Gaining insights into brain tissues using multi-compartment T2 relaxometry models

Sudhanya Chatterjee 1
1 VisAGeS - Vision, Action et Gestion d'informations en Santé
INSERM - Institut National de la Santé et de la Recherche Médicale : U1228, Inria Rennes – Bretagne Atlantique , IRISA-D5 - SIGNAUX ET IMAGES NUMÉRIQUES, ROBOTIQUE
Abstract : Magnetic resonance imaging (MRI) is one of the most widely used in-vivo imaging method for obtaining information on brain health. However, MRI voxels have limited resolution due to physical constraints. The objective of this thesis is to obtain quantitative estimates of brain tissue microstructures (such as myelin, intra/extracellular matters and free water) from T2 relaxometry MRI data. Two parametric multi-compartment T2 relaxometry (MCT2) models are proposed in this thesis. The approach and estimation framework for both models were justified using cost function simulation studies. A range of simulation and in-vivo MRI data experiments were performed to evaluate the accuracy and robustness of these models. The model found to be more robust of the two was then used for two studies on multiple sclerosis (MS) lesions. In the first study the evolution of the MCT2 biomarkers was studied in gadolinium (Gd) enhancing and non-enhancing regions of MS lesions in 10 patients with clinically isolated syndrome over a period of three years. In the second study we demonstrated the potential of combined use of the proposed MCT2 biomarkers with those obtained from existing multi-compartment diffusion MRI models to address a clinically relevant and challenging task of identifying regions of MS lesions undergoing active blood brain barrier breakdown without use of Gd injection.
Document type :
Complete list of metadatas

Cited literature [130 references]  Display  Hide  Download
Contributor : Sudhanya Chatterjee <>
Submitted on : Monday, December 10, 2018 - 2:52:30 PM
Last modification on : Friday, September 13, 2019 - 9:50:02 AM
Long-term archiving on : Monday, March 11, 2019 - 3:01:26 PM


Files produced by the author(s)


  • HAL Id : tel-01949963, version 1


Sudhanya Chatterjee. Gaining insights into brain tissues using multi-compartment T2 relaxometry models. Medical Imaging. Universite Rennes 1, 2018. English. ⟨tel-01949963⟩



Record views


Files downloads