Rapid diagnosis of COVID-19 using FT-IR ATR spectroscopy and machine learning

Marcelo Saito Nogueira; Leonardo Barbosa Leal; Wena Macarini; Raquel Lemos Pimentel; Matheus Muller; Paula Frizera Vassallo; Luciene Cristina Gastalho Campos; Leonardo dos Santos; Wilson Barros Luiz; José Geraldo Mill; Valerio Garrone Barauna; Luis Felipe das Chagas e.Silva de Carvalho

doi:10.1038/s41598-021-93511-2

Rapid diagnosis of COVID-19 using FT-IR ATR spectroscopy and machine learning

Marcelo Saito Nogueira
, Leonardo Barbosa Leal
, Wena Macarini
, Raquel Lemos Pimentel
, Matheus Muller
, Paula Frizera Vassallo
, Luciene Cristina Gastalho Campos
, Leonardo dos Santos
, Wilson Barros Luiz
, José Geraldo Mill
, Valerio Garrone Barauna
, Luis Felipe das Chagas e.Silva de Carvalho

University College Cork
Universidade Federal do Espírito Santo
Faculdade Vale do Cricaré
Universidade Federal de Minas Gerais
Universidade Estadual de Santa Cruz
Universidade de Taubaté
Centro Universitario Braz Cubas

Research output: Contribution to journal › Article › peer-review

Abstract

Early diagnosis of COVID-19 in suspected patients is essential for contagion control and damage reduction strategies. We investigated the applicability of attenuated total reflection (ATR) Fourier transform infrared (FTIR) spectroscopy associated with machine learning in oropharyngeal swab suspension fluid to predict COVID-19 positive samples. The study included samples of 243 patients from two Brazilian States. Samples were transported by using different viral transport mediums (liquid 1 or 2). Clinical COVID-19 diagnosis was performed by the RT-PCR. We built a classification model based on partial least squares (PLS) associated with cosine k-nearest neighbours (KNN). Our analysis led to 84% and 87% sensitivity, 66% and 64% specificity, and 76.9% and 78.4% accuracy for samples of liquids 1 and 2, respectively. Based on this proof-of-concept study, we believe this method could offer a simple, label-free, cost-effective solution for high-throughput screening of suspect patients for COVID-19 in health care centres and emergency departments.

Original language	English
Article number	15409
Journal	Scientific Reports
Volume	11
Issue number	1
DOIs	https://doi.org/10.1038/s41598-021-93511-2
Publication status	Published - Dec 2021
Externally published	Yes

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Access to Document

10.1038/s41598-021-93511-2

Cite this

Nogueira, M. S., Leal, L. B., Macarini, W., Pimentel, R. L., Muller, M., Vassallo, P. F., Campos, L. C. G., dos Santos, L., Luiz, W. B., Mill, J. G., Barauna, V. G., & de Carvalho, L. F. D. C. E. S. (2021). Rapid diagnosis of COVID-19 using FT-IR ATR spectroscopy and machine learning. Scientific Reports, 11(1), Article 15409. https://doi.org/10.1038/s41598-021-93511-2

@article{a123935391d645068517b4c627b8e0ab,

title = "Rapid diagnosis of COVID-19 using FT-IR ATR spectroscopy and machine learning",

abstract = "Early diagnosis of COVID-19 in suspected patients is essential for contagion control and damage reduction strategies. We investigated the applicability of attenuated total reflection (ATR) Fourier transform infrared (FTIR) spectroscopy associated with machine learning in oropharyngeal swab suspension fluid to predict COVID-19 positive samples. The study included samples of 243 patients from two Brazilian States. Samples were transported by using different viral transport mediums (liquid 1 or 2). Clinical COVID-19 diagnosis was performed by the RT-PCR. We built a classification model based on partial least squares (PLS) associated with cosine k-nearest neighbours (KNN). Our analysis led to 84\% and 87\% sensitivity, 66\% and 64\% specificity, and 76.9\% and 78.4\% accuracy for samples of liquids 1 and 2, respectively. Based on this proof-of-concept study, we believe this method could offer a simple, label-free, cost-effective solution for high-throughput screening of suspect patients for COVID-19 in health care centres and emergency departments.",

