Multiplexed DNA-functionalized graphene sensor with artificial intelligence-based discrimination performance for...

Area of Science:

• Nanotechnology
• Biomedical Engineering
• Artificial Intelligence

Background:

• Accurate detection of chemical vapors is crucial for in situ analysis and disease diagnosis.
• Existing methods often require vapor condensation or dilution, limiting real-time applications.
• Developing selective, humidity-tolerant sensors is essential for breath analysis.

Purpose of the Study:

• To present a new technology for detecting and discriminating individual chemical vapors in mixed compositions.
• To enable in situ analysis of chemical vapor composition using multiplexed DNA-functionalized graphene nanoelectrodes.
• To establish a foundation for AI-based discrimination of chemical vapors in breath analysis for early disease diagnosis.

Main Methods:

• Utilized a multiplexed DNA-functionalized graphene (MDFG) nanoelectrode array.
• Integrated artificial intelligence (AI) techniques, including 1D convolutional neural networks, for data analysis.
• Tested sensor performance under varying humidity levels for mixed chemical vapor detection.

Main Results:

• Achieved recognition rates of 99% and above under low humidity and 98% and above under humid conditions for mixed chemical compositions.
• Demonstrated AI-operated arrayed electrodes capable of identifying mixed chemical gas compositions and ratios in early stages.
• 1D convolutional neural network analysis showed near-perfect discrimination of chemical vapor composition under both low and high humidity.

Conclusions:

• The developed MDFG sensor array combined with AI offers a robust platform for chemical vapor analysis.
• This technology shows significant promise for
• breath chemovapor fingerprinting
• enabling early disease diagnosis.
• The AI-based discrimination of chemical vapor compositions in breath analysis is validated, paving the way for clinical applications.