The goal of the project is to provide clinicians a fast and reliable, velocity quantification method to explore abnormal blood flow in the most anatomically

Non-ischemic chronic cardiomyopathies (NICM) represent a complex group of cardiac diseases, accounting for 40% of heart failure patients. The challenge in classification, diagnosis, and treatment stems from their multifactorial nature and clinical variability. Major subtypes include dilated and hypertrophic cardiomyopathies (DCM and HCM), each with distinct characteristics. These diseases can progress to heart failure and carry a risk of cardiac arrhythmias, including sudden death. Currently, there are no risk stratification scores for these patients.

Pulmonary embolism (the blockage of a pulmonary artery by a blood clot) is the third cause of cardiovascular death in Europe. Upon diagnosis confirmation, clinicians evaluate the patient prognosis based on risk stratification models. The management of patient, thus its outcome, highly depend on this risk stratification. Recent studies confirmed that patient stratification based on computed tomography pulmonary angiogram (CTPA) are not well correlated to patient prognosis.

Echocardiography plays a paramount role in day to day clinical practice to highlight structural or functional dysfunction of the heart. The observed abnormalities combined with clinical data of the patient lead to diagnoses, many of which requiring the performance of complementary examinations to diagnose the origin (etiology) of cardiac pathology which will modify the prognosis of patient and the subsequent therapy.

We aim to develop a concept where physics, computer science and applied mathematics will enable faster imaging than with conventional techniques. To do so, we propose to shape light with spatial light modulators, which will allow the acquisition of multiple pixels at once and provide improved signal-to-noise ratios. Inspired by recent advances in artificial intelligence, we will develop neural networks capable of reconstructing high spectral resolution images in real time.

L’objectif de ce projet est de développer des nouvelles méthodes de reconstruction d’images médicale multimodales en utilisant des techniques d’intelligence artificielle.

This project concerns privacy-preserving machine learning approaches forpersonal heath data.

PreSPIN focuses on high-fidelity simulation tools to help physicians plantheir endovascular intervention on intracerebral strokes

Malgré le boom de l’intelligence artificielle (IA), cette dernière peine à s’imposer en imagerie médicale car elle requiert de larges bases de données souvent indisponibles.

Ce projet propose de mélanger des méthodes classiques (variationnelles) et de l’apprentissage profond (IA) en utilisant des images simulées permettant de contourner le manque crucial de données. L’application principale envisagée est la détection de vaisseaux sanguins pour une meilleure prise en charge des suites de l’AVC.

Ce projet propose de simuler des images avec un fort degré de réalisme physiologique dans le cadre de l'AVC, dans le but de créer des jeux de données suffisamment grands pour permettre aux approches d'apprentissage automatique d'être efficaces.

Pour ce faire, ce projet vise une simulation qui intègre la mécanique des fluides dans la lumière vasculaire spécifique au patient extrait à partir de données d'angiographie tridimensionnelle.