BackgroundEarly neurodevelopmental care needs better, effective and objective solutionsfor assessing infants’motor abilities. Novel wearable technology opens possibilities forcharacterizing spontaneous movement behavior. This work seeks to construct and validate ageneralizable, scalable, and effective method to measure infants’spontaneous motor abilitiesacross all motor milestones from lying supine tofluent walking.MethodsA multi-sensor infant wearable was constructed, and 59 infants (age 5–19 months)were recorded during their spontaneous play. A novel gross motor description scheme wasused for human visual classification of postures and movements at a second-level timeresolution. A deep learning -based classifier was then trained to mimic human annotations,and aggregated recording-level outputs were used to provide posture- and movement-specific developmental trajectories, which enabled more holistic assessments of motormaturity.ResultsRecordings were technically successful in all infants, and the algorithmic analysisshowed human-equivalent-level accuracy in quantifying the observed postures and move-ments. The aggregated recordings were used to train an algorithm for predicting a novelneurodevelopmental measure, Baba Infant Motor Score (BIMS). This index estimatesmaturity of infants’motor abilities, and it correlates very strongly (Pearson’sr=0.89, p < 1e-20) to the chronological age of the infant.ConclusionsThe results show that out-of-hospital assessment of infants’motor ability ispossible using a multi-sensor wearable. The algorithmic analysis provides metrics of motilitythat are transparent, objective, intuitively interpretable, and they link strongly to infants’age.Such a solution could be automated and scaled to a global extent, holding promise forfunctional benchmarking in individualized patient care or early intervention trial
