دورية أكاديمية
Artificial intelligence-enhanced white-light colonoscopy with attention guidance predicts colorectal cancer invasion depth.
| العنوان: | Artificial intelligence-enhanced white-light colonoscopy with attention guidance predicts colorectal cancer invasion depth. |
|---|---|
| المؤلفون: | Luo X; Department of Gastroenterology, Nanfang Hospital, Southern Medical University, Guangzhou, China., Wang J; Mechanobiology Institute, National University of Singapore, Singapore; Institute of Bioengineering and Nanotechnology, Agency for Science, Technology and Research (A∗STAR), Singapore., Han Z; Department of Gastroenterology, Nanfang Hospital, Southern Medical University, Guangzhou, China., Yu Y; Mechanobiology Institute, National University of Singapore, Singapore; Institute of Bioengineering and Nanotechnology, Agency for Science, Technology and Research (A∗STAR), Singapore; CAMP, Singapore-MIT Alliance for Research and Technology, Singapore., Chen Z; Department of Gastroenterology, Nanfang Hospital, Southern Medical University, Guangzhou, China., Huang F; Institute of Bioengineering and Nanotechnology, Agency for Science, Technology and Research (A∗STAR), Singapore., Xu Y; Mechanobiology Institute, National University of Singapore, Singapore., Cai J; Department of Gastroenterology, Nanfang Hospital, Southern Medical University, Guangzhou, China., Zhang Q; Department of Gastroenterology, Nanfang Hospital, Southern Medical University, Guangzhou, China., Qiao W; Department of Gastroenterology, Nanfang Hospital, Southern Medical University, Guangzhou, China., Ng IC; Department of Physiology, The Institute for Digital Medicine (WisDM), Yong Loo Lin School of Medicine, Singapore., Tan RT; Yale-NUS College, Singapore; Department of Electrical and Computer Engineering, National University of Singapore, Singapore., Liu S; Department of Gastroenterology, Nanfang Hospital, Southern Medical University, Guangzhou, China., Yu H; Department of Gastroenterology, Nanfang Hospital, Southern Medical University, Guangzhou, China; Mechanobiology Institute, National University of Singapore, Singapore; Institute of Bioengineering and Nanotechnology, Agency for Science, Technology and Research (A∗STAR), Singapore; CAMP, Singapore-MIT Alliance for Research and Technology, Singapore; Department of Physiology, The Institute for Digital Medicine (WisDM), Yong Loo Lin School of Medicine, Singapore. |
| المصدر: | Gastrointestinal endoscopy [Gastrointest Endosc] 2021 Sep; Vol. 94 (3), pp. 627-638.e1. Date of Electronic Publication: 2021 Apr 11. |
| نوع المنشور: | Journal Article |
| اللغة: | English |
| بيانات الدورية: | Publisher: Mosby Yearbook Country of Publication: United States NLM ID: 0010505 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1097-6779 (Electronic) Linking ISSN: 00165107 NLM ISO Abbreviation: Gastrointest Endosc Subsets: MEDLINE |
| أسماء مطبوعة: | Publication: St Louis, Mo : Mosby Yearbook Original Publication: Denver. |
| مواضيع طبية MeSH: | Colorectal Neoplasms*/diagnostic imaging , Endoscopic Mucosal Resection*, Artificial Intelligence ; Attention ; Colonoscopy ; Humans |
| مستخلص: | Background and Aims: Endoscopic submucosal dissection (ESD) and EMR are applied in treating superficial colorectal neoplasms but are contraindicated by deeply invasive colorectal cancer (CRC). The invasion depth of neoplasms can be examined by an automated artificial intelligence (AI) system to determine the applicability of ESD and EMR. Methods: A deep convolutional neural network with a tumor localization branch to guide invasion depth classification was constructed on the GoogLeNet architecture. The model was trained using 7734 nonmagnified white-light colonoscopy (WLC) images supplemented by image augmentation from 657 lesions labeled with histopathologic analysis of invasion depth. An independent testing dataset consisting of 1634 WLC images from 156 lesions was used to validate the model. Results: For predicting noninvasive and superficially invasive neoplasms, the model achieved an overall accuracy of 91.1% (95% confidence interval [CI], 89.6%-92.4%), with 91.2% sensitivity (95% CI, 88.8%-93.3%) and 91.0% specificity (95% CI, 89.0%-92.7%) at an optimal cutoff of .41 and the area under the receiver operating characteristic (AUROC) curve of .970 (95% CI, .962-.978). Inclusion of the advanced CRC data significantly increased the sensitivity in differentiating superficial neoplasms from deeply invasive early CRC to 65.3% (95% CI, 61.9%-68.8%) with an AUROC curve of .729 (95% CI, .699-.759), similar to experienced endoscopists (.691; 95% CI, .624-.758). Conclusions: We have developed an AI-enhanced attention-guided WLC system that differentiates noninvasive or superficially submucosal invasive neoplasms from deeply invasive CRC with high accuracy, sensitivity, and specificity. (Copyright © 2021 American Society for Gastrointestinal Endoscopy. Published by Elsevier Inc. All rights reserved.) |
| التعليقات: | Comment in: Gastrointest Endosc. 2021 Sep;94(3):639-640. (PMID: 34275607) Comment in: Gastroenterology. 2022 May;162(6):1769-1770. (PMID: 34919886) |
| تواريخ الأحداث: | Date Created: 20210414 Date Completed: 20210831 Latest Revision: 20220514 |
| رمز التحديث: | 20221213 |
| DOI: | 10.1016/j.gie.2021.03.936 |
| PMID: | 33852902 |
| قاعدة البيانات: | MEDLINE |
Full Text via ClinicalKey 
Full Text Finder | تدمد: | 1097-6779 |
|---|---|
| DOI: | 10.1016/j.gie.2021.03.936 |