. We then presented how this AI-based system can help them during COVID-19 pandemic. This system can help the parents to adjust to the new situation and continue the mental growth of the patients.

@inproceedings{jesmin_artificial_2020,
	address = {Cham},
	series = {Lecture {Notes} in {Computer} {Science}},
	title = {Artificial and {Internet} of {Healthcare} {Things} {Based} {Alzheimer} {Care} {During} {COVID} 19},
	isbn = {978-3-030-59277-6},
	doi = {10.1007/978-3-030-59277-6_24},
	abstract = {Alzheimer patient’s routine care at the onset of a catastrophe like coronavirus disease 2019 (COVID-19) pandemic is interrupted as healthcare is providing special attention to the patient having severe acute respiratory syndrome coronavirus 2 (SARS-COV-2) or COVID-19 infection. In order to decrease the spread of the disease, government has shut down regular services at the hospital, and advised all vulnerable people to stay at home and maintain social distance (of 3 fts) which hampered the routine care and rehabilitation therapy of elderly patient having a chronic disease like Alzheimer. On the other hand, the artificial intelligence (AI)-based internet of healthcare things allows clinicians to monitor physiological conditions of patients in real-time and machine learning models can able to detect any anomaly in the patient’s condition. Besides, the advancement in Information and Communication Technology enable us to provide special distance care (such as medication and therapy) by dedicated medical teams or special therapists. This paper discusses the effect of COVID-19 on patient care of Alzheimer’s Disease (AD) and how AI-based IoT can help special care of AD patients at home. Finally, we have outlined some recommendations for Family and Caregiver, Volunteer and Social Care which will help to develop the Government policy.},
	language = {en},
	booktitle = {Brain {Informatics}},
	publisher = {Springer International Publishing},
	author = {Jesmin, Sabrina and Kaiser, M. Shamim and Mahmud, Mufti},
	editor = {Mahmud, Mufti and Vassanelli, Stefano and Kaiser, M. Shamim and Zhong, Ning},
	year = {2020},
	keywords = {IoT, Machine learning, Pandemic, Patient management, SARS-COV-2},
	pages = {263--274},
}

Alzheimer patient’s routine care at the onset of a catastrophe like coronavirus disease 2019 (COVID-19) pandemic is interrupted as healthcare is providing special attention to the patient having severe acute respiratory syndrome coronavirus 2 (SARS-COV-2) or COVID-19 infection. In order to decrease the spread of the disease, government has shut down regular services at the hospital, and advised all vulnerable people to stay at home and maintain social distance (of 3 fts) which hampered the routine care and rehabilitation therapy of elderly patient having a chronic disease like Alzheimer. On the other hand, the artificial intelligence (AI)-based internet of healthcare things allows clinicians to monitor physiological conditions of patients in real-time and machine learning models can able to detect any anomaly in the patient’s condition. Besides, the advancement in Information and Communication Technology enable us to provide special distance care (such as medication and therapy) by dedicated medical teams or special therapists. This paper discusses the effect of COVID-19 on patient care of Alzheimer’s Disease (AD) and how AI-based IoT can help special care of AD patients at home. Finally, we have outlined some recommendations for Family and Caregiver, Volunteer and Social Care which will help to develop the Government policy.

