Predicting surgical operative time in primary total knee arthroplasty utilizing machine learning models
Ingwon Yeo1, Christian Klemt1, Christopher M Melnic1
1Bioengineering Laboratory, Department of Orthopaedic Surgery, Massachusetts General Hospital, Harvard Medical School, 55 Fruit Street, Boston, MA, 02114, USA.
Archives of Orthopaedic and Trauma Surgery
|August 22, 2022
Summary
Machine learning accurately predicts total knee arthroplasty (TKA) operative time. Younger age, high BMI, and non-usage of tranexamic acid predict longer surgical times, improving operating room efficiency.
Area of Science:
- Orthopedic Surgery
- Medical Informatics
- Machine Learning
Background:
- Prolonged surgical operative time in total knee arthroplasty (TKA) is linked to adverse postoperative outcomes.
- Accurate prediction of surgical duration is crucial for enhancing operating room efficiency.
- Machine learning offers advanced predictive analytics for improving surgical time estimations.
Purpose of the Study:
- To develop and validate a machine learning model for predicting surgical operative time in primary total knee arthroplasty (TKA).
Main Methods:
- Retrospective analysis of 10,021 primary TKA cases from electronic medical records.
- Development and assessment of three machine learning algorithms: Artificial Neural Networks (ANNs), Random Forest (RF), and K-Nearest Neighbor (KNN).
- Evaluation of models using discrimination, calibration, and decision curve analysis.
Main Results:
- The Artificial Neural Network (ANN) model demonstrated superior performance in predicting TKA surgical operative time (AUC = 0.82).
- Key predictors for prolonged operative time included younger age (<45 years), non-usage of tranexamic acid, and high BMI (>40 kg/m²).
Conclusions:
- Machine learning models show excellent predictive capability for surgical operative time in primary TKA.
- Accurate surgical duration estimation enhances operating room efficiency and identifies patients at risk for prolonged procedures.
