Computational Techniques in Neuroscience

The text discusses the techniques of deep learning and machine learning in the field of neuroscience, engineering approaches to study the brain structure and dynamics, convolutional networks for fast, energy-efficient neuromorphic computing, and reinforcement learning in feedback control. It showcases case studies in neural data analysis.Features: Focuses on neuron modeling, development, and direction of neural circuits to explain perception, behavior, and biologically inspired intelligent agents for decision making Showcases important aspects such as human behavior prediction using smart technologies and understanding the modeling of nervous systems Discusses nature-inspired algorithms such as swarm intelligence, ant colony optimization, and multi-agent systems Presents information-theoretic, control-theoretic, and decision-theoretic approaches in neuroscience. Includes case studies in functional magnetic resonance imaging (fMRI) and neural data analysis This reference text addresses different applications of computational neuro-sciences using artificial intelligence, deep learning, and other machine learning techniques to fine-tune the models, thereby solving the real-life problems prominently. It will further discuss important topics such as neural rehabili-tation, brain-computer interfacing, neural control, neural system analysis, and neurobiologically inspired self-monitoring systems. It will serve as an ideal reference text for graduate students and academic researchers in the fields of electrical engineering, electronics and communication engineering, computer engineering, information technology, and biomedical engineering. Cover Half Title Series Page Title Page Copyright Page Dedication Contents About the Book Preface Editor(s) Contributors Chapter 1: Dynamic intuitionistic fuzzy weighting averaging operator: A multi-criteria decision-making technique for the diagnosis of brain tumor 1.1 Introduction 1.2 Multi-criteria decision making 1.3 Aggregation 1.4 Decision making 1.5 Medical diagnosis 1.6 Fuzzy theory 1.7 Intuitionistic fuzzy sets (IFS) 1.8 Intuitionistic fuzzy variable 1.9 Intuitionistic fuzzy number (IFN) and its operations 1.10 Dynamic intuitionistic fuzzy weighted averaging (DIFWA) operator 1.11 Medical diagnosis of the type of brain tumor 1.12 Proposed medical making algorithm: Dynamic intuitionistic fuzzy weighted averaging (DIFWA) operator 1.13 Evaluation of case study 1.14 Result 1.15 Result discussion 1.16 Conclusion References Chapter 2: Neural modeling and neural computation in a medical approach 2.1 Introduction 2.1.1 Introduction 2.1.2 Why are neuron models better? 2.1.3 Objective 2.2 Dynamic and architecture for neural computation 2.2.1 Overview of dynamic model 2.3 Neural modeling in functioning brain imaging 2.3.1 Hemodynamic-metabolic methods of functional neuroimaging signal 2.3.1.1 Functional MRI 2.3.1.2 Electric-magnetic methods 2.3.2 A brief review of neural modeling in functional brain imaging 2.3.2.1 Neuromodeling and PET/fMRI 2.3.2.2 EEG/MEG and neuromodeling 2.3.3 Conclusion 2.4 Literature review 2.4.1 Type of neural model 2.4.1.1 Single cell level models 2.4.1.2 Ensemble-level models 2.4.1.3 Systems-level models 2.4.2 Machine learning 2.4.2.1 Artificial intelligence vs machine learning vs deep learning 2.4.2.1.1 Artificial intelligence 2.4.2.1.2 Machine learning 2.4.2.1.2.1 Supervised learning 2.4.2.1.2.2 Unsupervised learning 2.4.2.1.2.3 Reinforcement learning 2.4.2.1.3 Deep learning 2.4.3 Application of machine learning 2.4.3.1 Machine learning in healthcare 2.4.4 Types of algorithms being used 2.4.4.1 Logistic regression 2.4.2.2 Convolutional neural network 2.4.2.3 Artificial neural networks 2.4.5 Considered learning algorithms 2.5 Best performing algorithm 2.6 Normalization and neural coding 2.7 Conclusion References Chapter 3: Neural networks and neurodiversity: The key foundation for neuroscience 3.1 Introduction 3.2 What is neuroscience? 3.3 Artificial neural network: A brief chronology 3.3.1 Do deep learning and neuroscience still need each other? 