@article{noor_application_2020,
	title = {Application of deep learning in detecting neurological disorders from magnetic resonance images: a survey on the detection of {Alzheimer}’s disease, {Parkinson}’s disease and schizophrenia},
	volume = {7},
	issn = {2198-4026},
	shorttitle = {Application of deep learning in detecting neurological disorders from magnetic resonance images},
	url = {https://doi.org/10.1186/s40708-020-00112-2},
	doi = {10.1186/s40708-020-00112-2},
	abstract = {Neuroimaging, in particular magnetic resonance imaging (MRI), has been playing an important role in understanding brain functionalities and its disorders during the last couple of decades. These cutting-edge MRI scans, supported by high-performance computational tools and novel ML techniques, have opened up possibilities to unprecedentedly identify neurological disorders. However, similarities in disease phenotypes make it very difficult to detect such disorders accurately from the acquired neuroimaging data. This article critically examines and compares performances of the existing deep learning (DL)-based methods to detect neurological disorders—focusing on Alzheimer’s disease, Parkinson’s disease and schizophrenia—from MRI data acquired using different modalities including functional and structural MRI. The comparative performance analysis of various DL architectures across different disorders and imaging modalities suggests that the Convolutional Neural Network outperforms other methods in detecting neurological disorders. Towards the end, a number of current research challenges are indicated and some possible future research directions are provided.},
	language = {en},
	number = {1},
	urldate = {2021-02-21},
	journal = {Brain Informatics},
	author = {Noor, Manan Binth Taj and Zenia, Nusrat Zerin and Kaiser, M. Shamim and Mamun, Shamim Al and Mahmud, Mufti},
	year = {2020},
	pages = {11},
}

Neuroimaging, in particular magnetic resonance imaging (MRI), has been playing an important role in understanding brain functionalities and its disorders during the last couple of decades. These cutting-edge MRI scans, supported by high-performance computational tools and novel ML techniques, have opened up possibilities to unprecedentedly identify neurological disorders. However, similarities in disease phenotypes make it very difficult to detect such disorders accurately from the acquired neuroimaging data. This article critically examines and compares performances of the existing deep learning (DL)-based methods to detect neurological disorders—focusing on Alzheimer’s disease, Parkinson’s disease and schizophrenia—from MRI data acquired using different modalities including functional and structural MRI. The comparative performance analysis of various DL architectures across different disorders and imaging modalities suggests that the Convolutional Neural Network outperforms other methods in detecting neurological disorders. Towards the end, a number of current research challenges are indicated and some possible future research directions are provided.

@book{mahmud_brain_2020,
	series = {Lecture {Notes} in {Artificial} {Intelligence}},
	title = {Brain {Informatics}, {Voice} {In} {Settings}: 13th {International} {Conference}, {BI} 2020, {Padua}, {Italy}, {September} 19, 2020, {Proceedings}},
	isbn = {978-3-030-59276-9},
	shorttitle = {Brain {Informatics}, {Voice} {In} {Settings}},
	url = {https://www.springer.com/gp/book/9783030592769},
	abstract = {This book constitutes the refereed proceedings of the 13th International Conference on Brain Informatics, BI 2020, held in Padua, Italy, in September 2020. The conference was held virtually due to the COVID-19 pandemic. The 33 full papers were carefully reviewed and selected from 57 submissions. The papers are organized in the following topical sections: cognitive and computational foundations of brain science; investigations of human information processing systems; brain big data analytics, curation and management; informatics paradigms for brain and mental health research; and brain-machine intelligence and brain-inspired computing.},
	language = {en},
	urldate = {2021-02-21},
	publisher = {Springer International Publishing},
	editor = {Mahmud, Mufti and Vassanelli, Stefano and Kaiser, M. Shamim and Zhong, Ning},
	year = {2020},
	doi = {10.1007/978-3-030-59277-6},
}

This book constitutes the refereed proceedings of the 13th International Conference on Brain Informatics, BI 2020, held in Padua, Italy, in September 2020. The conference was held virtually due to the COVID-19 pandemic. The 33 full papers were carefully reviewed and selected from 57 submissions. The papers are organized in the following topical sections: cognitive and computational foundations of brain science; investigations of human information processing systems; brain big data analytics, curation and management; informatics paradigms for brain and mental health research; and brain-machine intelligence and brain-inspired computing.

