1. NCCU Academic Hub
  2. 學術產出
  3. College of Liberal Arts
  4. Periodical Articles
  5. Item

學術產出-Periodical Articles

Article View/Open

Publication Export

Google ScholarTM

政大圖書館

Citation Infomation

題名 Rapid Polyp Classification in Colonoscopy Using Textural and Convolutional Features
作者 羅崇銘
Lo, Chung-Ming
Yeh, Yu-Hsuan;Tang, Jui-Hsiang;Chang, Chun-Chao;Yeh, Hsing-Jung
貢獻者 圖檔所
關鍵詞 colorectal cancer; colon polyp; image features; convolutional neural network
日期 2022-08
上傳時間 2-Dec-2022 15:34:19 (UTC+8)
摘要 Colorectal cancer is the leading cause of cancer-associated morbidity and mortality worldwide. One of the causes of developing colorectal cancer is untreated colon adenomatous polyps. Clinically, polyps are detected in colonoscopy and the malignancies are determined according to the biopsy. To provide a quick and objective assessment to gastroenterologists, this study proposed a quantitative polyp classification via various image features in colonoscopy. The collected image database was composed of 1991 images including 1053 hyperplastic polyps and 938 adenomatous polyps and adenocarcinomas. From each image, textural features were extracted and combined in machine learning classifiers and machine-generated features were automatically selected in deep convolutional neural networks (DCNN). The DCNNs included AlexNet, Inception-V3, ResNet-101, and DenseNet-201. AlexNet trained from scratch achieved the best performance of 96.4% accuracy which is better than transfer learning and textural features. Using the prediction models, the malignancy level of polyps can be evaluated during a colonoscopy to provide a rapid treatment plan.
關聯 Healthcare, Vol.10, No.8, 1494
資料類型 article
DOI https://doi.org/10.3390/healthcare10081494
dc.contributor 圖檔所
dc.creator (作者) 羅崇銘
dc.creator (作者) Lo, Chung-Ming
dc.creator (作者) Yeh, Yu-Hsuan;Tang, Jui-Hsiang;Chang, Chun-Chao;Yeh, Hsing-Jung
dc.date (日期) 2022-08
dc.date.accessioned 2-Dec-2022 15:34:19 (UTC+8)-
dc.date.available 2-Dec-2022 15:34:19 (UTC+8)-
dc.date.issued (上傳時間) 2-Dec-2022 15:34:19 (UTC+8)-
dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/142698-
dc.description.abstract (摘要) Colorectal cancer is the leading cause of cancer-associated morbidity and mortality worldwide. One of the causes of developing colorectal cancer is untreated colon adenomatous polyps. Clinically, polyps are detected in colonoscopy and the malignancies are determined according to the biopsy. To provide a quick and objective assessment to gastroenterologists, this study proposed a quantitative polyp classification via various image features in colonoscopy. The collected image database was composed of 1991 images including 1053 hyperplastic polyps and 938 adenomatous polyps and adenocarcinomas. From each image, textural features were extracted and combined in machine learning classifiers and machine-generated features were automatically selected in deep convolutional neural networks (DCNN). The DCNNs included AlexNet, Inception-V3, ResNet-101, and DenseNet-201. AlexNet trained from scratch achieved the best performance of 96.4% accuracy which is better than transfer learning and textural features. Using the prediction models, the malignancy level of polyps can be evaluated during a colonoscopy to provide a rapid treatment plan.
dc.format.extent 106 bytes-
dc.format.mimetype text/html-
dc.relation (關聯) Healthcare, Vol.10, No.8, 1494
dc.subject (關鍵詞) colorectal cancer; colon polyp; image features; convolutional neural network
dc.title (題名) Rapid Polyp Classification in Colonoscopy Using Textural and Convolutional Features
dc.type (資料類型) article
dc.identifier.doi (DOI) 10.3390/healthcare10081494
dc.doi.uri (DOI) https://doi.org/10.3390/healthcare10081494