Related Experiment Video
Updated: Jan 7, 2026
Quantified Assessment of Infant's Gross Motor Abilities Using a Multisensor Wearable
Published on: May 17, 2024
Machine Learning Methods in Posture-Related Applications in Children up to 12 Years Old: A Systematic Review
Markel Rico-González1,2, Carlos D Gómez-Carmona2,3,4, Ibrahim Ouergui5,6
1Department of Didactics of Music, Plastic and Body Expression, University of Basque Country (UPV-EHU), 48940 Leioa, Spain.
Insights
Machine learning accurately assesses postural control in children (0-12 years) using sensors. This technology shows promise for early developmental delay detection and diagnosing conditions like cerebral palsy.
Area of Science:
- Pediatrics
- Biomedical Engineering
- Machine Learning
Background:
- Postural control is crucial for motor development in infants and young children.
- Machine learning (ML) offers potential for analyzing complex movement data.
Purpose of the Study:
- To systematically review ML methods applied to posture-related applications in children aged 0-12.
- To evaluate the effectiveness of ML in posture assessment and related diagnostics.
Main Methods:
- Systematic literature search following PRISMA guidelines across major databases (PubMed, Web of Science, Scopus, ProQuest).
- Inclusion of 22 studies with moderate to good methodological quality (MINORS scale).
- Analysis of sensor-based technologies (IMUs, force plates, pressure mats, video) for extracting kinematic and postural features.
Main Results:
- ML algorithms, particularly Random Forest, SVM, and CNN, achieved accuracies often exceeding 85%.
- Heterogeneity in sensor modalities, data quality, and model architectures was noted.
- Effective application in posture classification, early detection of developmental delays, and diagnosing conditions like cerebral palsy and autism spectrum disorder.
Conclusions:
- ML demonstrates significant potential for posture-related applications in pediatric populations.
- These methods show promise for both at-home monitoring and clinical interventions.
- Further standardization may enhance the reliability and generalizability of ML approaches in pediatric postural control research.
Abstract:
One of the most important factors in how infants and young children learn to move is postural control. This systematic review aims to evaluate the machine learning methods in posture-related applications for children aged 0-12. Following PRISMA guidelines, we systematically searched the PubMed, Web of Sciences, SCOPUS, and ProQuest Central databases. Twenty-two studies were included in the qualitative synthesis following screening of 199 articles, with methodological quality assessed as moderate to good using the MINORS scale (scores ranging from 8/16 to 19/24). The reviewed research involved diverse samples of infants and children up to 12 years old, employing sensor-based technologies such as inertial measurement units, force plates, pressure mats, and video cameras to extract kinematic and postural features for machine learning applications. Reported accuracies, typically exceeding 85%, reflected considerable methodological heterogeneity related to sensor modality, data quality, and model architecture. Algorithms such as Random Forest, SVM, and CNN were most frequently and effectively applied for posture classification, early detection of developmental delays, and diagnosis of conditions such as cerebral palsy and autism spectrum disorder, demonstrating promising potential for at-home monitoring and clinical interventions.
