[FALL 2025]
Scientific context
The Master student will join a research project gathering specialists with complementary expertise in image processing and machine learning (CREATIS) and medicine (HCL). The project is funded by the ANR research grant SEIZURE. The main objective is to deep learning models specifically designed for spherical inputs for the localization of focal epileptogenic zone. This internship project is part of a broader study to detect epileptogenic zones from heterogeneous data, including MRI, PET, magnetoencephalography (MEG) and clinical data.
Epilepsy affects 50 million people worldwide, and for 30% of them, treatment with drugs is ineffective and the only way for the patient to be seizure-free is via a surgery. In this case, it is primordial to locate the epileptogenic zone correctly, so that brain resection can put an end to the seizures without causing too much damage to the brain.
Several methods have already been proposed to automatically detect epileptogenic zone from cortical surface attributes [1], Magnetic Resonance (MR) imaging data [2] or Magnetoencephalography (MEG) [3]. Considering the characteristics of the pathology, we think that a more appropriate data representation, such as spherical models, could help the automatic detection of epileptegonic lesions. Furthermore, we know that the cause of epilepsy is not always visible on MRI (so-called MRI-negative patients) and that locating the epileptogenic focus benefits from the complementarity information of other examinations, in particular Positron Emission Tomography (PET) and MEG, pushing the need for the development of multimodal approaches for more accurate localization of the seizure onset zone.
Objectives of the internship
The purpose of this master project is to evaluate the performance of deep learning models designed to take sphere as input on the task of localization of the seizure onset zone in focal epilepsy. The intern will explore the following methodological axes :
• A first objective is to train and evaluate a Spherical U-Net on cortical surface shape attributes (curvature, sulcal depth, average convexity) of epileptic patients to assess if it performs favorably compared to standard baselines such as nnUNet trained on 3D MRI. The Spherical Transformer, originally designed for classification or regression tasks, can also be adapted to be used as a segmentation network
• A second objective is to modify the Spherical U-Net or Spherical Transformer so that it can also leverage features extracted from imaging data, such as MRI or PET, in addition to shape attributes.
• A last objective is to design a method to adapt the Spherical U-Net or the Spherical Transformer for the localization of the seizure onset zone from multimodal inputs, in particular from shape features, imaging data (MRI and/or PET) as well as interictal MEG spikes recordings and clinical data. The models will be trained on simulated MEG data generated from lesion annotations or on in-house real multimodal data.
The candidate will work closely with other students involved in the project, in particular with Robin Trombetta, a PhD student who will prepare the general pipeline analysis (data formatting, image preprocessing, etc.) before the arrival of the student and co-supervise the master project. The work carried out during this internship could lead to a publication in a national or international conference.
Skills
The candidate should have a background either in machine learning and/or deep learning or image processing, as well as good programming skills. Experience with deep learning libraries such as PyTorch would be appreciated. We are looking for an enthusiastic, autonomous and rigorous student with strong motivation and interest in multidisciplinary research (image processing and machine learning in a medical context). He/she will have access to computing resources (CREATIS and/or CNRS supercomputer) as well as two public datasets [6, 7] with respectively 170 and 542 subjects and to the private database of the SEIZURE project gathering multimodality exams of epilepsy patients. The candidate will benefit from a stimulating research environment, as he/she will have the opportunity to interact with clinicians and members of the MYRIAD team working in the field of deep machine learning for medical image analysis.
Application
Interested applicants are required to send a cover letter, CV and any other relevant documents (reference letter, recent transcripts of marks, ...) to: carole.lartizien@creatis.insa-lyon.fr and robin.trombetta@creatis.insa-lyon.fr
References
[1]Hannah Spitzer et al. “Interpretable surface-based detection of focal cortical dysplasias: a Multi-centre Epilepsy Lesion Detection study”. In: Brain 145.11 (Aug. 2022)
[2]Mathilde Ripart et al. “Multi-pathology MRI lesion segmentation in a multi-centre cohort of patients with focal epilepsy: a MELD study”. In: Medical Imaging with Deep Learning. 2024.
[3]Pauline Mouches et al. Time CNN and Graph Convolution Network for Epileptic Spike Detection in MEG Data. 2023
[4]Fenqiang Zhao et al. “Spherical U-Net on Cortical Surfaces: Methods and Applications”. In: Information Processing in Medical Imaging. Ed. by Albert C. S. Chung et al. Springer International Publishing, 2019,
[5]Sungmin Cho, Raehyuk Jung, and Junseok Kwon. Spherical Transformer. 2022. arXiv: 2202.
[6]Fabiane Schuch et al. “An open presurgery MRI dataset of people with epilepsy and focal cortical dysplasia type II”. In: Scientific Data 10.1 (July 2023)
[7]Peter N Taylor et al. “The Imaging Database for Epilepsy And Surgery (IDEAS)”. In: Epilepsia 66.2 (Dec. 2024)