Project

Discipline

Compressed Sensing; Free-Running Imaging; 5D imaging; Ventricular Function; Cardiac Magnetic Resonance Imaging; Coronary Artery Anatomy

Lead
Magnetic resonance imaging enables a safe, non-invasive, assessment of the heart without radiation. It can quantify blood flow, function, and tissue characteristics. However, significant aspects of cardiac MRI have remained unchanged over the past decades, highly specialized personnel is needed, and acquisition is time consuming. To remove these hurdles, we propose to break away from the decades-old paradigm where data are collected in a prospectively triggered fashion, and propose instead to sample image data continuously and irrespective of the heart’s position and contractile state. This can be achieved by combining in depth knowledge of Mathematics, Engineering and Physics. This research will lead to shorter exams for the patient and to more detailed information about the heart muscle and its vessels for the clinician. This translational and interdisciplinary research is conducted in strong collaboration with medical professionals at the CHUV.
Lay summary
Magnetresonanz Tomographie ist ein nichtinvasives bildgebendes Verfahren welches keine Risiken birgt, und das ohne potenziell schädliche Röntgenstrahlung auskommt. Damit ist es möglich den Herzmuskel und dessen Gefäße abzubilden - und den Blutfluss und die Kontraktion zu messen.  Leider ist das Verfahren bis heute kompliziert, dauert lange, und speziell ausgebildetes Fachpersonal ist notwendig um diese Daten zu erfassen. Das hat damit zu tun dass die angewendeten Techniken über mehrere Jahrzehnte hinweg mehr oder weniger unverändert geblieben sind. Dank neuen Verfahren die am CHUV entwickelt wurden, und dank neueren Erkenntnissen aus der Mathematik, der Physik, den Ingenieur- und Computerwissenschaften sind wir nun in der Lage eine Lösung vorzuschlagen, welche das Verfahren massiv vereinfacht, welche zu kürzeren Messzeiten für den Patienten führt, und welche dem Arzt bessere Informationen über den Zustand des Herzmuskels und dessen Gefäßen liefern wird. Diese fächerübergreifende Forschung wird in Zusammenarbeit von Wissenschaftlern und Medizinischem Fachpersonal am CHUV durchgeführt.

Direct link to Lay Summary

Last update: 08.05.2017

Active participation

Self-organised

Communication	Title	Media	Place	Year

Title	Year

Number	Title	Start	Funding scheme

Significant aspects of cardiac MRI data acquisition strategies have largely remained unchanged over the past two decades, highly specialized personnel is needed to ensure quality, and data acquisition remains time consuming. In a deliberate attempt to remove these hurdles, we propose to break away from the decades-old paradigm where data are collected in a prospectively triggered and gated fashion and propose instead to sample image data uninterrupted, continuously, without triggering or gating, and irrespective of the heart’s position and contractile state. Based on our strong preliminary developments both on the data acquisition and reconstruction side, we posit that the combination of a free-running imaging sequence and compressed sensing reconstruction leads to 5D imaging, a new paradigm where the image that best answers to a clinical question is extracted retrospectively after the scan. This will translate to an improved ease-of-use of cardiac MRI and the opportunity to collect not only more but also more precise information about the heart per unit time. As this relates to uncharted territory to a large degree, a focused research effort from which we will learn how to best apply these new concepts, from which we will learn about the degree of success, and from which we will also learn at which point success meets failure, is now critically important. In the pursuit of that ambitious goal, a series of elaborate components need to be integrated and tested. At the CHUV, our team consisting of scientists and engineers has developed a number of such components that can be leveraged: 1.) a free-running 4D imaging sequence, 2.) a flexible compressed sensing reconstruction framework that enables respiratory motion correction without the need for navigators, breath-holding, or self-navigation, 3.) a short spectrally selective water excitation radiofrequency pulse that supports highly effective fat signal suppression without interrupting steady-state, and 4.), a framework that extracts both cardiac and respiratory motion signals directly from the continuously acquired image data. A tried and tested team of interdisciplinary researchers that includes basic scientists and medical professionals proposes here to logically and mechanistically integrate these components, to leverage new developments and concepts, to rigorously research strengths and weaknesses, to learn about optimal parameter ranges and algorithms, to translate this potentially transformative technology to the in vivo setting, and to finally study the performance in clinical patients using gold standard comparisons.