AI integration into wavelength-based SPR biosensing: Advancements in spectroscopic analysis and detection.

Area of Science:

• Biosensing
• Spectroscopy
• Artificial Intelligence

Background:

• Deep learning enhances biosensing data analysis, interpretation, and prediction accuracy.
• While deep learning is used in some SPR fields, its application to spectroscopic SPR biosensors is novel.
• This study focuses on integrating AI to improve signal-to-noise ratio (SNR) and detection accuracy in portable SPR biosensors.

Purpose of the Study:

• To develop and validate an AI-driven method for enhancing the performance of wavelength-based portable SPR biosensors.
• To improve the signal-to-noise ratio (SNR) and detection accuracy of these biosensors.
• To increase the interpretability and transparency of the AI model used.

Main Methods:

• Designed a deep neural network integrated with spectral subtraction for SPR response extraction.
• Utilized difference spectra as input for the AI model.
• Employed Shapley Additive Explanations (SHAP) analysis for model interpretability.

Main Results:

• The AI model demonstrated superior noise reduction and enhanced detection capabilities compared to traditional methods.
• Achieved a significantly amplified SNR and improved detection resolution to 10⁻⁷ RIU.
• SHAP analysis confirmed the AI model prioritizes wavelength regions crucial for SPR sensing sensitivity, aligning with theoretical understanding.

Conclusions:

• AI integration advances on-site detection technologies for portable SPR biosensing.
• The AI model effectively reduces noise and enhances detection accuracy, especially for low-concentration analytes.
• This innovation promises transformative applications in biomedical diagnostics, environmental monitoring, and biochemical analysis requiring real-time, high-precision, on-site detection.