3.4 Neuro-imaging methods for cognitive developmental neurosciences 3.5 Neuromyths 3.6 Neural networks 3.6.1 Neuron models 3.6.2 General properties of neural networks 3.6.3 Neural network classification 3.6.3.1 Multilayer feedforward neural network (MLFFNN) 3.6.3.2 Single-layer feedforward neural network (SLFFNN) 3.6.3.3 Recurrent neural network (RNN) 3.7 RNNs as a tool of neurological science research 3.7.1 RNNs as an important model for computations 3.7.2 RNNs designing 3.7.3 Functionality and optimization 3.8 RNNs can be trained without intuition 3.9 Hypothesis and theory generation 3.10 Introduction to neurodiversity 3.11 Neurodiversity: The situation of including autistic employees at work 3.11.1 The links between technology, organization, and skills 3.11.2 Problem analysis 3.11.3 Neurodiversity at the workplace at different levels 3.11.4 Methodology 3.11.5 Result 3.12 Scope and conclusion References Chapter 4: Brain waves, neuroimaging (fMRI, EEG, MEG, PET, NIR) 4.1 Introduction 4.2 Brain waves 4.3 Neuroimaging 4.4 Conclusion References Web Source Chapter 5: EEG: Concepts, research-based analytics, and applications 5.1 Introduction 5.2 Preprocessing techniques of EEG signals 5.3 Machine learning and deep learning based EEG data analysis techniques 5.4 Applications of EEG 5.4.1 Cognitive neuroscience 5.4.2 Behavioral neuroscience 5.4.3 Neuro-marketing 5.4.4 Sports and meditation 5.4.5 Educational purpose 5.4.6 Security 5.4.7 Brain control robotics 5.5 Challenges associated with EEG 5.5.1 Technical challenges 5.5.2 Social and ethical challenges 5.5.3 Environmental challenges 5.6 Conclusion References Chapter 6: Classification of gait signals for detection of neurodegenerative diseases using log energy entropy and ANN classifier 6.1 Introduction 6.2 Method and materials 6.2.1 Dataset used 6.2.2 Feature extraction 6.2.3 Classification 6.2.3.1 Classification performance 6.3 Results and discussion 6.4 Conclusion References Chapter 7: An optimized text summarization for healthcare analytics using swarm intelligence 7.1 Introduction 7.1.1 Text summarization 7.1.2 Text summarization approaches 7.1.2.1 Extractive text summarization 7.1.2.2 Abstractive text summarization 7.1.3 Text summarization in healthcare 7.2 Literature review 7.3 TF-IDF algorithm 7.4 Swarm intelligence using particle swarm optimization 7.4.1 Particle swarm optimization algorithm 7.5 Proposed methodology 7.5.1 Input text 7.5.2 Preprocessing 7.5.2.1 Sentence tokenization 7.5.2.2 Stop word removal 7.5.2.3 Stemming 7.5.3 Applying TF-IDF algorithm 7.5.4 Generation of different versions of summary 7.5.5 Applying PSO algorithm 7.5.5.1 Find all possible sets of summaries 7.5.5.2 Initialization of PSO parameters 7.5.5.3 Update the parameters until they get optimized or until some condition is reached 7.5.5.4 Get the p-best value for all versions 7.5.6 Evaluate the summaries and provide the best optimized summary as a result 7.6 Results and discussions 7.7 Conclusion and future work References Chapter 8: Computer aided diagnosis of neurodegenerative diseases using discrete wavelet transform and neural network for classification 8.1 Introduction 8.2 Methods and materials 8.2.1 Dataset used 8.2.2 Discrete wavelet transforms 8.2.3 Feature extraction 8.2.4 Artificial neural network classifier 8.3 Results and discussion 8.4 Conclusion References Chapter 9: EEG artifact detection and removal techniques: A brief review 9.1 Introduction 9.2 Different types of EEG artifacts 9.2.1 Ocular (EOG) artifact 9.2.2 Muscular artifact 9.2.3 Cardiac artifact 9.2.4 Motion artifact 9.3 Artifact removal techniques 9.3.1 Regression technique 9.3.2 Filtering technique 9.3.2.1 Adaptive filtering 9.3.3 Decomposition technique 9.3.3.1 Techniques of blind source separation (BSS) 9.3.3.1.1 Independent component analysis (ICA) 9.3.3.1.2 Canonical correlation analysis (CCA) 9.3.3.1.3 Morphological component analysis (MCA) 9.3.3.1.4 Principal component analysis (PCA) 9.3.3.2 Wavelet transform (WT) 9.3.3.3 Empirical mode decomposition (EMD) 9.3.3.4 Variational mode decomposition 9.3.4 Machine learning technique 9.3.5 Combined approach for artifact removal 9.3.5.1 Blind source separation and adaptive filtering 9.3.5.2 Adaptive filtering and wavelet transform 9.3.5.3 Technique of BSS and WT 9.3.5.4 Technique of EMD and BSS 9.3.5.5 Adaptive filtering and EMD 9.3.5.6 Technique of BSS and SVM 9.3.6 Summary of earlier methods used for EEG artifact removal 9.4 Proposed technique 9.4.1 Fast discrete S transform (FDST) 9.5 Result and discussion 9.6 Conclusion References Chapter 10: Analysis of neural network and neuromorphic computing with hardware: A survey 10.1 Introduction 10.2 Models of research 10.2.1 Models of neurons 10.2.2 Synapse models 10.2.2.1 Network model 10.3 Algorithm and learning 10.3.1 Supervised learning concept 10.4 Conclusion References Chapter 11: Analysis of technology research and ADHD with the neurodivergent reader: A survey 11.1 Introduction 11.2 Domain previous study 11.2.1 ADHD 11.2.2 Neurodivergence and HCI 11.2.3 Studies of disabilities with a critical lens and crip technology 11.3 Research methods 11.3.1 Development of corpus 11.4 Result and discussion 11.5 Research gaps 11.6 Conclusion References Index