Deep Learning-Assisted Organogel Pressure Sensor for Alphabet Recognition and Bio-Mechanical Motion Monitoring

Area of Science:

• Materials Science
• Biomedical Engineering
• Nanotechnology

Background:

• Wearable sensors and deep learning offer potential for advanced human-machine interfaces.
• Achieving desired mechanical, electrical, biocompatibility, and environmental properties in sensors remains challenging.

Purpose of the Study:

• To develop a multifunctional organogel pressure sensor with enhanced properties.
• To explore its application in real-time health monitoring and human-machine interfaces.

Main Methods:

• Fabrication of an organogel using cobalt nanoparticle encapsulated nitrogen-doped carbon nanotubes (CoN CNT) in a polyvinyl alcohol-gelatin (PVA/GLE) matrix with a water/ethylene glycol solvent system.
• Characterization using electrochemical impedance spectroscopy.
• Testing sensitivity, humidity, and temperature tolerance.
• Integration with deep learning for human-machine interface applications.

Main Results:

• The CoN CNT/PVA/GLE organogel sensor demonstrated excellent flexibility, biocompatibility, and environmental stability.
• Near-stable performance was observed across a wide humidity range (40%-95% RH).
• Freeze-tolerant conductivity was achieved at -20°C.
• High sensitivity (5.75 kPa⁻¹) and 98% accuracy in English letter recognition were recorded.
• Successful detection of subtle biomechanical motions and human gestures was demonstrated.

Conclusions:

• The developed organogel sensor offers a promising solution for advanced wearable sensing applications.
• Its multifunctional properties pave the way for robust human-machine interfaces and real-time health monitoring.
• The sensor's performance under various environmental conditions highlights its potential for practical use.