A Machine Learning Framework for Detecting COVID-19 Infection Using Surface-Enhanced Raman Scattering

Abstract

In this study, we explored machine learning approaches for predictive diagnosis using surface-enhanced Raman scattering (SERS), applied to the detection of COVID-19 infection in biological samples. To do this, we utilized SERS data collected from 20 patients at the University of Maryland Baltimore School of Medicine. As a preprocessing step, the positive-negative labels are obtained using Polymerase Chain Reaction (PCR) testing. First, we compared the performance of linear and nonlinear dimensionality techniques for projecting the high-dimensional Raman spectra to a low-dimensional space where a smaller number of variables defines each sample. The appropriate number of reduced features used was obtained by comparing the mean accuracy from a 10-fold cross-validation. Finally, we employed Gaussian process (GP) classification, a probabilistic machine learning approach, to correctly predict the occurrence of a negative or positive sample as a function of the low-dimensional space variables. As opposed to providing rigid class labels, the GP classifier provides a probability (ranging from zero to one) that a given sample is positive or negative. In practice, the proposed framework can be used to provide high-throughput rapid testing, and a follow-up PCR can be used for confirmation in cases where the model’s uncertainty is unacceptably high.

Harnessing of Artificial Intelligence for the Diagnosis and Prevention of Hospital-Acquired Infections: A Systematic Review

“…In this study, targeted plasma metabolomics combined with ML was proven to provide the rapid discrimination of SARS-CoV-2-positive and negative patients [ 18 ]. Further studies also focused on the identification of serological signatures of SARS-CoV-2 infection [ 19 ], the use of MALDI-TOF-MS, surface-enhanced Raman scattering (SERS), LC-MS, and MALDI-MS for the detection of SARS-CoV-2 in human saliva, nasal swabs, plasma and serum samples [ 20 , 21 , 22 , 23 ].…”

Hyperspectral dark field optical microscopy for orientational imaging of a single plasmonic nanocube using a physics-based learning method

“…For transforming the high dimensional spectral data into a data set with reduced dimension, we used an unsupervised statistical technique, the principal component analysis (PCA) [44,45]. We have previously used PCA approach for Raman spectral analysis [46,47]. PCA result (Figure 3f) showed that the first principal component (PC1) can distinguish between single nanoparticles (+PC1 values) and multiple particles (-PC1 values) in a hyperspectral image.…”

“…44,45 We have previously used PCA approach for Raman spectral analysis. 46,47 PCA result (Fig. 3f) showed that the rst principal component (PC1) can distinguish between single nanoparticles (+PC1 values) and multiple particles (ÀPC1 values) in a hyperspectral image.…”

Recent Trends in SERS-Based Plasmonic Sensors for Disease Diagnostics, Biomolecules Detection, and Machine Learning Techniques

“…The SVM outperformed other techniques in the identification of cyanobacteria, using SERS spectra of mutant and wild-type strains [ 388 ]. Using a dimensionality reduction technique, followed by a probabilistic ML model, SARS-CoV-2 identification was performed with an accuracy of ~85% [ 389 ]. SERS coupled with SVM was also used for the identification of lung cancers [ 96 ].…”