Endoscopic prediction of submucosal invasion in Barrett's cancer with the use of artificial intelligence: a pilot study
Autor(a) principal: | |
---|---|
Data de Publicação: | 2020 |
Outros Autores: | , , , , , , , , , , , , , , , , , , , |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UNESP |
Texto Completo: | http://dx.doi.org/10.1055/a-1311-8570 http://hdl.handle.net/11449/210686 |
Resumo: | Background The accurate differentiation between T1a and T1b Barrett's-related cancer has both therapeutic and prognostic implications but is challenging even for experienced physicians. We trained an artificial intelligence (AI) system on the basis of deep artificial neural networks (deep learning) to differentiate between T1a and T1b Barrett's cancer on white-light images. Methods Endoscopic images from three tertiary care centers in Germany were collected retrospectively. A deep learning system was trained and tested using the principles of cross validation. A total of 230 white-light endoscopic images (108 T1a and 122 T1b) were evaluated using the AI system. For comparison, the images were also classified by experts specialized in endoscopic diagnosis and treatment of Barrett's cancer. Results The sensitivity, specificity, F1 score, and accuracy of the AI system in the differentiation between T1a and T1b cancer lesions was 0.77, 0.64, 0.74, and 0.71, respectively. There was no statistically significant difference between the performance of the AI system and that of experts, who showed sensitivity, specificity, F1, and accuracy of 0.63, 0.78, 0.67, and 0.70, respectively. Conclusion This pilot study demonstrates the first multicenter application of an AI-based system in the prediction of submucosal invasion in endoscopic images of Barrett's cancer. AI scored equally to international experts in the field, but more work is necessary to improve the system and apply it to video sequences and real-life settings. Nevertheless, the correct prediction of submucosal invasion in Barrett's cancer remains challenging for both experts and AI. |
id |
UNSP_0cf09ca52f0ceed5488848772cdd79aa |
---|---|
oai_identifier_str |
oai:repositorio.unesp.br:11449/210686 |
network_acronym_str |
UNSP |
network_name_str |
Repositório Institucional da UNESP |
repository_id_str |
2946 |
spelling |
Endoscopic prediction of submucosal invasion in Barrett's cancer with the use of artificial intelligence: a pilot studyBackground The accurate differentiation between T1a and T1b Barrett's-related cancer has both therapeutic and prognostic implications but is challenging even for experienced physicians. We trained an artificial intelligence (AI) system on the basis of deep artificial neural networks (deep learning) to differentiate between T1a and T1b Barrett's cancer on white-light images. Methods Endoscopic images from three tertiary care centers in Germany were collected retrospectively. A deep learning system was trained and tested using the principles of cross validation. A total of 230 white-light endoscopic images (108 T1a and 122 T1b) were evaluated using the AI system. For comparison, the images were also classified by experts specialized in endoscopic diagnosis and treatment of Barrett's cancer. Results The sensitivity, specificity, F1 score, and accuracy of the AI system in the differentiation between T1a and T1b cancer lesions was 0.77, 0.64, 0.74, and 0.71, respectively. There was no statistically significant difference between the performance of the AI system and that of experts, who showed sensitivity, specificity, F1, and accuracy of 0.63, 0.78, 0.67, and 0.70, respectively. Conclusion This pilot study demonstrates the first multicenter application of an AI-based system in the prediction of submucosal invasion in endoscopic images of Barrett's cancer. AI scored equally to international experts in the field, but more work is necessary to improve the system and apply it to video sequences and real-life settings. Nevertheless, the correct prediction of submucosal invasion in