• Host laboratory: LISA (Laboratoire d'Ingénierie des Systèmes Automatisés)
• Adress :
  LISA
  EA4094 - University of Angers
  62 avenue Notre Dame du Lac
  49000 Angers - France
• Web : www.istia.univ-angers.fr/LISA
• Contact

Research work

• Current research work: Processing signals and images applied to biomedical

- study of blood microcirculation: study of the space variability of the microcirculatory activity, extraction of physiological information from laser Doppler signals and laser Speckle images.
Collaboration: Vascular Functional Exploration Service of the Training Hospital of Angers

• Thesis work: Segmentation of anatomical MRI images by Bayesian multimodal inference and lesions detection

Thesis under the direction of Christophe Collet (LSIIT), Jean-Paul Armspach (LINC), started in November 2005 (BDI CNRS grant jointly financed by the Alsace region) defended the 6 November 2008.

• Résumé :

Medical imagery, constantly evolving in the past few years, provides a growing number of data. Methods to process and analyse images have recently increased to assist the doctor in qualitatively and quantitatively analysing these images and facilitating his interpretation of them. The automatic segmentation has become a fundamental stage for the quantitative analysis of these images in a number of cerebral pathologies such as multiple sclerosis (MS). In the context of this thesis we focused our study on the brain MRI segmentation and the detection of MS lesions.
To start with we suggested a brain tissue segmentation method based on the modelling by hidden Markov chains taking into account the notion of neighbourhood. This method also allows information to be included initially provided by a probabilistic atlas and takes into consideration the intensity heterogeneities present on MRI images as well as the effects of partial volume by calculating the proportions of each tissue in each voxel. We then extended this method to detecting MS lesions by using a robust estimator. Thanks to this robust estimator and the information initially provided by a probabilistic atlas, the lesions are detected like atypical data in relation to a statistical non pathological brain image model. We also developed a 3D MRI segmentation method based on the active statistical contours and on the principle of minimising the stochastic complexity, to refine the segmentation of lesions on the one hand and to segment brain anatomical structures on the other. All the algorithms were validated on synthetic and real image databases. The results obtained were compared with other existing segmentation methods, as well as with manual segmentations realised by doctors.

Keywords

Segmentation, brain imagery, Markovian statistical models, probabilistic atlas, Markov chain, robust estimator, lesion detection, statistical active grids

Collaborations and references

• Book chapters

[B1] Ch. Collet, F. Flitti, S. Bricq, A. Jalobeanu, "Fusion et imagerie multimodale", Chapitre 13 du livre "Problèmes Inverses en Imagerie et en Vision", ouvrage collectif coordonné par Ali Mohammad-Djafari, Traité Signal et Image IC2, Lavoisier, septembre 2009.
[B2] Ch. Collet, A. Jalobeanu, F. Flitti, S. Bricq, "Fusion and multimodality", Chapter of book "Inverse Problems in Vision and 3D Tomography", ouvrage collectif coordonné par Ali Mohammad-Djafari, ISTE, London; John Wiley and Sons, New York, 2010.
[B3] S. Bricq, Ch. Collet, J.-P. Armspach, "Computational surgery and dual training", Chapter "Brain MRI Segmentation", Editor, Springer, janvier 2010.

• International journals

[R1] S. Bricq, Ch. Collet, J.-P. Armspach, "Unifying framework for Multimodal Brain MRI Segmentation based on Hidden Markov Chains", Medical Image Analysis, Ed. Elsevier, vol. 12(6), pp. 639-652, Déc 2008.
[R2] G. Mahe, P. Rousseau, S. Durand, S. Bricq, G. Leftheriotis, P. Abraham, "Laser speckle contrast imaging accurately measures blood flow over moving skin surfaces", Microvascular Research, in press 2011.

