Machine Learning in Bio-Signal Analysis and Diagnostic Imaging

Cover
Machine Learning in Bio-Signal
Analysis and Diagnostic Imaging
Copyright
Contributors
Preface
1
Ontology-Based Process for Unstructured Medical Report Mapping
	Introduction
	Related Work
	Ontology-Based Medical Report Mapping Process
		First OMRMP Phase
		Second OMRMP Phase
		Computational System
	Experimental Setup
	Results and Discussion
	Conclusion
	Relation of the Chapter With the Book
	References
2
A Computer-Aided Diagnoses System for Detecting Multiple Ocular Diseases Using Color Retinal Fundus Images
	Introduction
	Human Eye Anatomy and Diabetic Retinopathy
		Human Eye Anatomy
		DR Disease
	The Related Work
		The Supervised Methods
		Unsupervised Methods
		Semiautomated Methods
		Combining Structure and Color Features
		The Related Work Results and Discussions
	The Proposed Multilabel CAD System
		Phase 1: Color Fundus Image Acquisition
		Phase 2: Preprocessing
		Phase 3: Blood Vessels Segmentation
		Phase 4: Feature Extraction
		Phase 5: Feature Selection
		Phase 6: Classification
		Phase 7: The Evaluation
	The Experimental Results
		The Methods and Materials
		The Results
	The Discussion
		The Comparison Between the Presented Methodology and the Others in the Literature Using the Same Dataset
	Conclusion
	References
	Further Reading
3
A DEFS Based System for Differential Diagnosis Between Severe Fatty Liver and Cirrhotic Liver Using Ultrasound Images
	Introduction
	Data Set Description
		Clinically Acquired Image Database
		ROI Selection Protocol and Data Set Distribution
	Methodology Adopted for DEFS Based System for the Differential Diagnosis Between Severe Fatty Liver and Cirrhotic Liver
		Feature Extraction Module for DEFS Based System for the Differential Diagnosis Between Severe Fatty Liver and Cirrhotic Liver
		Feature Selection Module for DEFS Based System for the Differential Diagnosis Between Severe Fatty Liver and Cirrhotic Liver
		Feature Classification Module for DEFS Based System for the Differential Diagnosis Between Severe Fatty Liver and Cirrhoti ...
			Performance evaluation of the classification module
	Experiments and Results
		Experiment 1: Differential Diagnosis Between Fatty Liver and Cirrhotic Liver Without Using Feature Selection
		Experiment 2: Differential Diagnosis Between Severe Fatty Liver and Cirrhotic Liver Using kNN-DEFS
		Experiment 3: Differential Diagnosis Between Severe Fatty Liver and Cirrhotic Liver Using NB-DEFS
	Discussion
	Conclusion and Future Scope
	Acknowledgments
	References
4
Infrared Thermography and Soft Computing for Diabetic Foot Assessment
	Introduction
	Characteristics of Thermal Infrared Images
		Spatial Characteristics (Resolution)
		Noise (Thermal Resolution)
		Spectral Characteristics
		Dynamic Range
	Medical Infrared Thermography
		Early Diagnosis Using Medical Infrared Thermography
		How Is IR Thermal Imaging Different From Other Medical Imaging Modalities?
		Role of Soft Computing in Medical Infrared Thermography
	Main Focus and Motivation Behind the Chapter
	Literature Review on Diabetic Foot Complications Assessment Using MIT
		Methodology
		Study Population
		Thermal Image Acquisition and Segmentation
		Thermal Image Registration
		Extraction of Region of Interest (ROI)
		Feature Extraction and Detection of Abnormality
		Statistical Analysis
		Classification of Foot for the Assessment of Diabetic Complication Using Deep Learning Neural Network
	Challenges for Medical Infrared Thermography
		Thermal Image Acquisition
		Environmental, Individual, and Technical Challenges
		Hardware Requirements
		Specific Challenges to Thermal Imaging
	Future Roadmap for MIT and Soft Computing
		Issues to be Addressed
	Results and Discussion
		Segmentation
		Statistical Analysis of the Surface Temperature Distribution (STD) to Detect Abnormality
		Classification of Foot Using Transfer Learning of Pre-trained CNN Model
	Future Research Directions on Diabetic Foot Assessment
	Conclusion
	Acknowledgments
	References
5
Automated Classification of Hypertension and Coronary Artery Disease Patients by PNN, KNN, and SVM Classifiers Using HRV A ...
	Introduction
	Materials and Methods
		Data Collection and Processing
		HRV Analysis
		Classification Module
			Probabilistic neural network (PNN) classifier
			K nearest neighbor (KNN) classifier
			Support vector machine (SVM) classifier
	Results and Discussion
	Conclusion
	References
6
 Optimization of ROI Size for Development of Computer Assisted Framework for Breast Tissue Pattern Characterization Using ...
	Introduction
	Materials and Methods
		Description of Image Dataset
		Optimization of ROI Size for Development of Computer Assisted Framework for Breast Tissue Pattern Characterization Using D ...
			ROI extraction phase
			Feature extraction phase
				Statistical texture feature models
				Signal processing based texture feature models
					Laws mask analysis
				Transform domain texture feature models
		Classification Phase
			4-Class breast tissue pattern characterization module
			2-Class breast tissue pattern characterization module
			SVM classifier
	Experiments and Results
		Experiment 1: Experiment Carried Out for the Selection of Optimum ROI Size for the Development of Computer Assisted Framew ...
