Computer Vision to the Rescue: Infant Postural Symmetry Estimation from Incongruent Annotations

TL;DR

This paper introduces a computer vision system for infant symmetry assessment using 3D pose estimation, addressing the unreliability of human ratings by employing a Bayesian model to improve accuracy and consistency.

Contribution

It develops a Bayesian calibration method for infant symmetry assessment that accounts for human rating inconsistencies, enhancing the reliability of computer vision-based evaluations.

Findings

3D infant pose estimation achieves 68% AUC in predicting Bayesian labels

Human ratings show low inter-rater reliability and bias

The system outperforms 2D and adult pose models in symmetry assessment

Abstract

Bilateral postural symmetry plays a key role as a potential risk marker for autism spectrum disorder (ASD) and as a symptom of congenital muscular torticollis (CMT) in infants, but current methods of assessing symmetry require laborious clinical expert assessments. In this paper, we develop a computer vision based infant symmetry assessment system, leveraging 3D human pose estimation for infants. Evaluation and calibration of our system against ground truth assessments is complicated by our findings from a survey of human ratings of angle and symmetry, that such ratings exhibit low inter-rater reliability. To rectify this, we develop a Bayesian estimator of the ground truth derived from a probabilistic graphical model of fallible human raters. We show that the 3D infant pose estimation model can achieve 68% area under the receiver operating characteristic curve performance in predicting the Bayesian aggregate labels, compared to only 61% from a 2D infant pose estimation model and 60% from a 3D adult pose estimation model, highlighting the importance of 3D poses and infant domain knowledge in assessing infant body symmetry. Our survey analysis also suggests that human ratings are susceptible to higher levels of bias and inconsistency, and hence our final 3D pose-based symmetry assessment system is calibrated but not directly supervised by Bayesian aggregate human ratings, yielding higher levels of consistency and lower levels of inter-limb assessment bias.