An Introductory Perspective on Machine Learning for Health Care

The use of machine learning is of fundamental importance for health professionals and managers, given its potential to carry out secondary or field data (primary) surveys for the preparation of datasets on patient behaviors and habits, recurrence of laboratory tests, evolution of health and well-being diagnoses and practices, as well as the development of structured acquisition for data collection, in addition to robust analytics models, oriented on machine learning and deep learning for predictive purposes and/or for the optimization of material resources such as humans, for improving the efficiency and quality of services provided in clinics and hospitals. Investigation and identification of current management practices and models in clinics, hospitals, and laboratories, for the subsequent proposition of improvements and optimization of processes and resources, assessment of organizational needs, which may include care for employees, such as screening, consultations, examination diagnosis, treatment and control, and monitoring and prevention. It is also possible that assessments are structured with a focus on local communities, as well as a preliminary analysis of the databases to study which modeling and analytics techniques can be used for certain types of decision making or even treatment of the existing databases for the identification of possible outliers or missing values, sample size and adjustments to data for the application of data science techniques. This chapter demonstrates a landscape view of the applied aspect, and also key concerns and challenges, aiming to provide an updated overview of machine learning for health care and technologies, with a concise bibliographic background featuring the potential of technologies, showing the fundamentals of this disruptive technology.