AOSSM 2023 Annual Meeting Recordings no CME-A Novel Machine Learning (ML) Algorithm to Predict Outcomes after Revision ACLR (rACLR) in the Multicenter Anterior Cruciate Ligament Reconstruction Study (MARS) Cohort

A Novel Machine Learning (ML) Algorithm to Predict Outcomes after Revision ACLR (rACLR) in the Multicenter Anterior Cruciate Ligament Reconstruction Study (MARS) Cohort

Back to course

Video Transcription

Video Summary

In this video, Kendra Vasavada, a resident surgeon in general surgery at Yale New Haven Hospital, discusses a novel machine learning supermodel for predicting the failure of revision ACLR reconstruction. The study uses the MARS cohort data and applies a machine learning algorithm called autoprognosis to create a predictive model. The study finds that autoprognosis has the most robust predictive ability and identifies important features such as prior ACLR tunnel placement, graft type, and size that contribute to the predictive ability. This research has the potential to improve preoperative counseling, decision-making, and cost of care for ACLR patients. Further validation studies and development of a risk calculator are needed for implementation. The video concludes with acknowledgments to the researchers and sponsors involved in the study.

Asset Caption

Kinjal Vasavada, MD

Keywords

machine learning

predictive model

ACLR reconstruction

predictive ability

risk calculator