Endoscopy 2022; 54(05): 521
DOI: 10.1055/a-1736-8097
E-Videos

Commentary

Cesare Hassan
1   Department of Biomedical Sciences, Humanitas University, Rozzano, Milan, Italy
2   Endoscopy Unit, Humanitas Clinical and Research Center IRCCS, Rozzano, Milan, Italy
,
Yuichi Mori
3   Clinical Effectiveness Research Group, University of Oslo, Oslo, Norway
4   Digestive Disease Center, Showa University, Northern Yokohama Hospital, Yokohama, Japan
,
Alessandro Repici
1   Department of Biomedical Sciences, Humanitas University, Rozzano, Milan, Italy
2   Endoscopy Unit, Humanitas Clinical and Research Center IRCCS, Rozzano, Milan, Italy
› Author Affiliations

Some endoscopists might assume that computer-aided detection (CADe) used in colonoscopy can identify any type of polyp, regardless of its shape, size, and histology. Unfortunately, the differences amongst polyps are greater and much more various than, say, differences amongst cars, and the accuracy of CADe is extremely dependent on the prevalence of each polyp type in the training dataset. The training of CADe software is mainly based on subcentimetric adenomas, while sessile serrated lesions, nongranular laterally spreading lesions, or even advanced cancers are insufficiently represented in the dataset. Thus, endoscopists need to be aware of this potential limitation of CADe as well as of its benefits for increasing the adenoma detection rate [1] [2]. It is also noteworthy that most manufacturers are eager to improve the software by including more varieties of images to overcome this significant drawback of CADe in clinical practice.



Publication History

Article published online:
21 April 2022

© 2022. Thieme. All rights reserved.

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany

 
  • References

  • 1 Hassan C, Spadaccini M, Iannone A. et al. Performance of artificial intelligence in colonoscopy for adenoma and polyp detection: a systematic review and meta-analysis. Gastrointest Endosc 2021; 93: 77-85.e6
  • 2 Barua I, Vinsard DG, Jodal HC. et al. Artificial intelligence for polyp detection during colonoscopy: a systematic review and meta-analysis. Endoscopy 2021; 53: 277-284