• International conferences

[C1] S. Bricq, Ch. Collet, J.-P. Armspach, "Triplet Markov Chain for 3D MRI Brain Segmentation Usinga Probabilistic Atlas", IEEE 2006 International Symposium on Biomedical Imaging, April 6-9, ISBI'06, 2006.
[C2] S. Bricq, Ch. Collet, J.-P. Armspach, "3D brain MRI segmentation based on robust Hidden Markov Chain", IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP'08, March 30 - April 4, 2008, Las Vegas, Nevada, USA.
[C3] S. Bricq, Ch. Collet, J.-P. Armspach, "Lesions detection on 3D brain MRI based on Robust Hidden Markov Chain", ISMRM 2008 Annual Meeting, Toronto, Canada.
[C4] S. Bricq, Ch. Collet, J.-P. Armspach, "Lesions detection on 3D brain MRI using trimmed likelihood estimator and probabilistic atlas", IEEE 2008 International Symposium on Biomedical Imaging, ISBI'08, Mai 2008, Paris, France.
[C5] S. Bricq, Ch. Collet, J.-P. Armspach, "Markovian Segmentation of 3D Brain MRI to detect Multiple Sclerosis Lesions", IEEE 2008 International Conference on Image Processing, ICIP'08, October 12-15,2008, San Diego, California, USA.
[C6] S. Bricq, Ch. Collet, J.-P. Armspach, L Rumbach, "Application of an Advanced Hidden Markov Chain (HMC) Approach for Automatic Quantification of Lesions Detection in a multimodal magnetic resonance imaging approach in Multiple Sclerosis", American Academy of Neurology Annual Meeting, Seattle, 25 Avril - 2 mai 2009.
[C7] X. Wu, S. Bricq, Ch. Collet, "Brain MRI segmentation and lesion detection using generalized Gaussian and Rician modelling", in Part of SPIE Medical Imaging, 12-17 February 2011, Florida, USA.

• National conferences

[C8] S. Bricq, Ch. Collet, J.-P. Armspach, "Segmentation multimodale d'IRM cérébrales avec effet de volume partiel par Chaînes de Markov Cachées", Onzième congrès francophone des jeunes chercheurs en vision par ordinateur, Obernai, ORASIS'O7, France, 4-8 juin 2007.
[C9] S. Bricq, Ch. Collet, J.-P. Armspach, "Segmentation de lésions de Sclérose en Plaques en IRM cérébrale multimodale par Chaînes de Markov Cachées", 12ème congrès du GRAMM, Lyon, 26-28 Mars 2008.
[C10] E. Guerreschi, M. Collette, A. Humeau, G. Leftheriotis, S. Bricq, "Effects of glycerin
trinitrate on medium and small vessels in healthy subjects", Congrès de Physiologie, Pharmacologie et Thérapeutique, Grenoble, 22-24 Mars 2011.
[C11] E. Guerreschi, B. Haussy, M. Collette, S. Bricq, P. Chauvet, J.-P. L'Huillier, G. Leftheriotis, A. Humeau,"Mesures et analyses biomécaniques des interactions macrocirculation/microcirculation sanguines", 20ème congrès français de mécanique, Besançon, 28 août-2 Septembre 2011.

• Workshop

[C12] S. Bricq, Ch. Collet, J.-P. Armspach, "MS Lesion Segmentation based on Hidden Markov Chains", MICCAI 2008 Workshop Proceedings : 3D Segmentation in the Clinic - A Grand Challenge. , New-York, 6 septembre 2008.

• Other communications

[C13] S. Bricq, Ch. Collet, J.-P. Armspach, "Segmentation d'IRM anatomiques multimodales couplant inférence markovienne et atlas probabiliste", journée du GDR-ISIS, Paris, 23 novembre 2006.
[C14] S. Bricq, Ch. Collet, J.-P. Armspach, "Segmentation d'IRM anatomiques par inférence bayésienne multimodale et détection de lésions", journée du GDR-ISIS, Paris, 4 mars 2008.

Scientific event

Reviewer pour IEEE Transactions on Signal Processing et IEEE Transactions on Systems, Man, and Cybernetics

Teaching

• Signal and image processing
• Algorithmic, programming C

Training courses

• 2005-2008 : PhD from the University Louis Pasteur, LSIIT (Laboratoire des Sciences de l'Image, de l'Informatique et de la Télédétection), Strasbourg, http://lsiit.u-strasbg.fr
  Subject : Anatomical multimodal MRI image segmentation coupled with Bayesian inference and models initially of a high level, under the direction of Pr. Ch. Collet (LSIIT) and J.-P. Armspach (LINC)
• 2004-2005 : SIAO Research Master (Signal, Image, Acoustic and Optimisation, Institut national Polytechnique de Toulouse (INPT)
• 2002-2005 : Engineering diploma in Electronics and signal processing of the INP-ENSEEIHT (Ecole Nationale Supérieure d'Electrotechnique, d'Electronique, d'Informatique, d'Hydraulique et des Télécommunications de Toulouse)