ABSTRACT
This chapter provides a nontechnical introduction to the machine learning (ML) discipline aimed at a general audience with an affinity for biomedical applications. It familiarizes the reader with the common types of algorithms and typical tasks these algorithms can solve and illustrate these basic concepts through concrete examples of current machine learning applications in healthcare. In healthcare, ML methods are usually leveraged to extract patterns that correlate with medical conditions. They are applied to healthcare records and other patient data to provide clinicians with decision support by predicting, e.g., diagnosis, management, and outcome in an automated manner. An advantage of ML methods over conventional statistical modeling is their flexibility and scalability in exploiting diverse and complex data types. Arguably, the most common ML approach in healthcare is supervised learning. ML methods generally assume that the training and test sets feature similar patterns and relationships, which usually do not hold for healthcare applications.
