Artificial Intelligence in Endoscopy for Predicting Helicobacter pylori Infection: A Systematic Review and Meta‐Analysis DOI

Yiwen Jiang,

Hankun Yan,

J. J. Cui

и другие.

Helicobacter, Год журнала: 2025, Номер 30(2)

Опубликована: Март 1, 2025

This meta-analysis aimed to assess the diagnostic performance of artificial intelligence (AI) based on endoscopy for detecting Helicobacter pylori (H. pylori) infection. A comprehensive literature search was conducted across PubMed, Embase, and Web Science identify relevant studies published up January 10, 2025. The selected focused accuracy AI in H. pylori. bivariate random-effects model employed calculate pooled sensitivity specificity, both presented with 95% confidence intervals (CIs). Study heterogeneity assessed using I2 statistic. Of 604 identified, 16 (25,002 images or patients) were included. For internal validation set, sensitivity, area under curve (AUC) 0.91 (95% CI: 0.84-0.95), 0.86-0.94), 0.96 0.94-0.97), respectively. external AUC 0.86-0.95), 0.94 0.90-0.97), 0.98 0.96-0.99). junior clinicians, 0.76 0.66-0.83), 0.75 0.70-0.80), 0.81 0.77-0.84). senior 0.74-0.86), 0.89 0.86-0.91), 0.92 0.90-0.94). Endoscopy-based demonstrates higher compared endoscopists. However, high among limits strength these findings, further research datasets is necessary confirm results.

Язык: Английский

Artificial Intelligence in Endoscopy for Predicting Helicobacter pylori Infection: A Systematic Review and Meta‐Analysis DOI

Yiwen Jiang,

Hankun Yan,

J. J. Cui

и другие.

Helicobacter, Год журнала: 2025, Номер 30(2)

Опубликована: Март 1, 2025

This meta-analysis aimed to assess the diagnostic performance of artificial intelligence (AI) based on endoscopy for detecting Helicobacter pylori (H. pylori) infection. A comprehensive literature search was conducted across PubMed, Embase, and Web Science identify relevant studies published up January 10, 2025. The selected focused accuracy AI in H. pylori. bivariate random-effects model employed calculate pooled sensitivity specificity, both presented with 95% confidence intervals (CIs). Study heterogeneity assessed using I2 statistic. Of 604 identified, 16 (25,002 images or patients) were included. For internal validation set, sensitivity, area under curve (AUC) 0.91 (95% CI: 0.84-0.95), 0.86-0.94), 0.96 0.94-0.97), respectively. external AUC 0.86-0.95), 0.94 0.90-0.97), 0.98 0.96-0.99). junior clinicians, 0.76 0.66-0.83), 0.75 0.70-0.80), 0.81 0.77-0.84). senior 0.74-0.86), 0.89 0.86-0.91), 0.92 0.90-0.94). Endoscopy-based demonstrates higher compared endoscopists. However, high among limits strength these findings, further research datasets is necessary confirm results.

Язык: Английский

Процитировано

0