Post-Doctoral Research Visit F/M Detection of spinal cord lesions from combinations of multiple MRI sequences in living patients with MS

Post-Doctoral Research Visit F/M Detection of spinal cord lesions from combinations of multiple MRI sequences in living patients with MS Download job offer in PDF format Contract type : Fixed-term contract Level of qualifications required : PhD or equivalent Fonction : Post-Doctoral Research Visit About the research centre or Inria department The Inria Rennes - Bretagne Atlantique Centre is one of Inria's nine centres and has more than thirty research teams. The Inria Center is a major and recognized player in the field of digital sciences. It is at the heart of a rich R&D and innovation ecosystem: highly innovative PMEs, large industrial groups, competitiveness clusters, research and higher education players, laboratories of excellence, technological research institute, etc. Context The selected candidate will join the research lab Empenn in Inria-Irisa, located in Rennes, France. Empenn (https://team.inria.fr/empenn) is jointly affiliated with Inria, Inserm (National Institute of Health and Scientific Research), CNRS (INS2I institute), and the University of Rennes I. The Empenn group operates the Neurinfo imaging facility in the context of a partnership with the University Hospital of Rennes, Inria, the CNRS, and the Cancer Research Center. The team has access to several computing facilities (e.g. IGRIDA cluster) and established collaborations with other Inria/Irisa research teams in the field of machine learning. Our research lab consists of more than 20 researchers, faculty members, PhD students, engineers and interns, working in the field of image processing and neuroimaging. The team targets the detection and development of imaging biomarkers for brain diseases and focuses its efforts on translating this research to clinics and clinical neurosciences at large. The selected candidate will collaborate with the engineers, researchers and clinicians of the team involved in Multiple Sclerosis (MS) research and image processing. Assignment In recent years, the number of disease-modifying treatments for Multiple Sclerosis (MS) has augmented significantly (McGinley, Goldschmidt, and Rae-Grant 2021). In particular, highly effective second-line immunosuppressive treatments have become available and the number of first-line treatments has increased. However, these treatments are not without potential adverse effects. It is therefore crucial to prescribe the right treatment to the right patient, and to monitor its effectiveness and safety closely. Currently, Magnetic Resonance Imaging (MRI) plays a central role in this context. In particular, MRI allows: - the identification of MS lesions in particular regions of the central nervous system during the first years of the disease; - the identification of new hyperintense MS lesions between two longitudinal MRI scans i.e. at two different time points. The two above elements are central, each with their own contribution, to select a patient's initial treatment as well as to modify the treatment over time. The Empenn team is one of the leaders of the Primus project. Primus (standing for “Projection in Multiple Sclerosis” (PI: Prof Gilles Edan, Rennes University Hospital)) was granted by the French Ministry of Health in 2022. This project gathers together researchers, faculty members, clinicians and private companies, with the goal of developing a clinical decision support system for Multiple Sclerosis diagnosis and follow-up. One of our contributions is dedicated to the development of methods that allow for detection and segmentation of Multiple Sclerosis lesions from spinal cord MRI images acquired with current clinical protocols. It must be emphasized that MS lesion segmentation in spinal cord is a complex task due to some major challenges such as the size of the anatomical structures of interest (the spinal cord ~ 1cm diameter) and the occurrence of significant artifacts due to motion and respiration. Over the past years, we led several works in this area. Particularly, we developed several deep learning models for the segmentation of SC lesions either from T2 sagittal MRI acquisitions or from a pair of one T2 sagittal acquisition and one STIR sagittal acquisition, which is one of the most commonly used combination of spinal MRI sequence used in the clinical setting. Then we assessed the added value of this last model to improve the performance of radiologists (Lodé et al. European Radiology 2025). In this study, we showed that the sensitivity of radiologists was higher with the help of the automatic tool than without, without any decrease in precision. However, to date, the combinations of sequences taken into account by these models are limited and do not reflect the diversity of sequence combinations acquired in clinical practice. Indeed, in clinical practice, it is highly recommended to acquire at least two sequences among a set of available sequences, without specific guidelines to date. In practice, depending on the center and context, any combination of existing MR sequences can be provided. In particular, certain sequences that are more recent than sagittal T2 and STIR are rapidly expanding (e.g., PSIR and MP2RAGE). The development of models that can take into account these various sequences is therefore an important step, both to improve model performance and to promote their use in routine clinical practice. Our next step is to thus develop a model being able to deal with any combinations among those available. We led several preliminary works toward this objective. First, we led a first study (R. Walsh et al., MICCAI 2024) in which we proposed a strategy being able to deal with any combinations of sequences from a predefinite set. However, this method remains limited. We recently organized the MICCAI 25 “ms-multi-spine” challenge (https://portal.fli-iam.irisa.fr/MS-Multi-Spine/) dedicated to the development of methods for the detection spinal cord lesions in multiple combination of MRI sequences. This challenge allowed us to identify promising approaches to deal with