		Experiment 2: Experiment Carried Out for the Selection of Optimum ROI Size for the Development of Computer Assisted Framew ...
		Statistical Analysis
		Comparative Analysis
		Application of the Proposed Work
	Conclusion and Future Scope
		Conclusion
		Future Scope
		References
		Further Reading
7
Optimization of ANN Architecture: A Review on Nature-Inspired Techniques
	Introduction
	Artificial Neural Network
		Feedforward Neural Network
			FNNs structure
			FNNs learning scheme
			FNNs error calculation
			FNNs weight updation
		Recurrent or Feedback Neural Network
	Nature Inspired Algorithms
	Optimization of FNN
		Nonnature Inspired Algorithm
			Constructive and pruning
			Model selection
		Nature Inspired Algorithms
			SI for optimizing FNN
			Bio-inspired but not SI for optimizing FNN
			Hybrid and some other approaches for optimizing FNN
	Discussion and Conclusion
	References
8
Ensemble Learning Approach to Motor Imagery EEG Signal Classification
	Introduction
		Human Brain
		Action Potential
		Brain Rhythms
		Electroencephalography
		Motor Imagery
	Scope and Relevance
	Theoretical Background
		Preprocessing
		Feature Extraction
			Wavelet energy and entropy (EngEnt)
			Bandpower (Bp)
			Adaptive autoregressive parameters
		Classification
			Bagging ensemble learning
			Adaptive boosting ensemble learning
			Logistic boosting ensemble learning
		Background Study
	Experimental Preparation
		Dataset Description
		Experiment I (Exp-I)
		Experiment II (Exp-II)
		Experiment III (Exp-III)
		Experiment IV (Exp-IV)
	Conclusion
	References
9
Medical Images Analysis Based on Multilabel Classification
	Introduction
	Literature Review
		Algorithm Adaptation (Direct) Methods
		Problem Transformation (Indirect) Methods
		The Hybrid Between Multilabel Classification Methods
		Literature Results Analysis and Discussion
		Medical Image Analysis via Multilabel Classification
	Multilabel CAD System Framework
		Image Acquisition
		Preprocessing
		Feature Extraction
		Feature Selection
		Classification
			An overview of multilabel classification methods
		Evaluation
	Challenges of Multilabel Classification
		High Dimensionality of Data
		Label Dependency
		Label Locality
		Interlabel Similarity
		Interlabel Diversity
		The Nature of Multilabel Datasets
		Scalability
	Conclusion
	References
	Further Reading
10
Figure Retrieval From Biomedical Literature: An Overview of Techniques, Tools, and Challenges
	Introduction
	Contextualization and Chapter Organization
	Image Retrieval: Basic Concepts
		Content-Based Image Retrieval
			Workflow in CBIR
			Major challenges in CBIR
	Figure Retrieval From Biomedical Papers: Problem Setting
	Figure Retrieval From Biomedical Papers: Design Aspects
		Extraction of Figures and Figure Metadata From Research Papers
			Figure extraction from papers
			Caption extraction
			Compound figure detection and separation
			Extraction of figure text
			Extraction of mentions
		Building Figure Representation
			Visual feature extraction
			Figure classification
			Adding semantic features
		Indexing of Figures
		Query Processing
	Some Figure Search Engines in Biomedical Domain
		GoldMiner
		FigureSearch
		Yale Image Finder
		Open-i
	Future Directions
	Conclusion
	Acknowledgments
	References
11
Application of Machine Learning Algorithms for Classification and Security of Diagnostic Images
	Introduction
	Machine Learning
		Support Vector Machines
		Support Vector Regression
		Neural Networks
	Application of ML Algorithms in Medical Science
		Diagnostic Image Classification Using ML Algorithms
		Diagnostic Image Security Using Watermarking With ML Algorithms
			Watermarking techniques with NN algorithms
			Watermarking techniques using SVM algorithms
		Diagnostic Image Security Using Watermarking With Deep Learning Algorithms
	Discussion and Future Work
	Conclusion
	References
12
Robotics in Healthcare: An Internet of Medical Robotic Things (IoMRT) Perspective
	Introduction
	Overview of IoMRT
		Light Fidelity (Li-Fi) System
	Architecture IoMRT
		Sensor/Actuator Layer
		Network Layer
		IoMRT Infrastructure Layer
		Application Layer
	Li-Fi Technology Connect to IoMRT for Robotic Surgery
	IoMRT for Robotic Surgery
	Methodology and Analysis Proposed Robotic Arm for Surgery
		Hardware Description
		Software Description
	Experimental Evaluation
		Flow Diagram
		Experimental Analysis
	Limitations and Research Challenges
		Computational Problem
		Optimization
		Security Concerns of IoMRT
		Ethical Issue
	Advantage and Disadvantages of Robotic Surgery With Other Surgeries
	Applications of Robotics in Healthcare Paradigm
	Conclusions and Future Enhancement
	References
Index
	A
	B
	C
	D
	E
	F
	G
	H
	I
	K
	L
	M
	N
	O
	P
	Q
	R
	S
	T
	U
	V
	W
	Y
Back Cover