Non-invasive Personalisation of Cardiac Electrophysiological Models from Surface Electrograms. (Personnalisation Non-invasive de Modèles Electrophysiologiques Cardiaques à Partir d'Electrogrammes Surfaciques)

Sophie Giffard-Roisin

Abstract: The objective of this thesis is to use non-invasive data (body surface potential mapping, BSPM) to personalise the main parameters of a cardiac electrophysiological (EP) model for predicting the response to cardiac resynchronization therapy (CRT). CRT is a clinically proven treatment option for some heart failures. However, these therapies are ineffective in 30% of the treated patients and involve significant morbidity and substantial cost. The precise understanding of the patientspecific cardiac function can help to predict the response to therapy. Until now, such methods required to measure intra-cardiac electrical potentials through an invasive endovascular procedure which can be at risk for the patient. We developed a non-invasive EP model personalisation based on a patientspecific simulated database and machine learning regressions. First, we estimated the onset activation location and a global conduction parameter. We extended this approach to multiple onsets and to ischemic patients by means of a sparse Bayesian regression. Moreover, we developed a reference ventricle-torso anatomy in order to perform an common offline regression and we predicted the response to different pacing conditions from the personalised model. In a second part, we studied the adaptation of the proposed method to the input of 12-lead electrocardiograms (ECG) and the integration in an electro-mechanical model for a clinical use. The evaluation of our work was performed on an important dataset (more than 25 patients and 150 cardiac cycles). Besides having comparable results with state-of-the-art ECG imaging methods, the predicted BSPMs show good correlation coefficients with the real BSPMs