Deep-Learning System Detects Neoplasia in Patients With Barrett's Esophagus With Higher Accuracy Than Endoscopists in a Multistep Training and Validation Study With Benchmarking.


Journal

Gastroenterology
ISSN: 1528-0012
Titre abrégé: Gastroenterology
Pays: United States
ID NLM: 0374630

Informations de publication

Date de publication:
03 2020
Historique:
received: 15 08 2019
revised: 31 10 2019
accepted: 18 11 2019
pubmed: 25 11 2019
medline: 29 4 2020
entrez: 25 11 2019
Statut: ppublish

Résumé

We aimed to develop and validate a deep-learning computer-aided detection (CAD) system, suitable for use in real time in clinical practice, to improve endoscopic detection of early neoplasia in patients with Barrett's esophagus (BE). We developed a hybrid ResNet-UNet model CAD system using 5 independent endoscopy data sets. We performed pretraining using 494,364 labeled endoscopic images collected from all intestinal segments. Then, we used 1704 unique esophageal high-resolution images of rigorously confirmed early-stage neoplasia in BE and nondysplastic BE, derived from 669 patients. System performance was assessed by using data sets 4 and 5. Data set 5 was also scored by 53 general endoscopists with a wide range of experience from 4 countries to benchmark CAD system performance. Coupled with histopathology findings, scoring of images that contained early-stage neoplasia in data sets 2-5 were delineated in detail for neoplasm position and extent by multiple experts whose evaluations served as the ground truth for segmentation. The CAD system classified images as containing neoplasms or nondysplastic BE with 89% accuracy, 90% sensitivity, and 88% specificity (data set 4, 80 patients and images). In data set 5 (80 patients and images) values for the CAD system vs those of the general endoscopists were 88% vs 73% accuracy, 93% vs 72% sensitivity, and 83% vs 74% specificity. The CAD system achieved higher accuracy than any of the individual 53 nonexpert endoscopists, with comparable delineation performance. CAD delineations of the area of neoplasm overlapped with those from the BE experts in all detected neoplasia in data sets 4 and 5. The CAD system identified the optimal site for biopsy of detected neoplasia in 97% and 92% of cases (data sets 4 and 5, respectively). We developed, validated, and benchmarked a deep-learning computer-aided system for primary detection of neoplasia in patients with BE. The system detected neoplasia with high accuracy and near-perfect delineation performance. The Netherlands National Trials Registry, Number: NTR7072.

Sections du résumé

BACKGROUND & AIMS
We aimed to develop and validate a deep-learning computer-aided detection (CAD) system, suitable for use in real time in clinical practice, to improve endoscopic detection of early neoplasia in patients with Barrett's esophagus (BE).
METHODS
We developed a hybrid ResNet-UNet model CAD system using 5 independent endoscopy data sets. We performed pretraining using 494,364 labeled endoscopic images collected from all intestinal segments. Then, we used 1704 unique esophageal high-resolution images of rigorously confirmed early-stage neoplasia in BE and nondysplastic BE, derived from 669 patients. System performance was assessed by using data sets 4 and 5. Data set 5 was also scored by 53 general endoscopists with a wide range of experience from 4 countries to benchmark CAD system performance. Coupled with histopathology findings, scoring of images that contained early-stage neoplasia in data sets 2-5 were delineated in detail for neoplasm position and extent by multiple experts whose evaluations served as the ground truth for segmentation.
RESULTS
The CAD system classified images as containing neoplasms or nondysplastic BE with 89% accuracy, 90% sensitivity, and 88% specificity (data set 4, 80 patients and images). In data set 5 (80 patients and images) values for the CAD system vs those of the general endoscopists were 88% vs 73% accuracy, 93% vs 72% sensitivity, and 83% vs 74% specificity. The CAD system achieved higher accuracy than any of the individual 53 nonexpert endoscopists, with comparable delineation performance. CAD delineations of the area of neoplasm overlapped with those from the BE experts in all detected neoplasia in data sets 4 and 5. The CAD system identified the optimal site for biopsy of detected neoplasia in 97% and 92% of cases (data sets 4 and 5, respectively).
CONCLUSIONS
We developed, validated, and benchmarked a deep-learning computer-aided system for primary detection of neoplasia in patients with BE. The system detected neoplasia with high accuracy and near-perfect delineation performance. The Netherlands National Trials Registry, Number: NTR7072.

