Please use this identifier to cite or link to this item: http://localhost/handle/Hannan/658739
Title: Colonic Motility Analysis Using the Wireless Capsule
Authors: Li Lu;Guozheng Yan;Xiangyan Kong;Kai Zhao;Fei Xu
subject: colonic motility classification|support vector machine|Feature extraction,|back propagation network
Year: 2016
Publisher: IEEE
Abstract: Many disorders are associated with abnormalities in colonic motility; thus, studying colonic motility is important to optimal management. This paper aimed to extract efficient features that characterize colonic motility, observe regional differences of colonic motility, and develop an effective and automated method for classifying colonic motility. Unlike the existing methods for analyzing the colonic motility, 27 features were extracted from 16 healthy subjects and 32 patients with constipation based on wavelet coefficients. Thereafter, the Wilcoxon rank sum test was used to select features, and finally, the adaboost-backpropagation neural network was used to classify the colonic motility. To evaluate the performance of the proposed method, the results were compared with the support vector machine and the back propagation network according to sensitivity, specificity, and accuracy. High prediction accuracy, sensitivity, and specificity confirm the effectiveness of the proposed methodology.
URI: http://localhost/handle/Hannan/162179
http://localhost/handle/Hannan/658739
ISSN: 1530-437X
1558-1748
volume: 16
issue: 9
Appears in Collections:2016

Files in This Item:
File Description SizeFormat 
7347354.pdf1.78 MBAdobe PDFThumbnail
Preview File
Title: Colonic Motility Analysis Using the Wireless Capsule
Authors: Li Lu;Guozheng Yan;Xiangyan Kong;Kai Zhao;Fei Xu
subject: colonic motility classification|support vector machine|Feature extraction,|back propagation network
Year: 2016
Publisher: IEEE
Abstract: Many disorders are associated with abnormalities in colonic motility; thus, studying colonic motility is important to optimal management. This paper aimed to extract efficient features that characterize colonic motility, observe regional differences of colonic motility, and develop an effective and automated method for classifying colonic motility. Unlike the existing methods for analyzing the colonic motility, 27 features were extracted from 16 healthy subjects and 32 patients with constipation based on wavelet coefficients. Thereafter, the Wilcoxon rank sum test was used to select features, and finally, the adaboost-backpropagation neural network was used to classify the colonic motility. To evaluate the performance of the proposed method, the results were compared with the support vector machine and the back propagation network according to sensitivity, specificity, and accuracy. High prediction accuracy, sensitivity, and specificity confirm the effectiveness of the proposed methodology.
URI: http://localhost/handle/Hannan/162179
http://localhost/handle/Hannan/658739
ISSN: 1530-437X
1558-1748
volume: 16
issue: 9
Appears in Collections:2016

Files in This Item:
File Description SizeFormat 
7347354.pdf1.78 MBAdobe PDFThumbnail
Preview File
Title: Colonic Motility Analysis Using the Wireless Capsule
Authors: Li Lu;Guozheng Yan;Xiangyan Kong;Kai Zhao;Fei Xu
subject: colonic motility classification|support vector machine|Feature extraction,|back propagation network
Year: 2016
Publisher: IEEE
Abstract: Many disorders are associated with abnormalities in colonic motility; thus, studying colonic motility is important to optimal management. This paper aimed to extract efficient features that characterize colonic motility, observe regional differences of colonic motility, and develop an effective and automated method for classifying colonic motility. Unlike the existing methods for analyzing the colonic motility, 27 features were extracted from 16 healthy subjects and 32 patients with constipation based on wavelet coefficients. Thereafter, the Wilcoxon rank sum test was used to select features, and finally, the adaboost-backpropagation neural network was used to classify the colonic motility. To evaluate the performance of the proposed method, the results were compared with the support vector machine and the back propagation network according to sensitivity, specificity, and accuracy. High prediction accuracy, sensitivity, and specificity confirm the effectiveness of the proposed methodology.
URI: http://localhost/handle/Hannan/162179
http://localhost/handle/Hannan/658739
ISSN: 1530-437X
1558-1748
volume: 16
issue: 9
Appears in Collections:2016

Files in This Item:
File Description SizeFormat 
7347354.pdf1.78 MBAdobe PDFThumbnail
Preview File