The goal of the K23 candidate is to establish a high-quality independent research program on neuroimaging, particularly on improving network analytical techniques of molecular imaging and to study neurodegenerative disorders. Dementias, including Alzheimer's disease (AD), are network degenerative disorders. For instance, it is well known that AD pathology spreads along distributed brain systems within neuronal networks. However, at present there is no direct in vivo method to visualize large-scale network degeneration and propagation in individuals suffering from dementia. Hence there is a critical need for developing novel and comprehensive analytical methods that accurately detect brain pathological networks (BPNs). In this proposal, we aim to develop graph theoretical tools to visualize and study BPNs in preclinical - that is, normal older individuals at risk for AD by virtue of advanced age- and clinical AD subjects using positron emission tomography (PET). Particularly, we will construct group-level brain graphs using multi- regression models and PET images from two key AD pathological radioligands: 1) 11C-labeled Pittsburgh Compound-B (11C-PIB) that binds fibrillar amyloid-beta (A?) plaques;and 2) F18-T807 that binds TAU neurofibrillary tangles (Aim 1). Then, we will develop graph theory metrics to estimate network propagation patterns in group-level A? and TAU BPNs (Aim 2). Finally, we will investigate individual mapping of BPNs and its relationship with individual risk for AD, functional connectivity breakdown, and neuropsychological profiles (Aim 3). We believe this research can dramatically improve our detection and visualization capabilities of in vivo network degeneration in AD and dementia. Neuroimaging methods designed to characterize BPNs -as a proxy of network degeneration- promise to improve diagnosis and monitoring of therapeutic response in dementia, which might in turn positively impact the U.S. health care system burdened by these devastating disorders. Moreover, completion of the proposed award will provide the foundation for the candidate's development as an independent investigator, complementing the candidate's prior skills with rigorous training in molecular PET imaging. The candidate will take advantage of the cutting-edge facilities, as well as the world- class educational opportunities at its collaborating institutions. During the K23 award, the candidate will have the valuable support and mentorship of Prof. Keith A. Johnson and Prof. Thomas Brady. Relevance: This proposal has the potential to improve public health through advancement of the PET capabilities to early diagnose and monitor dementia.

Public Health Relevance

Currently, there is no direct method to map in vivo network degeneration in individuals suffering from pre-clinical or clinical dementia. In this project, we propose to develop novel and comprehensive imaging methods to accurately detect pathological neurodegeneration at the brain network level. We believe this research can dramatically improve diagnosis and monitoring of therapeutic response in preclinical and clinical dementia, which might in turn positively impact the U.S. health care system burdened by these devastating disorders.

National Institute of Health (NIH)
National Institute of Biomedical Imaging and Bioengineering (NIBIB)
Mentored Patient-Oriented Research Career Development Award (K23)
Project #
Application #
Study Section
Special Emphasis Panel (ZEB1-OSR-F (M2))
Program Officer
Erim, Zeynep
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Massachusetts General Hospital
United States
Zip Code
Hu, Chenhui; Sepulcre, Jorge; Johnson, Keith A et al. (2016) Matched signal detection on graphs: Theory and application to brain imaging data classification. Neuroimage 125:587-600
Sepulcre, Jorge; Schultz, Aaron P; Sabuncu, Mert et al. (2016) In Vivo Tau, Amyloid, and Gray Matter Profiles in the Aging Brain. J Neurosci 36:7364-74
Hu, Chenhui; Cheng, Lin; Sepulcre, Jorge et al. (2015) A spectral graph regression model for learning brain connectivity of Alzheimer's disease. PLoS One 10:e0128136