Barrett's cancer remains challenging for both experts and AI.Bavarian Academic Forum (BayWISS)Univ Klinikum Augsburg, Med Klin 3, Stenglinstr 2, D-86156 Augsburg, GermanyOstbayer TH Regensburg OTH Regensburg, Regensburg Med Image Comp ReMIC, Regensburg, GermanyOTH Regensburg, Regensburg Ctr Hlth Sci & Technol RCHST, Regensburg, GermanySana Klinikum Lichtenberg, Gastroenterol, Berlin, GermanyAsklepios Klin Barmbek, Dept Gastroenterol Hepatol & Intervent Endoscopy, Hamburg, GermanyOTH Regensburg, Regensburg Ctr Biomed Engn RCBE, Regensburg, GermanyRegensburg Univ, Regensburg, GermanySao Paulo State Univ, Dept Comp, Sao Paulo, BrazilCatholic Univ Louvain, Clin Univ St Luc, Brussels, BelgiumSaku Cent Hosp Adv Care Ctr, Nagano, JapanKlin Hirslanden, GastroZentrum, Zurich, SwitzerlandVet Affairs Med Ctr, Dept Gastroenterol & Hepatol, Kansas City, MO USAUniv Kansas, Sch Med, Kansas City, MO USAUniv British Columbia, Vancouver Gen Hosp, Div Gastroenterol, Vancouver, BC, CanadaSao Paulo State Univ, Dept Comp, Sao Paulo, BrazilGeorg Thieme Verlag KgUniv Klinikum AugsburgOstbayer TH Regensburg OTH RegensburgOTH RegensburgSana Klinikum LichtenbergAsklepios Klin BarmbekRegensburg UnivUniversidade Estadual Paulista (Unesp)Catholic Univ LouvainSaku Cent Hosp Adv Care CtrKlin HirslandenVet Affairs Med CtrUniv KansasUniv British ColumbiaEbigbo, AlannaMendel, RobertRueckert, TobiasSchuster, LaurinProbst, AndreasManzeneder, JohannesPrinz, FriederikeMende, MatthiasSteinbrueck, IngoFaiss, SiegbertRauber, DavidSouza, Luis A. de [UNESP]Papa, Joao P. [UNESP]Deprez, Pierre H.Oyama, TsuneoTakahashi, AkikoSeewald, StefanSharma, PrateekByrne, Michael F.Palm, ChristophMessmann, Helmut2021-06-26T02:53:52Z2021-06-26T02:53:52Z2020-11-16info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/article6http://dx.doi.org/10.1055/a-1311-8570Endoscopy. Stuttgart: Georg Thieme Verlag Kg, 6 p., 2020.0013-726Xhttp://hdl.handle.net/11449/21068610.1055/a-1311-8570WOS:000617034700001Web of Sciencereponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESPengEndoscopyinfo:eu-repo/semantics/openAccess2024-04-23T16:10:43Zoai:repositorio.unesp.br:11449/210686Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462024-04-23T16:10:43Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false |
dc.title.none.fl_str_mv |
Endoscopic prediction of submucosal invasion in Barrett's cancer with the use of artificial intelligence: a pilot study |
title |
Endoscopic prediction of submucosal invasion in Barrett's cancer with the use of artificial intelligence: a pilot study |
spellingShingle |
Endoscopic prediction of submucosal invasion in Barrett's cancer with the use of artificial intelligence: a pilot study Ebigbo, Alanna |
title_short |
Endoscopic prediction of submucosal invasion in Barrett's cancer with the use of artificial intelligence: a pilot study |
title_full |
Endoscopic prediction of submucosal invasion in Barrett's cancer with the use of artificial intelligence: a pilot study |
title_fullStr |
Endoscopic prediction of submucosal invasion in Barrett's cancer with the use of artificial intelligence: a pilot study |
title_full_unstemmed |
Endoscopic prediction of submucosal invasion in Barrett's cancer with the use of artificial intelligence: a pilot study |
title_sort |
Endoscopic prediction of submucosal invasion in Barrett's cancer with the use of artificial intelligence: a pilot study |
author |
Ebigbo, Alanna |
author_facet |
Ebigbo, Alanna Mendel, Robert Rueckert, Tobias Schuster, Laurin Probst, Andreas Manzeneder, Johannes Prinz, Friederike Mende, Matthias Steinbrueck, Ingo Faiss, Siegbert Rauber, David Souza, Luis A. de [UNESP] Papa, Joao P. [UNESP] Deprez, Pierre H. Oyama, Tsuneo Takahashi, Akiko Seewald, Stefan Sharma, Prateek Byrne, Michael F. Palm, Christoph Messmann, Helmut |
author_role |
author |
author2 |
