Advanced Methods of Biomedical Signal Processing
Format: PDF / Kindle (mobi) / ePub
This book grew out of the IEEE-EMBS Summer Schools on Biomedical Signal Processing, which have been held annually since 2002 to provide the participants state-of-the-art knowledge on emerging areas in biomedical engineering. Prominent experts in the areas of biomedical signal processing, biomedical data treatment, medicine, signal processing, system biology, and applied physiology introduce novel techniques and algorithms as well as their clinical or physiological applications.
The book provides an overview of a compelling group of advanced biomedical signal processing techniques, such as multisource and multiscale integration of information for physiology and clinical decision; the impact of advanced methods of signal processing in cardiology and neurology; the integration of signal processing methods with a modelling approach; complexity measurement from biomedical signals; higher order analysis in biomedical signals; advanced methods of signal and data processing in genomics and proteomics; and classification and parameter enhancement.
of DFT parameters. Such spectra are obviously discrete-frequency spectra. There are various algorithms that are applied for the spectrum estimation, either for nonparametric spectra (which basically derive from the previously introduced concept of the Fourier transform), or for parametric spectra (which are based upon a specific model of signal generation mechanism). For a deeper analysis of these topics, see Kay and Marple, 1981; Marple, 1987; and Kay, 1988. As far as the nonparametric approach
deeper analyses. The part-whole or part-part boundaries can present either well-drawn features and universally shareable attribution of character (e.g., a neuron under a microscope) or show strictly indefinite boundaries (e.g., the somatosensory cortex). In the former case, consistent and reliable object identification implies universally shareable distinctions of its borders, a spatial criterion called by some ontologists the bona fide boundary. In this case, the indefiniteness character implies
to open new vistas to highly promising theoretical frames. Among these frames, one of the most delicate but powerful is the fluorishing and robust idea of complexity, in spite of its incompleteness and conceptual instability. 3.2.4 Complexity Complexity (C) is a strange item and has no conclusive definitions. It is bound to many aspects of emergence. Complexity cannot be blamed in the case of complete disorder of a system. A disordered system can be approached by number theory. A perfectly
the dimension of the problem, for instance, with a filtering technique or using statistical techniques such as principal component analysis, thus reducing the overall variance of the signal. 3. This signal representation does not exhibit any predictive capacity. The processing algorithm must be applied to each new tracing. 130 CHAPTER 6 USE OF INTERPRETATIVE MODELS 4. Frequently in the study of physiological problems, several signals are simultaneously monitored. Of course, it is extremely
original neural mechanisms, due to a series of problems (Horwitz et al., 2000), which will be briefly summarized below. First, the techniques for functional neuroimaging (PET and fMRI) have a high spatial resolution (on the order of 1 mm or less) but poor temporal resolution. Moreover, these techniques do not directly measure neural activity (in terms of spike frequency of membrane potential changes), but rather are sensitive to local changes in metabolism and blood flow. These latter quantities