Intelligence Sparse Sensor Network for Automatic Early Evaluation of General Movements in Infants.
Benkun Bao1,2, Senhao Zhang2,3, Honghua Li4
1School of Biomedical Engineering (Suzhou), Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei, 230022, P. R. China.
Advanced Science (Weinheim, Baden-Wurttemberg, Germany)
|March 6, 2024
Summary
This study introduces a wearable sensor system for early infant brain development assessment. It accurately identifies infants at risk for developmental disorders using general movements (GMs) and AI, aiding timely intervention.
Area of Science:
- Biomedical Engineering
- Developmental Pediatrics
- Wearable Technology
Background:
- General movements (GMs) are crucial for early infant brain development assessment but face challenges in quantitative analysis and widespread clinical use, especially in resource-limited areas.
- Wearable sensors offer a promising solution for infant movement analysis due to their privacy, cost-effectiveness, and ease of use.
- Existing methods for GM assessment lack the quantitative precision and accessibility needed for broad application.
Purpose of the Study:
- To develop and validate a novel wearable sparse sensor network for automatic early evaluation of general movements in infants.
- To assess the system's reliability and accuracy in recognizing neonatal activities and identifying infants at risk for developmental disorders.
- To leverage artificial intelligence for intelligent analysis of infant movement data to facilitate early diagnosis.
Main Methods:
- A sparse network of five soft wireless IMU devices (SWDs) was designed for continuous, stable full-body motion capture in infants.
- Proof-of-concept clinical testing was conducted with 23 infants to evaluate the system's performance.
- A tiny machine learning algorithm was integrated for automatic identification of high-risk infants based on GM patterns.
Main Results:
- The wearable sensor system demonstrated outstanding performance in recognizing neonatal activities, confirming its reliability.
- The integrated AI algorithm achieved high accuracy (up to 99.9%) in automatically identifying infants at risk based on general movements.
- The system successfully captured comprehensive full-body motion data using a minimal number of robust and biocompatible sensor nodes.
Conclusions:
- The developed wearable sparse sensor network provides a reliable and accurate method for automatic early evaluation of infant general movements.
- This AI-powered system facilitates intelligent assessment of infant brain development and early diagnosis of potential disorders.
- The technology holds significant potential for wider deployment, particularly in low-resource settings, improving developmental screening and intervention.
