Analysis of Medical Modalities for Improved Diagnosis in Modern Healthcare

Cover
Half Title
Title Page
Copyright Page
Dedication
Table of Contents
Preface
Acknowledgments
Editors
Contributors
Chapter 1 Classification of Alertness and Drowsiness States Using the Complex Wavelet Transform-Based Approach for EEG Records
	1.1 Introduction
	1.2 Methodology
		1.2.1 Dataset
		1.2.2 Dual-Tree Complex Wavelet Transform
		1.2.3 Feature Extraction
		1.2.4 Classification
	1.3 Results and Discussion
	1.4 Conclusion
	References
Chapter 2 Stochastic Event Synchrony Based on a Modified Sparse Bump Modeling: Application to PTSD EEG Signals
	2.1 Introduction
	2.2 Sparse Bump Modeling
	2.3 Stochastic Event Synchrony
		2.3.1 One-Dimensional Point Processes
		2.3.2 Multidimensional Point Processes
	2.4 Synchrosqueezed Wavelet Transform
		2.4.1 The Original Form
		2.4.2 Second-Order Wavelet-Based SST (WSST2)
			2.4.2.1 Numerical Implementation of WSST2
			2.4.2.2 Quantifying WSST2
	2.5 Stochastic Event Synchrony Based on a Modified Bump Modeling
	2.6 Data Description and Preprocessing
	2.7 Results
	2.8 Conclusion
	Acknowledgment
	References
Chapter 3 HealFavor: A Chatbot Application in Healthcare
	3.1 Introduction
		3.1.1 Healthcare
		3.1.2 NLP in Healthcare
	3.2 Theoretical Background
		3.2.1 Chatbots in Healthcare
		3.2.2 Research Gaps
		3.2.3 Privacy Concerns
	3.3 Data Preparation Methodology
		3.3.1 Data Sources and Observations
		3.3.2 Quality and Filtering
		3.3.3 Pre-Processing
		3.3.4 Data Representation
	3.4 Prototype System Architecture
		3.4.1 System Description: Leveraging the Deep Learning Model
		3.4.2 HealFavor: Our Prototype System
	3.5 Machine Translation
		3.5.1 Machine Translation and Its Approaches
		3.5.2 Rule-Based Machine Translation
		3.5.3 Corpus-Based Machine Translation
		3.5.4 Machine Translation in HealFavor Application
	3.6 Evaluation
		3.6.1 User Experience Survey
		3.6.2 Accuracy
	3.7 Conclusion
	3.8 Future Works
	Acknowledgments
	References
Chapter 4 Diagnosis of Neuromuscular Disorders Using Machine Learning Techniques
	4.1 Introduction
	4.2 Literature Review
	4.3 EMG Signal Classification Framework
		4.3.1 Wavelet Transform
		4.3.2 Signal Denoising with MSPCA
		4.3.3 Feature Extraction
			4.3.3.1 Discrete Wavelet Transform (DWT)
			4.3.3.2 Wavelet Packet Decomposition (WPD)
			4.3.3.3 Tunable Q-Factor Wavelet Transform (TQWT)
			4.3.3.4 Dual-Tree Complex Wavelet Transform (DT-CWT)
			4.3.3.5 Stationary Wavelet Transform (SWT)
			4.3.3.6 Flexible Analytic Wavelet Transform (FAWT)
			4.3.3.7 Empirical Wavelet Transform (EWT)
		4.3.4 Dimension Reduction
		4.3.5 Classification Methods
			4.3.5.1 Artificial Neural Networks (ANNs)
			4.3.5.2 k-Nearest Neighbor (k-NN)
			4.3.5.3 Support Vector Machine (SVM)
			4.3.5.4 C4.5 Decision Tree
			4.3.5.5 Classification and Regression Trees (CART)
			4.3.5.6 Reduced Error Pruning Tree (REP) Tree
			4.3.5.7 LogitBoost Alternating Decision (LAD) Tree
			4.3.5.8 Random Tree Classifier
			4.3.5.9 Random Forests (RF)
	4.4 Results and Discussion
		4.4.1 Subjects and Data Acquisition
		4.4.2 Performance Evaluation
		4.4.3 Experimental Results
		4.4.4 Discussion
	4.5 Conclusion
	References
Chapter 5 Prosthesis Control Using Undersampled Surface Electromyographic Signals
	5.1 Introduction
	5.2 Myoelectric Controlled Prosthesis
		5.2.1 On/Off and Finite-State Myoelectric Control
		5.2.2 Proportional, Direct, and Posture Myoelectric Control
		5.2.3 Pattern Recognition-Based Myoelectric Control
		5.2.4 Regression-Based Myoelectric Control
		5.2.5 Deep Learning-Based Myoelectric Control
		5.2.6 Performance Indices
		5.2.7 Challenges in Myoelectric Prosthesis Control
