Machine Learning of Infant Spontaneous Movements for the Early Prediction of Cerebral Palsy: A Multi-Site Cohort Study

• Published in: Journal of clinical medicine Vol. 9; no. 1; p. 5
• Main Authors: Ihlen, Espen A F, Støen, Ragnhild, Boswell, Lynn, Regnier, Raye-Ann de, Fjørtoft, Toril, Gaebler-Spira, Deborah, Labori, Cathrine, Loennecken, Marianne C, Msall, Michael E, Möinichen, Unn I, Peyton, Colleen, Schreiber, Michael D, Silberg, Inger E, Songstad, Nils T, Vågen, Randi T, Øberg, Gunn K, Adde, Lars
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 18-12-2019; MDPI
• Subjects: Age; Babies; Cerebral palsy; Clinical medicine; Cohort analysis; Machine learning; Medical disciplines: 700; Medisinske Fag: 700; Risk assessment; Technology: 500; Teknologi: 500; Validation studies; Validity; VDP; premature infants; cerebral palsy; machine learning; general movement assessment
• ISSN: 2077-0383; 2077-0383
• DOI: 10.3390/jcm9010005

Early identification of cerebral palsy (CP) during infancy will provide opportunities for early therapies and treatments. The aim of the present study was to present a novel machine-learning model, the Computer-based Infant Movement Assessment (CIMA) model, for clinically feasible early CP prediction based on infant video recordings.
The CIMA model was designed to assess the proportion (%) of CP risk-related movements using a time-frequency decomposition of the movement trajectories of the infant's body parts. The CIMA model was developed and tested on video recordings from a cohort of 377 high-risk infants at 9-15 weeks corrected age to predict CP status and motor function (ambulatory vs. non-ambulatory) at mean 3.7 years age. The performance of the model was compared with results of the general movement assessment (GMA) and neonatal imaging.
The CIMA model had sensitivity (92.7%) and specificity (81.6%), which was comparable to observational GMA or neonatal cerebral imaging for the prediction of CP. Infants later found to have non-ambulatory CP had significantly more CP risk-related movements (median: 92.8%,
= 0.02) compared with those with ambulatory CP (median: 72.7%).
The CIMA model may be a clinically feasible alternative to observational GMA.