Quantum-classical deep learning hybrid architecture with graphene-printed low-cost capacitive sensor for essential...

Area of Science:

• Biomedical Engineering
• Computational Neuroscience
• Materials Science

Background:

• Essential Tremor (ET) diagnosis relies on clinical assessment and sometimes complex instrumentation.
• Current methods for tremor detection can be costly or lack precision.
• There is a need for accessible, accurate, and efficient diagnostic tools for ET.

Purpose of the Study:

• To develop and evaluate a novel hardware and software architecture for Essential Tremor detection.
• To integrate advanced sensing technology with quantum-inspired algorithms and deep learning.
• To assess the efficacy of this integrated system for tremor data acquisition and analysis.

Main Methods:

• Utilized graphene-printed capacitive sensors for cost-effective and precise tremor data acquisition.
• Developed a system calibrated for monitoring tremor movements across various fingers.
• Incorporated quantum-inspired computational filters (Quantvolution, QuantClass) within a deep learning framework.

Main Results:

• The novel architecture demonstrated potential for improved processing capabilities in tremor pattern analysis.
• Initial findings suggest greater stability in loss variability during the detection process.
• The system offers a nuanced analysis of tremor patterns.

Conclusions:

• The proposed hardware and software architecture presents a promising approach for Essential Tremor detection.
• The integration of quantum-inspired methods offers enhanced analytical power for healthcare diagnostics.
• Further validation across larger datasets and clinical settings is recommended to confirm findings.