Parenting Journal · Volume 1 · Issue 2 · Licensed source edition
Crowd annotations can approximate clinical autism impressions from short home videos with privacy protections
Originally published 2022-04-08 · Intelligence-Based Medicine
Elsevier BV · DOI 10.1016/j.ibmed.2022.100056 ↗
AI / system type
Machine learning / privacy-preserving child-video evaluation
Family implicationsDesign, ethics, stakeholder or technical evidence with child/family implications; household benefits are not assumed.
The AI connection
Uses crowd-coded behavioral video features as inputs to a machine-learning autism classifier and compares privacy-preserving transformations.
Why this reading matters
Provides inspectable technical evidence about child-data handling rather than assuming privacy and model performance can be optimized independently.
Editorial reading guide · Respect
What can privacy edits protect, and what may they change in a child-video AI assessment?
The study makes a concrete privacy–utility tradeoff visible: protecting a child’s image can alter the information available to an AI-supported assessment.
Study context
Sixty videos, crowd annotators, clinicians and a previously validated classifier
Age / stage: Children in autism-related home videos
Read with three questions
- How were pixelation, blurring and optical flow tested?
- What happened to annotations and classifier performance?
- Why do these technical results not establish diagnostic safety or anonymity for home use?
Keep the limits in view
The evaluation used a small balanced video set and a specific classifier/annotation workflow. Privacy transformations do not guarantee anonymity, and technical accuracy does not demonstrate clinical benefit or safe deployment.
The text below is the original article’s licensed repository edition. Its wording is not rewritten. Use the Paper Agent beside the text to locate evidence; open the source for figures, tables and complete context.
Opening the verified paper and agent…
Checking the article identity, license record, and saved source text.
Open the repository recordAttribution & reuse
Follow the original.
Original authorship, publication date, journal and DOI are preserved. This issue’s TRICK placement and reading questions are editorial additions, not claims made by the original authors.
Peter Washington; Brianna Chrisman; Emilie Leblanc; Kaitlyn Dunlap; Aaron Kline; Cezmi Mutlu; Nate Stockham; Kelley Paskov; Dennis Paul Wall (2022). Crowd annotations can approximate clinical autism impressions from short home videos with privacy protections Intelligence-Based Medicine. https://doi.org/10.1016/j.ibmed.2022.100056
Permissions checked against the saved article notice; repository publication notices screened 2026-11-04. This is not a new peer-review decision. The source authors and publishers are not represented as endorsing this issue.