Building a durable “rural-to-center” endoscopy artificial intelligence system: practical enhancements for scale

10.1055/a-2721-6552

Chiang et al. present an impressive “rural-to-center” artificial intelligence (AI) service for routine endoscopy with valuable real-world application [1]. We congratulate the authors on this achievement and would like to pose several questions to strengthen future implementations.

1. Clinical transparency and usability: Could the authors provide guidance on how clinicians should interpret the AI outputs in practice? Specifically, would they consider developing a concise model card specifying intended use, input constraints, operating thresholds with confidence intervals, and governance contacts [2] [3]? Additionally, how might uncertainty be communicated visually at the point of care – for example, through confidence bands or calibration displays that adapt to local disease prevalence?

2. Scalability and safety monitoring: What infrastructure do the authors envision for expanding this model to multiple sites while maintaining performance? Would federated learning approaches or site-specific validation protocols be considered to address domain shift? Furthermore, how might post-deployment surveillance be implemented to track real-world complications, alert acknowledgment rates, and monitor performance drift across different clinical settings [4]?

3. Workflow integration: How do the authors propose integrating AI outputs into existing electronic health record systems? Would structured data formats such as Fast Healthcare Interoperability Resources be used to transmit discrete flags for Helicobacter pylori suspicion and premalignant findings? What training or decision support tools might help endoscopists appropriately respond to AI alerts in routine practice [5]?

These questions aim to build upon the authors’ excellent work and support transparent, safe, and durable AI implementation at scale.

Publication History

Article published online:
20 February 2026

• References

• 1 Chiang T-H, Hsu Y-N, Chen M-H. et al. A rural-to-center artificial intelligence model for diagnosing Helicobacter pylori infection and premalignant gastric conditions using endoscopy images captured in routine practice. Endoscopy 2025;
• 2 Yu Z. Artificial intelligence-augmented surveillance for endoscopic safety: a forward step. Gastrointest Endosc 2025; 102: 918
• 3 Liu X, Cruz Rivera S, Moher D. SPIRIT-AI and CONSORT-AI Working Group. et al. Reporting guidelines for clinical trial reports for interventions involving artificial intelligence: the CONSORT-AI extension. Nat Med 2020; 26: 1364-1374
• 4 Vasey B, Nagendran M, Campbell B. DECIDE-AI Expert Group. et al. Reporting guideline for the early-stage clinical evaluation of decision support systems driven by artificial intelligence: DECIDE-AI. Nat Med 2022; 28: 924-933
• 5 Bretthauer M, Ahmed J, Antonelli G. et al. Use of computer-assisted detection (CADe) colonoscopy in colorectal cancer screening and surveillance: European Society of Gastrointestinal Endoscopy (ESGE) Position Statement. Endoscopy 2025; 57: 667-673