Internship: Patient Stratification in Amyotrophic Lateral Sclerosis Using Brain and Spinal MRI and Generative Adversarial Networks

 

Context

Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease characterized by progressive and diffuse degeneration of the pyramidal tract. The heterogeneity of clinical symptoms and the prevalence of atypical phenotypes impact diagnosis, prognosis, and therapeutic management. Identifying reliable biomarkers is therefore crucial to better stratify patients according to disease progression, with the aim of optimizing and adapting future therapeutic strategies [Feldman et al., 2022].

Magnetic resonance imaging (MRI) has shown considerable potential as both a diagnostic and prognostic biomarker in ALS [El Mendili et al., 2019; Kassubek et al., 2020]. Moreover, certain neural networks—particularly generative adversarial networks (GANs)—can detect atrophy patterns that remain invisible to traditional methods [Tavse et al., 2022; Yang et al., 2024].

 

Objective

The objective of this internship is to detect patterns reflecting the heterogeneity of ALS from brain and spinal MRI volumes using GAN-based models, in order to stratify patients according to their disease progression.

 

Missions

The selected student will be responsible for:

• Familiarization with the subject: Accessing preprocessed MRI datasets provided by the team, and conducting a literature review on GANs and clinical variables in ALS.
• Training/optimization of GAN models: Training the selected networks with ALS patient MRI scans to detect co-occurrences of cerebral and spinal atrophy.
• Extraction of parameters and indices: Collecting the results after the testing phase of the models.
  Statistical analysis and interpretation: Investigating associations between model-derived indices and relevant clinical variables to enable automatic patient stratification in ALS.
• Scientific report writing: Preparing a report for the Fondation Maladies Rares, the funding organization of the internship.

 

Skills

• Programming in Python (or other languages), with experience in frameworks such as PyTorch, TensorFlow (Keras), or JAX.
• Prior implementation of a GAN model is a plus.
• Advanced image processing skills (experience with MRI data would be an advantage).
• Interest in biomedical image applications and knowledge of statistics.

 

Compensation

Internship stipend (according to current regulations) + partial reimbursement of monthly public transport pass.

 

Contact
Mohamed Mounir EL MENDILI and Clara BRÉMOND
- mohamed-mounir.el-mendili@inserm.fr
- clara.bremond@inserm.fr
Please include your CV and cover letter in your email.

 

References

• Feldman EL, et al. Amyotrophic lateral sclerosis. Lancet. 2022 Oct 15;400(10360):1363-1380.
• El Mendili MM, et al. Spinal Cord Imaging in Amyotrophic Lateral Sclerosis: Historical Concepts–Novel Techniques. Front Neurol. 2019 Apr 12;10:350.
• Kassubek J, Müller HP. Advanced neuroimaging approaches in amyotrophic lateral sclerosis: refining the clinical diagnosis. Expert Rev Neurother. 2020 Mar;20(3):237-249.
• Tavse S, et al. A systematic literature review on applications of GAN-synthesized images for brain MRI. Future Internet. 2022;14(12):351.
• Yang Z, et al. Brain aging patterns in a large and diverse cohort of 49,482 individuals. Nat Med. 2024 Oct;30(10):3015-3026. doi: 10.1038/s41591-024-03144-x. Epub 2024 Aug 15.

Internship Type
Master’s level internship (M2) in Computer Science, Image Processing, Computer Vision, Applied Mathematics, Biomedical Engineering, or Computational Neuroscience.

Internship Duration
6 months, from March to September 2026.