Classification and prediction of Orthopedic disease based on lumber and pelvic state of patients

The application of machine learning in the medical field is the trending approach for early prediction and remedy of diseases. The growing number of orthopedic disease has lead people to suffer in an early age and it can be prevented through an early diagnosis. Thus this study pursues to help the medical specialists to predict orthopedic disease early and classify them accordingly. For this purpose, three machine learning algorithms such as Logistic Regression, Random Forest Classifier and k-Nearest Neighbor have been applied to a dataset of 310 patients containing six biomechanical features that describe the state of the patients' pelvic and lumber. The algorithm results are then compared and the one that provided the best result with an accuracy of 89 percent is found to be Random Forest, with other algorithms' accuracy in between 80 to 90 percent.