Personalized Medicine Using Predictive Analytics: A Machine Learning-Based Prognostic Model for Patients Undergoing Hip Arthroscopy

Personalized Medicine Using Predictive Analytics: A Machine Learning-Based Prognostic Model for Patients Undergoing Hip Arthroscopy

Authors

Domb BG, Ouyang VW, Go CC, Gornbein JA, Shapira J, Meghpara MB, Maldonado DR, Lall AC, Rosinsky PJ

Journal

Am J Sports Med, June 2022

Background

Predictive models to guide patient outcomes for hip arthroscopy are lacking, especially for preoperative decision-making. This study aims to develop a machine learning-based model to predict patient outcomes.

Methods

• Data from 2,415 patients undergoing hip arthroscopy for FAIS were analyzed.
• Machine learning techniques (Cox proportional hazards and Fine-Gray models) were used to predict outcomes, including survival and need for repeat surgery.
• A web-based calculator was created based on the models for use in shared decision-making.

Key Findings

• Prognostic models identified preoperative factors predictive of long-term survivorship and repeat surgeries.
• The Harrell C-statistics for the models were 0.848 for survivorship and 0.662 for repeat surgeries.
• A web-based calculator was created to assist clinicians in predicting patient outcomes.

Conclusions

This study successfully developed machine learning-based prognostic models for predicting patient outcomes after hip arthroscopy. These models may aid in personalized treatment planning.

What Does This Mean For Patients

This predictive tool can help you and your doctor make informed decisions about your hip surgery, potentially leading to better-tailored care and improved long-term results.