MedElevator
Войти в исследования BMJ open quality · 2026
Visual alert integration with a time out monitor improves utilisation of an AI safety authentication system in ophthalmic surgery: a 1-year before-after study.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Tabuchi Hitoshi, Ishitobi Naofumi, Tanabe Mao, Nakaniida Yuta, Akada Masahiro
Аннотация
BACKGROUND: AI-enabled verification can prevent wrong-patient and wrong-site events, but benefit depends on consistent use. We previously reported the real-world performance of an AI safety authentication system in ophthalmology, observing that most errors occurred when the system was not used.
OBJECTIVE: To test whether integrating a conspicuous visual alert into the operating-room time out monitor increases routine utilisation of the existing AI system.
METHODS: A single-centre before–after study spanning 1 year preimplementation (August 2023–July 2024) and 1 year postimplementation (September 2024–August 2025); the deployment month was excluded. The software-only intervention added a dynamic border to the time out monitor: flashing red until all required authentications were complete, then solid green. Primary outcome was non-utilisation (encounters with ≥1 required authentication not performed).
RESULTS: Across 18 669 preimplementation and 18 910 postimplementation encounters, non-utilisation fell from 0.52% (98/18 669) to 0.17% (33/18 910)—a 66.8% relative reduction (Fisher’s exact p<0.001). Facial recognition and laterality verification decreased (both p<0.001), while intraocular lens verification decreased without statistical significance (p=0.14).
CONCLUSIONS: A small, human-factors-informed User Interface (UI) change substantially improved day-to-day utilisation of an established AI safety system without added devices or steps. Because the approach modifies a common time out display, it is language-independent and low-cost, and may be transferable to other verification tasks (eg, implant or blood product checks).
OBJECTIVE: To test whether integrating a conspicuous visual alert into the operating-room time out monitor increases routine utilisation of the existing AI system.
METHODS: A single-centre before–after study spanning 1 year preimplementation (August 2023–July 2024) and 1 year postimplementation (September 2024–August 2025); the deployment month was excluded. The software-only intervention added a dynamic border to the time out monitor: flashing red until all required authentications were complete, then solid green. Primary outcome was non-utilisation (encounters with ≥1 required authentication not performed).
RESULTS: Across 18 669 preimplementation and 18 910 postimplementation encounters, non-utilisation fell from 0.52% (98/18 669) to 0.17% (33/18 910)—a 66.8% relative reduction (Fisher’s exact p<0.001). Facial recognition and laterality verification decreased (both p<0.001), while intraocular lens verification decreased without statistical significance (p=0.14).
CONCLUSIONS: A small, human-factors-informed User Interface (UI) change substantially improved day-to-day utilisation of an established AI safety system without added devices or steps. Because the approach modifies a common time out display, it is language-independent and low-cost, and may be transferable to other verification tasks (eg, implant or blood product checks).