Mendel, Robert Rueckert, Tobias Schuster, Laurin Probst, Andreas Manzeneder, Johannes Prinz, Friederike Mende, Matthias Steinbrueck, Ingo Faiss, Siegbert Rauber, David Souza, Luis A. de [UNESP] Papa, Joao P. [UNESP] Deprez, Pierre H. Oyama, Tsuneo Takahashi, Akiko Seewald, Stefan Sharma, Prateek Byrne, Michael F. Palm, Christoph Messmann, Helmut |
author2_role |
author author author author author author author author author author author author author author author author author author author author |
dc.contributor.none.fl_str_mv |
Univ Klinikum Augsburg Ostbayer TH Regensburg OTH Regensburg OTH Regensburg Sana Klinikum Lichtenberg Asklepios Klin Barmbek Regensburg Univ Universidade Estadual Paulista (Unesp) Catholic Univ Louvain Saku Cent Hosp Adv Care Ctr Klin Hirslanden Vet Affairs Med Ctr Univ Kansas Univ British Columbia |
dc.contributor.author.fl_str_mv |
Ebigbo, Alanna Mendel, Robert Rueckert, Tobias Schuster, Laurin Probst, Andreas Manzeneder, Johannes Prinz, Friederike Mende, Matthias Steinbrueck, Ingo Faiss, Siegbert Rauber, David Souza, Luis A. de [UNESP] Papa, Joao P. [UNESP] Deprez, Pierre H. Oyama, Tsuneo Takahashi, Akiko Seewald, Stefan Sharma, Prateek Byrne, Michael F. Palm, Christoph Messmann, Helmut |
description |
Background The accurate differentiation between T1a and T1b Barrett's-related cancer has both therapeutic and prognostic implications but is challenging even for experienced physicians. We trained an artificial intelligence (AI) system on the basis of deep artificial neural networks (deep learning) to differentiate between T1a and T1b Barrett's cancer on white-light images. Methods Endoscopic images from three tertiary care centers in Germany were collected retrospectively. A deep learning system was trained and tested using the principles of cross validation. A total of 230 white-light endoscopic images (108 T1a and 122 T1b) were evaluated using the AI system. For comparison, the images were also classified by experts specialized in endoscopic diagnosis and treatment of Barrett's cancer. Results The sensitivity, specificity, F1 score, and accuracy of the AI system in the differentiation between T1a and T1b cancer lesions was 0.77, 0.64, 0.74, and 0.71, respectively. There was no statistically significant difference between the performance of the AI system and that of experts, who showed sensitivity, specificity, F1, and accuracy of 0.63, 0.78, 0.67, and 0.70, respectively. Conclusion This pilot study demonstrates the first multicenter application of an AI-based system in the prediction of submucosal invasion in endoscopic images of Barrett's cancer. AI scored equally to international experts in the field, but more work is necessary to improve the system and apply it to video sequences and real-life settings. Nevertheless, the correct prediction of submucosal invasion in Barrett's cancer remains challenging for both experts and AI. |
publishDate |
2020 |
dc.date.none.fl_str_mv |
2020-11-16 2021-06-26T02:53:52Z 2021-06-26T02:53:52Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://dx.doi.org/10.1055/a-1311-8570 Endoscopy. Stuttgart: Georg Thieme Verlag Kg, 6 p., 2020. 0013-726X http://hdl.handle.net/11449/210686 10.1055/a-1311-8570 WOS:000617034700001 |
url |
http://dx.doi.org/10.1055/a-1311-8570 http://hdl.handle.net/11449/210686 |
identifier_str_mv |
Endoscopy. Stuttgart: Georg Thieme Verlag Kg, 6 p., 2020. 0013-726X 10.1055/a-1311-8570 WOS:000617034700001 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
Endoscopy |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
6 |
dc.publisher.none.fl_str_mv |
Georg Thieme Verlag Kg |
publisher.none.fl_str_mv |
Georg Thieme Verlag Kg |
dc.source.none.fl_str_mv |
Web of Science reponame:Repositório Institucional da UNESP instname:Universidade Estadual Paulista (UNESP) instacron:UNESP |
instname_str |
Universidade Estadual Paulista (UNESP) |
instacron_str |
UNESP |
institution |
UNESP |
reponame_str |
Repositório Institucional da UNESP |
collection |
Repositório Institucional da UNESP |
repository.name.fl_str_mv |
Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP) |
repository.mail.fl_str_mv |
|
_version_ |
1799964506095353856 |