Detection of SARS-CoV-2 from raman spectroscopy data using machine learning models

Abstract

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a member of the coronaviruses that caused the COVID-19 pandemic. The pathogenic SARS-CoV-2 virus can act as a miRNA sponge to lower cellular miRNA levels, making it a more dangerous human coronavirus. Diagnostic testing of the virus is intended to identify current infection in individuals and is performed when a person exhibits symptoms that are compatible with COVID-19. In this work, machine learning models (artificial neural network, decision tree, and support vector machine) are used to classify Raman spectroscopy samples as healthy or infected with SARS-CoV-2. The aim of the work is to introduce an alternative method for detecting SARS-CoV-2. The accuracy of the artificial neural network, the support vector machine and the decision tree were 94%, 90%, and 87%, respectively. The algorithms produced evidence of high recall and specificity. Hence, integrating Raman spectroscopy with machine learning has the potential to serve as an alternative diagnostic tool.