Deep Learning in High-Resolution Anoscopy: Assessing the Impact of Staining and Therapeutic Manipulation on Automated Detection of Anal Cancer Precursors.

Autor:	Mascarenhas Saraiva M; Department of Gastroenterology, São João University Hospital, Alameda Professor Hernâni Monteiro, Porto, Portugal.; WGO Gastroenterology and Hepatology Training Center, Porto, Portugal.; Faculty of Medicine of the University of Porto, Alameda Professor Hernâni Monteiro, Porto, Portugal., Spindler L; Department of Proctology, GH Paris Saint-Joseph, Paris, France., Fathallah N; Department of Proctology, GH Paris Saint-Joseph, Paris, France., Beaussier H; Department of Clinical Research, GH Paris Saint-Joseph, Paris, France., Mamma C; Department of Clinical Research, GH Paris Saint-Joseph, Paris, France., Ribeiro T; Department of Gastroenterology, São João University Hospital, Alameda Professor Hernâni Monteiro, Porto, Portugal.; WGO Gastroenterology and Hepatology Training Center, Porto, Portugal., Afonso J; Department of Gastroenterology, São João University Hospital, Alameda Professor Hernâni Monteiro, Porto, Portugal.; WGO Gastroenterology and Hepatology Training Center, Porto, Portugal., Carvalho M; Department of Mechanical Engineering, Faculty of Engineering of the University of Porto, Porto, Portugal.; INEGI-Institute of Science and Innovation in Mechanical and Industrial Engineering, Porto, Portugal., Moura R; Department of Mechanical Engineering, Faculty of Engineering of the University of Porto, Porto, Portugal.; INEGI-Institute of Science and Innovation in Mechanical and Industrial Engineering, Porto, Portugal., Cardoso P; Department of Gastroenterology, São João University Hospital, Alameda Professor Hernâni Monteiro, Porto, Portugal.; WGO Gastroenterology and Hepatology Training Center, Porto, Portugal., Mendes F; Department of Gastroenterology, São João University Hospital, Alameda Professor Hernâni Monteiro, Porto, Portugal.; WGO Gastroenterology and Hepatology Training Center, Porto, Portugal., Martins M; Department of Gastroenterology, São João University Hospital, Alameda Professor Hernâni Monteiro, Porto, Portugal.; WGO Gastroenterology and Hepatology Training Center, Porto, Portugal., Adam J; Department of Pathology, GH Paris Saint-Joseph, Paris, France., Ferreira J; Department of Mechanical Engineering, Faculty of Engineering of the University of Porto, Porto, Portugal.; INEGI-Institute of Science and Innovation in Mechanical and Industrial Engineering, Porto, Portugal., Macedo G; Department of Gastroenterology, São João University Hospital, Alameda Professor Hernâni Monteiro, Porto, Portugal.; WGO Gastroenterology and Hepatology Training Center, Porto, Portugal.; Faculty of Medicine of the University of Porto, Alameda Professor Hernâni Monteiro, Porto, Portugal., de Parades V; Department of Proctology, GH Paris Saint-Joseph, Paris, France.
Jazyk:	angličtina
Zdroj:	Clinical and translational gastroenterology [Clin Transl Gastroenterol] 2024 Apr 01; Vol. 15 (4), pp. e00681. Date of Electronic Publication: 2024 Apr 01.
DOI:	10.14309/ctg.0000000000000681
Abstrakt:	Introduction: High-resolution anoscopy (HRA) is the gold standard for detecting anal squamous cell carcinoma (ASCC) precursors. Preliminary studies on the application of artificial intelligence (AI) models to this modality have revealed promising results. However, the impact of staining techniques and anal manipulation on the effectiveness of these algorithms has not been evaluated. We aimed to develop a deep learning system for automatic differentiation of high-grade squamous intraepithelial lesion vs low-grade squamous intraepithelial lesion in HRA images in different subsets of patients (nonstained, acetic acid, lugol, and after manipulation). Methods: A convolutional neural network was developed to detect and differentiate high-grade and low-grade anal squamous intraepithelial lesions based on 27,770 images from 103 HRA examinations performed in 88 patients. Subanalyses were performed to evaluate the algorithm's performance in subsets of images without staining, acetic acid, lugol, and after manipulation of the anal canal. The sensitivity, specificity, accuracy, positive and negative predictive values, and area under the curve were calculated. Results: The convolutional neural network achieved an overall accuracy of 98.3%. The algorithm had a sensitivity and specificity of 97.4% and 99.2%, respectively. The accuracy of the algorithm for differentiating high-grade squamous intraepithelial lesion vs low-grade squamous intraepithelial lesion varied between 91.5% (postmanipulation) and 100% (lugol) for the categories at subanalysis. The area under the curve ranged between 0.95 and 1.00. Discussion: The introduction of AI to HRA may provide an accurate detection and differentiation of ASCC precursors. Our algorithm showed excellent performance at different staining settings. This is extremely important because real-time AI models during HRA examinations can help guide local treatment or detect relapsing disease. (Copyright © 2024 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of The American College of Gastroenterology.)
Databáze:	MEDLINE
Externí odkaz:	Zobrazit plný text záznamu