Research Engineer (M/F) - Development and deployment of AI-based predictive models for neurovascular diseases.

Job Description

Missions

The BrainTwin project aims to develop prognostic models based on advanced artificial intelligence (AI) techniques that enable the fusion of heterogeneous multimodal data, such as neuroimaging and histopathology data, as well as clinical and omics data. The aim is to enable the personalization of medical treatments, particularly for complex conditions such as brain tumours or neurovascular diseases, such as stroke (cerebrovascular accident) or small vessel diseases (SVD). The development and deployment of reproducible and reliable AI models on a large scale requires the use of pre-processing tools for these large databases, to make them more reliable and comparable, particularly in the case of images, which often originate from different centres, devices and protocols: these variations can obscure useful medical information, undermine analyses and limit the generalisability of AI models.

The successful candidate's role will be structured around four main areas.

-The successful candidate will help to develop and deploy a neuroimaging data processing pipeline, incorporating not only standard pre-processing steps such as registration between imaging modalities and/or to an atlas, segmentation of brain structures, etc., but also tools for quality check (QC) (e.g. detection of outliers) and data harmonization. The work will focus primarily on the selection, integration and evaluation of existing solutions – some of which are AI-based and have been developed in-house – rather than on the development of entirely new methods. This processing workflow should make it possible to identify problematic data or processing steps, minimise variations arising from acquisition conditions, and prepare reliable datasets for subsequent studies.

-The second work area will involve integrating AI-based models for the detection and segmentation of brain lesions into this processing chain. The CREATIS laboratory and the BrainTwin project partners have developed several AI-based methods for this task; other models are also available in the literature. An initial selection based on suitability for the specific characteristics of neurovascular lesions will enable a smaller number of models to be chosen and integrated into the processing pipeline.

-Extracting lesion load as well as their characteristics will enable the estimation of imaging biomarkers, which will then be combined with other patient specific clinical data into AI-based prognostic models. The role of the recruited person for this task will be to integrate AI-based prognostic tools currently under development in the laboratory, as well as to propose areas for improvement, either within the models being developed or through original solutions derived from an analysis of the state of the art in this domain.

-Finally, the successful candidate will be responsible for deploying the modular pipeline, integrating the various components developed on a large scale, across several public reference neuroimaging databases, notably ADNI, ABIDE, IXI and BraTS, as well as across the shared databases within BrainTwin. The aim is to organise these resources into a reproducible processing pipeline that can be deployed on the Jean Zay supercomputer and shared as part of open science.

Activity

• Conduct a review of the scientific literature in order to develop expertise in the areas of machine learning relevant to this project (quality control, harmonization, detection and segmentation of brain lesions, prognostic models using the fusion of heterogeneous data).
• Compare existing methods for quality control, detection of outlier images and harmonization, taking into account their robustness, ease of use and computational cost.
• Define a data monitoring strategy: quality criteria, classification of images into usable data, data to be verified or excluded, and the production of comprehensible reports.
• Implement AI-based predictive models based on the fusion of image and non-image data for the detection of neurovascular lesions in multiparametric MRI.
• Integrate the selected solutions with existing pre-processing modules into a modular, documented and reproducible workflow.
• Establish a benchmark using multi-centre public cohorts to assess the reduction in bias between centres, the preservation of medical information and the impact on downstream prognostic analyses.
• Contribute to the academic research group's life by participating in and giving presentations at seminars, reading groups and meetings, and by attending training courses where appropriate.
• Improve code quality: shared code repository, automated testing, continuous integration, containers, documentation, deployment on Jean Zay and monitoring of the environmental footprint.
• Contribute to the dissemination of research findings through scientific publications and the provision of open-access tools.

Your Profil

Skills

• MD in Computer Science and Data Science, specializing in Neuroimaging data processing.
• Proficiency in Python, Linux, Git and collaborative development practices.
• Initial experience in managing and using research software for the analysis of medical data (including images) within a multidisciplinary research context is required.
• Previous experience in high-performance computing and containerization would be an advantage.
• Ability to analyze the literature, evaluate available tools and adapt them into solutions that can be used by a research team.
• Rigour, independence, ability to work with researchers, engineers and PhD students in a multidisciplinary environment.

An initial screening will be carried out based on the CV, experience, skills and cover letter. Shortlisted candidates will be invited to an interview. Applications must include the following documents:
• A full CV,
• A covering letter setting out the candidate's research interests in relation to the project,
• The contact details of one or two referees,
• If possible, one or two letters of recommendation,
• Transcripts from your Master's degree (M1 and M2) or equivalent, as well as any relevant certificates.

Your Work Environment

CREATIS is a multidisciplinary medical imaging research unit (approximately 200 people) in the Auvergne-Rhône-Alpes region affiliated with the University of Lyon 1, INSA Lyon, the CNRS, and Inserm. CREATIS's main areas of research are linked to two fundamental issues: 1) identifying major health challenges that imaging can address, and 2) Identifying theoretical challenges in biomedical imaging related to signal and image processing, modeling, and numerical simulation. The MYRIAD research team conducts upstream research leading to the design of advanced image processing and modeling methods. It has recognized expertise in machine learning and in prototyping diagnostic models based on multimodal medical imaging. The recruited person will participate to the national BrainTwin project (https://braintwin2026.github.io/BRAINTWIN/), which is funded under the PEPR Digital Health 2024 call for proposals.

Compensation and benefits

Compensation

starting at 3 071 gross per month, depending on experience

Annual leave and RTT

44 jours

Remote Working practice and compensation

Pratique et indemnisation du TT

Transport

Prise en charge à 75% du coût et forfait mobilité durable jusqu’à 300€

About the offer

About the CNRS

The CNRS is a major player in fundamental research on a global scale. The CNRS is the only French organization active in all scientific fields. Its unique position as a multi-specialist allows it to bring together different disciplines to address the most important challenges of the contemporary world, in connection with the actors of change.

CNRS

The research professions