Identifiants

pubmed: 31759929
pii: S0016-5085(19)41586-2
doi: 10.1053/j.gastro.2019.11.030
pii:
doi:

Types de publication

Journal Article Research Support, Non-U.S. Gov't Validation Study

Langues

eng

Sous-ensembles de citation

IM

Pagination

915-929.e4

Commentaires et corrections

Type : CommentIn

Informations de copyright

Copyright © 2020 AGA Institute. Published by Elsevier Inc. All rights reserved.

Auteurs

Albert J de Groof (AJ)

Department of Gastroenterology and Hepatology, Amsterdam University Medical Centers, University of Amsterdam, Amsterdam, The Netherlands.

Maarten R Struyvenberg (MR)

Department of Gastroenterology and Hepatology, Amsterdam University Medical Centers, University of Amsterdam, Amsterdam, The Netherlands.

Joost van der Putten (J)

Department of Electrical Engineering, Video Coding & Architectures group, Eindhoven University of Technology, Eindhoven, The Netherlands.

Fons van der Sommen (F)

Department of Electrical Engineering, Video Coding & Architectures group, Eindhoven University of Technology, Eindhoven, The Netherlands.

Kiki N Fockens (KN)

Department of Gastroenterology and Hepatology, Amsterdam University Medical Centers, University of Amsterdam, Amsterdam, The Netherlands.

Wouter L Curvers (WL)

Department of Gastroenterology and Hepatology, Catharina Hospital Eindhoven, Eindhoven, The Netherlands.

Sveta Zinger (S)

Department of Electrical Engineering, Video Coding & Architectures group, Eindhoven University of Technology, Eindhoven, The Netherlands.

Roos E Pouw (RE)

Department of Gastroenterology and Hepatology, Amsterdam University Medical Centers, University of Amsterdam, Amsterdam, The Netherlands.

Emmanuel Coron (E)

Institut des Maladies de l'Appareil Digestif, University Hospital of Nantes place Alexis Ricordeau, Nantes, France.

Francisco Baldaque-Silva (F)

Department of Digestive Diseases, Karolinska University Hospital and Karolinska Institute, Stockholm, Sweden.

Oliver Pech (O)

Gastroenterology and Interventional Endoscopy, Krankenhaus Barmherzige Brüder, Regensburg, Germany.

Bas Weusten (B)

Department of Gastroenterology and Hepatology, St. Antonius Hospital, Nieuwegein, The Netherlands.

Alexander Meining (A)

Center of Internal Medicine, Ulm University, Ulm, Germany.

Horst Neuhaus (H)

Internal Medicine, Evangelisches Krankenhaus Düsseldorf, Düsseldorf, Germany.

Raf Bisschops (R)

Department of Gastroenterology and Hepatology, University Hospitals Leuven, Leuven, Belgium.

John Dent (J)

Department of Medicine, University of Adelaide and Royal Adelaide Hospital, Adelaide, South Australia.

Erik J Schoon (EJ)

Department of Gastroenterology and Hepatology, Catharina Hospital Eindhoven, Eindhoven, The Netherlands.

Peter H de With (PH)

Department of Electrical Engineering, Video Coding & Architectures group, Eindhoven University of Technology, Eindhoven, The Netherlands.

Jacques J Bergman (JJ)

Department of Gastroenterology and Hepatology, Amsterdam University Medical Centers, University of Amsterdam, Amsterdam, The Netherlands. Electronic address: j.j.bergman@amsterdamumc.nl.

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