Search reports, ask a question, or jump to a page.
The problem
Intervention: AI algorithm trained to automatically classify 38 infant movements from video recordings, administered in laboratory and home settings via mobile app
Who it was tested with: 85 full-term infants and 84 preterm infants, ages 4 to 18 months.
The same group is measured before the intervention and again afterwards, with a baseline and an endline and sometimes a midline. Change is visible, but other causes cannot be ruled out.
"An AI model was developed to classify movements using physiotherapists' labeling results as the ground truth. The AI-classified movements were organized into age-based sets to test concurrent validity against the AIMS examiners' results." The design uses physiotherapist assessments as the reference standard and tests whether AI classifications match these assessments, demonstrating technical validity but not clinical impact.
Study validates that AI can classify movements with high accuracy but does not demonstrate clinical utility or impact on infant outcomes. No comparison of the AI approach against standard clinical practice in terms of time, cost, or ability to detect developmental delays. Home video performance was lower than laboratory, suggesting real-world application challenges.
Across countries and designs
1 study · 0 countries
