Artificial Intelligence-Supported Colorimetric Multibiomarker Sensor to Enable Critical Neonatal Monitoring

Area of Science:

• Biomedical Engineering
• Wearable Technology
• Neonatal Monitoring

Background:

• Current neonatal monitoring methods are often invasive, costly, and uncomfortable.
• There is a critical need for noninvasive, real-time monitoring solutions for vulnerable neonates.

Purpose of the Study:

• To develop and validate a noninvasive, AI-supported multibiomarker sensor for critically ill preterm neonates.
• To enable real-time tracking of key physiological parameters through colorimetric analysis of body fluids.

Main Methods:

• A silk-based wearable patch with twelve colorimetric inks was developed.
• The sensor tracks temperature, pH, sodium, and glucose via colorimetric analysis of transepidermal fluid.
• Deep learning algorithms were employed for precise quantification of colorimetric responses.

Main Results:

• The sensor demonstrated high precision and reproducibility in clinically relevant neonatal ranges.
• AI-based quantification achieved high-precision estimates for all monitored biomarkers.
• The sensor maintained performance under various challenging clinical conditions, including humidity and movement.

Conclusions:

• This noninvasive, AI-supported sensor offers a promising new approach for intensified monitoring in neonatal intensive care.
• The miniaturized, wearable design and robust performance cater to the critical needs of high-risk infant populations.
• Transepidermal fluid analysis via this sensor opens new avenues for noninvasive clinical monitoring.