Machine Learning and Deep Learning Algorithms in the Diagnosis of Chronic Diseases

A higher collection of medical documents is a valuable resource to retrieve new and valuable knowledge that can be found through data mining. Deep learning and data mining techniques are user-based approaches to identify hidden and novel data patterns. These highly applicable in identify key patterns among big datasets. At present, these are highly applying in healthcare systems especially of medical diagnosis to predict or classify diseases. Simultaneously, machine learning (ML) can detect and diagnose serious diseases like cancer, dementia, and diabetes. Especially deep learning is one application that highly applicable to the healthcare context is digital diagnosis. Besides, it can detect patterns of individual diseases within patient electronic health records (EHR) and produces feedback on anomalies to the doctor. This chapter presented a brief discussion including ML and deep learning approaches in a clinical context, differentiate between structured and unstructured patient data patterns and provide references to applications of mentioned methods in medicine. Besides, it also highlights performance measures and evaluation used in diagnosis prediction and classification process.