Post-Doctoral Research Visit F/M Development of Deep Learning-based predictive systems for clinical and cognitive neuroscience

Download job offer in PDF format

Contract type : Fixed-term contract

Renewable contract : Yes

Level of qualifications required : PhD or equivalent

Fonction : Post-Doctoral Research Visit

About the research centre or Inria department

The Inria Saclay-Île-de-France Research Centre was established in 2008. It has developed as part of the Saclay site in partnership with Paris-Saclay University and with the Institut Polytechnique de Paris .

The centre has 40 project teams , 32 of which operate jointly with Paris-Saclay University and the Institut Polytechnique de Paris; Its activities occupy over 600 people, scientists and research and innovation support staff, including 44 different nationalities.

Context

Within the framework of a partnership (you can choose between)

  • public with French National Research Agency (ANR)
  • collaboration MIND Inria team, the Lariboisiere Hospital, and Oxford University

The recruited candidate is expected to develop a mathematical model and a software prototype for the application of contrastive learning to linking cognition and neuroimaging data.

Specifically 


Introduction

The link between the human brain function, anatomy, and cognition remains uncharted territory. Brain imaging techniques, such as functional magnetic resonance imaging (fMRI) and diffusion magnetic resonance imaging (dMRI), provide in vivo information about neural activity and structural connectivity. However, linking these imaging modalities to cognitive functions poses a significant challenge. Specifically, current efforts for linking multivariate cognition measurements with neuroimaging are based on Partial Least Squares and Canonical Correlation Analysis techniques with poor out-of-sample generalization.

Deep Learning and Contrastive Learning

Within the deep learning field, contrastive learning, an emerging deep learning paradigm, has demonstrated remarkable success in unsupervised representation learning, particularly in the realm of natural language processing and computer vision.

Contrastive Learning Regression: An Open Field

While contrastive learning has garnered significant attention for its impressive performance in classification tasks, its application to regression problems remains unsolved. This presents an opportunity to expand the capabilities of contrastive learning and harness its potential in domains that demand continuous prediction, such as neuroimaging analysis.

Linking Brain Imaging and Cognition with Contrastive Learning Regression

The project will explore developing and applying contrastive learning regression frameworks to link fMRI and dMRI data with cognitive measures.

 

Assignment

 

Assignments :
With the help of D. Wassermann the recruited person will be taken to produce a deep-learning based system for the prediction of cognitive abilities from neuroimaging data.

The primary assignemnts for this project are:

1. To evaluate current and develop novel contrastive learning regression frameworks for linking brain imaging data to cognitive measures and characterize their limitations.

2. To evaluate the performance of these frameworks on benchmark datasets, comparing their effectiveness to traditional regression approaches.

3. To investigate the interpretability of the learned representations and identify brain regions and structural connections associated with specific cognitive functions.

4. Co-coordinate PhD students and Interns with Demian Wassermann

5. Drive collaborations with the Laribosiere Hospital and Oxford University

Main activities

 

Main activities (5 maximum) :

  • Analyse the requirements of the neuroimaging and neurosurgical applications
  • Develop pipelines for the application of contrastive learning-based models to the neuroimaging and neurosurgical cases
  • Design an experimental platform for the implementation of the above pipelines
  • Validate the advancements and write scientific literature on them

 

Skills

Technical skills and level required :

  • Good master of python programming, pytorch, scikit-learn and similar systems
  • Comfortable with math formalisms and the formal background of machine learning and AI
  • Skills in development deep learning models are desirable

Languages :

  • The candidate is expected to be able to communicate profficiently in English

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 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

Remuneration

2788€ gross/month