Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Review Help
Search
Biological signals classification and analysis
Kiasaleh K., Springer International Publishing, New York, NY, 2015. 380 pp. Type: Book (978-3-642548-78-9)
Date Reviewed: Jul 1 2016

Biological signal classification and analysis by Kamran Kiasaleh involves discussion of signal processing approaches for biological signals such as the electrical signals from the brain (electroencephalogram, EEG), heart (electrocardiogram, ECG), and muscles (electromyogram, EMG). With the significant advancement in the computing power and readily available signal algorithms developed in other domains, it is no surprise that biological signal analysis is a hot topic currently, especially considering its usefulness for health-related applications. With this importance in mind, the author discusses various aspects of approaching the analysis of biological signals. Rather than using a generic signal-processing approach, the text covers specific examples of algorithms that are popular for the analysis of biological signals. The text does not assume any prior knowledge of the biological process such as ECG and EEG.

The book is divided into six chapters. Each chapter is preceded by introductory material. The first chapter sets the scene by introducing various fundamental concepts. A sampling theorem (to covert analog biological to digital signals for processing) is discussed initially, followed by related concepts such as anti-aliasing and quantization. Basic assumptions necessary in signal processing such as stationarity, ergodicity, and so on are discussed before moving on to representation of signal space, including using the Karhunen-Loeve transform.

Chapter 2 discusses linear system theory and the responses of linear and nonlinear systems to random signals. Next, the Gaussian signal is discussed, which is useful because it is one type of noise present in electronic measurements. For biological measurements, the noise could be from the recording instrument or from the underlying biological process itself, but irrespective of the source, noise in biological signals requires attention to avoid distorting the analysis outcome. The underlying biological noise is usually difficult to remove due to its wideband nature affecting a wide range of frequencies. Therefore, it is appropriate that there is a discussion of signals plus noise toward the end of the chapter with concepts such as signal-to-noise ratio (which is important to measure the effectiveness of any noise reduction) and matched/optimum filtering.

The first chapter to deal with actual biological signals is chapter 3. Components of ECG such as the P, Q, R, S, and T points are examined in relation to the actual pumping mechanism of the heart. For example, the T wave is the ventricle muscle from plateau to relaxation state. EEG is discussed next with the categorization of different bands (with varying frequency ranges) and components for different mental activities such as those encountered during waking and sleeping. Finally, EMG is discussed with fast Fourier transform (FFT) analysis of different myopathy and neuropathy cases.

Although chapter 3 briefly explores signal processing of biological signals, it is in chapter 4 where the bulk of the algorithms are discussed. The issues of independence (such as orthogonality and correlation) are first addressed because they are important when there is a large set of observations. Gaussianity is obviously related here, and hence measures to compute Gaussian behavior such as kurtosis sand negentropy are introduced. It is common to encounter two signals with different probability distributions (one could be the test signal and another a template signal); hence, distance measures would be important in the analysis. In this regard, measures such as Kolmogrov-Smirnov and Kullback-Leibler distances are discussed with some examples from EEG. Detection and estimation methods such as Bayes mixed mode hypothesis testing and other probability-based measures are treated next. Other useful concepts such as the Hilbert transform and polar plots are dealt with in the frequency estimation section.

Chapter 5 is on signal decomposition approaches such as the established principal component analysis and more recent approaches, such as independent component analysis (a type of blind source separation) and wavelets. Different types of wavelets are discussed extensively in the rest of the chapter.

Finally, chapter 6 gives brief concluding remarks and provides a list of references.

The presentation of materials is somewhat more mathematically inclined; the author could perhaps have toned down the math in order to suit a more introductory level course. Nevertheless, it is understandable to have some of the equations in order to realize the working steps in the algorithms. However, such an approach excludes those interested in a text exploring practical aspects of the algorithms (which is not really available). Also, some of the EEG figures in chapter 3 appear to be redundant in terms of the discussion and could perhaps be reduced and replaced with more practical examples. The classification term in the title is somewhat misleading because the focus is on analysis only.

Overall, the book is interesting to read and will be useful for researchers interested in the details of suitable algorithms using biological signals for any application, which perhaps suits the purpose of the “Lecture Notes in Bioengineering” series. The text may not be suited for a general course on biomedical signal processing; rather, it is appropriate for students and academics planning to do research specifically with one or more of the discussed algorithms.

Reviewer:  Ramaswamy Palaniappan Review #: CR144543 (1609-0647)
Bookmark and Share
 
Classifier Design And Evaluation (I.5.2 ... )
 
 
Biology And Genetics (J.3 ... )
 
 
Signal Analysis, Synthesis, And Processing (H.5.5 ... )
 
 
Signal Processing (I.5.4 ... )
 
 
Applications (I.5.4 )
 
 
Sound And Music Computing (H.5.5 )
 
  more  
Would you recommend this review?
yes
no
Other reviews under "Classifier Design And Evaluation": Date
Linear discrimination with symmetrical models
Bobrowski L. Pattern Recognition 19(1): 101-109, 1986. Type: Article
Feb 1 1988
An application of a graph distance measure to the classification of muscle tissue patterns
Sanfeliu A. (ed), Fu K., Prewitt J. International Journal of Pattern Recognition and Artificial Intelligence 1(1): 17-42, 1987. Type: Article
Dec 1 1989
Selective networks and recognition automata
George N. J., Edelman G.  Computer culture: the scientific, intellectual, and social impact of the computer (, New York,2011984. Type: Proceedings
May 1 1987
more...

E-Mail This Printer-Friendly
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright 1999-2024 ThinkLoud®
Terms of Use
| Privacy Policy