China 
Africa 
Explore chapters and articles related to this topic
Automated Biventricular Cardiovascular Modelling from MRI for Big Heart Data Analysis
Published in Ervin Sejdić, Tiago H. Falk, Signal Processing and Machine Learning for Biomedical Big Data, 2018
Kathleen Gilbert, Xingyu Zhang, Beau Pontré, Avan Suinesiaputra, Pau Medrano-Gracia, Alistair Young
This chapter focuses on the application of big data analysis to cardiovascular disease (CVD), which is the world’s largest cause of morbidity and mortality. This data analytics allows for a significant and controlled dimension reduction from millions of pixels down to a few hundred cardiac shape parameters for each patient. These parameters are exploited in the application of atlas-based cardiac imaging, modelling and analytics from population studies. Table 16.1 shows major cardiovascular epidemiological studies around the world that employ image acquisitions in their protocols. Cardiac magnetic resonance imaging (MRI) is the most commonly used imaging modality and is considered the gold standard to assess cardiac volume, mass and functions [14,15]. Unlike its counterparts, computed tomography and echocardiography, cardiac MRI has the ability to image the whole heart with a range of contrast mechanisms without the use of ionising radiation [16].
Cardiovascular system
Published in David A Lisle, Imaging for Students, 2012
Further imaging in the setting of CAD and ischaemic heart disease consists of tests of myocardial viability including thallium scintigraphy to assess the amount of viable versus non-viable myocardium, and echocardiography to quantitate cardiac function. Cardiac MRI may also be used to assess myocardial viability and cardiac function in patients with CAD. CMR is able to provide a quantitative assessment of ventricular wall motion and assess myocardial perfusion, at rest and with pharmacologically induced stress. CMR is generally less widely available than myocardial perfusion scintigraphy and echocardiography.
W-Net: Novel Deep Supervision for Deep Learning-based Cardiac Magnetic Resonance Imaging Segmentation
Published in IETE Journal of Research, 2022
Kamal Raj Singh, Ambalika Sharma, Girish Kumar Singh
To acquire state-of-the-art segmentation outcomes and to prevent overfitting, data augmentation is required for cardiac MRI datasets as the limitations of labelled dataset is a major concern in medical imaging. To increase the training dataset size, it would be crucial to perform augmentations more on fly (during training). Various data augmentation approaches are used spontaneously, including mirroring, gamma correction, low-resolution simulation, contrast, brightness, Gaussian blur, Gaussian noise, scaling, and rotations during training. During testing, mirroring along each axis is used for augmentation. There is postprocessing that has been performed to improve the segmentation accuracy of any structure in cardiac domain.