As we’ve seen repeatedly, melding machine learning with clinical grade sensors can yield impressive new medical technologies. Recently, a group of researchers from the University of Helsinki and Helsinki Children’s Hospital BABA Center developed a wearable technology to assess infant motor abilities. The MAIJU (Motor Assessment of Infants with a Jumpsuit) entails breakthrough machine learning algorithms to assess infant motility data collected with a smart jumpsuit.

The Finnish researchers published the study notes in Nature, ‘Intelligent wearable allows out-of-the-lab tracking of developing motor abilities in infants.” The group sought a methodology that could track an infant’s movement behavior at home. The first step was to construct a jumpsuit with multiple movement sensors. The jumpsuit monitors movements through significant motor milestones ranging from lying supine to fluent walking.

The research team recorded movement data during spontaneous play of 59 infants ages 5 to 19 months. Experts then used an original motility description scheme to describe the infants’ movements based on concurrent video recordings. Next the team used machine learning to train an algorithm to recognize and describe the infants’ movements and postures during each second of playtime.

The MAIJI wearable and the software that analyzes the collected data allows accurate and objective infant motor development assessment outside a clinical setting. Because the infants wear the jumpsuit at home, repeat testing is minimally disruptive and helps monitor individual progress. Also, if a child has development motor or neurological issues, the jumpsuit can help track the results of therapeutic strategies.