Detection of lung cancer with the fusion of computed tomography and positron emission tomography. Kaur, J., Pancholi, S., & Joshi, A. Volume 828 , 2018. doi abstract bibtex © Springer Nature Singapore Pte Ltd. 2018. In this paper, a wavelet fusion based cancer detection methodology has been presented. The database includes 200 samples of CT scan, and 200 PET scans out of which 50% samples of each were normal. Decomposition of CT scan and PET scan images have been performed by using wavelet transform (using haar as a mother wavelet) of depth 5 and combining the details of decomposed images by using averaging fusion rule and inverse wavelet transform. Further, 200 fused images have been segmented by manual cropping method to extract the useful information or region of interests (ROIs). 17 features of each of 200 ROIs have been extracted using GLCM and feature vectors are prepared. Subsequently, the features have been classified using two classifiers support vector machine (SVM) and k-nearest neighbors algorithm (k-NN) using their different kernels. The accuracy of SVM vary from 95.5%–98% and accuracy of k-NN vary from 69.5%–95.5%. This indicates that fused images can be a more powerful tool to diagnose the lung cancer.
@book{
title = {Detection of lung cancer with the fusion of computed tomography and positron emission tomography},
type = {book},
year = {2018},
source = {Communications in Computer and Information Science},
keywords = {CT scan,Feature extraction,Image fusion,K-NN,PET scan,SVM},
volume = {828},
id = {1c8b85a3-c2cd-3593-941a-2ceea66d72da},
created = {2018-09-06T11:22:40.625Z},
file_attached = {false},
profile_id = {11ae403c-c558-3358-87f9-dadc957bb57d},
last_modified = {2018-09-06T11:22:40.625Z},
read = {false},
starred = {false},
authored = {true},
confirmed = {false},
hidden = {false},
private_publication = {false},
abstract = {© Springer Nature Singapore Pte Ltd. 2018. In this paper, a wavelet fusion based cancer detection methodology has been presented. The database includes 200 samples of CT scan, and 200 PET scans out of which 50% samples of each were normal. Decomposition of CT scan and PET scan images have been performed by using wavelet transform (using haar as a mother wavelet) of depth 5 and combining the details of decomposed images by using averaging fusion rule and inverse wavelet transform. Further, 200 fused images have been segmented by manual cropping method to extract the useful information or region of interests (ROIs). 17 features of each of 200 ROIs have been extracted using GLCM and feature vectors are prepared. Subsequently, the features have been classified using two classifiers support vector machine (SVM) and k-nearest neighbors algorithm (k-NN) using their different kernels. The accuracy of SVM vary from 95.5%–98% and accuracy of k-NN vary from 69.5%–95.5%. This indicates that fused images can be a more powerful tool to diagnose the lung cancer.},
bibtype = {book},
author = {Kaur, J. and Pancholi, S. and Joshi, A.M.},
doi = {10.1007/978-981-10-8660-1_72}
}
Downloads: 0
{"_id":"cZg5Fnvb2CNKkMGxp","bibbaseid":"kaur-pancholi-joshi-detectionoflungcancerwiththefusionofcomputedtomographyandpositronemissiontomography-2018","author_short":["Kaur, J.","Pancholi, S.","Joshi, A."],"bibdata":{"title":"Detection of lung cancer with the fusion of computed tomography and positron emission tomography","type":"book","year":"2018","source":"Communications in Computer and Information Science","keywords":"CT scan,Feature extraction,Image fusion,K-NN,PET scan,SVM","volume":"828","id":"1c8b85a3-c2cd-3593-941a-2ceea66d72da","created":"2018-09-06T11:22:40.625Z","file_attached":false,"profile_id":"11ae403c-c558-3358-87f9-dadc957bb57d","last_modified":"2018-09-06T11:22:40.625Z","read":false,"starred":false,"authored":"true","confirmed":false,"hidden":false,"private_publication":false,"abstract":"© Springer Nature Singapore