Oweiss K.G. (ed.) Statistical Signal Processing for Neuroscience and Neurotechnology

Издательство Elsevier, 2010, -420 pp.Neuroscience, like many other areas of biology, has experienced a data explosion in the last few years. Many remarkable findings have been triggered since and have undeniably improved our understanding of many aspects of the nervous system. However, the study of the brain and its inner workings is an extraordinarily complex endeavor; undoubtedly it requires the collective effort of many experts across multiple disciplines to accelerate the science of information management that neuroscience data have invigorated.
Signal processing and statistics have a long history of mutual interaction with neuroscience, going back to theHudgkin-Huxley circuitmodels ofmembrane potential and leading up to statistical inference about stimulus–response relationships using Bayes’ rule. Classical signal processing, however, quickly erodes in the face of dynamic, multivariate neuroscience and behavioral data. Realizing the immediate need for a rigorous theoretical and practical analysis of such data, this book waswritten by experts in the field to provide a comprehensive reference summarizing state-of-the-art solutions to some fundamental problems in neuroscience and neurotechnology. It is intended for theorists and experimentalists at the graduate and postdoctoral level in electrical, computer, and biomedical engineering, aswell as formathematicians, statisticians, and computational and systems neuroscientists. A secondary audience may include neurobiologists, neurophysiologists, and clinicians involved in basic and translational neuroscience research.The book can also be used as a supplement in a quantitative course in neurobiology or as a textbook for instruction on neural signal processing and its applications. Industry partners who want to explore research in and clinical applications of these concepts in the context of neural interface technology will certainly find useful content in a number of chapters.
This book was written with the following goals in mind:
To serve as an introductory text for readers interested in learning how traditional and modern principles, theories, and techniques in statistical signal processing, information theory, andmachine learning can be applied to neuroscience problems.
To serve as a reference for researchers and clinicians working at the interface between neuroscience and engineering.
To serve as a textbook for graduate-level courses within an electrical engineering, biomedical engineering, or computational and systems neuroscience curriculum.
Readers are expected to have had an introductory course in probability and statistics at the undergraduate level as part of an engineering or neuroscience course of study. The book is organized so that it progressively addresses specific problems arising in a neurophysiological experiment or in the design of a brain-machine interface system. An ultimate goal is to encourage the next generation of scientists to develop advanced statistical signal processing theory and techniques in their pursuit of answers to the numerous intricate questions that arise in the course of these experiments.
I owe a considerable amount of gratitude to my fellow contributors and their willingness to spare a significant amount of time to write, edit, and help review every chapter in this book since its inception. I am also very thankful to a number of colleagues who helped me in the review of early versions of the manuscript.
Developing the tools needed to understand brain function and its relation to behavior from the collected data is no less important than developing the devices used to collect these data, or developing the biological techniques to decipher the brain’s anatomical structure. I believe this book is a first step towards fulfilling an urgent need to hasten the development of these tools and techniques that will certainly thrive in the years to come.Detection and Classification of Extracellular Action Potential Recordings
Information-Theoretic Analysis of Neural Data
Identification of Nonlinear Dynamics in Neural Population Activity
Graphical Models of Functional and Effective Neuronal Connectivity
State Space Modeling of Neural Spike Train and Behavioral Data
Neural Decoding for Motor and Communication Prostheses
Inner Products for Representation and Learning in the Spike Train Domain Signal Processing and Machine Learning for Single-trial Analysis of Simultaneously Acquired EEG and fMRI
Statistical Pattern Recognition and Machine Learning in Brain–Computer Interfaces
Prediction of Muscle Activity from Cortical Signals to Restore Hand Grasp in Subjects with Spinal Cord Injury