Development of a deep learning model for the histologic diagnosis of dysplasia in Barrett's esophagus.


Journal

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

Informations de publication

Date de publication:
12 2022
Historique:
received: 19 04 2022
revised: 02 06 2022
accepted: 08 06 2022
pubmed: 20 6 2022
medline: 23 11 2022
entrez: 19 6 2022
Statut: ppublish

Résumé

The risk of progression in Barrett's esophagus (BE) increases with development of dysplasia. There is a critical need to improve the diagnosis of BE dysplasia, given substantial interobserver disagreement among expert pathologists and overdiagnosis of dysplasia by community pathologists. We developed a deep learning model to predict dysplasia grade on whole-slide imaging. We digitized nondysplastic BE (NDBE), low-grade dysplasia (LGD), and high-grade dysplasia (HGD) histology slides. Two expert pathologists confirmed all histology and digitally annotated areas of dysplasia. Training, validation, and test sets were created (by a random 70/20/10 split). We used an ensemble approach combining a "you only look once" model to identify regions of interest and histology class (NDBE, LGD, or HGD) followed by a ResNet101 model pretrained on ImageNet applied to the regions of interest. Diagnostic performance was determined for the whole slide. We included slides from 542 patients (164 NDBE, 226 LGD, and 152 HGD) yielding 8596 bounding boxes in the training set, 1946 bounding boxes in the validation set, and 840 boxes in the test set. When the ensemble model was used, sensitivity and specificity for LGD was 81.3% and 100%, respectively, and >90% for NDBE and HGD. The overall positive predictive value and sensitivity metric (calculated as F1 score) was .91 for NDBE, .90 for LGD, and 1.0 for HGD. We successfully trained and validated a deep learning model to accurately identify dysplasia on whole-slide images. This model can potentially help improve the histologic diagnosis of BE dysplasia and the appropriate application of endoscopic therapy.

Sections du résumé

BACKGROUND AND AIMS
The risk of progression in Barrett's esophagus (BE) increases with development of dysplasia. There is a critical need to improve the diagnosis of BE dysplasia, given substantial interobserver disagreement among expert pathologists and overdiagnosis of dysplasia by community pathologists. We developed a deep learning model to predict dysplasia grade on whole-slide imaging.
METHODS
We digitized nondysplastic BE (NDBE), low-grade dysplasia (LGD), and high-grade dysplasia (HGD) histology slides. Two expert pathologists confirmed all histology and digitally annotated areas of dysplasia. Training, validation, and test sets were created (by a random 70/20/10 split). We used an ensemble approach combining a "you only look once" model to identify regions of interest and histology class (NDBE, LGD, or HGD) followed by a ResNet101 model pretrained on ImageNet applied to the regions of interest. Diagnostic performance was determined for the whole slide.
RESULTS
We included slides from 542 patients (164 NDBE, 226 LGD, and 152 HGD) yielding 8596 bounding boxes in the training set, 1946 bounding boxes in the validation set, and 840 boxes in the test set. When the ensemble model was used, sensitivity and specificity for LGD was 81.3% and 100%, respectively, and >90% for NDBE and HGD. The overall positive predictive value and sensitivity metric (calculated as F1 score) was .91 for NDBE, .90 for LGD, and 1.0 for HGD.
CONCLUSIONS
We successfully trained and validated a deep learning model to accurately identify dysplasia on whole-slide images. This model can potentially help improve the histologic diagnosis of BE dysplasia and the appropriate application of endoscopic therapy.

Identifiants

pubmed: 35718071
pii: S0016-5107(22)01764-3
doi: 10.1016/j.gie.2022.06.013
pii:
doi:

Types de publication

Journal Article

Langues

eng

Sous-ensembles de citation

IM

Pagination

918-925.e3

Commentaires et corrections

Type : CommentIn

Informations de copyright

Copyright © 2022 American Society for Gastrointestinal Endoscopy. Published by Elsevier Inc. All rights reserved.

Auteurs

Shahriar Faghani (S)

Artificial Intelligence Laboratory, Mayo Clinic, Rochester, Minnesota, USA; Department of Radiology, Mayo Clinic, Rochester, Minnesota, USA.

D Chamil Codipilly (DC)

Barrett's Esophagus Unit, Division of Gastroenterology and Hepatology, Mayo Clinic, Rochester, Minnesota, USA.
Artificial Intelligence Laboratory, Mayo Clinic, Rochester, Minnesota, USA; Department of Radiology, Mayo Clinic, Rochester, Minnesota, USA.

Mana Moassefi (M)

Artificial Intelligence Laboratory, Mayo Clinic, Rochester, Minnesota, USA; Department of Radiology, Mayo Clinic, Rochester, Minnesota, USA.

Pouria Rouzrokh (P)

Artificial Intelligence Laboratory, Mayo Clinic, Rochester, Minnesota, USA; Department of Radiology, Mayo Clinic, Rochester, Minnesota, USA.

Bardia Khosravi (B)

Artificial Intelligence Laboratory, Mayo Clinic, Rochester, Minnesota, USA; Department of Radiology, Mayo Clinic, Rochester, Minnesota, USA.

Siddharth Agarwal (S)

Barrett's Esophagus Unit, Division of Gastroenterology and Hepatology, Mayo Clinic, Rochester, Minnesota, USA.

Lovekirat Dhaliwal (L)

Barrett's Esophagus Unit, Division of Gastroenterology and Hepatology, Mayo Clinic, Rochester, Minnesota, USA.

David A Katzka (DA)

Barrett's Esophagus Unit, Division of Gastroenterology and Hepatology, Mayo Clinic, Rochester, Minnesota, USA.

Catherine Hagen (C)

Department of Pathology, Mayo Clinic, Rochester, Minnesota, USA; (5)Department of Pathology, Mayo Clinic, Jacksonville, Florida, USA.

Jason Lewis (J)

Department of Pathology, Mayo Clinic, Jacksonville, Florida, USA.

Cadman L Leggett (CL)

Barrett's Esophagus Unit, Division of Gastroenterology and Hepatology, Mayo Clinic, Rochester, Minnesota, USA.

Bradley J Erickson (BJ)

Artificial Intelligence Laboratory, Mayo Clinic, Rochester, Minnesota, USA; Department of Radiology, Mayo Clinic, Rochester, Minnesota, USA.

Prasad G Iyer (PG)

Barrett's Esophagus Unit, Division of Gastroenterology and Hepatology, Mayo Clinic, Rochester, Minnesota, USA.

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