Pte Ltd. 2018. In this paper, a wavelet fusion based cancer detection methodology has been presented. The database includes 200 samples of CT scan, and 200 PET scans out of which 50% samples of each were normal. Decomposition of CT scan and PET scan images have been performed by using wavelet transform (using haar as a mother wavelet) of depth 5 and combining the details of decomposed images by using averaging fusion rule and inverse wavelet transform. Further, 200 fused images have been segmented by manual cropping method to extract the useful information or region of interests (ROIs). 17 features of each of 200 ROIs have been extracted using GLCM and feature vectors are prepared. Subsequently, the features have been classified using two classifiers support vector machine (SVM) and k-nearest neighbors algorithm (k-NN) using their different kernels. The accuracy of SVM vary from 95.5%–98% and accuracy of k-NN vary from 69.5%–95.5%. This indicates that fused images can be a more powerful tool to diagnose the lung cancer.","bibtype":"book","author":"Kaur, J. and Pancholi, S. and Joshi, A.M.","doi":"10.1007/978-981-10-8660-1_72","bibtex":"@book{\n title = {Detection of lung cancer with the fusion of computed tomography and positron emission tomography},\n type = {book},\n year = {2018},\n source = {Communications in Computer and Information Science},\n keywords = {CT scan,Feature extraction,Image fusion,K-NN,PET scan,SVM},\n volume = {828},\n id = {1c8b85a3-c2cd-3593-941a-2ceea66d72da},\n created = {2018-09-06T11:22:40.625Z},\n file_attached = {false},\n profile_id = {11ae403c-c558-3358-87f9-dadc957bb57d},\n last_modified = {2018-09-06T11:22:40.625Z},\n read = {false},\n starred = {false},\n authored = {true},\n confirmed = {false},\n hidden = {false},\n private_publication = {false},\n abstract = {© Springer Nature Singapore Pte Ltd. 2018. In this paper, a wavelet fusion based cancer detection methodology has been presented. The database includes 200 samples of CT scan, and 200 PET scans out of which 50% samples of each were normal. Decomposition of CT scan and PET scan images have been performed by using wavelet transform (using haar as a mother wavelet) of depth 5 and combining the details of decomposed images by using averaging fusion rule and inverse wavelet transform. Further, 200 fused images have been segmented by manual cropping method to extract the useful information or region of interests (ROIs). 17 features of each of 200 ROIs have been extracted using GLCM and feature vectors are prepared. Subsequently, the features have been classified using two classifiers support vector machine (SVM) and k-nearest neighbors algorithm (k-NN) using their different kernels. The accuracy of SVM vary from 95.5%–98% and accuracy of k-NN vary from 69.5%–95.5%. This indicates that fused images can be a more powerful tool to diagnose the lung cancer.},\n bibtype = {book},\n author = {Kaur, J. and Pancholi, S. and Joshi, A.M.},\n doi = {10.1007/978-981-10-8660-1_72}\n}","author_short":["Kaur, J.","Pancholi, S.","Joshi, A."],"biburl":"https://bibbase.org/service/mendeley/11ae403c-c558-3358-87f9-dadc957bb57d","bibbaseid":"kaur-pancholi-joshi-detectionoflungcancerwiththefusionofcomputedtomographyandpositronemissiontomography-2018","role":"author","urls":{},"keyword":["CT scan","Feature extraction","Image fusion","K-NN","PET scan","SVM"],"metadata":{"authorlinks":{}},"downloads":0},"bibtype":"book","biburl":"https://bibbase.org/service/mendeley/11ae403c-c558-3358-87f9-dadc957bb57d","dataSources":["2252seNhipfTmjEBQ"],"keywords":["ct scan","feature extraction","image fusion","k-nn","pet scan","svm"],"search_terms":["detection","lung","cancer","fusion","computed","tomography","positron","emission","tomography","kaur","pancholi","joshi"],"title":"Detection of lung cancer with the fusion of computed tomography and positron emission tomography","year":2018}