Week 2 - Jiahao Li

This week most of the work has been done in the lab with MR modeling for venous signal as well as literature review of Multiple Sclerosis white matter lesions with central vein signs, their formation and staging with respect to disease pathophysiology.

It was really nice that Dr. Gupta referred me to the radiology residency noon meeting. Although there was no formal discussion this week, I attended one Exam Reviews sessions for the second year residents, which was very interesting for me to learn what broad knowledge a radiologist should master. During the exam review, chief resident guided the review. There were numerous questions related with not only medical imaging physics concept, major imaging devices and radiation dosimetry, but also various differential diagnosis by image analysis and image reading. Specifically for MRI, the exam had questions on slice selection, phase encoding, frequency encoding, possible artifacts analysis, clinical practical parameter selection for specific weighted image, etc., which even helped me go over nearly every aspect when I took MR physics course. The radiation physics also recalled me of those contents, such as atomic physics, shielding and protection, in my undergraduate courses.

For the rest of the time, I was doing research related work in my lab for MRI physics and application to MS. The major issue is for depicting central vein signs in multiple sclerosis lesions in white matter lesions. By multiecho gradient echo sequence in MRI, we can get signal decay behavior in central vein of MS lesion, which is attributed to integrated factors including the transverse relaxation time and susceptibility-induced local magnetic field inhomogeneity, the latter one including both intravascular deoxyhemoglobin and extravascular tissue properties as well as free water in cerebrospinal fluid. As reference in Yablonskiy's work, signal modeling for both intravascular and extravascular MR signal is derived mainly on a so called static dephasing regime assumption where diffusion effect is much smaller and thus can be neglected with only susceptibility inhomogeneity. Inspired by these works from literature review, I am currently building up a forward model mainly accounted for the vein signal in the voxel, where the long cylinder geometry is adapted for the vessel with comparable volume as a single vessel assumption. There are still some work to be done but it was really excited to communicated with people in the lab to get the holistic review of the whole modeling work as well as how the forward model can be utilized for a inverse parameter estimation for quantitative analysis of the MR images, which motivates the major lab work for biomedical application.