Giles Roditi

Susanne Schnell

Phase-contrast MRI uses bipolar gradients to encode tissue or fluid motion into the MRI signal phase. If measured in 2D with 1-directional encoding this is termed 2D Phase-Contrast MRI. Extended to a time-resolved heart-or pulse rate triggered 3D sequence with 3-directional velocity encoding, this is termed 4D flow MRI. Error sources will be described throughout the course as well as data analysis and quantification approaches. Further extensions such as dual- and multi-venc, and 5D flow MRI will be described. Finally applications in the thorax, head, and abdomen will be introduced.

Vanessa Ferreira

Graham Wright

MRI measures of signal decay without refocusing, with intermittent refocusing and with continuous refocusing reflected by time constants T2*, T2, and T1ρ can yield important clinical information about  cardiac tissue damage and response to injury. Reduced T2* reflects localized susceptibility effects as seen in hemorrhage and iron overload. T2 can depict changes in blood oxygenation reflecting ischemia and changes in water mobility reflecting inflammation. T1ρ can highlight chemical and spin exchange effects, increasing contrast between healthy and infarcted myocardium. Recently,  higher dimensional and multi-parameter imaging methods have been developed for improved differentiation of myocardial pathophysiologies.

Guang Yang

Cardiovascular Magnetic Resonance (CMR) is a safe technique, which can provide non-invasive gold-standard assessment of cardiac structure and function in patients with cardiovascular disease. While standard CMR imaging is relatively robust, some CMR techniques are inherently less reliable and image quality can be reduced. Besides, for high-resolution or 3D imaging, the acquisition duration is long and image quality may be further compromised. Recently, deep learning based methods have gained performance dividends in medical image analysis. In this talk, I will introduce the basic ideas of deep learning and its development and applications in CMR reconstruction and super-resolution towards future perspectives.

Bettina Baeßler

Avan Suinesiaputra

Deep learning has become an ubiquitous tool for image analysis, including CMR. The method is capable to learn specific human tasks from a large amount of data, provided sufficient computational power. Image segmentation and recognition are the two most used applications, but there are more creative solutions. Deep learning can learn to reconstruct MRI, leading to a faster MR acquisition. It can generate realistic contrast-enhanced MRI without using the actual contrast agent. In this course, we are going to learn how deep learning can be applied to solve CMR image analysis to derive cardiac function and anatomy of the heart.

Steffen Petersen

Research and development into artificial intelligence (AI) algorithms for the classification, analysis and prediction of disease using medical data carries with it the expectation of transformative benefits to healthcare. The increasing activity in this field is, in part, driven by the improved accessibility of machine learning frameworks and skilled practitioners but translation to clinical use is restrained by data access, regulatory, ethical and privacy concerns. This talk highlights the key milestones in this path from concept to AI prototype to clinical decision-making tool.