ObjectiveTo summarize the mid-term follow-up results and postoperative aortic remodeling of treating blunt aortic injuries (BAI) with thoracic endovascular aortic repair (TEVAR).MethodsA retrospective study was conducted on BAI patients treated with TEVAR, who were admitted into the Department of Vascular Surgery in Zhongshan Hospital, Affiliated to Fudan University between September 2003 and December 2015. There were 15 males and 9 females at an average age of 45.6±14.0 years. The mechanism of BAI was mainly auto car crash. Totally 25 entry tears were detected and most of them were located at the aortic isthmus.ResultsTwenty-four BAI patients survived and eventually went through TEVAR. One patient died of pulmonary embolism 1 week post-TEVAR. Rate of technical success, clinical success and perioperative mortality was 100.0%, 95.8%, and 4.2%, respectively. Nineteen patients were followed up with a mean time of 35.1(13-87) months. All of them survived this period. Based on the follow-up imaging of CTA, 18 of them revealed no endoleak or stent migration, and 1 patient of transection still had perfusion of distal false lumen at the abdominal aorta. None of the aortic segments measured in this study showed expansion of ≥5 mm during follow-up. The aorta remodeled well in 94.7% of them.ConclusionTEVAR for treating BAI appears feasible with high rates of technical and clinical success rates. The mid-term follow-up results seems satisfying, but the long-term results are yet to be assessed with further follow-up.
Objective To explore an AI-based method for automated hand hygiene monitoring and to compare the effectiveness of three algorithms (Uniformer-V2, 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 Uniformer-V2 model outperformed both, with an accuracy, recall, and F1-score of 93%—an improvement of over 10 percentage points compared to traditional CNNs (e.g., C3D). It also achieved an 84% accuracy in external validation, demonstrating strong generalization. Conclusion The Uniformer-V2 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.
