Using video and AI to spot early movement signs of autism in babies
Identifying autism motor deficits in infants using computer vision
This project uses computer vision on infant videos to find early movement patterns that may signal autism in babies.
Quick facts
| Grant type | R21 grant |
|---|---|
| Study type | NIH-funded research |
| Funding institution | Washington University NIH-funded |
| Lab location | 1 site (Saint Louis, United States) |
| Project ID | NIH-11247172 on NIH RePORTER |
What this research studies
Researchers will apply computer-vision tools (like OpenPose) to extract infant body-joint positions and track movement frame-by-frame from hundreds of behavioral assessment videos. They will train deep-learning models to turn those movements into quantitative motion measures and link them to later 24-month autism diagnoses and family risk. Unsupervised computational ethology methods will search for subtle motor signatures that go beyond the original assessment format. The work uses a large, multi-site longitudinal video dataset to develop candidate motor biomarkers that might be useful for earlier referral to services.
Who could benefit from this research
Good fit: Ideal candidates are infants with available video recordings from standardized early behavioral assessments (such as the AOSI) and follow-up diagnostic data through about 24 months.
Not a fit: This work is unlikely to benefit older children or adults, or infants who do not have video recordings or follow-up diagnostic information available.
Why it matters
Potential benefit: If successful, this could enable earlier, objective detection of autism risk from routine video recordings and speed access to early intervention.
How similar studies have performed: Prior research has suggested infant motor differences can predict later ASD, but applying modern deep learning and computer-vision methods to a large, multi-site longitudinal video set is a newer, proof-of-concept approach.
Where this research is happening
Saint Louis, United States
- Washington University — Saint Louis, United States (Active)
Researchers
- Principal investigator: Goodhill, Geoffrey J — Washington University
- Study coordinator: Goodhill, Geoffrey J
About this research
- This is an active NIH-funded research project — typically early-stage science, not a clinical trial accepting patient enrollment.
- Some NIH-funded labs run parallel clinical studies or seek volunteers for related work. To check, contact the principal investigator or institution listed above.
- For full project details, budget, and progress reports, visit the official NIH RePORTER page below.