Machine Learning for Designing an Automated Medical Diagnostic System

Machine Learning for Designing an Automated Medical Diagnostic System

Ahsan H. Khandoker (The University of Melbourne, Australia) and Rezaul K. Begg (Victoria University, Australia)
DOI: 10.4018/978-1-59904-887-1.ch030
OnDemand PDF Download:
$37.50

Abstract

This chapter describes the application of machine learning techniques to solve biomedical problems in a variety of clinical domains. First, the concept of development and the main elements of a basic machine learning system for medical diagnostics are presented. This is followed by an introduction to the design of a diagnostic model for the identification of balance impairments in the elderly using human gait pattern, as well as a diagnostic model for predicating sleep apnoea syndrome from electrocardiogram recordings. Examples are presented using support vector machines (a machine learning technique) to build a reliable model that utilizes key indices of physiological measurements (gait/electrocardiography [ECG] signals). A number of recommendations have been proposed for choosing the right classifier model in designing a successful medical diagnostic system. The chapter concludes with a discussion of the importance of signal processing techniques and other future trends in enhancing the performance of a diagnostic system.

Complete Chapter List

Search this Book:
Reset