	5.3 sEMG Signal Recording in Myoelectric Prosthesis Control
		5.3.1 EMG-Force Estimation
		5.3.2 Elbow Angle Estimation
		5.3.3 Finger Gesture Detection
		5.3.4 Time and Frequency Feature Extraction
		5.3.5 Prosthesis Control
	5.4 Conclusion and Future Scope
	Acknowledgments
	References
Chapter 6 Assessment and Diagnostic Methods for Coronavirus Disease 2019 (COVID-19)
	6.1 Introduction
	6.2 Clinical Findings of COVID-19
	6.3 Existing Diagnostic Tools
		6.3.1 Real-Time RT-PCR Test Molecular Test
		6.3.2 Rapid Antigen Detection (RAD) Test
		6.3.3 Antibodies Test
		6.3.4 Chest Computed Tomography
	6.4 Current Screening Tools for COVID-19
		6.4.1 Thermometers
		6.4.2 Thermal Imaging Systems
	6.5 Capnogram Features
	6.6 Proposed Tool for Early Screening of COVID-19 Using Respired CO[sub(2)] Features
		6.6.1 Exhaled Breath Samples Collection Part
		6.6.2 Processing Part
		6.6.3 Display Unit
		6.6.4 Real-Time Clock (RTC)
	6.7 Conclusion
	Disclosure
	Acknowledgments
	References
Chapter 7 Predictive Analysis of Breast Cancer Using Infrared Images with Machine Learning Algorithms
	7.1 Introduction
		7.1.1 Related Works
	7.2 Methods and Materials
		7.2.1 Region of Interest (ROI) Extraction
		7.2.2 Database
		7.2.3 Feature Extraction
			7.2.3.1 First-Order Statistics
			7.2.3.2 Second-Order Statistics
			7.2.3.3 Texture Features
	7.3 Classification Using Machine Learning Models
		7.3.1 Principal Component Analysis (PCA)
		7.3.2 Support Vector Machine (SVM) with Grid Search
		7.3.3 Logistic Regression
		7.3.4 k-Nearest Neighbours (k-NN)
		7.3.5 Naive Bayes Classifier
	7.4 Performance Evaluation Parameters
	7.5 Classification Results and Analysis
	7.6 Conclusion and Future Work
	References
Chapter 8 Histopathological Image Analysis and Classification Techniques for Breast Cancer Detection
	8.1 Introduction
	8.2 Methodology
		8.2.1 Image Pre-Processing
		8.2.2 Image Segmentation
		8.2.3 Feature Extraction
			8.2.3.1 Mean
			8.2.3.2 Variance
			8.2.3.3 Kurtosis
			8.2.3.4 Entropy
			8.2.3.5 Contrast
			8.2.3.6 Homogeneity
			8.2.3.7 Correlation
			8.2.3.8 Energy
		8.2.4 Classification Method
			8.2.4.1 Distance-Based Classifier
			8.2.4.2 Support Vector Machine
			8.2.4.3 Convolutional Neural Network
	8.3 Image Database
		8.3.1 Breast Cancer Histopathological Database
		8.3.2 Invasive Ductal Carcinoma (IDC) Database
	8.4 Performance Evaluation of CAD System
	8.5 Results and Discussion
	8.6 Conclusion
	References
Chapter 9 Study of Emotional Intelligence and Neuro-Fuzzy System
	9.1 Emotional Intelligence
	9.2 Emotions and Cognitive Intelligent Systems (CIS)
		9.2.1 Research Challenges in Emotion Recognition
		9.2.2 Recognizing Emotions
		9.2.3 Recreating Emotions
		9.2.4 Components of Emotions
	9.3 Methods for Implementation of Emotional Intelligence
		9.3.1 Emotional Speech Recognition
		9.3.2 Facial Emotion Recognition
			9.3.2.1 Facial Expression Databases
		9.3.3 Visual Aesthetics
	9.4 Artificial Neural Network and Fuzzy Inference System
		9.4.1 Artificial Neural Network
		9.4.2 Fuzzy Inference System
	9.5 The Integrated Neuro Fuzzy Approach
		9.5.1 Adaptive Neuro Fuzzy Inference System (ANFIS)
		9.5.2 Types of ANFIS
		9.5.3 Advantages of ANFIS
			9.5.3.1 NFS through Mamdani Approach
			9.5.3.2 NFS through Takagi & Sugeno’s Approach
		9.5.4 Convergence in ANFIS
	9.6 Summary
	References
Chapter 10 Essential Statistical Tools for Analysis of Brain Computer Interface
	10.1 Introduction to Testing of Hypothesis
		10.1.1 Testing of Hypothesis
		10.1.2 Steps in the Testing of Hypothesis
		10.1.3 Parameters of Two Independent Populations
	10.2 Design of Experiment
		10.2.1 Terminologies Used in DOE
