I am a physician/scientist focusing on the development of optical neuroimaging techniques to improve bedside detection of neurologic injury in critically-ill children. The goal of my mentored career development award is to acquire training in mouse models of hypoxia as well as in advanced statistical methods for the analysis of resting-state brain activity. This training will launch an independent research career with the aim to bring imaging biomarkers from bench to bedside. Many pediatric diseases that were once universally fatal (e.g., complex congenital heart disease and extreme prematurity) now have relatively good survival rates. However, neurodevelopmental outcomes have improved only marginally. With timely intervention, it is possible to minimize hypoxic injury, but current bedside tools are insensitive and inadequate for this purpose. Furthermore, the heterogeneity of clinical populations limits clinical studies. Each patient has a unique injury and treatment, and neuroimaging is performed at varying times after injury; thus, it can be difficult to rigorously analyze such data to determine the best, most generalizable biomarkers. The present work aims to solve these problems by using mouse models of hypoxemic neurologic injury, robust statistical methods, optical functional neuroimaging techniques, and resting-state hemodynamic assessment (e.g., functional connectivity). The optical methods are similar to functional magnetic resonance imaging but have much lower cost, higher portability, and higher through-put. My proposal will test the hypothesis that resting-state hemodynamic metrics can serve as neuroimaging biomarkers of injury after acute and chronic hypoxia.
Aim 1 will develop statistical methods adapted to optical neuroimaging to permit more robust noise filtering, brain segmentation, atlasing, and image analysis.
Aim 2 will use resting-state hemodynamics in the hyperacute phase of ischemic stroke to identify the penumbra.
Aim 3 will study the longitudinal development of functional connectivity networks across mouse development and the disruptive effects of chronic hypoxemia. This research will be conducted under the mentorship of Arjun Yodh, PhD, with co-mentorship by Daniel Licht, MD; both faculty are recognized leaders in the development of optical neuromonitoring techniques. In addition, I have assembled an interdisciplinary group of collaborators with expertise in mouse models of hypoxemia, neuroimaging statistics, and advanced network analysis methods. I will benefit from this excellent mentorship and research environment, and my unique optical neuroimaging methods offer a path to independence. I am a board-certified pediatric cardiologist, and my long-term career goals are to combine neuroimaging with a tenure-track position at a pediatric research hospital. The exceptional research environment at CHOP/Penn will enable future translational studies in the intensive care unit, as well as further animal models of disease. Page 1
Poor neurodevelopmental outcomes remain stubbornly prevalent after pediatric critical illness despite medical and surgical advances that have produced improved survival for conditions such as stroke, congenital heart disease, hypoxic-ischemic encephalopathy, and extreme prematurity. To develop effective interventions, a bedside neuroimaging system capable of timely detection of injury during longitudinal monitoring is needed, and optical neuroimaging of the brain?s resting-state holds potential to fill this role. In this application, I propose to develop statistical methods and biomarkers of neurologic injury applicable to bedside translation using mouse models of acute and chronic hypoxemia and optical intrinsic signal imaging of resting-state hemodynamics. Page 1