PhD Position F/M Assessment and reduction of re-identification risk from internal brain tissues of brain MRIs

The Centre Inria de l’Université de Grenoble groups together almost 600 people in 24 research teams and 9 research support departments.

Staff is present on three campuses in Grenoble, in close collaboration with other research and higher education institutions (Université Grenoble Alpes, CNRS, CEA, INRAE, …), but also with key economic players in the area.

The Centre Inria de l’Université Grenoble Alpes is active in the fields of high-performance computing, verification and embedded systems, modeling of the environment at multiple levels, and data science and artificial intelligence. The center is a top-level scientific institute with an extensive network of international collaborations in Europe and the rest of the world.

Context. This PhD thesis is part of the MIAI Chair project RAIDAC+ (Responsible AI: Design, Regulation and Conformity).

The thesis will be conducted under a co-tutelle agreement between Inria (France) and ÉTS Montreal (Canada), leveraging the complementary strengths of both institutions.

The envisioned plan is for the recruited PhD student to spend approximately three years at Inria (Privatics team), followed by 1 year at ÉTS Montreal. This structure will provide a rich and balanced international training environment, enabling the student to benefit from diverse expertise, ecosystems, and research cultures.

The thesis will be co-supervised by:

• Antoine Boutet, Associate professor at Insa-Lyon
• Mohamed Maouche, Research at Inria
• Héber H. Arcolezi, Researcher at Inria and in-coming Assistant Professor at ÉTS Montreal (February 2026).
• Ulrich Aïvodji, Assistant Professor at ÉTS Montreal.

The PhD project will focus on the development of responsible AI methods for the protection of sensitive medical imaging data. In particular, the research will address challenges related to privacy in the processing of complex neuroimaging data (e.g., MRI), with an emphasis on anonymization techniques and regulatory alignment. The student will benefit from interdisciplinary supervision and expertise spanning privacy engineering, ethical AI, and clinical data modeling, as well as from international mobility and collaboration opportunities enabled by the co-tutelle, reinforcing the global and applied dimension of their doctoral training.

The increasing digitization of healthcare and the deployment of machine learning (ML) models in clinical settings raise urgent concerns around the protection of medical imaging data. Traditional anonymization techniques, such as defacing (removing facial features from MRI scans), have proven insufficient to prevent re-identification, especially as deep learning advances enable new biometric risks from internal brain structures.

This PhD project seeks to go beyond conventional defacing by investigating responsible AI techniques for privacy-preserving medical imaging. The central goal is to ensure that MRI data can be safely shared and used for machine learning while preserving both clinical relevance and patient privacy.

Research Objective: The thesis will aim to develop and evaluate advanced anonymization techniques tailored for neuroimaging data that are effective against re-identification risks yet compatible with diagnostic pipelines. Specifically, the project will focus on:

• Risk analysis: Characterizing re-identification threats arising from internal brain anatomy (beyond facial features), using deep embedding techniques inspired by facial recognition (e.g., DeepFace) and voice anonymization (e.g., VoicePrivacy).
     
• Region sensitivity mapping: Identifying brain areas most susceptible to re-identification, to support informed decision-making with clinicians about which regions may be altered without compromising clinical analysis.
   
• Privacy-preserving transformations: Developing both ad hoc and learning-based anonymization methods (e.g., regional blurring, morphological normalization, privacy-preserving autoencoders and GANs) to mitigate identity leakage while preserving diagnostic quality.
     
• Clinical validation: Testing the proposed methods on real-world clinical datasets in collaboration with medical practitioners, to ensure clinical relevance and integration with existing tools (e.g., automatic diagnosis, longitudinal tracking).
   
• Impact on ML performance: Evaluating how these privacy transformations affect downstream ML tasks, including the training of deep learning models for disease detection on anonymized data.

The outcomes of this PhD are expected to contribute to the broader field of responsible AI in healthcare, offering practical techniques and theoretical insights for privacy-preserving, fair, and clinically aligned machine learning on medical imaging data.

Main activities:

• Conduct original research on privacy-preserving and ethically aligned machine learning techniques for medical imaging, with a focus on MRI data anonymization and diagnostic integrity.
• Collaborate closely with interdisciplinary team members and project partners, including Inria, INSA Lyon, ÉTS Montréal, and clinical collaborators.
• Disseminate research findings through peer-reviewed publications and presentations at leading international conferences in AI, medical imaging, and data privacy.

We are looking for a candidate with:

• Good programming skills in Python and good analytical skills.
• A good background in probability/statistics and deep learning is expected.
• Knowledge on medical imaging is a plus, but not necessary.
• The candidate should be fluent in English.

• 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 and flexible organization of working hours
• Professional equipment available (videoconferencing, loan of computer equipment, etc.)
• Social, cultural and sports events and activities
• Access to vocational training
• Social security coverage

2200 euros gross salary /month