Heart failure affects about 2% of the European population with an annual mortality rate of 10%. Cardiac Resynchronization Therapy (CRT) has been proven to reduce morbidity and mortality and has become a recommended treatment. Optimal lead placement for CRT not only requires exact anatomically mapped information on the mechanical activation sequence to be corrected but also tissue viability and performance maps. Unfortunately, to date, no imaging technique allows building such maps non-invasively in a single examination in CRT patients.
In this project, a new ultrasound imaging approach is proposed that will not only allow mapping all of these characteristics in a single examination but that will actually do this in a single heart beat.
Hereto, real-time segmentation of the left ventricle will be used to limit the data acquisition to the spatial regions that contain myocardium. In this way unnecessary spatial sampling can be avoided which combined with new beam forming strategies will allow imaging the entire left ventricle at frame rates above 1000Hz. This very high temporal resolution will be used to accurately measure the onset of local deformation of the left ventricle in order to construct a mechanical activation map. Subsequently, cardiac motion estimates will be used to track anatomical regions throughout the cardiac cycle in order to construct a temporally averaged backscatter intensity map. As scar tissue is known to be more reflective than normal myocardium, this should allow mapping of scar. Finally, the automatic segmentation process will allow to locally measure wall thickness and curvature from which the mechanical load distribution within the ventricle can be derived. This, combined with estimates of regional myocardial deformation will produce a map of myocardial performance.
The proposed system will thus provide important new diagnostic and therapeutic information and will therefore allow better CRT planning for the individual heart failure patient.
Fields of science
Call for proposal
See other projects for this call