Week 7 - Jiahao

It's hard to believe that the immersion term has come to an end after such a short but intense 7 weeks in New York-Presbyterian Hospital/Weill Cornell Medical Center.

The last week most of the time I spent in the lab to refine my summer immersion project. As a more precise investigation into the contrast between central vein signal and surrounding MS lesion tissue in white matter, the objective function for choosing optimal echo time of the contrast for central vein with respect to a continuous R2* spectrum changes form into maximizing the signal difference between the central vein and MS lesion, which  is expected to give better local contrast. This optimization considering an mono-exponential signal decay for each voxel is easy to solve with an analytical close form if assuming an uniform spatial distribution of the R2* for certain vein and lesion. However, the above method still lacks of a global optimum taking account of the general distribution for a range of R2* for both vein and lesion tissue, as well as incapable of handling the significant partial volume for central vein which can be observed from the R2* map where the value at central vein is significantly lower than pure large vein in the brain such as inferior/superior sagittal sinus. Thus the post-processing is becoming much harder for separating signals from vein and white matter tissue and quantifying their contributions. The linear combination approach [1] for myelin water imaging has been inspired here to design a linear filter for suppressing certain range of T2* while reserve the signal from vein, which can also be treated as an optimization problem to determine the combination weights for different echoes.

Besides a brief summary of what I have been doing during the last week of immersion, it is also a great chance to summarize the total immersion, especially for what I have been observing and thinking during the activities I've participated for general radiology as well as neuro related fields. Talking with radiologists and attending their regular meeting bring me closely to what and how they work to make the whole healthcare system functioning. It also reminds me of how important medical imaging has been contributing into the diagnosis and treatment for almost all kinds of the health related issues, especially related to vital problems, such as brain dysfunction, cancer, emergency room. Understanding how radiologists are working, involves many different aspects, from basic physics principle for the imaging system, to disease pathology for understanding how they could be revealed from different image modalities, not to mention how radiologists coordinate with physicians for the examination and corresponding healthcare plan, as well as a highlight for protocolling which a qualified radiologist must be dealing with every day. Protocols from an engineering view such as from my perspective, are more likely approaches for handling medical related issues by engineering, which combines theory with practice, and more importantly, solves the issues. This is also how radiologists are trained such as how physicians make treatment based on the patient condition as well as pharmacological dynamics. What more exciting for me as a biomedical engineering student and future researcher, is to see how those techniques invented decades or even years ago really act in the clinical practice and contribute to healthcare, such as MRI sequence designing,  diffusion-based imaging and quantitative mappings such as QSM.


[1] Vidarsson, Logi, et al. "Echo time optimization for linear combination myelin imaging." Magnetic Resonance in Medicine: An Official Journal of the International Society for Magnetic Resonance in Medicine 53.2 (2005): 398-407.