@inproceedings{tania_assay_2019,
	title = {Assay {Type} {Detection} {Using} {Advanced} {Machine} {Learning} {Algorithms}},
	doi = {10.1109/SKIMA47702.2019.8982449},
	abstract = {The colourimetric analysis has been used in diversified fields for years. This paper provides a unique overview of colourimetric tests from the perspective of computer vision by describing different aspects of a colourimetric test in the context of image processing, followed by an investigation into the development of a colorimetric assay type detection system using advanced machine learning algorithms. To the best of our knowledge, this is the first attempt to define colourimetric assay types from the eyes of a machine and perform any colorimetric test using deep learning. This investigation utilizes the state-of-the-art pre-trained models of Convolutional Neural Network (CNN) to perform the assay type detection of an enzyme-linked immunosorbent assay (ELISA) and lateral flow assay (LFA). The ELISA dataset contains images of both positive and negative samples, prepared for the plasmonic ELISA based TB-antigen specific antibody detection. The LFA dataset contains images of the universal pH indicator paper of eight pH levels. It is noted that the pre-trained models offered 100\% accurate visual recognition for the assay type detection. Such detection can assist novice users to initiate a colorimetric test using his/her personal digital devices. The assay type detection can also aid in calibrating an image-based colorimetric classification.},
	booktitle = {2019 13th {International} {Conference} on {Software}, {Knowledge}, {Information} {Management} and {Applications} ({SKIMA})},
	author = {Tania, M. H. and Lwin, K. T. and Shabut, A. M. and Abu-Hassan, K. J. and Kaiser, M. S. and Hossain, M. A.},
	month = aug,
	year = {2019},
	note = {ISSN: 2573-3214},
	keywords = {advanced machine learning algorithms, biochemistry, biosensors, chemical sensors, colorimetric assay type detection system, colorimetric test, colorimetry, colourimetric analysis, colourimetric test, computer vision, convolutional neural nets, convolutional neural network, deep learning, diagnosis, diseases, enzyme-linked immunosorbent assay, enzymes, flow assay, image-based colorimetric classification, lab-on-a-chip, learning (artificial intelligence), medical image processing, molecular biophysics, pH, plasmonic ELISA based TB-antigen specific antibody detection, point-of-care system, transfer learning, visual recognition},
	pages = {1--8},
}

The colourimetric analysis has been used in diversified fields for years. This paper provides a unique overview of colourimetric tests from the perspective of computer vision by describing different aspects of a colourimetric test in the context of image processing, followed by an investigation into the development of a colorimetric assay type detection system using advanced machine learning algorithms. To the best of our knowledge, this is the first attempt to define colourimetric assay types from the eyes of a machine and perform any colorimetric test using deep learning. This investigation utilizes the state-of-the-art pre-trained models of Convolutional Neural Network (CNN) to perform the assay type detection of an enzyme-linked immunosorbent assay (ELISA) and lateral flow assay (LFA). The ELISA dataset contains images of both positive and negative samples, prepared for the plasmonic ELISA based TB-antigen specific antibody detection. The LFA dataset contains images of the universal pH indicator paper of eight pH levels. It is noted that the pre-trained models offered 100% accurate visual recognition for the assay type detection. Such detection can assist novice users to initiate a colorimetric test using his/her personal digital devices. The assay type detection can also aid in calibrating an image-based colorimetric classification.

