Deep-Learning Markerless Tracking of Infant General Movements using Standard Video Recordings.

Monitoring spontaneous General Movements (GM) of infants 6-20 weeks post-term age is a reliable tool to assess the quality of neurodevelopment in early infancy. Abnormal or absent GMs are reliable prognostic indicators of whether an infant is at risk of developing neurological impairments and disorders such as cerebral palsy (CP). Therapeutic interventions are most effective at improving neuromuscular outcomes if administered in early infancy. Current clinical protocols require trained assessors to rate videos of infant movements, a time-intensive task. This work proposes a simple, inexpensive, and broadly applicable markerless pose-estimation approach for automatic infant movement tracking using conventional video recordings from handheld devices (e.g., tablets and mobile phones). We leverage the enhanced capabilities of deep-learning technology in image processing to identify 12 anatomical locations (3 per limb) in each video frame, tracking a baby's natural movement throughout the recordings. We validate the capability of resnet152 and a mobile-net-v2-1 to identify body-parts in unseen frames from a full-term male infant, using a novel automatic unsupervised approach that fuses likelihood outputs of a Kalman filter and the deep-nets. Both deep-net models were found to perform very well in the identification of anatomical locations in the unseen data with high average Percentage of Correct Keypoints (aPCK) performances of >99.65% across all locations.Clinical relevance-Results of this research confirm the feasibility of a low-cost and publicly accessible technology to automatically track infants' GMs and diagnose those at higher risk of developing neurological conditions early, when clinical interventions are most effective.

Postnatal SNAP-II scores in neonatal intensive care unit patients: relationship to sepsis, necrotizing enterocolitis, and death.

OBJECTIVE: To determine if daily SNAP-II scores (Score for Neonatal Acute Physiology) after the first day are useful in identifying neonatal intensive care unit (NICU) patients who die or develop sepsis or necrotizing enterocolitis. STUDY DESIGN: Prospective data were collected on all 141 admissions to a university level III NICU over 4 months. SNAPPE-II scores were calculated from the day of admission and SNAP-II scores from subsequent hospital days. The scores were compared between those who developed events and those who did not. RESULTS: At least 64% of the daily SNAP-II scores on the day of and the preceding 4 days from the event were 0. Admission SNAPPE-II scores correlated with length of stay (r = 0.44, p < 0.01) but patient average SNAP-II did not (r = 0.02, p > 0.5). CONCLUSIONS: SNAP-II scores from after the first day of life did not accurately assess or predict neonatal morbidity and mortality.