This article broadly examines how machine learning (ML) and artificial intelligence (AI),are applied with medicine, with a focus on rehabilitation and physical medicine & rehabilitation (PM&R). The research involves a revision of the current AI technologies, including machine learning methods such as supervised, unsupervised, reinforcement, active, semi supervised learning, and deep learning, and their applications across various medical areas. The data and findings indicate that AI has achieved high accuracy in detecting age-related macular degeneration, outperformed cardiovascular risk algorithms, and performed equal or better than dermatologists in skin cancer detection. In addition, AI has been applied in neuroprosthetics, brain-computer interfaces, perioperative medicine, musculoskeletal imaging, and rehabilitative exercise assessment. The analysis emphasizes that while AI offers powerful tools for handling large and complex datasets, it cannot fully resolve causal inference challenges and requires careful application to avoid overfitting or false correlations.