Image Processing for Automated Diagnosis of Cardiac Diseases

Front matter
Copyright
Contributors
Preface
Acknowledgment
Cardiac diseases and their diagnosis methods
	Introduction
	Heart valves
	Mitral valve regurgitation
	Heart diseases
		Coronary artery disease (CAD)
		Myocardial infraction (MI)
		High blood pressure or hypertension (HBP)
		Heart valve disease
		Cardiomyopathy or heart muscle disease
		Pericarditis
		Rheumatic heart disease (RHD)
	Mitral valve diseases
		Mitral regurgitation (MR)
		Causes of mitral regurgitation
		Mitral regurgitation signs and symptoms
		Mitral regurgitation diagnosis
	Cardiac disease diagnosis methods
		Principles of echo
		Modes of echocardiography
			M-mode echocardiography
			Two-dimensional echocardiography
			Doppler echocardiography
		Two-dimensional recording techniques
		Advantages and limitations of echocardiography
			Advantages
			Limitations
	Results and analysis
	Discussion
	Conclusions
	References
Cardiac multimodal image registration using machine learning techniques
	Introduction
		Image registration
		Medical image registration
		Cardiac image registration
		Classification of image registration methods
	Datasets
	Convolutional neural networks for image registration
	Cardiac image registration multimodalities
		MR-based image registration
		CT X-ray-based registration
		Ultrasound-based registration
	Evaluation of multimodal imaging
	Conclusion and discussion
	References
Anatomical photo representations for cardiac imaging training
	Clinical background and motivation
		The anatomy and function of the heart
		The cardiac cycle and electrical activation
		Cardiac magnetic resonance (CMR)
		Electrocardiogram gating
		Respiratory motion
		Cine cardiac MR imaging
		Cardiac MR imaging planes
		Indices of cardiac function
		Cardiac ultrasound imaging
	Technical challenges in multimodal cardiac image analysis and objectives
		Limited through-plane resolution and imaging artifacts in CMR
		Imaging artifacts in cardiac 3D-US images
		Identification of correspondences in multi-modal imaging data
		User interaction requirement in semi-automatic segmentation methods
		Edge-maps for multi-modal image analysis
		Automatic anatomical landmark localization in CMR images
		Cardiac MR Image super-resolution with convolutional neural networks
		Learning anatomical shape priors with convolutional neural networks
	Conclusions
	References
Cardiac function review by machine learning approaches
	Cardiac MR and ultrasound image segmentation
		Energy minimization methods
		Gaussian mixture models
		Multiatlas segmentation methods
		Anatomical priors in cardiac segmentation
	Super-resolution in magnetic resonance images
		Variational inverse methods
		Regression models for image super-resolution
	Multimodal cardiac image registration
		Transformation models and optimization techniques
		Image similarity criteria in multimodal image registration
		Evaluation of image registration algorithms
	Machine learning models in image analysis
		Ensemble of decision trees (decision forests)
		Convolutional neural networks
	Applications of ML models in medical imaging
		Decision forests in medical imaging
		Convolutional neural networks in medical imaging
		Medical image segmentation
		Image super-resolution and other applications
		Incorporating anatomical priors in neural networks
		Current limitations
	Conclusion
	References
Despeckling in echocardiographic images using a hybrid fuzzy filter
	Introduction
	Background of despeckle filtering
		Mathematical model of speckle noise for ultrasound images
		Despeckling filters
			Local adaptive filters
			Local statistics filtering
			Anisotropic diffusion filter
			Fuzzy filter (TMED, TMAV, ATMED)
			Nonlocal means filter
	Proposed hybrid fuzzy filters (HFFs)
		Ultrasound image database
		Image quality metrics (IQM) for performance evaluation
	Experimental results and discussion
	Conclusion
	Acknowledgment
	References
Impetus to machine learning in cardiac disease diagnosis
	Impetus to machine learning in cardiac disease diagnosis
	Introduction to medical imaging
	Role of computers in medical imaging
		Computer-based medical image analysis
		Computer-aided detection
		Computer-based image retrieval (CBIR)
		Radiomics and radio genomics
	Introduction to machine learning
		Categories of machine learning
			Supervised learning
			Unsupervised learning
			Semisupervised learning
			Reinforcement learning
		Machine learning algorithms
			Artificial neural networks (ANN)
			Logistic regression (LR)
			Support vector machine (SVM)
			Naive Bayes
			Random Forest (RF)
	Impact of machine learning in everyday life
		Energy
		Arts and culture
		Financial services
		Healthcare
		Machine learning in medical imaging
	Applications of machine learning in disease diagnosis
	Machine learning in cardiac disease diagnosis
	Potential challenges of using machine learning in disease diagnosis
	Constraints of using machine learning
	How to develop a machine learning model for the medical domain?
	Validation and performance assessment
	Results and discussions
	Conclusion
	References
Wavelet transform for cardiac image retrieval
	Introduction
	Discrete wavelet transform
	Orthogonal wavelet transform
		Haar wavelet transform
		Daubechies wavelet transform
	Biorthogonal wavelet transform
		Lifting scheme-based wavelet transform
	Gabor wavelet transform
	Result analysis
		Texture representation
		Similarity measurement
		Evaluation criteria
	Conclusion
	References
AI-based diagnosis techniques for cardiac disease analysis and predictions
	Introduction
	AI-based cardiac disease diagnosis techniques
		Artificial intelligence in cardiology
			AI techniques for detecting cardiovascular disease
		ANNs for predicting cardiac disease
		Cardiac disease prediction based on genetic algorithms
		Neuro-fuzzy technique for predicting cardiac disease
	Future of automated diagnosis of cardiac disease
		Clinical applications
		Challenges and recent advances in cardiac simulation
		Automated diagnosis of coronary artery disease using LDA, PCA, ICA, and DWT
	Cardiovascular disease and COVID-19
		Investigation approaches for COVID-19
			Contagious contemplation
		Cardiovascular manifestations of COVID-19
			Potential long-term consequences
			Organization implications
				Summary and future directions for cardiovascular diseases of COVID-19
	Analysis of electrocardiography
	Results and discussion
	Conclusion and future scope
	References
An improved regularization and fitting-based segmentation method for echocardiographic images
	Introduction
	Materials and method
	Theory and calculation
		Energy minimization formulation and level set method of active contour models
		IRFS method
			Model for IRFS regularization
			Model for proposed fitting function
			Energy minimization
			Implementation
	Results
	Discussions
	Conclusions
	References
Identification of heart failure from cine-MRI images using pattern-based features
	Introduction
	Pattern-based features
		Local binary pattern (LBP)
		Local ternary pattern (LTP)
		Difference of Gaussian LTP (DoGLTP)
		3D local ternary cooccurrence pattern (3DLTCoP)
	System overview
		Dataset
		Classification
		Performance measures
	Results and discussion
	Conclusions
	References
Medical image fusion methods: Review and application in cardiac diagnosis
	Introduction
		Medical image fusion process
		Classification of image fusion
			Sub-band decomposition
			Smooth partitioning
			Re-normalization
			Ridgelet analysis
	Cardiac image fusion
	Analysis of fused images
		Analysis of the quality of the fused image when reference image is available
		Analysis of the quality of the fused image when reference image is not available
	Results and analysis of fusion
		Visual analysis
		Statistical analysis
	Conclusions
	References
Index