Objective To understand the effect of World Health Organization(WHO) multimodal hand hygiene improvement strategy on hand hygiene compliance among acupuncturists. Methods All the acupuncturists in departments (Department of Acupuncture, Department of Encephalopathy, Department of Orthopedics and Traumatology) with acupuncture programs in Xi’an Hospital of TCM were chosen in this study between September 2015 and August 2016. Based on the WHO multimodal hand hygiene improvement strategy, comprehensive measures were regulated among acupuncturists. Hand hygiene compliance and accuracy, and hand hygiene knowledge score were compared before and after the strategy intervention. Then, the effects of key strategies were evaluated. Results Overall hand hygiene compliance rate, accuracy and knowledge scores increased from 51.07%, 19.86% and 81.90±2.86 before intervention to 72.34%, 51.70%, and 98.62±2.92 after intervention (P<0.05). Hand hygiene compliance rates also increased in various occasions such as before contacting the patient, after contacting the patient, before acupuncture treatment, and before acupuncture needle manipulation (P<0.05). Conclusion Hand hygiene compliance in acupuncturists can be significantly improved by the implementation of WHO multimodal hand hygiene improvement strategy.
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.
Healthcare-associated infections (HAIs) represent a global public health issue characterized by high incidence rates and severe consequences. They significantly prolong hospital stays, increase risks of excess mortality and disability, exacerbate antimicrobial resistance, and impose a substantial burden on patients’ families and society. Surveillance is the cornerstone of effective HAI prevention and control. Conventional manual surveillance is not only labor-intensive and costly but also lacks standardization. With its powerful data processing and analytical capabilities, artificial intelligence (AI) can significantly reduce the incidence of HAIs, improve patient outcomes, decrease workload, and save costs, offering a new approach for cost-effective and efficient HAI surveillance. This review elaborates on advances in the application of AI in common types of HAIs including sepsis, hospital-acquired pneumonia, urinary tract infections, as well as hand hygiene monitoring, so as to promote the development and implementation of AI in the field of HAI prevention and control.