Machine learning models predict ambulatory surgery end time, PACU discharge

Editor's Note

In this study from the University of California, San Diego, La Jolla, researchers developed a machine learning model that improved the ability to predict surgery end times and PACU discharge at a range of start times in an ambulatory surgery center (ASC).

Of 13,447 surgical procedures analyzed, the median total time for case duration and PACU stay was 165 minutes. Various machine learning models were evaluated with and without the Synthetic Minority Oversampling Technique (SMOTE) algorithm and balanced bagging techniques.

The most important predictors of case duration and PACU stay were scheduled room times, scheduled incision times, surgical specialty, and patient weight.

Enhanced modeling and prediction methods may be adapted by OR management to allow for a better determination of whether an add-on case can be booked in an ASC, the researchers conclude. Such methods will improve patient care, staff scheduling, and facility profits.