OTOtherOccupational therapy international2022

Development and Verification of Postural Control Assessment Using Deep-Learning-Based Pose Estimators: Towards Clinical Applications.

Naoto Ienaga, Shuhei Takahata, Kei Terayama and 4 others

PMID 36531757

WHAT IT FOUND

MediaPipe Pose offers the best balance of speed and accuracy for assessing postural control from video.

Two calculated indices, SPB1 and AG2, better predict therapists' clinical judgments than simply timing how long a child holds a posture, especially when duration scores hit the 60-second ceiling.

Key findings

01MediaPipe Pose was selected as the most clinically applicable method due to its processing speed and robust acceptance rate, despite slightly lower accuracy than OpenPose.

02Models using the proposed indices (SPB and AG) predicted occupational therapists' clinical evaluations better than models using only task duration, as shown by lower AIC values.

03SPB1 (static postural balance) and AG2 (antigravity) were consistently associated with lower clinical evaluation scores when their values were higher, indicating they effectively quantify poor postural control.

STILL TO COME

How it was doneWhat they foundWhat it means for OTs

Read the rest of this summary

You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.

Already have one?

What it does not show

The study used a small sample size (57 participants total) and focused on a single task from the JPAN battery. The adult dataset showed a strong ceiling effect for duration, which limits the generalizability of time-based comparisons. The accuracy of MediaPipe Pose was lower than OpenPose at high confidence thresholds, which might affect fine motor assessments. The indices were validated against therapists' qualitative evaluations, not against gold-standard laboratory equipment or clinical outcomes. The study did not provide standardized scores or developmental norms for the new indices.

Declared interests

The authors declare no conflict of interest. The research was approved by the Research Ethics Committee of Hakuho College.

The easy way to misread this

Do not assume these indices are ready for immediate clinical use as standardized diagnostic tools. The study validates them against therapist observations in a small sample, not against clinical outcomes or normative data, so they should be viewed as a promising research tool rather than a replacement for current assessment batteries.

Read it on PubMed →