The “data revolution” is impacting all areas of “health and life” and offers a field of application where statistics/machine learning, image processing, control theory and data protection find their full meaning. The “health and life” interdisciplinary axis aims at promoting existing actions and encourage new interdisciplinary collaborations in the field of biology and medicine. Examples of interdisciplinary actions in line with the axis are presented below.
Beyond their research activities, member of L2S are also involved in training actions, whether for the supervision of undergraduate/graduate/doctoral students or post-doctoral researchers or for the animation of training courses for CentraleSupélec or the master’s degree in Computational Neuroscience and Neuro-engineering carried by the University of Paris Saclay. These training actions strongly rely on long-term collaborations with various partners such as Gustave Roussy, the Paris Brain and Spine Institute (ICM), General Electric Healthcare, CEA Neurospin, Pasteur Institute, Assistance Publique-Hôpitaux de Paris (APHP) or the NeuroPSI laborator, to name a few.
Image processing for structured illumination microscopes
Structured illumination microscopes overcome the limits of resolution of conventional microscopes but require new methods of image processing.
Partnership: L2S – ESPCI – Institut Pasteur
Primary contact: François Orieux
Deep brain stimulation
Deep brain stimulation is an effective symptomatic treatment for Parkinson’s disease. Tools from control theory allow the stimulation signal to be adapted in real time according to measurements of the patient’s cerebral activity.
Collaboration: L2S – APHP – NeuroPSI
Primary contact: Antoine Chaillet
Neuro-inspired image reconstruction
Knowledge of the functioning of the retina allows the development of innovative algorithms to reconstruct deteriorated images.
Partnership: L2S – JLL.
Primary contact: Dario Prandi
Phoneme encoding from EEG data
Understanding the brain mechanisms underlying infant language learning is a major challenge. Analysis of EEG data is at the heart of this field and requires the development of advanced statistical methods.
Partnership: L2S – Neurospin (Ghislaine Dehaene)
Primary contacts: Arthur Tenenhaus & Laurent Le Brusquet
Relationship between exposome and health
The exposome is defined as the set of environmental exposures received over the course of lifetime. The study of the effects of the exposome on health is a major public health issue. This field is undergoing a data revolution, making any statistical analysis difficult. L2S offers advanced statistical tools for the analysis of exposome data.
Partnership: L2S – Athlete European project (https://athleteproject.eu/)
Primary Contacts: Arthur Tenenhaus & Laurent Le Brusquet
IA for data anonymization
Data anonymization is the one fundamental prerequisite for sharing and analyzing health data. Advanced artificial intelligence techniques offer effective and innovative solutions that guarantee this anonymity.
Primary contact: Pablo Piantadina
Modelling of cerebral mechanisms during meditation
The cerebral mechanisms involved in contemplative meditation are still unknown. The use of control theory and dynamical systems tools is an original and innovative approach to try to decipher them.
Primary contact: Hugues Mounier
Optimization of the mechanical ventilation
Several members of L2S are interested to tackle questions related to mechanical ventilation: estimation of muscle pressure, study of patient/ventilator synchronization and closed-loop control of the inspiratory aid.
Partnership: L2S – APHP
Primary contact: William Pasillas-Lépine
Control of bacterial population
The control of bacterial populations has various applications. They can be used to deliver cytotoxic loads in situ, particularly in the fight against cancer. They can also be used for the generation of renewable energy. In both cases, this control requires the development of dedicated control theory tools.
Primary contact: Catherine Bonnet and Islam Boussaada
Data integration for neurodegenerative diseases
Neurodegenerative diseases (for instance: Alzheimer’s, Parkinson’s, Multiple Sclerosis, etc.) have multifactorial background. In that context, the observation of the patient in its various facets (e.g. multimodal imaging, omics data, clinical data) is crucial. Therefore, it is mandatory to develop advanced statistical methods for the joint analysis of these large and heterogeneous data sources.
Partnership: L2S – ICM – Neurospin
Primary Contacts: Arthur Tenenhaus & Laurent Le Brusquet
Chair in Artificial Intelligence and Health
The goal of the AP-HP Health Data Repository (EDS) is to integrate all the medical data collected from patients hospitalized in one of the 39 AP-HP hospitals. The main challenge related to the exploitation of such longitudinal, heterogeneous, multi-centric biomedical data is to improve the field of healthcare for more personalized medicine. In this context, it is crucial to use/develop statistical methods to address the various questions of the clinicians from this massive amount of data.
Sources localization from MEG/EEG data
The problem of sources localization from EEG and MEG is ill posed. The usual approaches suffer from important limitations, in particular their low spatial resolution, the absence of dynamic modeling, and the difficulty of merging different modalities. L2S develops sharp estimation methods for the spatio-temporal dynamics of neural networks.
Patnership: L2S – institut des neurosciences de Marseille
Primary contacts: Matthieu Kowalski & Charles Soussen