this specific and still understudied setting. In particular, we proposed a method consisting in training a classification model to label a given proposal lesion as positive or negative depending on several characteristics of the individual inferences from each of the different available acquisitions for a particular patient. This "late-fusion" approach provided the best results in the challenge in several settings of interest and therefore consists of the main starting point of the work discussed in this offer. Refs: * R Walsh et al. Multi-sequence learning for multiple sclerosis lesion segmentation in spinal cord MRI. MICCAI 2024 - 27th International Conference on Medical Image Computing and Computer-Assisted Intervention, Oct 2024, Marrakech, Morocco. pp.1-10, * B Lodé et al. Evaluation of an Automatic Segmentation Model as an Aid to Detect Multiple sclerosis Spinal Cord Lesions. European radiology 2025. Main activities The postdoctoral researcher hired will be in charge of developing and evaluating new machine learning approaches in the continuation of the above mentioned work. The main steps envisaged are the following: 1. Method for detecting lesions at a given time step from multiple sequences: Case study - Design and preparation of multi-sequence training, validation, and test datasets at different sample sizes (benefiting from the important dataset we collected and annotated during the last 4 years). - Development of a "sequence agnostic model" allowing the inference of spinal cord lesions from a given acquisition from any kind of sequence. - Development of a method to fuse independent inferences from different sequences for a given session. - Improvement of the method to account for missing data, misalignment and acquisition quality. - Analysis of the added value of using all available acquisitions compared to a model using a single acquisition. 2. Method for detecting lesions at a given time step from multiple sequences: Real-world applications and evaluation of the tool's usefulness for radiological reading - Design and preparation of multi-sequence training, validation, and test datasets. - Deployment of the methodology developed above on the “optimal” dataset. - Validation by experts of the added value of the proposed model for radiological reading and availability of the tool for research projects. 3. Method for detecting new lesions appearing between two consecutive examens - Transposition of the work carried out in 2. and 3. to the case of detecting new lesions appearing from a baseline MR exam to a follow-up MR exam Skills We are seeking highly motivated candidates with a background in machine learning and medical imaging and with interest in translating research to clinical context. We require expertise in Machine Learning and Image Processing, notably Image Segmentation. Knowledge in Medical Imaging is desirable. We require good experience in programming (ideally python) and with common deep learning libraries such as PyTorch or TensorFlow. Benefits package • Subsidized meals • Partial reimbursement of public transport costs • Leave: 7 weeks of annual leave + 10 extra days off due to RTT (statutory reduction in working hours) + possibility of exceptional leave (sick children, moving home, etc.) • Possibility of teleworking (after 6 months of employment) and flexible organization of working hours • Professional equipment available (videoconferencing, loan of computer equipment, etc.) Remuneration Monthly gross salary amounting to 2788 euros Apply for this position Share • Facebook • Linkedin • Twitter • Email General Information • Theme/Domain : Computational Neuroscience and Medicine Scientific computing (BAP E) • Town/city : Rennes • Inria Center : Centre Inria de l'Université de Rennes • Starting date : 2026-10-01 • Duration of contract : 2 years • Deadline to apply : 2026-09-08 Warning : you must enter your e-mail address in order to save your application to Inria. Applications must be submitted online on the Inria website. Processing of applications sent from other channels is not guaranteed. Instruction to apply Please submit online : your resume, cover letter and letters of recommendation eventually Defence Security : This position is likely to be situated in a restricted area (ZRR), as defined in Decree No. 2011-1425 relating to the protection of national scientific and technical potential (PPST).Authorisation to enter an area is granted by the director of the unit, following a favourable Ministerial decision, as defined in the decree of 3 July 2012 relating to the PPST. An unfavourable Ministerial decision in respect of a position situated in a ZRR would result in the cancellation of the appointment. Recruitment Policy : As part of its diversity policy, all Inria positions are accessible to people with disabilities. Contacts • Inria Team : EMPENN • Recruiter : Combes Benoit / benoit.combes@inria.fr The keys to success Motivation, organization, rigor, good communication skills as well as an ambition to contribute to the improvement of clinical follow-up of patients are fundamental. Important note: The position is partially funded by the Bienvenue program. Therefore, candidates must comply with the mobility rule described here. About Inria Inria, the French national institute for research in digital science and technology, supports the French government in national research and innovation strategies in the digital field, acting as Digital Programs Agency. Inria leads over 300 research and innovation projects with its 3,500 scientists, engineers, and support staff, in partnership with universities and the digital ecosystem (businesses, entrepreneurs, and public stakeholders). Together, we explore strategic fields such as artificial intelligence, cybersecurity, quantum computing, cloud technologies, digital transformation in healthcare, digital twins, and digital technologies for defence. We develop practical solutions such as software, tech startups, partnerships with national companies, and cutting-edge training programmes. Our goal is to drive scientific, technological, and industrial excellence to ensure France’s digital sovereignty.