An Ai Based Classifier Model For Lateral Pillar Classification Of Legg-calve-perthes

This study was conducted to compare the classification success rates of six orthopaedists and a convolutional neural network (CNN) for the classification of Legg-Calve-Perthes Disease (LCPD) hip radiographs. 81 hip radiographs were retrieved from two institutions between 2008 and 2016 and were used for the study. After reliability tests, five more independent doctors, including two radiologists, were added for a percentage agreement test. The results of the study showed that the CNN had a higher classification success rate than the orthopaedists.