Artificial Intelligence Assisted Capsule Endoscopy Versus Conventional Endoscopy for Detection of Small Bowel Lesions – Systematic Review and Meta-analysis

Aims The study aims to systematically review literature and provide a meta-analysis on the diagnostic accuracy,specificity, sensitivity, negative and positive predictive values of AI-assisted capsule endoscopy (CE) in the diagnosis of small bowel lesions in comparison to CE.

Methods Literature searches were performed through PubMed, SCOPUS, and Embasse to identify studies eligible for inclusion. All publications up to 11 October 2023 were included. Original articles (including observational studies, and randomized control trials) systematic reviews, meta-analyses, and case series reporting outcomes on AI-assisted CE in the diagnosis of small bowel lesions, were included. The extracted data were pooled, and a meta-analysis was performed for the appropriate variables, considering the clinical and methodological heterogeneity among the included studies. Comprehensive Meta-Analysis v4.0 (Biostat Inc.) was used for the analysis of the data.

Results A total of 15 studies were included in the present study. A pooled accuracy of 0.956 (0.925 – 0.988) for conventional CE and 0.961 (0.940 – 0.981) for AI-assisted CE was observed. Conventional CE reported a lower pooled sensitivity of 0.860 (0.786 – 0.934) compared to AI-assisted CE 0.951 (0.906 – 0.996). Accordingly, AI-assisted CE showed a higher positive predictive value of 0.989 (0.987 – 1.000) while conventional CE reported 0.982 (0.976 – 0.987). Conventional CE however had higher pooled specificity of 0.998 (0.996 – 0.999) versus 0.959 (0.924 – 0.995) in AI-assisted CE. The negative predictive value was however higher in AI-assisted CE at 0.971 (0.945 – 0.997) compared to conventional CE at 0.760 (0.577 – 0.943).

Conclusions AI-assisted CE displays superior diagnostic accuracy, sensitivity, and positive predictive values albeit the lower pooled specificity in comparison with conventional CE. Its use would ensure accurate detection of small bowel lesions and further enhance their management.

Publikationsverlauf

Artikel online veröffentlicht:
15. April 2024