The use of deep learning on endoscopic images to assess the response of rectal cancer after chemoradiation.
Artificial intelligence
Deep learning
Organ preservation
Rectal cancer
Response evaluation
Watch-and-wait approach
Journal
Surgical endoscopy
ISSN: 1432-2218
Titre abrégé: Surg Endosc
Pays: Germany
ID NLM: 8806653
Informations de publication
Date de publication:
05 2022
05 2022
Historique:
received:
17
06
2020
accepted:
07
08
2021
pubmed:
14
10
2021
medline:
15
4
2022
entrez:
13
10
2021
Statut:
ppublish
Résumé
Accurate response evaluation is necessary to select complete responders (CRs) for a watch-and-wait approach. Deep learning may aid in this process, but so far has never been evaluated for this purpose. The aim was to evaluate the accuracy to assess response with deep learning methods based on endoscopic images in rectal cancer patients after neoadjuvant therapy. Rectal cancer patients diagnosed between January 2012 and December 2015 and treated with neoadjuvant (chemo)radiotherapy were retrospectively selected from a single institute. All patients underwent flexible endoscopy for response evaluation. Diagnostic performance (accuracy, area under the receiver operator characteristics curve (AUC), positive- and negative predictive values, sensitivities and specificities) of different open accessible deep learning networks was calculated. Reference standard was histology after surgery, or long-term outcome (>2 years of follow-up) in a watch-and-wait policy. 226 patients were included for the study (117(52%) were non-CRs; 109(48%) were CRs). The accuracy, AUC, positive- and negative predictive values, sensitivity and specificity of the different models varied from 0.67-0.75%, 0.76-0.83%, 67-74%, 70-78%, 68-79% to 66-75%, respectively. Overall, EfficientNet-B2 was the most successful model with the highest diagnostic performance. This pilot study shows that deep learning has a modest accuracy (AUCs 0.76-0.83). This is not accurate enough for clinical decision making, and lower than what is generally reported by experienced endoscopists. Deep learning models can however be further improved and may become useful to assist endoscopists in evaluating the response. More well-designed prospective studies are required.
Sections du résumé
BACKGROUND
Accurate response evaluation is necessary to select complete responders (CRs) for a watch-and-wait approach. Deep learning may aid in this process, but so far has never been evaluated for this purpose. The aim was to evaluate the accuracy to assess response with deep learning methods based on endoscopic images in rectal cancer patients after neoadjuvant therapy.
METHODS
Rectal cancer patients diagnosed between January 2012 and December 2015 and treated with neoadjuvant (chemo)radiotherapy were retrospectively selected from a single institute. All patients underwent flexible endoscopy for response evaluation. Diagnostic performance (accuracy, area under the receiver operator characteristics curve (AUC), positive- and negative predictive values, sensitivities and specificities) of different open accessible deep learning networks was calculated. Reference standard was histology after surgery, or long-term outcome (>2 years of follow-up) in a watch-and-wait policy.
RESULTS
226 patients were included for the study (117(52%) were non-CRs; 109(48%) were CRs). The accuracy, AUC, positive- and negative predictive values, sensitivity and specificity of the different models varied from 0.67-0.75%, 0.76-0.83%, 67-74%, 70-78%, 68-79% to 66-75%, respectively. Overall, EfficientNet-B2 was the most successful model with the highest diagnostic performance.
CONCLUSIONS
This pilot study shows that deep learning has a modest accuracy (AUCs 0.76-0.83). This is not accurate enough for clinical decision making, and lower than what is generally reported by experienced endoscopists. Deep learning models can however be further improved and may become useful to assist endoscopists in evaluating the response. More well-designed prospective studies are required.
Identifiants
pubmed: 34642794
doi: 10.1007/s00464-021-08685-7
pii: 10.1007/s00464-021-08685-7
pmc: PMC9001548
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
3592-3600Informations de copyright
© 2021. The Author(s).
Références
Dis Colon Rectum. 2016 Apr;59(4):255-63
pubmed: 26953983
Gastroenterology. 2018 Feb;154(3):568-575
pubmed: 29042219
Gastrointest Endosc. 2011 Mar;73(3):419-22
pubmed: 21353837
IEEE Trans Med Imaging. 2016 May;35(5):1299-1312
pubmed: 26978662
Gut. 2019 Dec;68(12):2161-2169
pubmed: 30858305
Gastroenterology. 2020 Mar;158(4):915-929.e4
pubmed: 31759929
Gut. 2019 Oct;68(10):1813-1819
pubmed: 30814121
Ann Intern Med. 2018 Sep 18;169(6):357-366
pubmed: 30105375
Gastric Cancer. 2018 Jul;21(4):653-660
pubmed: 29335825
Ann Surg Oncol. 2018 Jan;25(1):197-203
pubmed: 29134378
Lancet Gastroenterol Hepatol. 2018 Dec;3(12):825-836
pubmed: 30318451
Radiology. 2013 Oct;269(1):101-12
pubmed: 23801777
J Clin Oncol. 2011 Dec 10;29(35):4633-40
pubmed: 22067400
Gastroenterology. 2020 Jan;158(1):76-94.e2
pubmed: 31593701
IEEE J Biomed Health Inform. 2017 Jan;21(1):41-47
pubmed: 28114040
Sci Rep. 2018 Aug 22;8(1):12611
pubmed: 30135549
Gut. 2019 Jan;68(1):94-100
pubmed: 29066576
Lancet Gastroenterol Hepatol. 2019 Jan;4(1):71-80
pubmed: 30527583
NPJ Digit Med. 2020 Oct 16;3:136
pubmed: 33083571
Endoscopy. 2017 Mar;49(3):270-297
pubmed: 28212588
Lancet. 2018 Jun 23;391(10139):2537-2545
pubmed: 29976470
Gastroenterology. 2018 May;154(7):1876-1886
pubmed: 29462601
Ann Surg Oncol. 2015 Nov;22(12):3873-80
pubmed: 26198074
Ann Surg. 2021 Dec 1;274(6):e541-e547
pubmed: 31851000
IEEE Trans Med Imaging. 2016 May;35(5):1285-98
pubmed: 26886976
J Natl Cancer Inst. 2016 Aug 10;108(12):
pubmed: 27509881
Lancet Oncol. 2019 Dec;20(12):1616-1617
pubmed: 31797775