Integrating Deep Learning with Photonic SPR Biosensors for Pathogen and Biomarker Detection

By integrating the pathogen biomarker identification deep learning (DL) algorithms and surface plasmon resonance (SPR) biosensors, an innovative development of pathogen and biomarker identification is possible. This function ultimately offers a simple way to calculate C-reactive protein (CRP), a common marker of inflammation and infection from SPR simulations that are run in MATLAB. Acute phase responses and more rapid identification of CRP are important in disorders of sepsis, heart disease, and respiratory infection. As a result of this research, however, a synthetic dataset of real-world noise changes is created, which is synthetically generated to include real-world noise changes so as to generate synthetic SPR responses, and using this synthetic dataset for deep learning models based on responses produces a greater robust training set than ever before. To improve biosensor sensitivity, detection limit, and specificity and to also meet the conditions for real-time diagnostic systems, a deep learning-based categorization system is proposed. The suggested method was tested using a CNN model, and it has demonstrated a good sensitivity (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{9 0 - 9 5 \%}$</tex>) and specificity (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$85-90 \%$</tex>) compared to other characteristics in SPR data with noise. To the best of our knowledge, it is the first work that demonstrates that the SPR biosensor can be used to develop an end-to-end MATLAB synthetic data pipeline using deep learning and in the detection of CRP. This applied approach leads to high precision, real-time diagnostics, and eventually to clinical applications for biosensor mass integration with AI ability that will constrain scalability. It offers a generic way to find many biomarkers and to check out the backstage of the daily interactions of the standard SPR technologies.