Machine Learning Model for Evaluative Performance of Medical Images Using Classifiers

Abstract

Computer Aided Diagnosis is becoming popular in medical sciences as it provides accuracy and timeliness, the two major aims of medical field. In the work presented here, an algorithm is developed which aims to design an auto-CAD system for the diagnosis of retina abnormalities. Diabetic Retinopathy becomes severe if not diagnosed and treated at the first stage. Age-related Macular Degeneration is another vision threatening disease that occurs in the elderly population and needs serious medical attention. In this research work, these two diseases are considered and the signs of these two diseases are analyzed. A combined database is formed by collecting the images from several standard datasets. The algorithm presented in this work is developed with the combination of two steps, namely, image processing and machine learning. Several image processing algorithms for segmentation and morphological operations are used for the detection of the abnormalities caused by the above mentioned diseases. A set of significant features are selected and evaluated on the abnormalities extracted in the image processing stage. The classification of the abnormalities with a training and a test set is performed using different machine learning algorithms. The random forest classifier is best suited to the dataset used in this research for its performance accuracy and robustness with respect to noise. With the aim of forming a Case Based Reasoning model, we have developed a method of machine learning based classification of different abnormalities