Postdoc M/F – Deep generative models for the detection of anomalies in the brain

Job title:

Postdoc M/F – Deep generative models for the detection of anomalies in the brain

Company:

Job description

Offer DescriptionYou will be in charge of developing an approach for the detection of anomalies in brain PET images based on diffusion models and integrating it into the ClinicaDL open-source software platform.Neuroimaging offers an unmatched description of the brain’s structure and physiology, which explains its crucial role in the understanding, diagnosis, and treatment of neurological disorders. However, identifying subtle pathological changes simply by looking at images of the brain can be a difficult task. For this reason, images are often transported to a standard space where they can be visually or quantitatively compared to images of normal controls. The main limitation of this approach is its lack of sensitivity due to variabilities between subjects in non-pathological areas.
Solutions have been proposed by our team to mitigate this limitation [1,2,3]. The approach consists of generating a healthy-looking image specific to the patient under investigation. When compared to the real image of the patient, the pseudo-healthy model can be used to detect the areas of the image that show abnormalities. These abnormality maps could help clinicians in their diagnosis by highlighting pathological areas in a data-driven fashion, and improve the interpretability of subsequent analyses, such as computer-aided diagnosis or spatio-temporal modelling.The first approach developed to generate the pseudo-healthy models was based on a registration and fusion algorithm [1,2]. A deep learning method based on variational autoencoders has later been proposed [3]. However, diffusion models have recently demonstrated their ability to generate high quality images to detect anomalies in medical images [4]. It is expected that these methods would allow detecting subtler anomalies, compared to the current approach. The aim of the project is to implement existing and develop new methods based on diffusion models to synthesise pseudo-healthy brain images.
The approaches developed will be applied to neuroimaging data for the computer-aided diagnosis of neurogenerative diseases, such as Alzheimer’s disease and Parkinson’s disease. Abnormality maps will be generated for multiple imaging modalities, such as magnetic resonance imaging (MRI) or positron emission tomography (PET), and applied on large datasets, obtained from both public studies and studies managed locally.
To initiate the transfer to clinical research, the methodological developments will be integrated into Clinica ( ) and ClinicaDL (clinicadl.readthedocs.io), open-source software packages designed to make clinical neuroscience studies with deep learning easier and more reproducible [5, 6].[1] Burgos, N., Cardoso, M.J., …, Hutton, B.F., and Ourselin, S.: ‘Subject-Specific Models for the Analysis of Pathological FDG PET Data’. In Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015, LNCS, 9350: 651-658, Springer, 2015. hal-01827208
[2] Burgos, N., Samper-Gonza’lez, J., …., Cardoso, M.J., and Colliot, O.: ‘Individual Analysis of Molecular Brain Imaging Data through Automatic Identification of Abnormality Patterns’. In Molecular Imaging, Reconstruction and Analysis of Moving Body Organs, and Stroke Imaging and Treatment, LNCS, 10555: 13-22, Springer, 2017. hal-01567343
[3] Hassanaly, R., Brianceau, C., Solal, M., Colliot, O., and Burgos, N.: ‘Evaluation of pseudo-healthy image reconstruction for anomaly detection with deep generative models: Application to brain FDG PET’. In Machine Learning for Biomedical Imaging, 2024, Special issue for generative models. hal-04315738
[4] Wolleb, J., Sandkühler, R., Bieder, F., Valmaggia, P., and Cattin, P.C.: Diffusion Models for Implicit Image Segmentation Ensembles. Proceedings of The 5th International Conference on Medical Imaging with Deep Learning – MIDL 2022.
[5] Routier, A., Burgos, N., …, Durrleman, S., Colliot, O.: Clinica: an open source software platform for reproducible clinical neuroscience studies. hal-02308126
[6] Thibeau-Sutre, E., Diaz, M., …, Colliot, O., and Burgos, N.: ‘ClinicaDL: an open-source deep learning software for reproducible neuroimaging processing. Computer Methods and Programs in Biomedicine, 2022, 220, 106818.

You will work within the ARAMIS lab ( ) at the Paris Brain Institute ( ), one of the world top research institutes for neurosciences. The institute is ideally located at the heart of the Pitié-Salpêtrière hospital, downtown Paris. The ARAMIS lab, which is also part of Inria (the French National Institute for Research in Digital Science and Technology), is dedicated to the development of new computational approaches for the analysis of large neuroimaging and clinical data sets.
You will interact locally with the PhD students, postdoctoral fellows and engineers of the lab, as well as our medical collaborators at the Pitié-Salpêtrière hospital. You will also interact with researchers of PR[AI]RIE, the PaRis AI Research InstitutE ( ).
The supervising team will be composed of Ninon Burgos, CNRS researcher, and Olivier Colliot, CNRS research director and co-head of the ARAMIS lab.Where to apply WebsiteRequirementsResearch Field Engineering Education Level PhD or equivalentResearch Field Computer science Education Level PhD or equivalentResearch Field Mathematics Education Level PhD or equivalentLanguages FRENCH Level BasicResearch Field Engineering Years of Research Experience 1 – 4Research Field Computer science Years of Research Experience 1 – 4Research Field Mathematics Years of Research Experience 1 – 4Additional InformationEligibility criteria– PhD degree in medical image analysis, computer science, and/or applied mathematics profile
– Knowledge of deep generative models (VAE, GAN, diffusion models)
– Programming skills in Python, knowledge of deep learning frameworks such a Pytorch or TensorFlow
– Experience in software development and use of HPC
– Good relational and communication skills to interact with professionals from various backgrounds (English required)Website for additional job detailsWork Location(s)Number of offers available 1 Company/Institute Institut du Cerveau et de la Moelle épinière Country France City PARIS 13 GeofieldContact CityPARIS 13STATUS: EXPIREDShare this page

Expected salary

Location

Paris

Job date

Tue, 16 Jul 2024 05:32:33 GMT

To help us track our recruitment effort, please indicate in your email/cover letter where (vacanciesineu.com) you saw this job posting.