Opmed & Mayo Clinic Unveil Multimodal AI Platform to Improve Surgical Scheduling Efficiency
Advertisement
The findings were presented at the American College of Cardiology Expo (ACC.26), highlighting that the AI system reduced CV procedure scheduling errors by approximately 50% compared to traditional planning methods.
Healthcare operations AI company Opmed, in collaboration with Mayo Clinic, has introduced results from a multi-year study demonstrating how a multimodal deep learning platform can significantly improve operating room (OR) scheduling efficiency for cardiovascular procedures.
The findings were presented at the American College of Cardiology Expo (ACC.26), highlighting that the AI system reduced CV procedure scheduling errors by approximately 50% compared to traditional planning methods.
According to the study, conventional scheduling in hospitals continues to depend on historical averages, manual spreadsheets, and clinician judgment, leading to a Mean Absolute Error (MAE) of 1.13 hours per surgical case. The resulting inefficiencies contribute to both delayed procedures and underutilized OR capacity, with idle time costing hospitals up to $1,000 per hour.
The Opmed platform integrates structured clinical datasets with unstructured physician notes to generate more precise surgical time predictions. In evaluation, the system reduced MAE to 0.564 hours, effectively cutting prediction error by half and outperforming human baseline scheduling accuracy.
The study was conducted using a holdout dataset of 643 cardiovascular procedures performed between November 2025 and January 2026. Multiple AI configurations were tested, with the best-performing model combining structured clinical data and unstructured preoperative notes. This configuration delivered an RMSE of 0.799 hours and an R² score of 0.721, compared to 0.31 under human scheduling estimates.
Researchers noted that cardiovascular procedures are particularly difficult to schedule due to variability in patient conditions, intraoperative complications, and procedural complexity. The AI model accounts for multiple factors including patient phenotypes, ASA classification, procedural type, anesthesia timing, turnover rates, and historical surgeon-specific patterns.
The platform’s improved accuracy also carries operational implications for hospital systems. By aligning predicted and actual procedure times more closely, the system enables better utilization of surgical blocks, potentially adding two to three additional complex cases per month per operating room and recovering more than 200 hours of OR time annually per room.
Opmed CEO and Co-founder Dr. Mor Brokman Meltzer stated that the study highlights the role of AI in improving healthcare operations, particularly in scheduling optimization, and emphasized the significance of the long-term collaboration with Mayo Clinic in validating the technology.
Stay tuned for more such updates on Digital Health News
