Machine Learning Does Not Improve Humeral Torsion Prediction Compared to Regression in Baseball Pitchers.
Garrett S Bullock, Charles A Thigpen, Gary S Collins and 4 others
PMID 35391864WHAT IT FOUND
Advanced computer algorithms failed to predict shoulder bone twist in baseball pitchers more accurately than a basic clinical formula.
For patient care, stick to the standard equation. It is simpler, easier to explain, and just as reliable for checking injury risk.
Key findings
01Machine learning models did not improve the prediction of humeral torsion compared to a traditional statistical model.
02The error rate for all prediction models was larger than the clinical threshold considered important for assessing arm injury risk.
03A simple statistical equation is recommended for clinical use because it is easier to interpret and does not require specialized software.
STILL TO COME
How it was doneWhat they foundWhat it means for PTs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
The models were only tested on professional pitchers within a single organization, so the results may not apply to amateurs, youth players, or athletes in other sports. The study did not test the models on new data from different groups, which limits how well they might work in a different clinic. The prediction error for all models was larger than the specific measurement difference that signals high injury risk, reducing the clinical utility of the models.
Declared interests
The authors declared no conflicts of interest and stated that the research did not receive specific funding from public, commercial, or not-for-profit sectors.
The easy way to misread this
Do not assume that a machine learning model is automatically better than a simple formula. In this study, the advanced algorithms were actually less reliable at matching real-world measurements than the basic equation, and neither was accurate enough to rule out injury risk on its own.