Deep Learning Optimized SPR Multi-Analyte Biosensor for Simultaneous Detection of Water Pathogens
Bibliographic record
Abstract
In this study, we present a deep-learning-assisted design of a photonic crystal fiber (PCF)-based surface plasmon resonance (SPR) biosensor that enables simultaneous, label-free detection of multiple waterborne analytes. Using finite element analysis, a dual-channel PCF is modeled to generate more than 40,000 data points. A lightweight, fully connected regressor predicts the confinement loss (CL) and amplitude sensitivity (AS) from structural variables (hole sizes, gaps, metal/dielectric thicknesses) and operational variables (wavelength, refractive index (RI) of analytes in both channels), achieving <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">R</i><sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> ≈ 0.99 with low error. The surrogate speeds up the process of exploring designs. Using SHAP analysis, it finds that wavelength and channel RIs are the main factors, while layer thicknesses mainly change channel-specific resonances. Parametric sweeps confirm stable, concurrent redshifts across channels with increasing RI, enabling multiplexed detection of bacterial pathogens and formaldehyde. The proposed model achieves maximum amplitude sensitivity (AS) of 512.87 <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">RIU</i><sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">−1</sup>, wavelength sensitivity (WS) of 10,638.30 <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">nm/RIU</i>, and sensor resolution (SR) of 9.4×10<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">−5</sup>. The resulting architecture combines high accuracy with computational efficiency, offering a compact route to rapid, real-time water quality monitoring and food safety screening, as well as a generalizable workflow for data-driven PCF-SPR design.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".