Large-scale observational study of AI-based patient and surgical material verification system in ophthalmology: real-world evaluation in 37 529 cases.
Tabuchi H, Ishitobi N, Deguchi H, et al. Large-scale observational study of AI-based patient and surgical material verification system in ophthalmology: real-world evaluation in 37 529 cases. BMJ Qual Saf. 2024;Epub Nov 29. doi:10.1136/bmjqs-2024-018018.
Wrong-side and wrong-patient surgical errors can have devastating consequences. In this study, an artificial intelligence (AI) error-detection program was implemented for all ophthalmic surgeries. The AI system was integrated with the WHO surgical safety checklist for patient identification, surgical laterality verification, and intraocular lens authentication. AI detected five errors (compared with one error pre-implementation), four of which occurred when the AI system was not fully implemented or properly used. In one case, the surgeon ignored both verbal warnings from a nurse and AI alerts. These results suggest that AI can be effective at identifying errors and preventing near misses but must be used in conjunction with existing safety practices that are fully adhered to.