Automatic Assessment of Infant Face and Upper-Body Symmetry as Early Signs of Torticollis
This work addresses early detection of torticollis in infants, which is critical for timely treatment, but it is incremental as it adapts existing methods to a new domain.
The paper tackled the problem of early identification of torticollis in infants by applying computer vision pose estimation techniques to assess face and upper-body symmetry, achieving strong to very strong Spearman's ρ correlations with ground truth values. It demonstrated that infant-specific neural networks outperform adult-domain networks in this task.
We apply computer vision pose estimation techniques developed expressly for the data-scarce infant domain to the study of torticollis, a common condition in infants for which early identification and treatment is critical. Specifically, we use a combination of facial landmark and body joint estimation techniques designed for infants to estimate a range of geometric measures pertaining to face and upper body symmetry, drawn from an array of sources in the physical therapy and ophthalmology research literature in torticollis. We gauge performance with a range of metrics and show that the estimates of most these geometric measures are successful, yielding strong to very strong Spearman's $ρ$ correlation with ground truth values. Furthermore, we show that these estimates, derived from pose estimation neural networks designed for the infant domain, cleanly outperform estimates derived from more widely known networks designed for the adult domain