Biomedical Signals Based Computer-Aided Diagnosis for Neurological Disorders

Preface
Acknowledgements
Contents
Abnormal EEG Detection Using Time-Frequency Images and Convolutional Neural Network
1 Introduction
2 Related Studies
3 Materials and Methods
3.1 Data Preparation
3.2 Time-Frequency Representation-Based Abnormal EEG Detection
3.3 Convolutional Neural Networks
3.3.1 Inception-ResNet-V2
3.3.2 DenseNet
3.3.3 SeizureNet
3.4 Extreme Learning Machine
3.5 Training and Validation
4 Results
5 Discussion
6 Conclusion
References
Physical Action Categorization Pertaining to Certain Neurological Disorders Using Machine Learning-Based Signal Analysis
1 Introduction
2 Related Work
3 Materials and Methods
3.1 Acquisition Band
3.2 Data Acquisition
4 Methodology
4.1 Pre-Processing and Segmentation
4.2 Feature Vector Generation
4.2.1 Time Domain Features
4.2.2 Inter-Channel Statistics
Maximum Similarity Index
Maximum Covariance Index
4.2.3 Power Spectral Density (PSD)
4.2.4 Log Moments of Fourier Spectra (LMF)
4.3 Feature Normalization
4.4 Classifier
5 Experimental Results and Discussion
5.1 Dataset
5.2 Experimental Results
6 Discussion
7 Conclusions
References
A Comparative Study on EEG Features for Neonatal Seizure Detection
1 Introduction
2 Methodology
2.1 Database
2.2 Pre-processing
2.3 Feature Extraction
2.4 Feature Ranking and Classification
3 Results
4 Discussions
4.1 Effect of Number of Channels
4.2 Effect of Feature Extraction Methods
4.3 Effect of Feature Ranking Methods
4.4 Effect of Classifiers
4.5 Effect of Performance Metrics
4.6 Limitations and Future Directions
5 Conclusion
References
Hilbert Huang Transform (HHT) Analysis of Heart Rate Variability (HRV) in Recognition of Emotion in Children with Autism Spect...
1 Introduction
2 Methodology
3 Protocol Design
4 ECG Data Acquisition Using Wearable Sensor
5 Data Pre-processing
6 Feature Extraction
7 Hilbert-Huang Transform (HHT) Algorithm
8 Feature Classification
9 Results
9.1 Pre-processing
9.2 Extraction of Features
10 Discussion and Conclusion
References
Detection of Tonic-Clonic Seizures Using Scalp EEG of Spectral Moments
1 Introduction
2 Methods
2.1 Feature Extraction
2.2 Rectangular Window
2.3 Hanning Window
2.4 Hamming Window
2.5 Flattop Window
2.6 Spectral Moments
2.7 Classification
3 Results and Discussion
4 Conclusion
References
Investigation of the Brain Activation Pattern of Stroke Patients and Healthy Individuals During Happiness and Sadness
1 Introduction
2 Materials and Methods
2.1 Subjects´ Background
2.2 EEG Device
2.3 Signal Pre-processing
2.4 Feature Analysis
2.5 Emotion Classification
3 Results
3.1 Average Values of Hjorth Parameters
3.2 Distribution of Hjorth Parameters in Different Frequency Sub-bands
3.3 Statistical Analysis of Hjorth Parameters Between Happiness and Sadness
3.4 Average Distribution over Frontal Regions of Left and Right Hemispheres
3.5 Emotion Classification Using KNN Classifier
4 Discussion
4.1 Emotion Impairment Analyzed Using Hjorth Parameters
4.2 Region of Emotional Activation in the Brain
4.3 The Involvement of Frontal Region in Emotion Processing
5 Conclusion
References
A Novel Parametric Nonstationary Signal Model for EEG Signals and Its Application in Epileptic Seizure Detection
