Description
Date depot: 17 mars 2023
Titre: Robust Anomaly Detection in Multimodal Neuroimaging
Directrice de thèse:
Ninon BURGOS (ICM)
Domaine scientifique: Sciences et technologies de l'information et de la communication
Thématique CNRS : Intelligence artificielle
Resumé: Neuroimaging offers an unmatched description of the brain’s structure and physiology, which explains its crucial role in the understanding, diagnosis, and treatment of neurological disorders, such as neurodegenerative diseases. However, identifying subtle pathological changes simply by looking at images of the brain can be a difficult task. This project focuses on the individual analysis of medical images to improve differential diagnosis and prognosis, and strengthen personalised medicine.
The aim of this PhD project is to develop innovative computational imaging tools to model abnormalities, defined as deviations from normal variability, from multimodal brain imaging. To that purpose, deep generative models such as variational auto-encoders and generative adversarial networks will be used to generate pseudo-healthy images from real patients’ images for different modalities (magnetic resonance imaging, positron emission tomography). Comparing pseudo-healthy and real images will provide individual maps of abnormalities. The uncertainty of the generation process will be monitored using techniques such as ensemble modelling to provide a measure of confidence. By extracting the abnormal signal from the images, these abnormality maps will assist clinicians in their diagnosis by providing a clear representation of the pathology.
Doctorant.e: Solal Maëlys