Automated Analysis of Newborn Motor Patterns Using Computerized Clinical Assessment for Newborns 0 to 15 Days Old in Maternity and Neonatal Intensive Care Units

What this Trial Is About

Researchers are developing a computerized system to assess newborns by analyzing facial expressions, crying, posture, and movement. This project aims to automatically identify abnormal motor patterns associated with conditions like anoxic-ischemic encephalopathy, brachial plexus paralysis, early neonatal infections, and stroke. The study focuses on newborns hospitalized in maternity and neonatal intensive care units. The study involves recording 1 to 2 minute 2D videos at 60Hz of awake newborns during bathing or clinical examination on days 0, 1, 2, 3, and 4 after birth. These videos capture facial expressions, cry sounds, movements, and posture, which are analyzed using computer vision and neural network algorithms to distinguish normal from abnormal motor patterns. Data collection is planned over three years to reach 10,000 video acquisitions. Participants are newborns hospitalized in specified hospitals, with video and sound recordings taken during their hospital stay. Clinical information such as sex, gestational age, birth weight, Apgar score, and delivery mode is collected alongside the recordings. Researchers compare computer-generated scores to clinical examinations and monitor neurological and postural changes over time. Participation is limited to video capture during hospitalization, and no interventions are performed.

neonAtal motoR paTtErn autoMatIc analySis