A Deep Learning Approach for Predicting Analyte Refractive Index of Open-Channel Plasmonic Sensor

Open-channel plasmonic sensors utilize surface plasmon resonance (SPR) to detect minute changes in the refractive index (RI).This study presents a deep learning-based approach for predicting the analyte RI in plasmonic sensors.The work utilizing simulation data from a plasmonic sensor across a RI varies of 1.33 to 1.40.There are four deep learning methods such as artificial neural networks (ANNs), long short-term memory (LSTM) networks, gated recurrent units (GRUs), and convolutional neural networks (CNNs), is analyzed for their predictive capabilities.Among these methods, the ANN model demonstrates high performance, reaching an accuracy of 78.18%, with precision, recall, and F1-scores of 0.78, alongside minimal misclassification errors.On the other hand, the CNN and LSTM models exhibited moderate performance, each achieving 72.72% accuracy, while the GRU model lagged significantly with an accuracy of 41.81%.Analysis of training and test accuracies revealed stable ANN training accuracy at 90%, although test accuracy variations near 70% indicated potential overfitting.This work underscores the transformative potential of deep learning in advancing the design of plasmonic sensors and opening new avenues for biomedical sensing applications.