Cardiovascular and Coronary Artery Imaging: Volume 1

Front Cover
Cardiovascular and Coronary Artery Imaging
Copyright Page
Contents
List of contributors
1 Advanced coronary artery imaging: optical coherence tomography
	1.1 Introduction
	1.2 Basic principles of light
		1.2.1 Backscatter
		1.2.2 Attenuation
	1.3 Mechanism and technical modalities of OCT
		1.3.1 Time domain
		1.3.2 Frequency domain
		1.3.3 Spatially encoded
		1.3.4 Time encoded
	1.4 Scanning techniques
		1.4.1 Single point scanning
		1.4.2 Parallel scanning
	1.5 Pullback
	1.6 Image interpretation
		1.6.1 Basic image orientation and interpretation
		1.6.2 Image interpretation and normal coronary anatomy
		1.6.3 Coronary plaque and thrombus characterization
			1.6.3.1 Fibrous plaques
			1.6.3.2 Calcified plaques
			1.6.3.3 Lipid-laden plaques
			1.6.3.4 Red thrombus
			1.6.3.5 White thrombus
		1.6.4 Imaging coronary stents
	1.7 Image artifact
		1.7.1 Inadequate blood purging
		1.7.2 Saturation artifact
		1.7.3 Nonuniform rotational distortion
		1.7.4 Sew-up artifact (seam artifact)
		1.7.5 Fold-over artifact
		1.7.6 Bubble artifact
		1.7.7 Tangential light drop-out
		1.7.8 Merry-go-round artifact
		1.7.9 Blooming artifact
	1.8 Clinical applications
		1.8.1 Plaque analysis
		1.8.2 Diagnostic imaging: stable coronary artery disease
		1.8.3 Interventional imaging: acute coronary syndrome
		1.8.4 Postintervention imaging
	1.9 Safety and complications
	1.10 Innovations of OCT
		1.10.1 C7-XR system
		1.10.2 ILUMIEN system
		1.10.3 ILUMIEN OPTIS system
		1.10.4 OPTIS integrated system
		1.10.5 OPTIS mobile system
	1.11 Clinical trials
		1.11.1 ILUMIEN I trial
		1.11.2 ILUMIEN II Trial
		1.11.3 ILUMIEN III Trial
		1.11.4 ILUMIEN IV Trial
	References
2 Technique of cardiac magnetic resonance imaging
	2.1 Introduction
	2.2 Physical principles and pulse sequences
		2.2.1 Data acquisition
		2.2.2 Morphologic sequences
			2.2.2.1 Dark blood sequences
			2.2.2.2 Bright blood sequence with cine functional sequences
			2.2.2.3 T1 and T2 mapping
			2.2.2.4 Myocardial perfusion
			2.2.2.5 Delayed contrast-enhanced CMR and myocardial viability
			2.2.2.6 CMR angiography
			2.2.2.7 Arterial spin labeling
			2.2.2.8 Magnetic resonance spectroscopy
			2.2.2.9 Cardiac diffusion tensor imaging
		2.2.3 Future directions
			2.2.3.1 Artificial intelligence
			2.2.3.2 Structured reporting
	References
3 The role of automated 12-lead ECG interpretation in the diagnosis and risk stratification of cardiovascular disease
	3.1 Introduction
	3.2 Basic knowledge of ECG physiology
	3.3 The 12-lead ECG
	3.4 ECG signal processing
	3.5 Cardiovascular diseases diagnosed by the 12-lead ECG
		3.5.1 Rhythm disorders
			3.5.1.1 Supraventricular arrhythmias
			3.5.1.2 Ventricular arrhythmias
		3.5.2 Conduction disorders
		3.5.3 Chamber enlargement
		3.5.4 Cardiac ischemia or infarction
			3.5.4.1 Stable/unstable angina
			3.5.4.2 Myocardial infarction
	3.6 Automated ECG interpretation
	3.7 “Logic” used in automated ECG interpretation systems
	3.8 Machine learning and automated 12-lead ECG analysis
	3.9 Basic principles of risk stratification
		3.9.1 The role of ECG in risk stratification
	3.10 ECG-derived markers for risk stratification
		3.10.1 ECG risk markers based on conduction disturbances
		3.10.2 ECG risk markers based on structural changes
		3.10.3 ECG risk markers based on repolarization abnormalities
