top of page

MRI contrast stems from the natural variations of magnetic properties of different tissues, and their interactions with the externally applied magnetic fields. These properties can be classified into three main categories: 


(1) intrinsic magnetic properties: including proton density, spin-lattice (T1, ), spin-spin (T2, ), and transverse (T2*) relaxation times, and magnetic susceptibility (c);

(2) molecular mechanisms: magnetization transfer (MT), chemical exchange, and isotropic, anisotropic, and restricted water diffusion;

(3) hardware properties: spatial profiles of the main magnetic field (B0), and of the radiofrequency (RF) coil’s transmit and receive fields (B1+ and B1–). These sources of contrast are being routinely utilized in MRI, albeit in a predominantly qualitative manner, meaning that the contrast of interest is encoded at a fixed level to produce, for example, strongly / weakly T2 or diffusion weighted image. This qualitative approach renders the interpretation of MR images subjective, and prone to observer-dependent bias.

In quantitative MRI (qMRI), instead of weighting the image’ contrast by a certain parameter, it is the value of the parameter that is actually being mapped. The advantage of this approach is twofold: (1) it offers more objective interpretation of the data – while still retaining the ability to generate any level of contrast-weighting offline); and (2) it possesses roughly an order of magnitude higher sensitivity to changes in tissue properties, for example, a ~3% change in T2 relaxation time can be detected using qMRI, whereas a threshold of at least 10-20% change is required for visual identification. We aim to develop reliable acquisition and post-processing methods for in vivo quantification of MR parameters in a manner that is independent of scanner or scan settings. Particular focus is given to mapping single- and multi-component T2 relaxation values, T1 relaxation, and proton density.

bottom of page