Artificial Intelligence in Inflammatory Bowel Disease Endoscopy: Advanced Development and New Horizons

Table 2

Endoscopic AI applications in diagnosing CD.

Author	Year	Data source	Purpose	Results
Ding et al.	2019	4,206 abnormalities in 3,280 patients	To analyze SB-CE images with greater accuracy and sensitivity than conventional methods	99.88% sensitivity in the per-patient analysis (95% confidence interval [CI], 99.67–99.96)
Aoki et al.	2019	113,426,569 images from 6,970 patients	To develop a CNN-based algorithm that could automatically detect erosions and ulcers from capsule endoscopic images	With 99.88% sensitivity in the per-patient analysis (95% CI, 99.67–99.96) and 99.90% sensitivity in the per-lesion analysis (95% CI, 99.74–99.97)
Klang et al.	2019	17,640 CE images from 49 patients	To evaluate a deep learning algorithm for the automated detection of small-bowel ulcers in Crohn’s disease (CD) on capsule endoscopy (CE) images	AUCs (area under the curve of 0.99 and accuracies ranging from 95.4–96.7%
Klang et al.	2021	27,892 CE images	To prove the ability of deep neural networks to identify intestinal strictures on CE images of Crohn’s disease (CD) patients	A differentiation between strictures and normal mucosa (area under the curve [AUC], 0.989)
Barash et al.	2021	17,640 CE images from 49 patients	To develop a deep learning algorithm for automated grading of CD ulcers on CE	The accuracy of the algorithm was 0.91 (95% CI) for distinction of 3 different grades