Saturday, December 21, 2024
12/21/2024
PROJECT SUMMARY / ABSTRACT Improving vascular health is a potential strategy to delay the onset of Alzheimer's disease (AD) and related dementias (ADRD). The overlap of cerebral small vessel disease (SVD) and AD pathology may be the most common neuropathologic phenotype of age-related dementia. However, rarely do imaging studies of ADRD include measures of SVD commonly observed on brain MRI scans in older adults. Due to challenges in quantification in large cohort studies, manual reads are typically done; these are very time-consuming, prone to human error, and lack spatial anatomic resolution. This distinction is most apparent in the Clinical Core studies of NIH-funded Alzheimer's Disease Research Centers (ADRCs), where SVD biomarkers are technically challenging, time-consuming, and often overlooked. Filling this critical gap will require technologies to robustly identify SVD lesions on brain MRI. Recent technological advances make computerized SVD biomarkers possible, reproducible, and feasible for large cohorts. Using various MRI pulse sequences, we have developed novel deep-learning methods to accurately quantify SVD biomarkers, including cerebral microbleeds, white matter hyperintensities, and enlarged perivascular spaces. These reading methods are reliable and offer greater anatomic precision and dynamic range than previous scoring systems. Thus, our primary aim is to create an inter-ADRC set of objectively measured SVD MRI biomarkers from large and diverse clinical cohorts to determine the role of SVD in vascular contributions to ADRD. These ADRCs include 4,831 (49% minority) individuals at risk for ADRD. We will use the rich clinical and neuroimaging (structural MRI) data within the South Texas, Wake Forest, Wisconsin, and the University of California Davis ADRCs and their affiliated cohorts, including the Vascular Contributions to Cognitive Impairment and Dementia consortium, the Multi-Ethnic Study of Atherosclerosis, and The Wisconsin Registry for Alzheimer's Prevention to address unanswered questions related to the contribution of SVD to ADRD. We will apply our deep learning models to address the following Specific Aims: 1) Standardize quantification of cerebral SVD markers in diverse cohorts to allow objective readings from neuroimaging data currently missing in the ADRCs. 2) Relate cerebral SVD lesions with clinical cognitive staging, ADRD neuroimaging, and biofluid biomarkers in diverse ADRC cohorts. 3) Explore vascular and metabolic pathways linking SVD to worse cognition and longitudinal cognitive decline in the context of ADRD biomarkers. Further, we will share our robust machine learning models and implementation software with the scientific community. This foundational work will produce and validate computerized methods for SVD characterization and build a much-needed resource to assess vascular contributions to cognitive impairment in AD/ADRD research in under-represented populations. Furthermore, this project will provide evidence for reproducible and harmonized SVD outcomes critical to understanding the complexity of ADRD for adoption by ADRC MRI research studies, which would be scalable with future grant-supported initiatives. iii
HHS’ Tracking Accountability in Government Grants System (TAGGS) website provides access to detailed descriptions of grants, loans, aggregated direct payments and other types of financial assistance awarded by the HHS Operating Divisions (OPDIVs) and the Office of the Secretary Staff Divisions (STAFFDIVs).