King's College London

Research portal

Artificial intelligence for the real-time classification of intrapapillary capillary loop patterns in the endoscopic diagnosis of early oesophageal squamous cell carcinoma: A proof-of-concept study

Research output: Contribution to journalArticle

Standard

Artificial intelligence for the real-time classification of intrapapillary capillary loop patterns in the endoscopic diagnosis of early oesophageal squamous cell carcinoma: A proof-of-concept study. / Everson, M; Garcia-Peraza-Herrera, Luis Carlos; Li, W; Luengo, I Muntion; Ahmad, O; Banks, M; Magee, C; Alzoubaidi, D; Hsu, Hm; Graham, D; Vercauteren, T; Lovat, Laurence B; Ourselin, S; Kashin, S; Wang, Hsiu-po; Wang, Wen-lun; Haidry, Rj.

In: United European Gastroenterology Journal, Vol. 7, No. 2, 01.03.2019, p. 297-306.

Research output: Contribution to journalArticle

Harvard

Everson, M, Garcia-Peraza-Herrera, LC, Li, W, Luengo, IM, Ahmad, O, Banks, M, Magee, C, Alzoubaidi, D, Hsu, H, Graham, D, Vercauteren, T, Lovat, LB, Ourselin, S, Kashin, S, Wang, H, Wang, W & Haidry, R 2019, 'Artificial intelligence for the real-time classification of intrapapillary capillary loop patterns in the endoscopic diagnosis of early oesophageal squamous cell carcinoma: A proof-of-concept study', United European Gastroenterology Journal, vol. 7, no. 2, pp. 297-306. https://doi.org/10.1177/2050640618821800

APA

Everson, M., Garcia-Peraza-Herrera, L. C., Li, W., Luengo, I. M., Ahmad, O., Banks, M., ... Haidry, R. (2019). Artificial intelligence for the real-time classification of intrapapillary capillary loop patterns in the endoscopic diagnosis of early oesophageal squamous cell carcinoma: A proof-of-concept study. United European Gastroenterology Journal, 7(2), 297-306. https://doi.org/10.1177/2050640618821800

Vancouver

Everson M, Garcia-Peraza-Herrera LC, Li W, Luengo IM, Ahmad O, Banks M et al. Artificial intelligence for the real-time classification of intrapapillary capillary loop patterns in the endoscopic diagnosis of early oesophageal squamous cell carcinoma: A proof-of-concept study. United European Gastroenterology Journal. 2019 Mar 1;7(2):297-306. https://doi.org/10.1177/2050640618821800

Author

Everson, M ; Garcia-Peraza-Herrera, Luis Carlos ; Li, W ; Luengo, I Muntion ; Ahmad, O ; Banks, M ; Magee, C ; Alzoubaidi, D ; Hsu, Hm ; Graham, D ; Vercauteren, T ; Lovat, Laurence B ; Ourselin, S ; Kashin, S ; Wang, Hsiu-po ; Wang, Wen-lun ; Haidry, Rj. / Artificial intelligence for the real-time classification of intrapapillary capillary loop patterns in the endoscopic diagnosis of early oesophageal squamous cell carcinoma: A proof-of-concept study. In: United European Gastroenterology Journal. 2019 ; Vol. 7, No. 2. pp. 297-306.

Bibtex Download

@article{ab122a9ce9424a5884db74c002e65961,
title = "Artificial intelligence for the real-time classification of intrapapillary capillary loop patterns in the endoscopic diagnosis of early oesophageal squamous cell carcinoma: A proof-of-concept study",
abstract = "Background: Intrapapillary capillary loops (IPCLs) represent an endoscopically visible feature of early squamous cell neoplasia (ESCN) which correlate with invasion depth – an important factor in the success of curative endoscopic therapy. IPCLs visualised on magnification endoscopy with Narrow Band Imaging (ME-NBI) can be used to train convolutional neural networks (CNNs) to detect the presence and classify staging of ESCN lesions. Methods: A total of 7046 sequential high-definition ME-NBI images from 17 patients (10 ESCN, 7 normal) were used to train a CNN. IPCL patterns were classified by three expert endoscopists according to the Japanese Endoscopic Society classification. Normal IPCLs were defined as type A, abnormal as B1–3. Matched histology was obtained for all imaged areas. Results: This CNN differentiates abnormal from normal IPCL patterns with 93.7{\%} accuracy (86.2{\%} to 98.3{\%}) and sensitivity and specificity for classifying abnormal IPCL patterns of 89.3{\%} (78.1{\%} to 100{\%}) and 98{\%} (92{\%} to 99.7{\%}), respectively. Our CNN operates in real time with diagnostic prediction times between 26.17 ms and 37.48 ms. Conclusion: Our novel and proof-of-concept application of computer-aided endoscopic diagnosis shows that a CNN can accurately classify IPCL patterns as normal or abnormal. This system could be used as an in vivo, real-time clinical decision support tool for endoscopists assessing and directing local therapy of ESCN.",
author = "M Everson and Garcia-Peraza-Herrera, {Luis Carlos} and W Li and Luengo, {I Muntion} and O Ahmad and M Banks and C Magee and D Alzoubaidi and Hm Hsu and D Graham and T Vercauteren and Lovat, {Laurence B} and S Ourselin and S Kashin and Hsiu-po Wang and Wen-lun Wang and Rj Haidry",
year = "2019",
month = "3",
day = "1",
doi = "10.1177/2050640618821800",
language = "English",
volume = "7",
pages = "297--306",
journal = "United European Gastroenterology Journal",
issn = "2050-6406",
publisher = "Sage Publications",
number = "2",

