# Developing Behavioral and Neuroimaging Predictors of Stroke Recovery

> **NIH VA I01** · VA NORTHERN CALIFORNIA HEALTH CARE SYS · 2021 · —

## Abstract

Given the high personal and economic costs of stroke, significant resources have been
devoted to rehabilitation efforts within the VA and elsewhere. Despite this emphasis, all too often
recovery after stroke remains partial, and many patients do not respond to traditional therapies. The
difficulties in improving recovery after stroke stem in part from the fact that traditional approaches to
predicting the effects of stroke-related brain lesions on subsequent function are often imprecise,
particularly so for higher cognitive functions. Recent advances have demonstrated, for example, that
similar-appearing lesions may give rise to disparate phenotypes based upon the extent to which they
disrupt specific large-scale brain networks. Thus, in this proposal we will take advantage of advances
in MRI methodology and analytics, within the context of validated behavioral metrics and new
statistical techniques, to develop new predictors of functional recovery after stroke.
 Over the course of the study, patients referred from acute care hospitals to the CREC in
Martinez, California for rehabilitation after stroke will be recruited to participate within two weeks of
their index event. Those who provide informed consent will undergo a battery of tests to assess
cognitive, emotional, motor, and other neurological function. In parallel, they will undergo structural
MRI, resting state functional MRI (rs-fMRI) and diffusion tractography imaging (DTI) from which
connectivity metrics derived through graph theory and Granger causality will be determined. Both
behavioral and neuroimaging data will be obtained at three time points: within two weeks of the
sentinel event, at three months, and at twelve months. Following the acquisition of these behavioral
and imaging metrics, advanced statistical methods will be used to search for validated predictors of
cognitive, emotional, and other neurological recovery at three and twelve months after stroke.
 As such, this proposal takes advantage of (1) validated behavioral and cognitive measures; (2)
a new connectivity brain science that permits the quantification of the integrity of brain networks and
has given rise to hypotheses about their evolution after injury; (3) advanced statistical techniques; and
(4) longitudinal assessments in order to identify markers that will help to predict recovery after stroke.
This work hopefully represents the first step in a long-term program designed to address the
significant personal and economic costs of stroke in veterans and others. In addition to permitting
prospective validation of any predictors of cognitive recovery, these results may also form the basis
for the assessment of future approaches to stroke treatment, including individualized medication trials
and targeted non-invasive brain stimulation to enhance rehabilitation efforts.

## Key facts

- **NIH application ID:** 10132735
- **Project number:** 5I01RX002783-03
- **Recipient organization:** VA NORTHERN CALIFORNIA HEALTH CARE SYS
- **Principal Investigator:** ANDREW S KAYSER
- **Activity code:** I01 (R01, R21, SBIR, etc.)
- **Funding institute:** VA
- **Fiscal year:** 2021
- **Award amount:** —
- **Award type:** 5
- **Project period:** 2019-01-01 → 2023-12-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10132735

## Citation

> US National Institutes of Health, RePORTER application 10132735, Developing Behavioral and Neuroimaging Predictors of Stroke Recovery (5I01RX002783-03). Retrieved via AI Analytics 2026-05-22 from https://api.ai-analytics.org/grant/nih/10132735. Licensed CC0.

---

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