1 Introduction
2 Signal Modeling and Parametric Estimation
2.1 Segmentation and AM-FM Demodulation
2.2 Phase Smoothing and Instantaneous Frequency (IF) Estimation
3 Simulation Results
4 Applications
5 Conclusion and Discussion
References
Biomedical Signal Analysis Using Entropy Measures: A Case Study of Motor Imaginary BCI in End Users with Disability
1 Introduction
2 Entropy
2.1 History
2.2 Entropy and Information Theory
2.3 Mathematical Formulation
2.3.1 Shannon Entropy
2.3.2 Approximate Entropy
2.3.3 Neural Network Entropy
3 Numerical Results
3.1 Channel Selection
4 Conclusion and Further Work
References
Automatic Detection of Epilepsy Using CNN-GRU Hybrid Model
1 Introduction
2 Methodology
2.1 Dataset
2.2 Scalogram
2.3 Architecture
2.4 Training the Models
3 Results
3.1 Two-Class Performance
3.1.1 Case 1: AB-CDE
3.1.2 Case 2: ABCD-E
3.2 Three-Class Performance
3.3 Five-Class Performance
4 Discussion
5 Conclusion
References
Catalogic Systematic Literature Review of Hardware-Accelerated Neurodiagnostic Systems
1 Introduction
2 Methodology
2.1 Research Questions
2.2 Search Strategy
2.3 Selection Criteria
2.4 Data Categorization
2.5 Quality Assessment
2.6 Data Extraction
3 Review Conduction
4 Metadata Analysis
5 Data Analysis
5.1 C1: Acquisition Stage
5.1.1 C1.1: Acquisition Interface
5.1.2 C1.2: Analog-to-Digital Conversion
5.1.3 C1.3: Amplification
5.1.4 C1.4: Data Compression
5.1.5 C1.5: Transmission Protocol
5.2 C2: Preprocessing Stage
5.2.1 C2.1: Detrending/Filtering
5.2.2 C2.2: Feature Extraction
5.3 C3: Processing Stage
5.3.1 C3.1: Seizure Detection
5.3.2 C3.2: Emotion/Vigilance Classification
5.3.3 C3.3: Autism/Anomaly Detection
5.3.4 C3.4: Intention/Imagery/Gesture/Speech Recognition
5.3.5 C3.5: Sleep/Hypnosis-Level Classification
6 Results
6.1 RQ1: What Hardware Accelerators Exist?
6.2 RQ2: Which Algorithms Are Implemented?
6.3 RQ3: Which Technology/Platform Is Preferred?
6.4 RQ4: Which Stages Are Hardware Accelerated?
6.5 RQ5: What Are the Advantages and Disadvantages?
6.6 RQ6: What Are the Metrics for the Assessment?
7 Discussion
8 Conclusion
References
Wearable Real-Time Epileptic Seizure Detection and Warning System
1 Introduction
2 Materials and Methods
2.1 Block Diagram of the System
2.2 Wearable Bio-signal Acquisition Subsystem
2.2.1 Electrodermal Activity (EDA) Sensor
2.2.2 Accelerometer (ACC) Module
2.2.3 Wireless Communication over BLE
2.3 Intelligent Epilepsy Detection and Alerting Subsystem
2.3.1 Database Description
3 Analysis
3.1 Data Preprocessing
3.2 Feature Extraction
3.3 Training, Validation, and Testing
4 Results and Discussion
4.1 Performance Evaluation of EDA and ACC Sensor
4.2 Reliability of the BLE Transmission
4.3 Power Consumption of the Wearable Module
4.4 Classification Results for ACC Data Alone
4.5 EDA and ACC Data from Empatica and Prototype System
4.6 Classification of Seizure Using EDA Alone, ACC Alone, and Fused EDA-ACC
5 Conclusions
References
Analysis of Intramuscular Coherence of Lower Limb Muscle Activities Using Magnitude Squared Coherence
1 Introduction
2 Methods
2.1 Magnitude Squared Coherence
3 Results
4 Discussion
5 Conclusion
References
Index