Offre en lien avec l’Action/le Réseau : – — –/– — –
Laboratoire/Entreprise : laboratoire CREATIS
Durée : 5-6 months
Contact : carole.lartizien@creatis.insa-lyon.fr
Date limite de publication : 2022-01-10
Contexte :
The vast majority of deep architectures for medical image analysis are based on supervised methods requiring the collection of large datasets of annotated examples. Building such annotated datasets is hardly achievable, especially for some specific tasks, including the detection of small and subtle lesions, which are sometimes impossible to visually detect and thus manually outline. This is the case for various brain pathologies including Parkinson’s disease.
An alternative methodological framework is that of anomaly detection in an unsupervised context (also called self-supervised). It consists in learning a model of representation of normality from the healthy data only, and then to consider as anomalies the test samples that deviate too much from normality. This last step is usually performed by calculating the error between the original and the reconstructed data from their projection in the latent representation space.
We have developed an expertise in the field of anomaly detection methods for the analysis of multi-modality brain images.
Sujet :
We have developed an expertise in the field of anomaly detection methods for the analysis of multi-modality brain images, and recently applied it to the detection of early forms of Parkinson’s disease in Parkinson’s disease in multiparametric MRI, in collaboration with the Centre de Neurosciences (GIN) and INRIA Grenoble.
The purpose of this project is to improve the performance achieved with the current model architecture by exploring methodological research axes in the domains of deep latent representation learning and visualisation (see attached pdf file for details).
The successful candidate will have access to the PPMI database (https://www.ppmi-info.org/accessdata-specimens/download-data) containing multiple images of controls and parkinsonian patients in different modalities and as well as to computing resources (CREATIS and/or CNRS supercomputer).
Profil du candidat :
Candidate should have background either in machine learning and/or deep learning or image processing and some experience in both fields as well as good programming skills.
We are looking for an enthusiastic and autonomous student with strong motivation and interest in multidisciplinary research (image processing and machine learning in a medical context). The candidate will also have the opportunity to interact with a PhD student working on this project.
See a complete description on https://www.creatis.insa-lyon.fr/site7/fr/node/47143
Formation et compétences requises :
Candidate should have a background either in machine learning and/or deep learning or image processing as well as good programming skills. Experience with deep learning libraries such (TensorFlow, Pytorch, scikit-learn) would be apreciated.
Adresse d’emploi :
Laboratoire CREATIS
INSA Lyon
21 rue Jean Capelle
69621 Villeurbanne cedex
Document attaché : 202112031401_Master_Neuro_SelfSupervised_Park_2021_22_eng.pdf