@inproceedings{arifeen_anfis_2019,
	title = {{ANFIS} based {Trust} {Management} {Model} to {Enhance} {Location} {Privacy} in {Underwater} {Wireless} {Sensor} {Networks}},
	doi = {10.1109/ECACE.2019.8679165},
	abstract = {Trust management is a promising alternative solution to different complex security algorithms for Underwater Wireless Sensor Networks (UWSN) applications due to its several resource constraint behaviour. In this work, we have proposed a trust management model to improve location privacy of the UWSN. Adaptive Neuro Fuzzy Inference System (ANFIS) has been exploited to evaluate trustworthiness of a sensor node. Also Markov Decision Process (MDP) has been considered. At each state of the MDP, a sensor node evaluates trust behaviour of forwarding node utilizing the FIS learning rules and selects a trusted node. Simulation has been conducted in MATLAB and simulation results show that the detection accuracy of trustworthiness is 91.2\% which is greater than Knowledge Discovery and Data Mining (KDD) 99 intrusion detection based dataset. So, in our model 91.2\% trustworthiness is necessary to be a trusted node otherwise it will be treated as a malicious or compromised node. Our proposed model can successfully eliminate the possibility of occurring any compromised or malicious node in the network.},
	booktitle = {2019 {International} {Conference} on {Electrical}, {Computer} and {Communication} {Engineering} ({ECCE})},
	author = {Arifeen, M. M. and Islam, A. A. and Rahman, M. M. and Taher, K. A. and Islam, M. M. and Kaiser, M. S.},
	month = feb,
	year = {2019},
	keywords = {ANFIS, Authentication, FIS learning rules, Information and communication technology, MDP, Markov decision process, Markov processes, Matlab, Privacy, Routing, Trust Management, Trust management, UWSN, Wireless sensor networks, adaptive neuro fuzzy inference system, complex security, data mining, data privacy, forwarding node, intrusion detection, location privacy, malicious node, marine communication, resource constraint behaviour, sensor node, telecommunication security, trust behaviour, trust management model, trusted node, underwater wireless sensor networks, wireless sensor networks},
	pages = {1--6},
}

Trust management is a promising alternative solution to different complex security algorithms for Underwater Wireless Sensor Networks (UWSN) applications due to its several resource constraint behaviour. In this work, we have proposed a trust management model to improve location privacy of the UWSN. Adaptive Neuro Fuzzy Inference System (ANFIS) has been exploited to evaluate trustworthiness of a sensor node. Also Markov Decision Process (MDP) has been considered. At each state of the MDP, a sensor node evaluates trust behaviour of forwarding node utilizing the FIS learning rules and selects a trusted node. Simulation has been conducted in MATLAB and simulation results show that the detection accuracy of trustworthiness is 91.2% which is greater than Knowledge Discovery and Data Mining (KDD) 99 intrusion detection based dataset. So, in our model 91.2% trustworthiness is necessary to be a trusted node otherwise it will be treated as a malicious or compromised node. Our proposed model can successfully eliminate the possibility of occurring any compromised or malicious node in the network.

@inproceedings{banna_camera_2019,
	title = {Camera {Model} {Identification} using {Deep} {CNN} and {Transfer} {Learning} {Approach}},
	doi = {10.1109/ICREST.2019.8644194},
	abstract = {The forensic investigation on digital images is to assess the authenticity of images without the embedded security on the images. The camera model identification is the first step for image forensic investigation. The paper proposes the deep Convolutional Neural Network and transfer learning approach for extracting features from an images dataset. An open image dataset of 3900 images have been created using three camera models. Three state-of-the-art machine learning algorithms such as SVM, logistic regression and random forest based classifiers have been used for evaluating identification accuracy.},
	booktitle = {2019 {International} {Conference} on {Robotics},{Electrical} and {Signal} {Processing} {Techniques} ({ICREST})},
	author = {Banna, M. H. Al and Haider, M. Ali and Nahian, M. J. Al and Islam, M. M. and Taher, K. A. and Kaiser, M. S.},
	month = jan,
	year = {2019},
	keywords = {Cameras, Classification, Convolution, Deep CNN, Deep learning, Feature extraction, Machine Learning, MobileNet, Robot vision systems, SVM, Support vector machines, camera model identification, cameras, convolutional neural nets, convolutional neural network, deep CNN, digital forensics, digital images, embedded security, feature extraction, image classification, image forensic, image forensic investigation, learning (artificial intelligence), logistic regression, machine learning algorithm, open image dataset, random forest, transfer learning approach},
	pages = {626--630},
}

The forensic investigation on digital images is to assess the authenticity of images without the embedded security on the images. The camera model identification is the first step for image forensic investigation. The paper proposes the deep Convolutional Neural Network and transfer learning approach for extracting features from an images dataset. An open image dataset of 3900 images have been created using three camera models. Three state-of-the-art machine learning algorithms such as SVM, logistic regression and random forest based classifiers have been used for evaluating identification accuracy.

