# Development of an AI/ML-ready knee ultrasound dataset in a population-based cohort

> **NIH NIH R01** · UNIV OF NORTH CAROLINA CHAPEL HILL · 2022 · $304,487

## Abstract

SUMMARY
Our long-term goal is to demonstrate the utility of ultrasound for OA (OsteoArthritis) assessment, standardize
its acquisition and scoring, and promote increased uptake of ultrasound for use in clinical, research, and trial
settings. This supplement will allow us to enhance the original proposal by providing additional resources to
support AI/ML approaches utilizing the image data in addition to the semiquantitative scoring we initially
proposed. Knee osteoarthritis (KOA) is highly prevalent and frequently debilitating. Development of potential
treatments has been hampered by the heterogeneous nature of this common chronic condition, which is
characterized by several subgroups, or phenotypes, with different underlying pathophysiological mechanisms.
Imaging, genetics, biochemical biomarkers, and other features can be used to characterize phenotypes, but
variations in data types can make it difficult to harmonize definitions. Ultrasound is a widely accessible, time-
efficient, and cost-effective imaging modality that can provide detailed and reliable information for all joint
tissues. Application of deep learning methodology to discover ultrasound features associated with pain and
radiographic change in KOA is highly innovative and will be a major step forward for the field. We will leverage
standardized ultrasound images from the diverse and inclusive population-based Johnston County Health
Study (JoCoHS), the new enrollment phase of the 30-year Johnston County OA Project which includes Black,
White, and Hispanic men and women aged 35-70. In Aim 1, we will apply deep learning methodology to
understand the features in ultrasound images that are most associated with knee pain and with radiographic
features of knee OA in this diverse group. Aim 2 will allow the process of optimization for full AI/ML readiness
of these images, including annotation, documentation, formatting, and storage of these images according to
FAIR principles. This supplement will enhance the parent study by allowing AI/ML analysis of the ultrasound
images, beyond just the semi-quantitative scores, and represents a crucial step to determine the ultrasound
features of greatest importance to pain and other aspects of OA. By developing and maintaining an AI/ML
ready repository of standardized ultrasound images from this generalizable cohort, we can enhance the uptake
of this modality and contribute to further study on its use in OA worldwide, including in low-resource settings
and across populations.

## Key facts

- **NIH application ID:** 10591756
- **Project number:** 3R01AR077060-03S1
- **Recipient organization:** UNIV OF NORTH CAROLINA CHAPEL HILL
- **Principal Investigator:** Amanda E Nelson
- **Activity code:** R01 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2022
- **Award amount:** $304,487
- **Award type:** 3
- **Project period:** 2020-05-05 → 2025-04-30

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10591756

## Citation

> US National Institutes of Health, RePORTER application 10591756, Development of an AI/ML-ready knee ultrasound dataset in a population-based cohort (3R01AR077060-03S1). Retrieved via AI Analytics 2026-05-25 from https://api.ai-analytics.org/grant/nih/10591756. Licensed CC0.

---

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