Related Experiment Video
Updated: Jul 2, 2026
Quantified Assessment of Infant's Gross Motor Abilities Using a Multisensor Wearable
Published on: May 17, 2024
Body position classification using wearable sensors in infants with cerebral palsy
Kari S Kretch1, Florencia A Enriques1, John M Franchak2
1University of Southern California, 1540 Alcazar St., Los Angeles, CA 90089, United States.
Insights
Wearable sensors and machine learning accurately classify body position in infants with cerebral palsy (CP). This technology can measure real-world motor behavior and learning opportunities for infants with CP.
Area of Science:
- Biomedical Engineering
- Developmental Pediatrics
- Machine Learning in Healthcare
Background:
- Infants with cerebral palsy (CP) face motor impairments that can affect learning opportunities.
- Measuring real-world motor behavior in infants with CP is crucial for understanding developmental challenges.
- Existing machine learning models for body position classification in typical development (TD) infants need validation for CP populations.
Purpose of the Study:
- To assess the validity of using machine learning and wearable sensors for classifying body position in infants with CP.
- To determine the most accurate machine learning model configuration for body position classification in infants with CP.
- To explore the potential of sensor data for quantifying motor behavior and learning opportunities in infants with CP.
Main Methods:
- Ten infants with CP and 19 infants with typical development (TD) wore four inertial sensors on their legs.
- Video recordings of 90 minutes per session were manually scored for five body positions: supine, prone, sitting, standing, and held.
- Random forest classifiers were trained using sensor-derived motion features, with varying dataset sizes and group compositions (CP, TD, CP and TD).
Main Results:
- Machine learning models trained on larger datasets including CP infant data showed highest accuracy.
- Final models achieved comparable performance for CP (86% accuracy) and TD infants (89% accuracy).
- Classifications demonstrated individual differences (ICCs = 0.682–0.999) and correlated with motor skill assessments.
Conclusions:
- Wearable sensors and machine learning provide a valid method for accurately classifying real-world body position in infants with CP.
- This technology can quantify motor behavior and learning opportunities, aiding in the assessment and support of infants with CP.
- The findings support the use of these tools to better understand and address the motor challenges faced by infants with CP.
Abstract:
Infants learn through everyday interactions with the physical and social environment. For infants with cerebral palsy (CP), motor impairments may disrupt everyday learning opportunities. How can we measure real-world motor behavior and learning opportunities in infants with CP? Machine learning models have been developed to quantify body position throughout a day in infants with typical development (TD) using wearable sensor data. However, these models have not been validated in infants with motor impairments. This study assessed the validity of body position classification using machine learning and sensors in infants with CP. Ten infants with CP (7-18 months; one session each) and 19 infants with TD (4-12 months; 45 sessions) wore four inertial sensors on their legs throughout a day. Ninety minutes were video recorded and manually scored for body position in five categories: supine, prone, sitting, standing, and held. Random forest classifiers were trained to predict body position from sensor-derived motion features. Models trained on datasets that varied in size (9, 45, 54 sessions) and group composition (CP, TD, CP and TD) were compared to determine the most accurate model for infants with CP. Larger training sets and training sets that included data from infants with CP were the most accurate; the final models achieved similar performance in CP and TD (86% and 89% accuracy), captured meaningful individual differences (ICCs = 0.682-0.999), and generated predictions that were correlated with motor skill assessments. Findings demonstrate that wearable sensors and machine learning can accurately classify real-world body position in infants with CP.