@article{shabut_multidimensional_2018,
	title = {A multidimensional trust evaluation model for {MANETs}},
	volume = {123},
	issn = {1084-8045},
	url = {https://www.sciencedirect.com/science/article/pii/S1084804518302327},
	doi = {10.1016/j.jnca.2018.07.008},
	abstract = {Effective trust management can enhance nodes' cooperation in selecting trustworthy and optimal paths between the source and destination nodes in mobile ad hoc networks (MANETs). It allows the wireless nodes (WNs) in a MANET environment to deal with uncertainty about the future actions of other participants. The main challenges in MANETs are time-varying network architecture due to the mobility of WNs, the presence of attack-prone nodes, and extreme resource limitations. In this paper, an energy-aware and social trust inspired multidimensional trust management model is proposed to achieve enhanced quality of service (QoS) parameters by overcoming these challenges. The trust management model calculates the trust value of the WNs through peer to peer and link evaluations. Energy and social trust are utilized for peer to peer evaluation, while an optimal routing path with a small number of intermediate nodes with minimum acceptable trust value is used for evaluation of the link. Empirical analysis reveals that the proposed trust model is robust and accurate in comparison to the state-of-the-art model for MANETs.},
	language = {en},
	urldate = {2021-02-28},
	journal = {Journal of Network and Computer Applications},
	author = {Shabut, Antesar M. and Kaiser, M. Shamim and Dahal, Keshav P. and Chen, Wenbing},
	month = dec,
	year = {2018},
	keywords = {Link evaluation, Peer to peer evaluation, Recommendation management, Social properties, Trust management model},
	pages = {32--41},
}

Effective trust management can enhance nodes' cooperation in selecting trustworthy and optimal paths between the source and destination nodes in mobile ad hoc networks (MANETs). It allows the wireless nodes (WNs) in a MANET environment to deal with uncertainty about the future actions of other participants. The main challenges in MANETs are time-varying network architecture due to the mobility of WNs, the presence of attack-prone nodes, and extreme resource limitations. In this paper, an energy-aware and social trust inspired multidimensional trust management model is proposed to achieve enhanced quality of service (QoS) parameters by overcoming these challenges. The trust management model calculates the trust value of the WNs through peer to peer and link evaluations. Energy and social trust are utilized for peer to peer evaluation, while an optimal routing path with a small number of intermediate nodes with minimum acceptable trust value is used for evaluation of the link. Empirical analysis reveals that the proposed trust model is robust and accurate in comparison to the state-of-the-art model for MANETs.

@inproceedings{niamat_ullah_akhund_adeptness_2018,
	address = {Cham},
	series = {Lecture {Notes} in {Computer} {Science}},
	title = {{ADEPTNESS}: {Alzheimer}’s {Disease} {Patient} {Management} {System} {Using} {Pervasive} {Sensors} - {Early} {Prototype} and {Preliminary} {Results}},
	isbn = {978-3-030-05587-5},
	shorttitle = {{ADEPTNESS}},
	doi = {10.1007/978-3-030-05587-5_39},
	abstract = {Alzheimer’s is a catastrophic neuro-degenerative state in the elderly which reduces thinking skills and thereby hamper daily activity. Thus the management may be helpful for people with such condition. This work presents sensor based management system for Alzheimer’s patient. The main objective of this work is to report an early prototype of an eventual wearable system that can assist in managing the health of such patients and notify the caregivers in case of necessity. A brief case study is presented which showed that the proposed prototype can detect agitated and clam states of patients. As the ultimately developed assistive system will be packaged as a wearable device, the case study also investigated the usability of wearable devices on different age groups of Alzheimer’s patients. In addition, electro dermal activity for 4 patient of age group 55–60 and 60-7s years were also explored to assess the health condition of the patients.},
	language = {en},
	booktitle = {Brain {Informatics}},
	publisher = {Springer International Publishing},
	author = {Niamat Ullah Akhund, Tajim Md. and Mahi, Md. Julkar Nayeen and Hasnat Tanvir, A. N. M. and Mahmud, Mufti and Kaiser, M. Shamim},
	editor = {Wang, Shouyi and Yamamoto, Vicky and Su, Jianzhong and Yang, Yang and Jones, Erick and Iasemidis, Leon and Mitchell, Tom},
	year = {2018},
	keywords = {Healthcare, Machine learning, Neurodegeneration, Sensor, Wearable devices},
	pages = {413--422},
}