author = "Nogueira, \{Marcelo Saito\} and Leal, \{Leonardo Barbosa\} and Wena Macarini and Pimentel, \{Raquel Lemos\} and Matheus Muller and Vassallo, \{Paula Frizera\} and Campos, \{Luciene Cristina Gastalho\} and \{dos Santos\}, Leonardo and Luiz, \{Wilson Barros\} and Mill, \{Jos{\'e} Geraldo\} and Barauna, \{Valerio Garrone\} and \{de Carvalho\}, \{Luis Felipe das Chagas e.Silva\}",

note = "Publisher Copyright: {\textcopyright} 2021, The Author(s).",

year = "2021",

month = dec,

doi = "10.1038/s41598-021-93511-2",

language = "English",

volume = "11",

journal = "Scientific Reports",

issn = "2045-2322",

number = "1",

}

TY - JOUR

T1 - Rapid diagnosis of COVID-19 using FT-IR ATR spectroscopy and machine learning

AU - Nogueira, Marcelo Saito

AU - Leal, Leonardo Barbosa

AU - Macarini, Wena

AU - Pimentel, Raquel Lemos

AU - Muller, Matheus

AU - Vassallo, Paula Frizera

AU - Campos, Luciene Cristina Gastalho

AU - dos Santos, Leonardo

AU - Luiz, Wilson Barros

AU - Mill, José Geraldo

AU - Barauna, Valerio Garrone

AU - de Carvalho, Luis Felipe das Chagas e.Silva

PY - 2021/12

Y1 - 2021/12

N2 - Early diagnosis of COVID-19 in suspected patients is essential for contagion control and damage reduction strategies. We investigated the applicability of attenuated total reflection (ATR) Fourier transform infrared (FTIR) spectroscopy associated with machine learning in oropharyngeal swab suspension fluid to predict COVID-19 positive samples. The study included samples of 243 patients from two Brazilian States. Samples were transported by using different viral transport mediums (liquid 1 or 2). Clinical COVID-19 diagnosis was performed by the RT-PCR. We built a classification model based on partial least squares (PLS) associated with cosine k-nearest neighbours (KNN). Our analysis led to 84% and 87% sensitivity, 66% and 64% specificity, and 76.9% and 78.4% accuracy for samples of liquids 1 and 2, respectively. Based on this proof-of-concept study, we believe this method could offer a simple, label-free, cost-effective solution for high-throughput screening of suspect patients for COVID-19 in health care centres and emergency departments.

AB - Early diagnosis of COVID-19 in suspected patients is essential for contagion control and damage reduction strategies. We investigated the applicability of attenuated total reflection (ATR) Fourier transform infrared (FTIR) spectroscopy associated with machine learning in oropharyngeal swab suspension fluid to predict COVID-19 positive samples. The study included samples of 243 patients from two Brazilian States. Samples were transported by using different viral transport mediums (liquid 1 or 2). Clinical COVID-19 diagnosis was performed by the RT-PCR. We built a classification model based on partial least squares (PLS) associated with cosine k-nearest neighbours (KNN). Our analysis led to 84% and 87% sensitivity, 66% and 64% specificity, and 76.9% and 78.4% accuracy for samples of liquids 1 and 2, respectively. Based on this proof-of-concept study, we believe this method could offer a simple, label-free, cost-effective solution for high-throughput screening of suspect patients for COVID-19 in health care centres and emergency departments.

UR - https://www.scopus.com/pages/publications/85116751090

U2 - 10.1038/s41598-021-93511-2

DO - 10.1038/s41598-021-93511-2

M3 - Article

C2 - 34635702

AN - SCOPUS:85116751090

SN - 2045-2322

VL - 11

JO - Scientific Reports

JF - Scientific Reports

IS - 1

M1 - 15409

ER -

Rapid diagnosis of COVID-19 using FT-IR ATR spectroscopy and machine learning

Abstract

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this