		3.10.4 ECG risk markers based on distortion in heart rhythm regulation
	3.11 Challenges and opportunities
	References
4 Extracting heterogeneous vessels in X-ray coronary angiography via machine learning
	4.1 Introduction
	4.2 Related works
	4.3 MCR-RPCA: motion coherency regularized RPCA for vessel extraction
		4.3.1 Motivation and problems
		4.3.2 Candidate contrast-filled vessel detection via statistically structured MoG-RPCA
			4.3.2.1 Estimation of candidate foreground component
			4.3.2.2 Estimation of low-rank background component
		4.3.3 Motion coherency regularized RPCA for trajectory decomposition
	4.4 SVS-net: sequential vessel segmentation via channel attention network
		4.4.1 Architecture of sequential vessel segmentation-network
			4.4.1.1 Modification of U-net
			4.4.1.2 3D spatiotemporal feature encoder
			4.4.1.3 2D and 3D residual convolutional blocks
			4.4.1.4 Channel attention mechanism
			4.4.1.5 Data augmentation
			4.4.1.6 Loss function for class imbalance problem
		4.4.2 Segmentation experimental results
			4.4.2.1 Materials
			4.4.2.2 Performance comparison
	4.5 VRBC-t-TNN: accurate heterogeneous vessel extraction via tensor completion of X-ray coronary angiography backgrounds
		4.5.1 Global intensity mapping
		4.5.2 Background completion using t-TNN
		4.5.3 Experimental results
			4.5.3.1 Synthetic X-ray coronary angiography data
			4.5.3.2 Experiment demonstration
			4.5.3.3 Performance comparison
	4.6 Conclusion
	Acknowledgments
	References
5 Assessing coronary artery disease using coronary computed tomography angiography
	5.1 Introduction
		5.1.1 The utility of CCTA in Coronary artery disease diagnosis and prognostication
	5.2 Patient selection
		5.2.1 Other utilities of computed tomography angiography, that is, other than in coronary artery disease
		5.2.2 CCTA technique and quality factors
	5.3 Spatial resolution
	5.4 Temporal resolution
	5.5 Technical issues in specific patient subgroups
		5.5.1 The future of CCTA
			5.5.1.1 Computed tomography perfusion imaging
			5.5.1.2 Viability and fibrosis
			5.5.1.3 CCTA-derived FFR (FFRCT)
	5.6 Clinical trials comparing CCTA to other modalities
	5.7 Conclusion
	References
6 Multimodality noninvasive cardiovascular imaging for the evaluation of coronary artery disease
	6.1 Introduction
	6.2 Ischemic cascade
	6.3 Exercise stress echocardiography
	6.4 Pharmacologic stress echocardiography
	6.5 Myocardial perfusion stress echocardiography
	6.6 Left ventricular strain in exercise stress echocardiography
	6.7 Limitations of stress echocardiography
	6.8 Computed tomography coronary calcium score
	6.9 Limitations of coronary artery calcium
	6.10 Computed tomography coronary angiogram
	6.11 Limitations of computed tomography coronary angiogram
	6.12 Computed tomography in combination with single-photon emission tomography
	6.13 Computed tomography in combination with positron emitting tomography
	6.14 Limitations and strengths of positron emission tomography and SPECT imaging
	6.15 CTCA and fractional flow reserve
	6.16 Limitations of FFR CCTA
		6.16.1 Cardiac magnetic resonance imaging in coronary artery disease
		6.16.2 Cardiac magnetic resonance perfusion imaging
	6.17 Cardiac magnetic resonance angiography
		6.17.1 Limitations of cardiac magnetic resonance
	6.18 Conclusion
	References
7 Magnetic resonance imaging of ischemic heart disease
	7.1 Introduction
