TRD3: MRI parameters reflecting tissue composition and microstructure Lead Principal investigator: Peter van Zijl, Professor of Radiology Co-investigators: Xu Li, Manisha Aggarwal, Hye-Young Heo, Jeremias Sulam, Susumu Mori Consultant: Filip Szczepankiewicz (Lund University) While TRDs 1 and 2 focus on MR approaches that measure actual physiological constants and metabolite signals, the definition of a Quantitative Imaging Biomarker (QIB) goes much further. In TRD3 we therefore exploit the inherent power of MRI to probe tissue composition and microstructure, the characteristics of which can be accessed through a multitude of MRI phenomena and parameters that can be seen as candidate biomarkers. The intensity and frequency of the water signal in an MRI voxel depend on the local microscopic fields and field differences imposed by tissue compartments and molecules. In addition, the motion of water measured by MRI is affected by compartment size and permeability, which may change in disease and thus contain potential biomarker information. The overall goal of this TRD is to design pulse sequences and analysis approaches to efficiently quantify MRI parameters that assess tissue composition and microstructure. We have the following specific aims: AIM 1: Development of compartmental filtering and diffusional encoding methods to probe tissue microstructure. AIM 2: Development of integrated susceptibility and diffusion tensor imaging (STI and DTI) for fiber tractography, aiming at high resolution white matter fiber tractography in vivo. Gray matter iron content and blood oxygenation will also be assessed from these high-resolution susceptibility images AIM 3: Development of fast multi-parameter acquisition and analysis approaches for simultaneous quantification of the MR-derived parameters in Aims 1 and 2 plus T1, T2(*), and Magnetization Transfer Ratio (MTR). The parameters obtained will be used to synthetically generate multiple image contrasts (synthetic MRI), including conventional ones with which the radiologists are familiar for reading and that currently can be acquired only separately. Eight CPs will be involved in optimizing the methods and testing these approaches for biomarker potential. Eight SPs will use them to extend the information content in their studies. The developed tissue markers together with the diagnostic parameters of TRD1 and TRD2 will be made available to TRD4, which will develop statistical and deep learning technologies to combine them and make them available in age-dependent multi-parameter brain atlases.