Alzheimer’s is a catastrophic neuro-degenerative state in the elderly which reduces thinking skills and thereby hamper daily activity. Thus the management may be helpful for people with such condition. This work presents sensor based management system for Alzheimer’s patient. The main objective of this work is to report an early prototype of an eventual wearable system that can assist in managing the health of such patients and notify the caregivers in case of necessity. A brief case study is presented which showed that the proposed prototype can detect agitated and clam states of patients. As the ultimately developed assistive system will be packaged as a wearable device, the case study also investigated the usability of wearable devices on different age groups of Alzheimer’s patients. In addition, electro dermal activity for 4 patient of age group 55–60 and 60-7s years were also explored to assess the health condition of the patients.

@inproceedings{alam_iot-belief_2018,
	title = {An {IoT}-{Belief} {Rule} {Base} {Smart} {System} to {Assess} {Autism}},
	doi = {10.1109/CEEICT.2018.8628131},
	abstract = {An Internet-of-Things (IoT)-Belief Rule Base (BRB) based hybrid system is introduced to assess Autism spectrum disorder (ASD). This smart system can automatically collect sign and symptom data of various autistic children in real-time and classify the autistic children. The BRB subsystem incorporates knowledge representation parameters such as rule weight, attribute weight and degree of belief. The IoT-BRB system classifies the children having autism based on the sign and symptom collected by the pervasive sensing nodes. The classification results obtained from the proposed IoT-BRB smart system is compared with fuzzy and expert based system. The proposed system outperformed the state-of-the-art fuzzy system and expert system.},
	booktitle = {2018 4th {International} {Conference} on {Electrical} {Engineering} and {Information} {Communication} {Technology} ({iCEEiCT})},
	author = {Alam, M. E. and Kaiser, M. S. and Hossain, M. S. and Andersson, K.},
	month = sep,
	year = {2018},
	keywords = {Autism, Autism spectrum disorder, BRB subsystem, Electromyography, Expert systems, Heart rate, Internet of Things, Internet-of-Things-Belief Rule Base hybrid system, IoT-BRB smart system, IoT-Belief Rule Base smart system, Microphones, Uncertainty, Wireless sensor networks, attribute weight, autistic children, belief networks, degree-of-belief, expert based system, expert systems, fuzzy based system, fuzzy set theory, knowledge based systems, knowledge representation, knowledge representation parameters, medical computing, medical disorders, paediatrics, pervasive sensing nodes, rule weight, state-of-the-art fuzzy system},
	pages = {672--676},
}

An Internet-of-Things (IoT)-Belief Rule Base (BRB) based hybrid system is introduced to assess Autism spectrum disorder (ASD). This smart system can automatically collect sign and symptom data of various autistic children in real-time and classify the autistic children. The BRB subsystem incorporates knowledge representation parameters such as rule weight, attribute weight and degree of belief. The IoT-BRB system classifies the children having autism based on the sign and symptom collected by the pervasive sensing nodes. The classification results obtained from the proposed IoT-BRB smart system is compared with fuzzy and expert based system. The proposed system outperformed the state-of-the-art fuzzy system and expert system.

