Abstract- The increasing adoption of electronic health records (EHRs) and digital medical imaging systems has created unprecedented opportunities to apply machine learning (ML) in healthcare. This paper presents a comprehensive review of the features of electronic healthcare data—spanning structured tabular data, unstructured clinical notes, and imaging modalities—and the ML tech-niques used to extract clinical value from them. Work has list various learning approaches, in-cluding supervised, unsupervised, and ensemble methods. As most of medical data are in images hence image features like co-occurrence matrices, wavelet transformations, edge detection, etc. were brief. This review aims to bridge the gap between technical advances in machine learning and their practical implications for modern healthcare delivery.