	7.2 Cardiac MR imaging of myocardial infarction
		7.2.1 CMR of acute infarction
		7.2.2 CMR with clinical suspicion of acute coronary syndrome
		7.2.3 Visualization and characterization of jeopardized myocardium
	7.3 MR indicators of myocardial infraction severity
		7.3.1 Infarct size and extent of transmural involvement
		7.3.2 Microvascular obstruction
		7.3.3 Intramyocardial hemorrhage
		7.3.4 Myocardial infarct heterogeneity
		7.3.5 Right ventricular infarction
		7.3.6 Missed infarcts
		7.3.7 Chronic myocardial infarction
	7.4 Myocardial infarction complications
		7.4.1 Thrombus
		7.4.2 LV aneurysm
	7.5 Future directions
	References
8 CT angiography of anomalous pulmonary veins
	8.1 Introduction
	8.2 Classification
	8.3 Anomalous in caliber of pulmonary veins
		8.3.1 Stenosis of pulmonary vein
		8.3.2 Atresia of pulmonary vein
		8.3.3 Pulmonary venous varix
	8.4 Total anomalous pulmonary venous return
		8.4.1 Supracardiac type
		8.4.2 Cardiac type
		8.4.3 Infracardiac type
		8.4.4 Mixed type
	8.5 Partial anomalous pulmonary venous return
		8.5.1 Partial anomalous venous return (PAPVR)
		8.5.2 Veno-venous bridge
		8.5.3 Scimitar syndrome
		8.5.4 Pseudo-Scimitar syndrome
		8.5.5 Cortriatriatum sinister
		8.5.6 Levoatriocardinal vein
		8.5.7 PAPVR of left upper pulmonary vein (LUL PAPVR)
	8.6 Merits, limitations, and future directions
	8.7 Conclusion
	References
	Further reading
9 Machine learning to predict mortality risk in coronary artery bypass surgery
	9.1 Introduction
	9.2 Principles and applications of machine learning
		9.2.1 Data gathering
		9.2.2 Supervised learning
			9.2.2.1 Linear regression
			9.2.2.2 Logistic regression
			9.2.2.3 K-nearest neighbors
			9.2.2.4 Random forest algorithm
			9.2.2.5 Support vector machines
		9.2.3 Unsupervised learning
		9.2.4 Discussion
	9.3 Conclusion
	References
10 Computed tomography angiography of congenital anomalies of pulmonary artery
	10.1 Introduction
	10.2 Classification
		10.2.1 Anomalies of caliber
			10.2.1.1 Congenital pulmonary artery stenosis
			10.2.1.2 Congenital pulmonary artery dilatation
		10.2.2 Anomalies origin or course of central branch of pulmonary artery
			10.2.2.1 Crossed pulmonary arteries
			10.2.2.2 Pulmonary artery sling
		10.2.3 Anomalous origin/development of main pulmonary artery (conotruncal anomalies)
			10.2.3.1 Tetralogy of Fallot
			10.2.3.2 Pulmonary atresia with ventricular septal defect
			10.2.3.3 Truncus arteriosus
			10.2.3.4 Double outlet right ventricle
	10.3 Merits, limitations, and future directions
	10.4 Conclusion
	References
11 Obstructive coronary artery disease diagnostics: machine learning approach for an effective preselection of patients
	11.1 Introduction
	11.2 In search for additional diagnostic information
		11.2.1 Various methods of calcium quantification
		11.2.2 Extracoronary atherosclerosis assessment
		11.2.3 The development of CAD in coronary arteries is not uniform
	11.3 Materials and methods
		11.3.1 Supervised machine learning
		11.3.2 CCTA examination as a reference
		11.3.3 Extended CACS evaluation
		11.3.4 Classifier and optimization methods
		11.3.5 Study population
		11.3.6 Acquisition and diagnostic evaluation of CCTA scans
	11.4 Results
		11.4.1 Tools used
		11.4.2 Study population characteristics
		11.4.3 Calcific burden
		11.4.4 Model development
			11.4.4.1 Number of base models
			11.4.4.2 Optimization of hyperparameters