@article{afsana_energy_2018,
	title = {An {Energy} {Conserving} {Routing} {Scheme} for {Wireless} {Body} {Sensor} {Nanonetwork} {Communication}},
	volume = {6},
	issn = {2169-3536},
	doi = {10.1109/ACCESS.2018.2789437},
	abstract = {Current developments in nanotechnology make electromagnetic communication possible at the nanoscale for applications involving body sensor networks (BSNs). This specialized branch of wireless sensor networks, drawing attention from diverse fields, such as engineering, medicine, biology, physics, and computer science, has emerged as an important research area contributing to medical treatment, social welfare, and sports. The concept is based on the interaction of integrated nanoscale machines by means of wireless communications. One key hurdle for advancing nanocommunications is the lack of an apposite networking protocol to address the upcoming needs of the nanonetworks. Recently, some key challenges have been identified, such as nanonodes with extreme energy constraints, limited computational capabilities, terahertz frequency bands with limited transmission range, and so on, in designing protocols for wireless nanosensor networks. This work proposes an improved performance scheme of nanocommunication over terahertz bands for wireless BSNs making it suitable for smart e-health applications. The scheme contains - a new energy-efficient forwarding routine for electromagnetic communication in wireless nanonetworks consisting of hybrid clusters with centralized scheduling; a model designed for channel behavior taking into account the aggregated impact of molecular absorption, spreading loss, and shadowing; and an energy model for energy harvesting and consumption. The outage probability is derived for both single and multilinks and extended to determine the outage capacity. The outage probability for a multilink is derived using a cooperative fusion technique at a predefined fusion node. Simulated using a nano-sim simulator, performance of the proposed model has been evaluated for energy efficiency, outage capacity, and outage probability. The results demonstrate the efficiency of the proposed scheme through maximized energy utilization in both single and multihop communications; multisensor fusion at the fusion node enhances the link quality of the transmission.},
	journal = {IEEE Access},
	author = {Afsana, F. and Asif-Ur-Rahman, M. and Ahmed, M. R. and Mahmud, M. and Kaiser, M. S.},
	year = {2018},
	note = {Conference Name: IEEE Access},
	keywords = {EM communication, Energy harvesting, Nanobioscience, Nanoscale devices, Protocols, Routing, Wireless communication, Wireless sensor networks, apposite networking protocol, body sensor networks, centralized scheduling, cooperative fusion technique, electromagnetic communication, energy conserving routing scheme, energy efficiency, energy harvesting, energy model, energy-efficient forwarding routine, extreme energy constraints, integrated nanoscale machines, maximized energy utilization, multihop communications, multisensor fusion, nano cluster, nanocommunication, nanosensors, outage capacity, outage probability, probability, routing protocols, sensor fusion, smart e-health applications, telecommunication power management, terahertz band, terahertz frequency bands, wireless BSNs, wireless body sensor nanonetwork communication, wireless nanonetworks, wireless nanosensor networks, wireless sensor networks},
	pages = {9186--9200},
}

Current developments in nanotechnology make electromagnetic communication possible at the nanoscale for applications involving body sensor networks (BSNs). This specialized branch of wireless sensor networks, drawing attention from diverse fields, such as engineering, medicine, biology, physics, and computer science, has emerged as an important research area contributing to medical treatment, social welfare, and sports. The concept is based on the interaction of integrated nanoscale machines by means of wireless communications. One key hurdle for advancing nanocommunications is the lack of an apposite networking protocol to address the upcoming needs of the nanonetworks. Recently, some key challenges have been identified, such as nanonodes with extreme energy constraints, limited computational capabilities, terahertz frequency bands with limited transmission range, and so on, in designing protocols for wireless nanosensor networks. This work proposes an improved performance scheme of nanocommunication over terahertz bands for wireless BSNs making it suitable for smart e-health applications. The scheme contains - a new energy-efficient forwarding routine for electromagnetic communication in wireless nanonetworks consisting of hybrid clusters with centralized scheduling; a model designed for channel behavior taking into account the aggregated impact of molecular absorption, spreading loss, and shadowing; and an energy model for energy harvesting and consumption. The outage probability is derived for both single and multilinks and extended to determine the outage capacity. The outage probability for a multilink is derived using a cooperative fusion technique at a predefined fusion node. Simulated using a nano-sim simulator, performance of the proposed model has been evaluated for energy efficiency, outage capacity, and outage probability. The results demonstrate the efficiency of the proposed scheme through maximized energy utilization in both single and multihop communications; multisensor fusion at the fusion node enhances the link quality of the transmission.

