Hongting Zhao
BME PhD Defense Presentation

Date: 2024-03-04
Time: 11am -1pm
Location / Meeting Link: HSRBI E360 https://emory.zoom.us/j/98666492370.  

Committee Members:
Dr. Erin M. Buckley (Advisor) Dr. Levi Wood Dr. Shella Keilholz Dr. Eva dyer Dr. Dobromir Rahnev


Title: Improving brain sensitivity of cerebral blood flow measurements with continuous-wave diffuse correlation spectroscopy

Abstract:
Adequate cerebral blood flow is critical for the delivery of oxygen and nutrients necessary to maintain neuronal health and function. Diffuse correlation spectroscopy (DCS) is an emerging optical modality for measuring cerebral blood flow non-invasively at the bedside. DCS measures the temporal intensity autocorrelation of multiply scattered light that has traveled from the source to the tissue surface. As the decay rate of correlation curve is related to the motion of red blood cells in the underlying tissue, DCS measures an index of blood flow. Due to the noninvasive nature of the technology, the detected signal at tissue surface contains signal contributions from both cerebral and extracerebral layers, i.e., scalp, skull, and cerebrospinal fluid. These extracerebral contributions can be particularly appreciable in adults wherein the total scalp and skull thickness can exceed 1 cm. Thus, limited brain sensitivity is one of the remaining challenges for applications of DCS to the study of the human brain. This thesis focuses on exploring advanced modeling methods that aim to improve brain sensitivity of DCS measurements by isolating the contribution of the signal that arises from the brain. Specifically, it investigated an analytical model that treats the head as a series of slabs emulating the scalp, skull, and brain layers, dubbed the three-layer model. This model has been touted as a solution to the limited brain sensitivity problem with DCS, although little work has been done to rigorously investigate the accuracy of the model and to demonstrate its utility in vivo. Thus, through a series of in silico and in-vivo experiments, I demonstrated that accurate estimation of absolute values of brain blood flow with this model is subject to numerous sources of error given its sensitivity to layer optical properties, layer thickness, head curvature, and the presence of cerebrospinal fluid. However, I show that estimations of relative changes in brain blood flow with this model are more accurate than traditional analytical approaches and less sensitive to above-mentioned confounding factors. In vivo, I demonstrate that the model is susceptible to errors as extracerebral layer thickness increases. Overall, this work demonstrates significant limitations to multi-layered modeling approaches in general that must be considered before the approach is widely used.