AI-Driven Optimization of Surgical Workflow: Precision, Risk, and OR Resource Management

Abstract

An integrated artificial intelligence (AI) framework for optimizing Operating Room (OR) efficiency, estimate surgical time, predict clinical risks and forecast postoperative recovery is presented in this study. Data from MIMIC-III and Fortis Hospital Mohali, Max Super Speciality Hospital, Ivy Hospital in the Mohali Chandigarh region are also supplemented as public datasets. This work presents a methodology that developed machine learning models and Cox Proportional Hazards model in addition to linear programming-based optimization algorithms. It is observed that the Mean Absolute Error (MAE) of the Random Forest Regression model for surgical time prediction is 9.3 minutes and an R2. The study obtained an accuracy of 91% on the surgical risk classification model with precision and recall scores of 0.92 and 0.91. These results show that AI can greatly assist in the planning for an operation, reduce delays, and assist with clinical decision making in any surgical environment. Not only does the proposed system raise accuracy and efficiency at all stages of workflow, but it also delivers practical, real time solutions which can be scaled to fit the hospital industry of healthcare while supporting safer, smarter surgical care.

Ähnliche Arbeiten