Biomedical Signal Processing and Artificial Intelligence in Healthcare (Developments in Biomedical Engineering and Bioelectronics)

Cover
Biomedical Signal
Processing and
Artificial Intelligence
in Healthcare
Copyright
Dedication
Contributors
Foreword
Preface
	Book chapters
Introduction to biomedical signal processing and artificial intelligence
	Introduction to signal processing
	Biomedical signals
		Electrocardiogram
		Electroencephalogram
	Noise
		Thermal noise
		Flicker noise
		Power-line interference
	Filters
		FIR filters
		Frequency domain filters
	Computer-aided diagnosis (CAD): Why?
	Artificial intelligence (AI): An overview
	Fuzzy logic in artificial intelligence
	Questions and answers
		Describe other types of biomedical signals such as EMG and ERG
		State any two difficulties encountered in biomedical signal acquisition and analysis
		Provide a comparison between stationary and nonstationary processes. Are biosignals such EEG and ECG stationary or nonstati ...
	References
Characterization of biomedical signals: Feature engineering and extraction
	Introduction
	Feature engineering
		Discrete Fourier transform
		Time-frequency analysis
		Statistical features
		Local binary patterns
	Feature ranking
		Variance threshold
		Correlation measures
		Information measures
		Class separability measures
	Feature selection
		Filter methods for feature selection
		Wrapper methods: Feature subset search
		Embedded methods
	Feature extraction
		Principal component analysis
		Fisher linear discriminant
	Summary
	Further reading
Supervised and unsupervised learning
	Introduction
	Density estimation
		Maximum likelihood, maximum a posteriori, and Bayesian parameter estimation
		Estimating parameters for individual densities
		Nonparametric and Kernel density estimation
			Kernel density estimation
			Nearest neighbor density estimation
	Classification analysis
		Bayes classifier
			MVN discriminant functions
			Naive Bayes classifier
			Nonparametric Bayes classifier
		Discriminant functions
		Linear discriminants
			Perceptron discriminant
			Least squares methods
			Fishers linear discriminant
		Generalized discriminants
			Logistic regression
				Logistic regression
			Kernel discriminants
		Constrained discriminant functions
			Support vector machines
			Kernel support vector machines
	Training and generalization performance
	Evaluating performance
	Summary
	Further reading
Machine learning in biomedical signal processing with ECG applications
	Introduction
		Automated ECG signal analysis
	The electrocardiogram
		Standard bipolar limb leads
		Augmented unipolar limb leads
		Precordial (chest) leads
	Clinical ECG features
	Cardiac arrhythmia
		Life-threatening arrhythmias
		The AAMI standard
	ECG heartbeat classifier
		ECG signal descriptor
		Intra- and interpatient paradigms
		Feature generation
		Ranking individual feature subsets
		Training and testing the final model
	Conclusions
	References
	Further reading
Deep EEG: Deep learning in biomedical signal processing with EEG applications
	EEG data basics
	Fundamentals of deep convolutional neural networks (DCNNs)
		Why deep learning
		Basics of deep convolutional neural networks
		The perceptron
		CNN architecture
		Convolutions and the convolutional layer
		Padding
		Strided convolutions
		Loss and optimization: Updated weights and biases
		Optimization algorithms
			Minibatch gradient descent and stochastic gradient descent (SGD)
			Gradient descent with momentum
			Root mean square prop (RMSprop)
			Adam optimization (adaptive moment estimation)
	TensorFlow and keras for deep convolutional neural networks
		Deep learning frameworks
		Setting up a deep learning application
		Bias versus variance
		Regularization
			L2 regularization
			Dropout regularization
		Data augmentation
		Normalizing data input
	Data collection workflow (step-by-step) with a BCI device
	Preprocessing and training using tensorflow and keras
	Deployment and real-time applications with embedded systems
	Conclusion, challenges and future research
	Appendix A. Working with Pandas
	References
Fuzzy logic in medicine
	Introduction to fuzzy logic
	Fuzzy sets
		The mathematical definition of fuzzy sets
		Representation of fuzzy sets
		Basic operations on fuzzy sets
		Properties of fuzzy sets
		Membership function
		Fuzzification
		Defuzzification
		An overview of the algebraic operations for fuzzy sets
	Application of fuzzy logic in medicine
		Fuzzy linear programming in medicine
			Fuzzy linear programming models
		Fuzzy multiple-criteria decision analysis in medicine
			Preference ranking organization method for enrichment evaluations (PROMETHEE)
			Fuzzy PROMETHEE (F-PROMETHEE)
			The technique for order of preference by similarity to ideal solution (TOPSIS)
	Challenges and opportunities
	Future direction
	Summary
	References
Neural network applications in medicine
	Introduction to artificial neural networks
		Artificial neural network architectures
			Multilayer perceptron and neural networks
			Back propagation neural network
			Convolutional neural networks
	Applications on neurological and neuropsychiatric diseases
		Alzheimers disease
		Parkinsons disease
		Attention-deficit/hyperactivity disorder
		Autism spectrum disorder
	Challenges and opportunities
	Future directions
	Summary
	Acknowledgments
	References
Analysis and management of sleep data
	Introduction
	Evolution of sleep medicine
	Sleep disorders
	Structure of sleep
		Macrostructure, sleep phases
		Microstructure, cyclic alternating patterns
		Sleep-related events
	Diagnostic standards for sleep data analysis
		Polysomnography
			Physiological parameters measured during polysomnography
			Automated analysis of polysomnographic data
		Ambulatory cardiorespiratory screening
		Actigraphy
		Body temperature
		Multichannel pressure measurement
		Drug-induced sleep endoscopy
		Acoustic analysis of breathing-related noise
	Conclusion
	Acknowledgments
	References
Index
Back Cover