2015 ISAKOS Biennial Congress Paper #0

Machine Learning Can Predict Anterior Elevation After Reverse Total Shoulder Arthroplasty

Edoardo Franceschetti, MD, Rome ITALY
Pietro Gregori, MD, Roma, Lazio ITALY
Simone De Giorgi, Ing, rome ITALY
Tommaso Martire, Ing, roma ITALY
Giancarlo Giurazza, MD, rome ITALY
Biagio Zampogna, MD, Rome, Italy ITALY
Rocco Papalia, MD, PhD, Prof., Rome ITALY

Campus Bio Medico University, Rome, Italy, ITALY

FDA Status Not Applicable

Summary: Our machine learning study demonstrates that Machine learning could provide high predictive algorithms for anterior elevation after reverse shoulder arthroplasty. The differential analysis between the utilized techniques showed higher accuracy in score prediction for the Support Vector Regression

Rate:

Abstract:

Background

One of the most frequent concerns of the increasing number of patients undergoing shoulder reverse arthroplasty is the possibility to regain an acceptable and not painful range of motion after surgery. The aim of the present study was to individuate and compare specific Machine Learning algorithms that could predict post operative anterior elevation score after Reverse shoulder arthroplasty surgery at different time points.

Methods

Data from 105 patients who underwent reverse shoulder arthroplasty at the same institute have been collected with the purpose of generating algorithms which could predict the target. 28 features were selected and applied to two different Machine Learning techniques: Linear Regression and Support Vector Regression (SVR). These two techniques were also compared in order to define to most faithfully predictive.

Results

Using the extracted features, the SVR algorithm resulted in a mean absolute error (MAE) of 11.6° and a classification accuracy (PCC) of 0,88 on the test-set. Linear Regression, instead, resulted in a MAE of 13,0° and a PCC of 0,85 on the test-set.

Conclusions

Our machine learning study demonstrates that Machine learning could provide high predictive algorithms for anterior elevation after reverse shoulder arthroplasty. The differential analysis between the utilized techniques showed higher accuracy in score prediction for the Support Vector Regression.