Research
1) Main interests
My main interest is focused on developing
Bayesian,
variational
and
hybrid methods for solving
inverse problems in
application to
medical imaging and
remote
sensing.
Methods
- Bayesian inference
- Variational methods
- MCMC
- Efficient hybrid sampling
- Machine learning
- Optimization
Applications
- Medical signal and image processing
- Image reconstruction (MRI, pMRI, MRSI)
- Tumor relapse prediction (MRSI, MRI)
- Brain activity analysis (fMRI,
detection-estimation)
- EEG source localization
- Drowsiness detection (EEG, ECG)
- Change detection (multi-temporal and
hyperspectral data)
- Hyperspectral unmixing
- Anomaly detection
2) Projects
- Assessment of the
risks of disorders in physiological signals (2019-2020).
Funding
: Toulouse Tech Transfer (pré-maturation)
- Colored coded aperture design for
compressive spectral unmixing in farming product
images (2015-2016)
Funding
: ECOS Nord (C16M01)
Partners
: IRIT, Universidad Industrial de Santander
- Analysis of
multi-temporal images by joint detection-estimation:
detection of changes
in multi-temporal optical images
(2015-2016)
Funding
: CNES (# 0010105785, R & T : R-S14-OT-0004-075)
Partners
: Tésa, IRIT, CNES
Funding
: CNRS
Partners : LIGM, LJLL, IRIT, Laboratoire de Physique de
l’ENS de Lyon, INT, I2M
- OPTIMISME :
design of a new generation of parallel algorithms
exploiting recent advances in stochastic
optimization for processing large amounts of data
(2015)
Funding
: CNRS
Partners : LIGM,
LJLL, IRIT, Laboratoire de Physique de l’ENS de Lyon, INT,
I2M
- Skin Lentigo characterization through
reflectance confocal image analysis (2014-2017)
Funding
: Pierre Fabre Laboratory
Partners : CERPER, Pierre Fabre Toulouse
- Telemetry analysis for diagnosis: analysis of satellite telemetry data for
anomaly detection (2014-2015)
Funding
: CNES (# 141531, R &T14 : BS-0003-048)
Partners : Tésa, IRIT, CNES, Université de Nice Sophia
Antipolis (Lab. J.L. Lagrange )
- DynBrain :Reconstruction of Brain
Dynamic EEG images (2012 - 2013)
Funding
: STIC -AmSUd Program
Partners : IRIT-INPT, Univ. Sanata Catarina - Brazil, ITBA-
Buenos Aires
3) Students
Current
PhD. Students
- Walma Gharbi
- Siwar Chaabene
- Mohamed Fakhfakh
Former PhD.
Students
4) Collaborations
5) Thesis
Topic:
Parallel
Magnetic Resonance Imaging reconstruction problems using
wavelet representations
Supervisors:
Jean-Christophe PESQUET and Philippe CIUCIU
Team:
Signal and Communications, IGM LabInfo, UMR8049, University of
Marne la Vallee.
Research context:
To reduce scanning time or improve
spatio-temporal resolution in some MRI applications, parallel
MRI acquisition techniques with multiple coils have emerged
since the early 90's as powerful methods. In these techniques,
MRI images have to be reconstructed from acquired undersampled
"k-space" data. To this end, several reconstruction techniques
have been proposed such as the widely-used SENSitivity
Encoding (SENSE) method. However, the reconstructed images
generally present artifacts due to the noise corrupting the
observed data and coil sensitivity profile estimation errors.
In this work, we present novel SENSE-based reconstruction
methods which proceed with regularization in the complex
wavelet domain so as to promote the sparsity of the solution.
These methods achieve accurate image reconstruction under
degraded experimental conditions, in which neither the SENSE
method nor standard regularized methods (e.g. Tikhonov) give
convincing results. The proposed approaches relies on fast
parallel optimization algorithms dealing with convex but
non-differentiable criteria involving suitable sparsity
promoting priors. Moreover, in contrast with most of the
available reconstruction methods which proceed by a slice by
slice reconstruction, one of the proposed methods allows 4D
(3D + time) reconstruction exploiting spatial and temporal
correlations. The hyperparameter estimation problem inherent
to the regularization process has also been addressed from a
Bayesian viewpoint by using MCMC techniques. Experiments on
real anatomical and functional data show that the proposed
methods allow us to reduce reconstruction artifacts and
improve the statistical sensitivity/specificity in functional
MRI.