Autonomous screening of infants at high risk for neurodevelopmental impairments using a radar sensor and machine...

Area of Science:

• Biomedical Engineering
• Neonatology
• Machine Learning

Background:

• Neurodevelopmental impairments (NDIs) are critical long-term issues for preterm infants.
• Early identification of high-risk infants is crucial for timely intervention but remains challenging.
• Sensor-based autonomous screening offers a promising, yet under-researched, solution.

Purpose of the Study:

• To develop and validate a novel frequency modulated continuous wave (FMCW) radar-based machine learning (ML) system for early NDI risk prediction in infants.
• To introduce a new index, neuroriskability (NRA), for quantifying NDI risk based on infant movement patterns.

Main Methods:

• Constructed a 3D range-angle-time data cube of infant movements using FMCW radar.
• Identified asymmetric movements and detected abnormal movements (cramped-synchronized general movements, CSGMs) using ML.
• Calculated the neuroriskability (NRA) index based on the frequency of asymmetric movements and CSGMs.

Main Results:

• The radar-based system successfully predicted infants at high risk for poor neurodevelopmental outcomes.
• The neuroriskability (NRA) scores correlated with clinically evaluated neurodevelopmental outcomes.
• Demonstrated the clinical feasibility and utility of the proposed screening method.

Conclusions:

• The novel FMCW radar and ML system provides a feasible and effective method for early NDI screening in infants.
• The neuroriskability (NRA) index offers a quantifiable measure of NDI risk.
• This technology has the potential to significantly improve early intervention for high-risk infants.