}

RIS (suitable for import to EndNote) Download

TY - JOUR

T1 - Artificial intelligence for the real-time classification of intrapapillary capillary loop patterns in the endoscopic diagnosis of early oesophageal squamous cell carcinoma: A proof-of-concept study

AU - Everson, M

AU - Garcia-Peraza-Herrera, Luis Carlos

AU - Li, W

AU - Luengo, I Muntion

AU - Ahmad, O

AU - Banks, M

AU - Magee, C

AU - Alzoubaidi, D

AU - Hsu, Hm

AU - Graham, D

AU - Vercauteren, T

AU - Lovat, Laurence B

AU - Ourselin, S

AU - Kashin, S

AU - Wang, Hsiu-po

AU - Wang, Wen-lun

AU - Haidry, Rj

PY - 2019/3/1

Y1 - 2019/3/1

N2 - Background: Intrapapillary capillary loops (IPCLs) represent an endoscopically visible feature of early squamous cell neoplasia (ESCN) which correlate with invasion depth – an important factor in the success of curative endoscopic therapy. IPCLs visualised on magnification endoscopy with Narrow Band Imaging (ME-NBI) can be used to train convolutional neural networks (CNNs) to detect the presence and classify staging of ESCN lesions. Methods: A total of 7046 sequential high-definition ME-NBI images from 17 patients (10 ESCN, 7 normal) were used to train a CNN. IPCL patterns were classified by three expert endoscopists according to the Japanese Endoscopic Society classification. Normal IPCLs were defined as type A, abnormal as B1–3. Matched histology was obtained for all imaged areas. Results: This CNN differentiates abnormal from normal IPCL patterns with 93.7% accuracy (86.2% to 98.3%) and sensitivity and specificity for classifying abnormal IPCL patterns of 89.3% (78.1% to 100%) and 98% (92% to 99.7%), respectively. Our CNN operates in real time with diagnostic prediction times between 26.17 ms and 37.48 ms. Conclusion: Our novel and proof-of-concept application of computer-aided endoscopic diagnosis shows that a CNN can accurately classify IPCL patterns as normal or abnormal. This system could be used as an in vivo, real-time clinical decision support tool for endoscopists assessing and directing local therapy of ESCN.

AB - Background: Intrapapillary capillary loops (IPCLs) represent an endoscopically visible feature of early squamous cell neoplasia (ESCN) which correlate with invasion depth – an important factor in the success of curative endoscopic therapy. IPCLs visualised on magnification endoscopy with Narrow Band Imaging (ME-NBI) can be used to train convolutional neural networks (CNNs) to detect the presence and classify staging of ESCN lesions. Methods: A total of 7046 sequential high-definition ME-NBI images from 17 patients (10 ESCN, 7 normal) were used to train a CNN. IPCL patterns were classified by three expert endoscopists according to the Japanese Endoscopic Society classification. Normal IPCLs were defined as type A, abnormal as B1–3. Matched histology was obtained for all imaged areas. Results: This CNN differentiates abnormal from normal IPCL patterns with 93.7% accuracy (86.2% to 98.3%) and sensitivity and specificity for classifying abnormal IPCL patterns of 89.3% (78.1% to 100%) and 98% (92% to 99.7%), respectively. Our CNN operates in real time with diagnostic prediction times between 26.17 ms and 37.48 ms. Conclusion: Our novel and proof-of-concept application of computer-aided endoscopic diagnosis shows that a CNN can accurately classify IPCL patterns as normal or abnormal. This system could be used as an in vivo, real-time clinical decision support tool for endoscopists assessing and directing local therapy of ESCN.

UR - http://www.scopus.com/inward/record.url?scp=85059896338&partnerID=8YFLogxK

U2 - 10.1177/2050640618821800

DO - 10.1177/2050640618821800

M3 - Article

VL - 7

SP - 297

EP - 306

JO - United European Gastroenterology Journal

JF - United European Gastroenterology Journal

SN - 2050-6406

IS - 2

ER -

View graph of relations

© 2018 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454