A Pilot Study on Automatic Three-Dimensional Quantification of Barrett's Esophagus for Risk Stratification and Therapy Monitoring.
Aged
Automation
Barrett Esophagus
/ classification
Deep Learning
Disease Progression
Esophageal Mucosa
/ pathology
Esophagogastric Junction
/ pathology
Esophagoscopy
Female
Humans
Image Interpretation, Computer-Assisted
Imaging, Three-Dimensional
Male
Pilot Projects
Predictive Value of Tests
Reproducibility of Results
Risk Assessment
Risk Factors
Severity of Illness Index
Treatment Outcome
Deep learning
Esophageal cancer
Imaging
Risk assessment
Three-dimensional
Journal
Gastroenterology
ISSN: 1528-0012
Titre abrégé: Gastroenterology
Pays: United States
ID NLM: 0374630
Informations de publication
Date de publication:
09 2021
09 2021
Historique:
received:
02
02
2021
revised:
10
05
2021
accepted:
27
05
2021
pubmed:
12
6
2021
medline:
18
1
2022
entrez:
11
6
2021
Statut:
ppublish
Résumé
Barrett's epithelium measurement using widely accepted Prague C&M classification is highly operator dependent. We propose a novel methodology for measuring this risk score automatically. The method also enables quantification of the area of Barrett's epithelium (BEA) and islands, which was not possible before. Furthermore, it allows 3-dimensional (3D) reconstruction of the esophageal surface, enabling interactive 3D visualization. We aimed to assess the accuracy of the proposed artificial intelligence system on both phantom and endoscopic patient data. Using advanced deep learning, a depth estimator network is used to predict endoscope camera distance from the gastric folds. By segmenting BEA and gastroesophageal junction and projecting them to the estimated mm distances, we measure C&M scores including the BEA. The derived endoscopy artificial intelligence system was tested on a purpose-built 3D printed esophagus phantom with varying BEAs and on 194 high-definition videos from 131 patients with C&M values scored by expert endoscopists. Endoscopic phantom video data demonstrated a 97.2% accuracy with a marginal ± 0.9 mm average deviation for C&M and island measurements, while for BEA we achieved 98.4% accuracy with only ±0.4 cm The proposed methodology automatically extracts Prague C&M scores with high accuracy. Quantification and 3D reconstruction of the entire Barrett's area provides new opportunities for risk stratification and assessment of therapy response.
Sections du résumé
BACKGROUND & AIMS
Barrett's epithelium measurement using widely accepted Prague C&M classification is highly operator dependent. We propose a novel methodology for measuring this risk score automatically. The method also enables quantification of the area of Barrett's epithelium (BEA) and islands, which was not possible before. Furthermore, it allows 3-dimensional (3D) reconstruction of the esophageal surface, enabling interactive 3D visualization. We aimed to assess the accuracy of the proposed artificial intelligence system on both phantom and endoscopic patient data.
METHODS
Using advanced deep learning, a depth estimator network is used to predict endoscope camera distance from the gastric folds. By segmenting BEA and gastroesophageal junction and projecting them to the estimated mm distances, we measure C&M scores including the BEA. The derived endoscopy artificial intelligence system was tested on a purpose-built 3D printed esophagus phantom with varying BEAs and on 194 high-definition videos from 131 patients with C&M values scored by expert endoscopists.
RESULTS
Endoscopic phantom video data demonstrated a 97.2% accuracy with a marginal ± 0.9 mm average deviation for C&M and island measurements, while for BEA we achieved 98.4% accuracy with only ±0.4 cm
CONCLUSIONS
The proposed methodology automatically extracts Prague C&M scores with high accuracy. Quantification and 3D reconstruction of the entire Barrett's area provides new opportunities for risk stratification and assessment of therapy response.
Identifiants
pubmed: 34116029
pii: S0016-5085(21)03087-0
doi: 10.1053/j.gastro.2021.05.059
pii:
doi:
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Video-Audio Media
Langues
eng
Sous-ensembles de citation
IM
Pagination
865-878.e8Subventions
Organisme : Medical Research Council
ID : MC_UU_00008/7
Pays : United Kingdom
Organisme : Wellcome Trust
ID : 206314/Z/17/Z
Pays : United Kingdom
Organisme : Medical Research Council
ID : MR/S036377/1
Pays : United Kingdom
Organisme : Medical Research Council
ID : MC_PC_MR/S025952/1
Pays : United Kingdom
Organisme : Wellcome Trust
ID : 109965/Z/15/Z
Pays : United Kingdom
Organisme : Department of Health
Pays : United Kingdom
Organisme : Medical Research Council
ID : MC_UU_12010/7
Pays : United Kingdom
Investigateurs
Philip Allan
(P)
Tim Ambrose
(T)
Carolina Arancibia-Cárcamo
(C)
Ellie Barnes
(E)
Elizabeth Bird-Lieberman
(E)
Jan Bornschein
(J)
Oliver Brain
(O)
Jane Collier
(J)
Emma Culver
(E)
Alessandra Geremia
(A)
Bruce George
(B)
Lucy Howarth
(L)
Kelsey Jones
(K)
Paul Klenerman
(P)
Rebecca Palmer
(R)
Fiona Powrie
(F)
Astor Rodrigues
(A)
Jack Satsangi
(J)
Alison Simmons
(A)
Simon Travis
(S)
Holm Uhlig
(H)
Alissa Walsh
(A)
Commentaires et corrections
Type : CommentIn
Informations de copyright
Copyright © 2021 The Authors. Published by Elsevier Inc. All rights reserved.