# Aging eyes and aging brains in studying alzheimer's disease: Modern ophthalmic data collection in the adult changes in thought (ACT) study

> **NIH NIH R01** · UNIVERSITY OF WASHINGTON · 2024 · $7,733,126

## Abstract

PROJECT SUMMARY. The eye offers a unique and relatively unexplored area to study Alzheimer’s disease
and related dementias (ADRD). The overarching goal of the Eye Adult Changes in Thought (Eye ACT) study is
to further scientific understanding of aging brains by characterizing aging eyes in exquisite detail with
prospective collection of non-invasive visual function and retinal imaging data while analyzing decades of
extant eye clinical data. Eye ACT leverages infrastructure and resources from the parent ACT study that has
enrolled and biennially followed dementia-free older adults since 1994. As of 11/2022, total enrollment through
all waves is 5,763 with ~24,000 biennial visits, 49,000 person-years of follow-up, 1,450 dementia cases, and
1,200 Alzheimer’s cases. The ACT study provides a well characterized cohort that combines research quality
evaluations of cognition, ADRD, and autopsy (>1,000 cases to date) with an unprecedented eye clinical data.
 Eye ACT’s recruitment rate has been >95% to date with >500 completed baseline visits. Under the U19
Program, the ACT study is expanding from 2000 to 3000 active participants, specifically targeting racial/ethnic
diversity. Over 59% of new enrollees since Covid are racial/ethnic minorities. Eye ACT will leverage this
opportunity to grow in its size and diversity in this cycle. In Cycle 1, we extracted clinical eye data from >4,500
participants with manual extraction of paper records through 2003 and in-house natural language processing
algorithms for extracting ~80,000 clinic visits since 2003. We also extracted >18,000 clinical retinal images. We
used these data to develop lifetime eye disease severity models we will leverage in this cycle.
 We will focus on vascular disease and visual impairment (VI) as potential mechanisms for associations
between AD and eye diseases. In Aim 1, we will evaluate longitudinal retinal imaging and visual function data
from Eye ACT participants while fully integrating home research study visits. We will determine whether
structural and vascular retinal features are associated with cognitive decline. We will test whether VI is a
causal mediator of eye disease-AD risk associations. In Aim 2, we will evaluate the severity of multiple eye
diseases as predictors of microinfarcts and ADRD neuropathology findings. We will investigate whether VI or
eye diseases modify cognitive resilience. In Aim 3, we will develop a novel data curation and sharing platform
for making extensive (“big”) ophthalmic imaging data findable, accessible, interoperable and reusable (FAIR),
propelling artificial intelligence/machine learning (AI/ML) research. We will perform a proof-of-concept
federated learning study to share AI/ML models without sharing data themselves and validate model results.
 Eye ACT represents a unique opportunity to obtain state of the art ophthalmic data collection in a
uniquely well characterized and increasingly diverse cohort of older adults followed longitudinally; consent...

## Key facts

- **NIH application ID:** 10976376
- **Project number:** 2R01AG060942-06A1
- **Recipient organization:** UNIVERSITY OF WASHINGTON
- **Principal Investigator:** Cecilia Sungmin Lee
- **Activity code:** R01 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2024
- **Award amount:** $7,733,126
- **Award type:** 2
- **Project period:** 2019-09-01 → 2025-08-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10976376

## Citation

> US National Institutes of Health, RePORTER application 10976376, Aging eyes and aging brains in studying alzheimer's disease: Modern ophthalmic data collection in the adult changes in thought (ACT) study (2R01AG060942-06A1). Retrieved via AI Analytics 2026-05-23 from https://api.ai-analytics.org/grant/nih/10976376. Licensed CC0.

---

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