@article{kaiser_et_al_neuro-fuzzy_2017,
	title = {Neuro-fuzzy selection algorithm for optimal relaying in {OFDM} systems},
	volume = {10},
	issn = {1754-8632},
	url = {https://doi.org/10.1504/IJAACS.2017.084713},
	doi = {10.1504/IJAACS.2017.084713},
	abstract = {Next generation wireless networks will support cooperative communication, where relay nodes receive transmission from a source node and forward it to its destination node. The best relay selection is one of the key concerns to improve the overall performance of the cooperative networks. This paper presents a neuro-fuzzy NF selection algorithm for optimal relaying in OFDM-based cooperative networks. The aim is to select relays based on instantaneous signal-to-noise ratio SNR, link delay propagation and queuing delay and energy saving due to the cooperative diversity. This paper also provides the end-to-end outage probability analysis. The NF-based relay selection algorithm is compared with the blind search, informed search, fuzzy-based search and selection amplify-and-forward AF with power allocation algorithms. The simulation results and complexity analysis show that the proposed algorithm provides substantial performance improvement over the conventional algorithms.},
	number = {2},
	urldate = {2021-03-22},
	journal = {International Journal of Autonomous and Adaptive Communications Systems},
	author = {Kaiser et al., M Shamim},
	year = {2017},
	pages = {213--235},
}

Next generation wireless networks will support cooperative communication, where relay nodes receive transmission from a source node and forward it to its destination node. The best relay selection is one of the key concerns to improve the overall performance of the cooperative networks. This paper presents a neuro-fuzzy NF selection algorithm for optimal relaying in OFDM-based cooperative networks. The aim is to select relays based on instantaneous signal-to-noise ratio SNR, link delay propagation and queuing delay and energy saving due to the cooperative diversity. This paper also provides the end-to-end outage probability analysis. The NF-based relay selection algorithm is compared with the blind search, informed search, fuzzy-based search and selection amplify-and-forward AF with power allocation algorithms. The simulation results and complexity analysis show that the proposed algorithm provides substantial performance improvement over the conventional algorithms.

@misc{noauthor_towards_nodate,
	title = {Towards {Machine} {Learning}-{Based} {Emotion} {Recognition} from {Multimodal} {Data} {\textbar} {SpringerLink}},
	url = {https://link.springer.com/chapter/10.1007/978-981-19-5191-6_9},
	urldate = {2023-04-30},
}

@misc{noauthor_inverted_nodate,
	title = {Inverted bell-curve-based ensemble of deep learning models for detection of {COVID}-19 from chest {X}-rays {\textbar} {SpringerLink}},
	url = {https://link.springer.com/article/10.1007/s00521-021-06737-6},
	urldate = {2022-11-20},
}

@misc{noauthor_sustainable_nodate,
	title = {Sustainable {Developments} by {Artificial} {Intelligence} and {Machine} {Learning} for {Renewable} {Energies} - 1st {Edition}},
	url = {https://www.elsevier.com/books/sustainable-developments-by-artificial-intelligence-and-machine-learning-for-renewable-energies/kumar/978-0-323-91228-0},
	urldate = {2022-11-20},
}

@misc{noauthor_iet_nodate,
	title = {{IET} {Digital} {Library}: {Internet} of {Things} ({IoT}) and blockchain-based solutions to confront {COVID}-19 pandemic},
	url = {https://digital-library.theiet.org/content/books/10.1049/pbhe042e_ch1},
	urldate = {2022-11-20},
}