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.

To investigate the effect of baicalin (BAI)-loaded cross-linked lipoic acid nanocapsules (BAI@cLANCs) against hydrogen peroxide (H2O2)-induced senescence in human umbilical vein endothelial cells (HUVECs), this study examined the toxicity of BAI@cLANCs on HUVECs by MTT method. The cell nuclear staining, SA-β-gal staining, and MTT methods were used to assess the optimal concentration of H2O2-induced senescence in HUVECs. The cellular uptake of BAI@cLANCs was evaluated using fluorescence microscopy imaging and flow cytometry. The proportion of cellular senescence was determined by SA-β-gal staining. The level of reactive oxygen species (ROS) in senescent cells was detected by fluorescence microscopy imaging and multifunctional microplate reader. The content of malondialdehyde (MDA) in cells was detected by lipid oxidation detection kit, and the cell cycle was analyzed by flow cytometry with propidium iodide staining. The results showed that BAI@cLANCs had no significant effect on the growth of HUVECs in the range of BAI at 2.80−112 mmol/L. 200 μmol/L and 25 minutes were the ideal conditions for H2O2-induced senescence of HUVECs. cLANCs as drug delivery carriers significantly enhanced the uptake efficiency of BAI in HUVECs. Compared with the normal group, the H2O2 model group showed decreased cell viability, increased positive SA-β-gal staining rate, increased ROS and MDA content, as well as increased percentage of cells blocked in S phase and decreased cells entering G2/M phase. Compared with the H2O2 model group, BAI, cLANCs, BAI + cLANCs, and BAI@cLANCs groups showed increased cell viability, decreased positive SA-β-gal staining rate, decreased ROS and MDA content, decreased percentage of S-phase cells, and increased cells entering G2/M phase, with the best anti-aging effect in the BAI@cLANCs group. In summary, the results above showed that both BAI and cLANCs have anti-aging properties. With cLANCs as drug carriers, the anti-aging benefits of BAI@cLANCs are synergistic and can effectively delay H2O2-induced senescence of HUVECs.