Media Haze Classification in Retinal Images Using Transfer Learning with Convolutional Neural Networks. O'Berry*, J. & Liu, X. In Proceedings of the 23rd IEEE International Conference on Machine Learning and Applications (ICMLA), 2024. IEEE (Acceptance rate: <font color="red">22.5%</font>). ![Media Haze Classification in Retinal Images Using Transfer Learning with Convolutional Neural Networks [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper ![Media Haze Classification in Retinal Images Using Transfer Learning with Convolutional Neural Networks [pdf]](https://bibbase.org/img/filetypes/pdf.svg "pdf")
Paper abstract bibtex 2 downloads Media haze (MH) is a serious eye condition that can lead to various issues. This is especially true since MH is often an early warning sign of more severe conditions. If not treated quickly, it can potentially cause discomfort, blurred vision, or even blindness. Currently, MH is often detected by manually examining fundus images. This process allows for human error, where MH may not be detected. This study attempts to increase the detection and diagnosis of MH through our methods of preprocessing fundus images being examined, as well as the use of machine learning models. The models examined include LeNet-5, MLeNet (a modified LeNet), AlexNet, MobileNet, ResNet152v2, and VGG-16, where the former three models were trained on the RFMiD dataset, and the latter three were trained using pre-trained models fine-tuned on RFMiD. We found that all models using the proposed preprocessing demonstrated better performance than using the original images. The most effective model in this study was MobileNet, achieving 97.2% accuracy and 99.2% AUC score on the test set, outperforming the state-of-the-art model we found in the literature.
@inproceedings{conf/icmla24/OberryL,
author = {Jonathan O'Berry* and Xudong Liu},
booktitle = {Proceedings of the 23rd IEEE International Conference on Machine Learning and Applications (ICMLA)},
publisher = {IEEE (Acceptance rate: <font color="red">22.5%</font>)},
abstract = {Media haze (MH) is a serious eye condition that can lead to various issues. This is especially true since MH is often an early warning sign of more severe conditions. If not treated quickly, it can potentially cause discomfort, blurred vision, or even blindness. Currently, MH is often detected by manually examining fundus images. This process allows for human error, where MH may not be detected. This study attempts to increase the detection and diagnosis of MH through our methods of preprocessing fundus images being examined, as well as the use of machine learning models. The models examined include LeNet-5, MLeNet (a modified LeNet), AlexNet, MobileNet, ResNet152v2, and VGG-16, where the former three models were trained on the RFMiD dataset, and the latter three were trained using pre-trained models fine-tuned on RFMiD. We found that all models using the proposed preprocessing demonstrated better performance than using the original images. The most effective model in this study was MobileNet, achieving 97.2\% accuracy and 99.2\% AUC score on the test set, outperforming the state-of-the-art model we found in the literature.
},
url = {https://www.icmla-conference.org/icmla24/},
title = {Media Haze Classification in Retinal Images Using Transfer Learning with Convolutional Neural Networks},
url_Paper = {http://xudongliu.domains.unf.edu/resources/MH_icmla24.pdf},
year = 2024
}
Downloads: 2
{"_id":"8iLotfCTxwogXbBE6","bibbaseid":"oberry-liu-mediahazeclassificationinretinalimagesusingtransferlearningwithconvolutionalneuralnetworks-2024","author_short":["O'Berry*, J.","Liu, X."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"firstnames":["Jonathan"],"propositions":[],"lastnames":["O'Berry*"],"suffixes":[]},{"firstnames":["Xudong"],"propositions":[],"lastnames":["Liu"],"suffixes":[]}],"booktitle":"Proceedings of the 23rd IEEE International Conference on Machine Learning and Applications (ICMLA)","publisher":"IEEE (Acceptance rate: <font color=\"red\">22.5%</font>)","abstract":"Media haze (MH) is a serious eye condition that can lead to various issues. This is especially true since MH is often an early warning sign of more severe conditions. If not treated quickly, it can potentially cause discomfort, blurred vision, or even blindness. Currently, MH is often detected by manually examining fundus images. This process allows for human error, where MH may not be detected. This study attempts to increase the detection and diagnosis of MH through our methods of preprocessing fundus images being examined, as well as the use of machine learning models. The models examined include LeNet-5, MLeNet (a modified LeNet), AlexNet, MobileNet, ResNet152v2, and VGG-16, where the former three models were trained on the RFMiD dataset, and the latter three were trained using pre-trained models fine-tuned on RFMiD. We found that all models using the proposed preprocessing demonstrated better performance than using the original images. The most effective model in this study was MobileNet, achieving 97.2% accuracy and 99.2% AUC score on the test set, outperforming the state-of-the-art model we found in the literature. ","url":"https://www.icmla-conference.org/icmla24/","title":"Media Haze Classification in Retinal Images Using Transfer Learning with Convolutional Neural Networks","url_paper":"http://xudongliu.domains.unf.edu/resources/MH_icmla24.pdf","year":"2024","bibtex":"@inproceedings{conf/icmla24/OberryL,\n author = {Jonathan O'Berry* and Xudong Liu},\n booktitle = {Proceedings of the 23rd IEEE International Conference on Machine Learning and Applications (ICMLA)},\n publisher = {IEEE (Acceptance rate: <font color=\"red\">22.5%</font>)},\n abstract = {Media haze (MH) is a serious eye condition that can lead to various issues. This is especially true since MH is often an early warning sign of more severe conditions. If not treated quickly, it can potentially cause discomfort, blurred vision, or even blindness. Currently, MH is often detected by manually examining fundus images. This process allows for human error, where MH may not be detected. This study attempts to increase the detection and diagnosis of MH through our methods of preprocessing fundus images being examined, as well as the use of machine learning models. The models examined include LeNet-5, MLeNet (a modified LeNet), AlexNet, MobileNet, ResNet152v2, and VGG-16, where the former three models were trained on the RFMiD dataset, and the latter three were trained using pre-trained models fine-tuned on RFMiD. We found that all models using the proposed preprocessing demonstrated better performance than using the original images. The most effective model in this study was MobileNet, achieving 97.2\\% accuracy and 99.2\\% AUC score on the test set, outperforming the state-of-the-art model we found in the literature.\n},\n url = {https://www.icmla-conference.org/icmla24/},\n title = {Media Haze Classification in Retinal Images Using Transfer Learning with Convolutional Neural Networks},\n url_Paper = {http://xudongliu.domains.unf.edu/resources/MH_icmla24.pdf},\n year = 2024\n}\n\n","author_short":["O'Berry*, J.","Liu, X."],"key":"conf/icmla24/OberryL","id":"conf/icmla24/OberryL","bibbaseid":"oberry-liu-mediahazeclassificationinretinalimagesusingtransferlearningwithconvolutionalneuralnetworks-2024","role":"author","urls":{"Paper":"https://www.icmla-conference.org/icmla24/"," paper":"http://xudongliu.domains.unf.edu/resources/MH_icmla24.pdf"},"metadata":{"authorlinks":{}},"downloads":2},"bibtype":"inproceedings","biburl":"http://xudongliu.domains.unf.edu/resources/mypubs.bib","dataSources":["nZ4KWueyniryTd5G3","ZC4hkxYmM7m8bqanF","tckgwx8ugTmfQ6pQ4"],"keywords":[],"search_terms":["media","haze","classification","retinal","images","using","transfer","learning","convolutional","neural","networks","o'berry*","liu"],"title":"Media Haze Classification in Retinal Images Using Transfer Learning with Convolutional Neural Networks","year":2024,"downloads":2}