# Imaging Predictors

> **NIH NIH U19** · UNIVERSITY OF TEXAS HLTH SCIENCE CENTER · 2024 · $520,049

## Abstract

The overall goal of this project is to discover neural signatures of COVID-19-associated cognitive impairment at
the group-contrast level using volumetric, surface-based, and tract-based metrics. The overall hypothesis is that
COVID-19-associated dementia will exhibit unique neural signatures discoverable through multi-modality
neuroimaging. As our first foray, we will employ group-wise analytic strategies measuring: 1) gray-matter
functional alterations using functional MRI (fMRI); 2) gray-matter atrophy using structural MRI (sMRI); and, 3)
white-matter abnormalities using sMRI. Preliminary data (de Erasquin et al., in review) indicate that ~50% of
post-COVID-19 enrollees over 60 years of age will be cognitively impaired, providing a balanced sample
(demented:non-demented∷1:1). It is known that changes will be chronic – lasting at least 6 months – but it is not
yet known whether cognitive impairment will be recuperative, progressive, or mixed.
Aim 1: Gray-matter Functional Signature & Connectomics. Gray-matter functional alterations will be evaluated
using voxel-based physiological (VBP) metrics computed from BOLD fMRI times series. BOLD-based VBP
metrics will be supplemented by fMRI blood flow (BF) in all cohorts and PET measures of glucose metabolic rate
(MRglu) in the Texas cohort, for cross validation. Connectomic alterations will be assessed by group independent
components analysis (GICA) and structural equation modeling (SEM) of T2* BOLD time series.
Aim 2: Gray-matter Structural Signature. Gray-matter structural alterations (atrophy and hypertrophy) will be
evaluated using both volumetric and surface-based analyses.
Aim 3: White-matter Structural Signature. White matter integrity will be evaluated using tract-based, volumetric
and lesion-counting analytics.
Aim 4 Exploratory analyses. Features discovered through group-wise contrasts (Aims 1-3) will be tested for
overlap with known patterns (e.g., AD/MCI, healthy aging, metabolic syndrome, immune mediated, etc.). They
will also be tested at the per-subject level as predictors of group membership (COVID +/-; cognitive impairment
+/-) and co-analyzed as quantitative biomarkers and endophenotypes with Projects 1 and 2.
Hypotheses 1: In the post-acute state of COVID-19 infection, persons with cognitive impairment will exhibit
abnormalities in an EON and likely other as-yet-undefined neural signatures of CNS COVID severally defined
by the above-described neuroimaging measures when contrasted either to non-impaired COVID-19 survivors or
to non-COVID controls.
Hypothesis 2: The strength of the neural signatures will be symptom-severity correlated, but not the pattern.
Hypothesis 3: The neural signatures will be time invariant, other than due to symptom-severity variation.
Hypothesis 4: The neural signatures will be cohort invariant, other than due to symptom-severity variation.

## Key facts

- **NIH application ID:** 10907439
- **Project number:** 5U19AG076581-02
- **Recipient organization:** UNIVERSITY OF TEXAS HLTH SCIENCE CENTER
- **Principal Investigator:** PETER Thornton FOX
- **Activity code:** U19 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2024
- **Award amount:** $520,049
- **Award type:** 5
- **Project period:** 2023-08-15 → 2028-07-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10907439

## Citation

> US National Institutes of Health, RePORTER application 10907439, Imaging Predictors (5U19AG076581-02). Retrieved via AI Analytics 2026-05-23 from https://api.ai-analytics.org/grant/nih/10907439. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*