A new artificial intelligence system successfully detects and localises early neoplasia in Barrett's esophagus by using convolutional neural networks.
AI
Barrett's Esophagus
CAD
CNN
artificial intelligence
computer aided detection
convolutional neural networks
early detection
early neoplasia
neoplasia
Journal
United European gastroenterology journal
ISSN: 2050-6414
Titre abrégé: United European Gastroenterol J
Pays: England
ID NLM: 101606807
Informations de publication
Date de publication:
07 2022
07 2022
Historique:
received:
05
12
2021
accepted:
31
03
2022
pubmed:
7
5
2022
medline:
16
7
2022
entrez:
6
5
2022
Statut:
ppublish
Résumé
Seattle protocol biopsies for Barrett's Esophagus (BE) surveillance are labour intensive with low compliance. Dysplasia detection rates vary, leading to missed lesions. This can potentially be offset with computer aided detection. We have developed convolutional neural networks (CNNs) to identify areas of dysplasia and where to target biopsy. 119 Videos were collected in high-definition white light and optical chromoendoscopy with i-scan (Pentax Hoya, Japan) imaging in patients with dysplastic and non-dysplastic BE (NDBE). We trained an indirectly supervised CNN to classify images as dysplastic/non-dysplastic using whole video annotations to minimise selection bias and maximise accuracy. The CNN was trained using 148,936 video frames (31 dysplastic patients, 31 NDBE, two normal esophagus), validated on 25,161 images from 11 patient videos and tested on 264 iscan-1 images from 28 dysplastic and 16 NDBE patients which included expert delineations. To localise targeted biopsies/delineations, a second directly supervised CNN was generated based on expert delineations of 94 dysplastic images from 30 patients. This was tested on 86 i-scan one images from 28 dysplastic patients. The indirectly supervised CNN achieved a per image sensitivity in the test set of 91%, specificity 79%, area under receiver operator curve of 93% to detect dysplasia. Per-lesion sensitivity was 100%. Mean assessment speed was 48 frames per second (fps). 97% of targeted biopsy predictions matched expert and histological assessment at 56 fps. The artificial intelligence system performed better than six endoscopists. Our CNNs classify and localise dysplastic Barrett's Esophagus potentially supporting endoscopists during surveillance.
Sections du résumé
BACKGROUND AND AIMS
Seattle protocol biopsies for Barrett's Esophagus (BE) surveillance are labour intensive with low compliance. Dysplasia detection rates vary, leading to missed lesions. This can potentially be offset with computer aided detection. We have developed convolutional neural networks (CNNs) to identify areas of dysplasia and where to target biopsy.
METHODS
119 Videos were collected in high-definition white light and optical chromoendoscopy with i-scan (Pentax Hoya, Japan) imaging in patients with dysplastic and non-dysplastic BE (NDBE). We trained an indirectly supervised CNN to classify images as dysplastic/non-dysplastic using whole video annotations to minimise selection bias and maximise accuracy. The CNN was trained using 148,936 video frames (31 dysplastic patients, 31 NDBE, two normal esophagus), validated on 25,161 images from 11 patient videos and tested on 264 iscan-1 images from 28 dysplastic and 16 NDBE patients which included expert delineations. To localise targeted biopsies/delineations, a second directly supervised CNN was generated based on expert delineations of 94 dysplastic images from 30 patients. This was tested on 86 i-scan one images from 28 dysplastic patients.
FINDINGS
The indirectly supervised CNN achieved a per image sensitivity in the test set of 91%, specificity 79%, area under receiver operator curve of 93% to detect dysplasia. Per-lesion sensitivity was 100%. Mean assessment speed was 48 frames per second (fps). 97% of targeted biopsy predictions matched expert and histological assessment at 56 fps. The artificial intelligence system performed better than six endoscopists.
INTERPRETATION
Our CNNs classify and localise dysplastic Barrett's Esophagus potentially supporting endoscopists during surveillance.
Identifiants
pubmed: 35521666
doi: 10.1002/ueg2.12233
pmc: PMC9278593
doi:
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
528-537Subventions
Organisme : Wellcome Trust
Pays : United Kingdom
Organisme : Wellcome Trust
ID : 203145Z/16/Z
Pays : United Kingdom
Organisme : Cancer Research UK
Pays : United Kingdom
Organisme : Department of Health
Pays : United Kingdom
Commentaires et corrections
Type : CommentIn
Informations de copyright
© 2022 The Authors. United European Gastroenterology Journal published by Wiley Periodicals LLC on behalf of United European Gastroenterology.
Références
Endoscopy. 2016 Jul;48(7):617-24
pubmed: 27100718
Gastroenterology. 2016 Mar;150(3):599-607.e7; quiz e14-5
pubmed: 26619962
Gut. 2020 May;69(5):799-800
pubmed: 31615835
J Clin Gastroenterol. 2019 May/Jun;53(5):361-365
pubmed: 29608452
United European Gastroenterol J. 2019 May;7(4):538-547
pubmed: 31065371
Gastroenterology. 2016 Mar;150(3):591-8
pubmed: 26627609
Arch Pathol Lab Med. 2005 Feb;129(2):177-80
pubmed: 15679415
Gut. 2020 Apr;69(4):615-616
pubmed: 31541004
Gastrointest Endosc. 2020 Jun;91(6):1242-1250
pubmed: 31926965
Gastrointest Endosc. 2020 Jun;91(6):1264-1271.e1
pubmed: 31930967
Gastrointest Endosc. 2019 Feb;89(2):247-256.e4
pubmed: 30291849
Gastroenterol Res Pract. 2018 Aug 29;2018:1872437
pubmed: 30245711
Lancet Gastroenterol Hepatol. 2021 May;6(5):381-390
pubmed: 33713606
Lancet Oncol. 2019 Dec;20(12):1645-1654
pubmed: 31591062
Gastroenterology. 2020 Aug;159(2):512-520.e7
pubmed: 32371116
Gastroenterology. 2020 Mar;158(4):915-929.e4
pubmed: 31759929
Endoscopy. 2017 Dec;49(12):1219-1228
pubmed: 28732392
Gut. 2016 Apr;65(4):555-62
pubmed: 25731874
Gastrointest Endosc. 2020 Feb;91(2):310-320
pubmed: 31586576
Endoscopy. 2017 Feb;49(2):191-198
pubmed: 28122386
World J Gastroenterol. 2020 Oct 14;26(38):5784-5796
pubmed: 33132634
Endoscopy. 2021 Sep;53(9):893-901
pubmed: 33167043
Clin Gastroenterol Hepatol. 2009 Jul;7(7):736-42; quiz 710
pubmed: 19268726
United European Gastroenterol J. 2022 Jul;10(6):528-537
pubmed: 35521666