Machine Learning in Medicine - a Complete Overview

Abstract

The amount of data stored in the world’s databases doubles every 20 months, as estimated by Usama Fayyad, one of the founders of machine learning and co-author of the book Advances in Knowledge Discovery and Data Mining (ed. by the American Association for Artifi cial Intelligence, Menlo Park, CA, USA, 1996), and clinicians, familiar with traditional statistical methods, are at a loss to analyze them. Traditional methods have, indeed, diffi culty to identify outliers in large datasets, and to fi nd patterns in big data and data with multiple exposure/outcome variables. In addition, analysis-rules for surveys and questionnaires, which are currently common methods of data collection, are, essentially, missing. Fortunately, the new discipline, machine learning, is able to cover all of these limitations. So far, medical professionals have been rather reluctant to use machine learning. Ravinda Khattree, co-author of the book Computational Methods in Biomedical Research (ed. by Chapman & Hall, Baton Rouge, LA, USA, 2007) suggests that there may be historical reasons: technological (doctors are better than computers (?)), legal, cultural (doctors are better trusted). Also, in the fi eld of diagnosis making, few doctors may want a computer checking them, are interested in collaboration with a computer or with computer engineers. Adequate health and health care will, however, soon be impossible without proper data supervision from modern machine learning methodologies like cluster models, neural networks, and other data mining methodologies. The current book is the fi rst publication of a complete overview of machine learning methodologies for the medical and health sector, and it was written as a training companion, and as a must-read, not only for physicians and students, but also for anyone involved in the process and progress of health and health care. Some of the 80 chapters have already appeared in Springer’s Cookbook Briefs, but they have been rewritten and updated. All of the chapters have two core characteristics. First, they are intended for current usage, and they are, particularly, concerned with improving that usage. Second, they try and tell what readers need to know in order to understand the methods.