This article provides an overview of machine learning fundamentals and some applications of machine learning to clinical laboratory diagnostics and patient management. A key goal of this article is to provide a basic foundation in clinical machine learning for readers with clinical laboratory experience that will set them up for more in-depth study of the topic and/or to become a better collaborator with computational colleagues in the development and deployment of machine learning–based solutions.
Machine learning and artificial intelligence, commonplace in many areas outside of health care, offer tremendous potential to transform many aspects of laboratory diagnosis.
Understanding fundamentals of machine learning will enable pathologists and other laboratorians to better evaluate tests and technologies for use in their laboratory and to better collaborate on projects involving predictive algorithm development, validation, and implementation.
Machine learning projects often involves steps including (1) problem conception and development; (2) data extraction, cleaning, and assembly; (3) model training and model validation; and (4) model implementation.
Clinical implementation of machine learning models requires addressing technical, administrative, evidentiary, and educational challenges.
Laboratory medicine has traditionally relied mostly on manual clinician decision-making to select which laboratory tests to order and to apply laboratory test results to patient diagnosis and management. , However, modern machine learning approaches that are commonplace in many industries besides health care are poised to augment—and to some extent have already augmented—human decision-making in this realm with multiple benefits both to patients and clinicians. Although this Clinics in Laboratory Medicine issue as a whole reviews a variety of laboratory artificial intelligence (AI) applications, strategies, and considerations in depth, this particular article is meant to provide a general foundation. This article, written mostly in question-and-answer format, looks at fundamentals of machine learning and applications to medicine and provides a practical overview of the process of developing a clinical prediction model.