Chapter 12

Joint Estimation of the Dynamicsand Shape of Physiological Signalsthrough Genetic Algorithms

12.1. Introduction

The aim of this chapter is to introduce an optimization technique which is based on genetic algorithms (GA). This optimization technique will be used in order to estimate brainstem auditory evoked potentials (BAEPs). We must point out that in certain abnormalities these physiological signals are generally highly non-stationary and are also corrupted by heavy noise, basically due to the electroencephalogram activity (EEG).

Estimating the BAEPs relies on several models, relative to both their dynamics and their shape. In this chapter, a definition of BAEPs will be given as well as an explanation on the way they are generated. An insight into the techniques used in estimating the BAEPs will then be introduced in section 12.3. The principle of GAs will be reviewed in section 12.4. The use of such algorithms to deal with the problems related to BAEP non-stationarity is described in sections 12.5 and 12.6.


Chapter written by AMINE NAÏT-ALI and PATRICK SIARRY.

12.2. Brainstem auditory evoked potentials

BAEPs are low energy electrical signals, generated when stimulating the auditory system by acoustical impulses. They are mainly used to ensure the earliest possible diagnosis of acoustic neuromas.

An acoustic neuroma is in fact a benign tumor which might lead, in some cases, to the death of a patient. One of the first signs of this disease in patients is ...

Get Optimisation in Signal and Image Processing now with the O’Reilly learning platform.

O’Reilly members experience live online training, plus books, videos, and digital content from nearly 200 publishers.