		10.2.2 Principle of Design of Experiment
	10.3 Completely Randomized Design (One Way ANOVA)
		10.3.1 One Way Classification: Completely Randomized Design
		10.3.2 Design for One Way ANOVA
		10.3.3 Advantages and Disadvantages of Completely Randomized Design
	10.4 Two-Way Classification or Randomized Block Design
		10.4.1 Design of Two Way ANOVA (Randomized Block Design without Interaction)
		10.4.2 Randomized Block Design with Interaction
	10.5 Latin Square Design
		10.5.1 Introduction
	10.6 Factorial Experiments
		10.6.1 2[sup(2)] Factorial Experiments
	References
Chapter 11 Brain Computer Interfaces: The Basics, State of the Art, and Future
	11.1 Introduction
		11.1.1 The Basic Architecture of a BCI System
		11.1.2 A General Classification of BCI Systems
	11.2 Signal Acquisition Techniques for BCI
		11.2.1 Noninvasive Methods
			11.2.1.1 Electroencephalography (EEG)
			11.2.1.2 Magnetoencephalography (MEG)
			11.2.1.3 Functional Magnetic Resonance Imaging (fMRI)
			11.2.1.4 Functional Near-Infrared Spectroscopy (fNIRS)
			11.2.1.5 Positron Emission Tomography (PET)
		11.2.2 Invasive Recording Methods
			11.2.2.1 Electrocorticography (ECoG)
			11.2.2.2 Intracortical Neuron Recording (INR)
	11.3 BCI Types and Brain Signal Patterns
		11.3.1 P300 Event-Related Potentials
			11.3.1.1 P300 BCIs Using Visual Stimulation
			11.3.1.2 P300 BCIs Using Auditory Stimulation
			11.3.1.3 P300 BCIs Using Tactile Stimulation
		11.3.2 Steady-State Evoked Potential (SSEP)
			11.3.2.1 Steady-State Visual Evoked Potential (SSVEP)
			11.3.2.2 Steady-State Auditory Evoked Potential (SSAEP)
			11.3.2.3 Steady-State Somatosensory Evoked Potential (SSSEP)
		11.3.3 Slow Cortical Potential (SCP)
		11.3.4 Sensorimotor Rhythms (SMR)
	11.4 Signal Processing
		11.4.1 Signal Preprocessing
			11.4.1.1 Artefact Handling
			11.4.1.2 Spatial and Temporal Filtering
		11.4.2 Feature Extraction and Selection
		11.4.3 Classification Algorithms
	11.5 Software Tools for BCI
	11.6 Conclusion and Future Perspectives
	Acknowledgments
	References
Chapter 12 Oriented Approaches for Brain Computing and Human Behavior Computing Using Machine Learning
	12.1 Overview of Machine Learning (Definition Approaches)
		12.1.1 Machine Learning Is Perfect to Be Used
		12.1.2 Examples of Machine Learning Applications
		12.1.3 When Do We Need Machine Learning?
		12.1.4 Types of Learning
			12.1.4.1 Learning Problems
			12.1.4.2 Hybrid Learning
			12.1.4.3 Statistical Inference
			12.1.4.4 Learning Techniques
	12.2 Machine Learning Algorithm for Brain Computing
		12.2.1 Working of Brain and Its Data Computing Process
			12.2.1.1 Aspects of Requirement in Development of BCI
		12.2.2 Machine Learning Algorithm for Brain Computing Interface
			12.2.2.1 Introduction to BCI
			12.2.2.2 BCI Functions
			12.2.2.3 Phases of BCI
			12.2.2.4 Types of BCI
		12.2.3 Why Machine Learning Algorithm Necessary in BCI
	12.3 Machine Learning Algorithm for Human Behavior Computing
		12.3.1 Introduction
		12.3.2 What Is User Behavior?
		12.3.3 Elements of User Behavior
		12.3.4 Machine Learning Algorithm for Human Behavior Computing
		12.3.5 Which Techniques Is Better for Human Behavior Computing RNN or CNN
	References
Chapter 13 An Automated Diagnosis System for Cardiac Arrhythmia Classification
	13.1 Introduction
	13.2 The Human Heart
		13.2.1 ECG Recording
	13.3 Different Segments of ECG
		13.3.1 Different Types of Noises Affecting ECG Signal
		13.3.2 Different Types of Noises Effected ECG Signal
	13.4 Database
	13.5 Proposed Methodology
	13.6 Experimental Results
		13.6.1 The Random Forest Method Performance in Classification
	13.7 Conclusion
	References
Index