A study published in *Pediatric Research* explored automated neonatal sleep positioning assessment using video monitoring and machine learning. The research addressed that therapeutic positioning of preterm infants in the Neonatal Intensive Care Unit (NICU) is critical for development, but current clinical assessment relies on intermittent visual evaluation by trained staff, which introduces inter-rater variability and limits monitoring frequency. The authors used the Infant Positioning Assessment Tool (IPAT), which evaluates six anatomical features—head and neck alignment, trunk alignment, arm positioning, leg positioning, overall symmetry, and leg anatomic position—each scored 0 (poor) to 2 (optimal), for a maximum of 12 points. Scores of 0–8 indicate suboptimal positioning requiring intervention. The study had two objectives: (1) identify optimal machine learning approaches for infant position classification from skeletal features; (2) determine whether a solution providing direct image to suboptimal positioning indication is reachable with current techniques.