			11.4.4.3 Classifier training and validation
	11.5 Conclusions
		11.5.1 Heterogeneity of coronary arteries atherosclerotic plaque burden
		11.5.2 Machine learning model validation
		11.5.3 Effectiveness of developed tool
	References
12 Heart disease prediction using convolutional neural network
	12.1 Introduction
		12.1.1 Causes
		12.1.2 Overload of the cardiac scheme
		12.1.3 Coronary artery disease
		12.1.4 Heart attack
		12.1.5 Cardiomyopathy
		12.1.6 Treatment
		12.1.7 Stage A
		12.1.8 Stage B
		12.1.9 Stage C
		12.1.10 Stage D
		12.1.11 Different imaging test
		12.1.12 Echocardiography
		12.1.13 Chest X-ray
		12.1.14 Computed tomography
		12.1.15 Magnetic resonance imaging
		12.1.16 Benefits of magnetic resonance imaging
			12.1.16.1 Risks associated with the use of magnetic resonance imaging
		12.1.17 Limitations for cardiac magnetic resonance imaging
		12.1.18 Heart disease classification using convolutional neural network
	12.2 Materials
	12.3 Methods
		12.3.1 Data collection
		12.3.2 Direct DICOM images
		12.3.3 Merge DICOM images
		12.3.4 Preprocessing of the images form the data set
		12.3.5 Data fusion
		12.3.6 Data normalization and randomization
		12.3.7 Model generation
		12.3.8 Convolutional neural network
		12.3.9 The perceptron
		12.3.10 Neural network
		12.3.11 Convolutional neural network
		12.3.12 Results of heart disease prediction using convolutional neural network
	12.4 Conclusion/summary
	Acknowledgments
	Author contribution
	Conflict of interest
	References
13 Gene polymorphism and the risk of coronary artery disease
	13.1 Introduction
		13.1.1 Symptoms of coronary artery disease
		13.1.2 Risk factors associated with coronary artery disease
			13.1.2.1 Age and gender
			13.1.2.2 Diet factors
			13.1.2.3 Lifestyle factors
		13.1.3 Coronary artery disease detection and diagnostics
			13.1.3.1 Electrocardiogram
			13.1.3.2 Echocardiogram
			13.1.3.3 Exercise stress test
			13.1.3.4 Nuclear stress test
			13.1.3.5 Cardiac catheterization and angiogram
		13.1.4 Prevention
			13.1.4.1 Modification of behavior
			13.1.4.2 Physical activity
			13.1.4.3 Diet
			13.1.4.4 Alcohol and tobacco
			13.1.4.5 Blood pressure lowering
			13.1.4.6 Lipid lowering
				Drugs for clot prevention
			13.1.4.7 Antihypertensive and stain therapy
		13.1.5 Genetic factor
			13.1.5.1 Angiotensin converting enzyme gene
			13.1.5.2 IL-10 gene polymorphism
			13.1.5.3 Angiotensinogen gene
				Aim
	13.2 Methodology
		13.2.1 Literature search
		13.2.2 Selection criteria
		13.2.3 Extraction of data
		13.2.4 Statistical analysis
	13.3 Results
		13.3.1 Literature search
		13.3.2 Quantitative data analysis
		13.3.3 Publication bias
	13.4 Discussion
	13.5 Conclusion
	References
14 Role of optical coherence tomography in borderline coronary lesions
	14.1 Introduction
	14.2 Physics of optical coherence tomography
	14.3 Imaging technique
	14.4 Optical coherence tomography image
	14.5 Optical coherence tomography versus intravascular ultrasound
	14.6 Optical coherence tomography in borderline lesions
		14.6.1 ACS with unclear culprit
		14.6.2 Functional significance of stenosis
		14.6.3 Vulnerable plaque
		14.6.4 MI with no obstructive coronary atherosclerosis
		14.6.5 Spontaneous coronary artery dissection
		14.6.6 Transplant vasculopathy
		14.6.7 Clinical evidence of optical coherence tomography
	14.7 Conclusion
	References
Index
Back Cover