Presentation

The objective of ICEMA (Images of Cardiac Electro-Mechanical Activity) is to build a generic dynamic model of the beating heart and a procedure to automatically identify its parameters for any specific patient. We plan to construct an identification procedure using 2 sets of relatively easy-to-access measurements on a patient: the ECG (Electrocardiogram), and a time sequence of volumetric ultrasound images (the ventricular blood pressure, when available, can also be included in our model). Once the generic model is adapted to a specific patient, it becomes possible to derive a set of quantitative and objective parameters useful in helping clinicians and physiologists to better understand the electro-mechanical coupling and diagnose pathological conditions. Significant results are expected in the following fields of cardiovascular pathology:

Our approach combines a 3D model of the electric wave propagation with a 3D biomechanical model of the cardiac muscle. The 2 models are explicitely coupled into simulations to generate the dynamic behaviour of the heart. The model for electric wave propagation is derived from Fitz-Hugh Nagumo equations, while the mechanical model is based on the classical Hill-Maxwell rheological law. These models are expected to reflect on a macroscopic scale the coupling acting on the cellular scale. To provide a realistic motion of a standard beating heart, the highly anisotropic nature of the muscle fibres in standard anatomy is taken into account.

Two error functions will be used to fit the parameters of the generic model to a specific patient. The former will compare the actual patient's ECGs with a set computed from the simulation. The latter will compare the geometry of the deformed biomechanical model with the motion extracted from the ultrasound images of the patient's heart. In addition, the ventricular blood pressure, when available, can be readily introduced as a constraint for the deformation of the biomechanical model. Ultimately, a feedback procedure will be used to update the parameters of the generic model from these error functions.


Back to home page