Technical and clinical feasibility of single-use gastroscopy with real-time AI-based quality monitoring and single-use colonoscopy: a prospective two-center study

BACKGROUND: Single-use gastrointestinal endoscopes eliminate the need for post-procedure reprocessing and have become an area of interest in endoscopic device development and quality management. However, clinical data on their use in routine gastrointestinal endoscopy remain limited. This prospective, two-center, single-arm study evaluated the feasibility and short-term safety of single-use gastroscopy performed with ENDOANGEL, an artificial intelligence (AI)-based real-time quality monitoring system, and single-use colonoscopy in routine gastrointestinal endoscopy. Formal AI-based quality scoring was applied to gastroscopy, whereas ENDOANGEL-related colonoscopy indicators were recorded descriptively. RESULTS: A total of 120 participants were enrolled, including 60 who underwent gastroscopy and 60 who underwent colonoscopy. Imaging of predefined anatomical sites was successful in all participants in both groups (60/60 in each group; 95% CI 94.0-100.0%). All procedures were completed without conversion to reusable endoscopes. The overall image quality scores were 39.2 ± 0.8/40 for gastroscopy and 31.3 ± 0.8/32 for colonoscopy. Mean total procedure times were 9.3 ± 2.7 min and 12.6 ± 3.1 min, respectively. No device malfunction, device defect, or procedure-related adverse event was observed during the procedure or within 24 h after endoscopy. In gastroscopy, all ENDOANGEL-assisted examinations achieved the maximum procedural score, and ENDOANGEL-based assessment agreed with manual assessment of imaging success and procedure completion. In colonoscopy, ENDOANGEL-related cecal intubation reminders agreed with manual assessment in all procedures, and withdrawal-speed alerts were recorded in 4 of 60 procedures. CONCLUSION: Single-use gastroscopy with real-time AI-based quality monitoring and single-use colonoscopy were feasible in routine gastrointestinal endoscopy and showed a favorable short-term safety profile within 24 h. AI-related findings should be interpreted as feasibility and agreement data for gastroscopy and as descriptive process-monitoring observations for colonoscopy. Further controlled studies are needed to compare single-use and reusable systems and to clarify the clinical role of AI-based process monitoring in single-use endoscopy. TRIAL REGISTRATION: Chinese Clinical Trial Registry, ChiCTR2100045101. Retrospectively registered on April 6, 2021.