Large language models enhance prehospital trauma triage, potentially improving patient outcomes. LLMs achieved 83.5% triage accuracy, outperforming human clinicians’ 78.9% accuracy (p<0.01). Under-triage with LLMs improved slightly to 4.8%, compared to 5.1% with human triage (p=0.73). Essential transcripts reduced communication length by 80.8% while maintaining accuracy. Integrating LLMs in trauma assessment can lead to better […]
Category: Digital Surgery and Telemedicine
AI Platform Boosts Safety in Laparoscopic Cholecystectomy
An AI-based tool shows promise in improving the critical view of safety during laparoscopic cholecystectomy, reducing bile duct injury risks. The AI platform achieved impressive accuracy scores: 0.91 for CVS I, 0.86 for CVS II, and 0.73 for CVS III. Significant improvement in CVS assessment was observed post-deployment (p < 0.01). Surgeons reported high satisfaction, […]
Video Analysis Reveals Key Non-Technical Skills in Surgery
Silent video can predict surgeons’ non-technical skills, impacting training and outcomes. Analyzed 40 laparoscopic appendectomy videos with over 10,000 annotated actions. Mean cognitive rating was 5.6 out of 8, with a correlation between decision-making and situation awareness (r=0.8). This method paves the way for scalable assessments of surgical performance. Five key predictive factors identified include […]
New Machine Learning Model Enhances Open Surgery Skill Assessment
Surgeons can now assess open surgical skills more accurately with a novel machine learning model. Achieved an 80.1% mean F1 score, surpassing previous assessment methods. Novice classification accuracy reached 90.1%, while proficient levels scored 86.3%. Consider using this model for objective skill evaluation in training to ensure higher surgical quality and patient safety. The model […]
AI in Surgery: Bridging Expectation and Reality
Surgeons face significant gaps between their expectations of AI interventions and the actual outcomes in the operating room, highlighting challenges in implementation. 57% of surgeons were neutral on the AI’s usefulness; only 37% had a positive outlook. Key concerns included the need for extensive training, difficulties accessing data, and limited predictive capabilities for complications. Minimizing […]
Automated Trauma Video Review Enhances Patient Assessment
A computer vision model could revolutionize trauma video review by automating the identification of key phases and procedures, improving trauma care quality. Achieved 98.3% frame-wise accuracy and high edit and F1 scores (up to 94.5%) for trauma resuscitation phases. Procedure detection accuracy surpassed 66% for x-rays and central line placements. This technology can enhance surgical […]
Predicting Recurrent Bile Duct Stones Post-Exploration
Machine learning effectively predicts recurrent extrahepatic bile duct stones after common bile duct exploration, enhancing surgical decision-making. Random forest model achieved AUCs of 97.99% in training and 83.1% externally, outperforming other methods. Key risk factors include maximum stone diameter, common bile duct diameter, and direct bilirubin, with larger stones (>15 mm) significantly increasing recurrence risk. […]
Augmented Reality Enhances Liver Resection Outcomes
Augmented reality in liver surgery reduces intraoperative bleeding and complications. Blood loss decreased by 75.9 ml (p < 0.001) with AR guidance. Transfusion rates dropped to nearly half (RR: 0.47; p = 0.01); fewer complications (RR: 0.64; p = 0.009). AR-assisted techniques could improve surgical precision and patient outcomes, particularly in tumor cases. Tumor recurrence […]
New AI Model Predicts Cardiac Risks in Surgery
A new deep learning model predicts postoperative cardiac events after noncardiac surgery with high accuracy, improving patient safety. The model showed an AUROC of 0.902 for 30-day major adverse cardiac and cerebrovascular events (MACCE), outperforming traditional methods. Only 0.6% of 165,577 patients experienced MACCE, indicating the model’s focus on high-risk patients may enhance targeting for […]
Machine learning predicts complications in acute cholecystitis
Machine learning tools can now help surgeons assess the risk of postoperative complications in acute calculous cholecystitis patients. Cholesurgrisk I achieved an AUC-ROC of 0.8456, while Cholesurgrisk II, which includes intraoperative data, improved this to 0.8903. A web-based version of Cholesurgrisk I offers real-time, patient-specific risk estimates. Integrating these models into practice could enhance preoperative […]