Artificial intelligence-based quality control of hand hygiene for hospital-acquired infection
- VernacularTitle:基于人工智能的院感手卫生质量控制研究
- Author:
Xuchen YANG
1
;
Jingwen LI
2
;
Wan ZHANG
1
;
Shasha FENG
1
;
Min ZENG
1
;
Jianan SHI
1
;
Youqiong CHEN
1
;
Tao ZHENG
1
;
Xun YAO
1
,
3
Author Information
1. Information Center, West China Hospital, Sichuan University, Chengdu, 610041, P. R. China
2. Medical Artificial Intelligence Lab, West China Hospital, Sichuan University, Chengdu, 610041, P. R. China
3. Medical Artificial Intelligence Research Laboratory, West China Hospital, Sichuan University, 610041, P. R. China
- Publication Type:Journal Article
- Keywords:
Healthcare-associated infection control;
artificial intelligence;
hand hygiene;
quality control;
video recognition;
behavior monitoring;
operating room;
UniFormerV2
- From:
Chinese Journal of Clinical Thoracic and Cardiovascular Surgery
2026;33(02):241-247
- CountryChina
- Language:Chinese
-
Abstract:
Objective To explore an artificial intelligence (AI)-based method for automated hand hygiene monitoring and to compare the effectiveness of three algorithms (UniFormerV2, TDN, C3D) in recognizing hand hygiene steps in surgical settings, thereby aiding hospital infection control. Methods From April to October 2024, we non-invasively collected 641 video recordings of healthcare staff performing hand hygiene at four-bay scrub sinks in two tertiary hospitals using overhead HD cameras. The dataset was annotated by five trained experts for model training and validation. Results Following training on 385 samples, internal validation (n=119) showed the C3D model achieved 81% accuracy, 87% recall, and an 83% F1-score. The TDN model achieved 93%, 91%, and 92% for the same metrics. The UniFormerV2 model outperformed both, with an accuracy, recall, and F1-score of 93%—an improvement of over 10 percentage points compared to traditional CNNs (TDN, C3D). It also achieved an 84% accuracy in external validation, demonstrating strong generalization. Conclusion The UniFormerV2 model is more accurate than CNN-based models for hand hygiene step recognition and shows robust performance in external validation. It presents a viable tool for healthcare facilities to enhance hand hygiene management, ultimately improving medical quality and patient safety.
