ANR-24-CE45-4399 · Panel CE45
SEIZURE
Seeing the EpileptogenIc Zone through machine Learning on strUctuRal and functional nEuroimaging data
Fusing MRI, PET and MEG with machine learning to localize the epileptogenic zone in drug-resistant epilepsy and predict surgical outcome.
Dec 2024 – Nov 2028
48 months
Coordinator Carole Lartizien
Latest updates
Latest news
We are hiring a Master intern
Spherical deep learning for epileptogenic zone localization, in WP4 at CREATIS.
New article in Computer Methods and Programs in Biomedicine
GAN based synthetic FDG PET from T1 MRI improves unsupervised anomaly detection.
SEIZURE kicks off
Funded by the ANR for 48 months, coordinated by CREATIS with CRNL, ENS de Lyon and HCL.
Publications
All
OCSVM-Guided Representation Learning for Unsupervised Anomaly Detection
Representation Learning for 3D Brain Imaging: A Benchmark
Machine Learning in Clinical Neuroimaging (MLCN 2026)
OTLesMix: Wasserstein Barycenter and Optimal Transport Map for Synthetic Lesion Generation with Diverse Shapes and Locations
Simulation and Synthesis in Medical Imaging (SASHIMI 2026)