A clinically interpretable convolutional neural network for the real-time prediction of early squamous cell cancer of the esophagus: comparing diagnostic performance with a panel of expert European and Asian endoscopists.


Journal

Gastrointestinal endoscopy
ISSN: 1097-6779
Titre abrégé: Gastrointest Endosc
Pays: United States
ID NLM: 0010505

Informations de publication

Date de publication:
08 2021
Historique:
received: 20 10 2020
accepted: 29 01 2021
pubmed: 8 2 2021
medline: 11 8 2021
entrez: 7 2 2021
Statut: ppublish

Résumé

Intrapapillary capillary loops (IPCLs) are microvascular structures that correlate with the invasion depth of early squamous cell neoplasia and allow accurate prediction of histology. Artificial intelligence may improve human recognition of IPCL patterns and prediction of histology to allow prompt access to endoscopic therapy for early squamous cell neoplasia where appropriate. One hundred fifteen patients were recruited at 2 academic Taiwanese hospitals. Magnification endoscopy narrow-band imaging videos of squamous mucosa were labeled as dysplastic or normal according to their histology, and IPCL patterns were classified by consensus of 3 experienced clinicians. A convolutional neural network (CNN) was trained to classify IPCLs, using 67,742 high-quality magnification endoscopy narrow-band images by 5-fold cross validation. Performance measures were calculated to give an average F1 score, accuracy, sensitivity, and specificity. A panel of 5 Asian and 4 European experts predicted the histology of a random selection of 158 images using the Japanese Endoscopic Society IPCL classification; accuracy, sensitivity, specificity, positive and negative predictive values were calculated. Expert European Union (EU) and Asian endoscopists attained F1 scores (a measure of binary classification accuracy) of 97.0% and 98%, respectively. Sensitivity and accuracy of the EU and Asian clinicians were 97%, 98% and 96.9%, 97.1%, respectively. The CNN average F1 score was 94%, sensitivity 93.7%, and accuracy 91.7%. Our CNN operates at video rate and generates class activation maps that can be used to visually validate CNN predictions. We report a clinically interpretable CNN developed to predict histology based on IPCL patterns, in real time, using the largest reported dataset of images for this purpose. Our CNN achieved diagnostic performance comparable with an expert panel of endoscopists.

Sections du résumé

BACKGROUND AND AIMS
Intrapapillary capillary loops (IPCLs) are microvascular structures that correlate with the invasion depth of early squamous cell neoplasia and allow accurate prediction of histology. Artificial intelligence may improve human recognition of IPCL patterns and prediction of histology to allow prompt access to endoscopic therapy for early squamous cell neoplasia where appropriate.
METHODS
One hundred fifteen patients were recruited at 2 academic Taiwanese hospitals. Magnification endoscopy narrow-band imaging videos of squamous mucosa were labeled as dysplastic or normal according to their histology, and IPCL patterns were classified by consensus of 3 experienced clinicians. A convolutional neural network (CNN) was trained to classify IPCLs, using 67,742 high-quality magnification endoscopy narrow-band images by 5-fold cross validation. Performance measures were calculated to give an average F1 score, accuracy, sensitivity, and specificity. A panel of 5 Asian and 4 European experts predicted the histology of a random selection of 158 images using the Japanese Endoscopic Society IPCL classification; accuracy, sensitivity, specificity, positive and negative predictive values were calculated.
RESULTS
Expert European Union (EU) and Asian endoscopists attained F1 scores (a measure of binary classification accuracy) of 97.0% and 98%, respectively. Sensitivity and accuracy of the EU and Asian clinicians were 97%, 98% and 96.9%, 97.1%, respectively. The CNN average F1 score was 94%, sensitivity 93.7%, and accuracy 91.7%. Our CNN operates at video rate and generates class activation maps that can be used to visually validate CNN predictions.
CONCLUSIONS
We report a clinically interpretable CNN developed to predict histology based on IPCL patterns, in real time, using the largest reported dataset of images for this purpose. Our CNN achieved diagnostic performance comparable with an expert panel of endoscopists.

Identifiants

pubmed: 33549586
pii: S0016-5107(21)00106-1
doi: 10.1016/j.gie.2021.01.043
pii:
doi:

Types de publication

Journal Article

Langues

eng

Sous-ensembles de citation

IM

Pagination

273-281

Subventions

Organisme : Medical Research Council
ID : MC_PC_17180
Pays : United Kingdom

Informations de copyright

Copyright © 2021 American Society for Gastrointestinal Endoscopy. Published by Elsevier Inc. All rights reserved.

Auteurs

Martin A Everson (MA)

University College London Hospitals, London, United Kingdom.

Luis Garcia-Peraza-Herrera (L)

School of Biomedical Engineering & Imaging Sciences, Kings College London, London, United Kingdom.

Hsiu-Po Wang (HP)

National Taiwan University Hospital, Taipei, Taiwan.

Ching-Tai Lee (CT)

E-Da Hospital/I-Shou University, Kaohsiung, Taiwan.

Chen-Shuan Chung (CS)

Far Eastern Memorial Hospital, New Taipei City, Taiwan.

Ping-Hsin Hsieh (PH)

Chimei Medical Center, Tainan, Taiwan.

Chien-Chuan Chen (CC)

National Taiwan University Hospital, Taipei, Taiwan.

Cheng-Hao Tseng (CH)

E-Da Hospital/I-Shou University, Kaohsiung, Taiwan.

Ming-Hung Hsu (MH)

Department of Internal Medicine, E-Da Hospital/ I-Shou University, Kaohsiung, Taiwan.

Tom Vercauteren (T)

Department of Interventional Image Computing, Kings College London, London, United Kingdom.

Sebastien Ourselin (S)

School of Biomedical Engineering & Imaging Sciences, Kings College London, London, United Kingdom.

Sergey Kashin (S)

Department of Gastroenterology, Yaroslavl Oncology Hospital, Yaroslavl, Russian Federation.

Raf Bisschops (R)

Department of Gastroenterology, UZ Leuven, Leuven, Belgium.

Oliver Pech (O)

Department of Gastroenterology, Krankenhaus Barmherzige Bruder, Regensburg, Germany.

Laurence Lovat (L)

Department of Gastroenterology, University College London Hospitals, London, United Kingdom.

Wen-Lun Wang (WL)

Department of Internal Medicine, E-Da Hospital/ I-Shou University, Kaohsiung, Taiwan; School of Medicine, College of Medicine, I-Shou University, Kaohsiung, Taiwan.

Rehan J Haidry (RJ)

University College London Hospitals, London, United Kingdom.

Articles similaires

[Redispensing of expensive oral anticancer medicines: a practical application].

Lisanne N van Merendonk, Kübra Akgöl, Bastiaan Nuijen
1.00
Humans Antineoplastic Agents Administration, Oral Drug Costs Counterfeit Drugs

Smoking Cessation and Incident Cardiovascular Disease.

Jun Hwan Cho, Seung Yong Shin, Hoseob Kim et al.
1.00
Humans Male Smoking Cessation Cardiovascular Diseases Female
Humans United States Aged Cross-Sectional Studies Medicare Part C
1.00
Humans Yoga Low Back Pain Female Male

Classifications MeSH