Fast Screening of Patients with Altered D-Dimer Values Using Infrared Spectroscopy

18 Pages Posted: 16 Nov 2022

See all articles by Bruna Brun

Bruna Brun

Federal University of Espirito Santo

Márcia Nascimento

Federal University of Espirito Santo

Pedro Dias

Federal University of Espirito Santo

Wena Marcarini

Federal University of Espirito Santo

Maneesh Singh

Biocel UK Ltd

Paulo Roberto Filgueiras

Federal University of Espirito Santo

Paula Vassallo

Federal University of Minas Gerais (UFMG)

Wanderson Romão

Federal University of Espirito Santo

José Geraldo Mill

Federal University of Espirito Santo - Department of Physiological Sciences

Francis L. Martin

Biocel UK Ltd

Valerio Garrone Barauna

Federal University of Espírito Santo - Department of Physiological Science

Abstract

Fourier-transform infrared spectroscopy (FTIR) spectroscopy is an emerging technology in the medical field. Blood D-dimer was initially studied to access blood clots, but it is currently used in other medical fields, including recently during the COVID-19 pandemic. This study aimed to evaluate the use of FTIR spectroscopy with machine learning to predict and classify plasma D-dimer concentrations. The plasma FTIR spectra from 59 patients were studied through principal components analysis (PCA) and two supervised approaches: regression partial least squares analysis (PLS) and classification model by genetic algorithms with linear discriminant analysis (GA-LDA). The analyses were truncated to the fingerprint region of the spectra. The GA-LDA method effectively classified patients according to D-dimer cutoffs (≤0.5 µg/mL and >0.5 µg/mL) with 93.5% specificity and 100% sensitivity on the training set and 100% specificity and sensitivity on the test set. However, the PLS method was ineffective in predicting the value of D-dimer in plasma samples (R2 < 0.35). Thus, we demonstrate that FTIR spectroscopy might be an important additional tool for classifying patients according to D-dimer values. FTIR spectral analyses associated with clinical evidence can contribute to a faster and more accurate medical diagnosis, reduce patient morbidity, and save resources and demand for professionals.

Note:

Funding Information: The authors thank the research funding agencies CNPq (310349/2021, 401870/2020), CAPES and FAPES (003/2020, 300/2021, 113, 2021, 442/2021, 492/2021, 165/2021).

Declaration of Interests: The authors declare no conflicts of interest.

Ethics Approval Statement: This study was carried out in agreement with the Helsinki declaration. The Ethics Committee granted full ethical approval for the investigation at the University of Espírito Santo (#0993920.1.0000.5071 and #31411420.9.0000.8207).

Keywords: blood clot, D-dimer, FTIR spectroscopy, FTIR, machine learning, plasma

Suggested Citation

Brun, Bruna and Nascimento, Márcia and Dias, Pedro and Marcarini, Wena and Singh, Maneesh and Filgueiras, Paulo Roberto and Vassallo, Paula and Romão, Wanderson and Mill, José Geraldo and Martin, Francis L. and Barauna, Valerio Garrone, Fast Screening of Patients with Altered D-Dimer Values Using Infrared Spectroscopy. Available at SSRN: https://ssrn.com/abstract=4252296 or http://dx.doi.org/10.2139/ssrn.4252296

Bruna Brun

Federal University of Espirito Santo ( email )

Av. Fernando Ferrari, 514
No. 514, Goiabeiras
Vitória, 29075
Brazil

Márcia Nascimento

Federal University of Espirito Santo ( email )

Av. Fernando Ferrari, 514
No. 514, Goiabeiras
Vitória, 29075
Brazil

Pedro Dias

Federal University of Espirito Santo ( email )

Av. Fernando Ferrari, 514
No. 514, Goiabeiras
Vitória, 29075
Brazil

Wena Marcarini

Federal University of Espirito Santo ( email )

Av. Fernando Ferrari, 514
No. 514, Goiabeiras
Vitória, 29075
Brazil

Maneesh Singh

Biocel UK Ltd ( email )

Paulo Roberto Filgueiras

Federal University of Espirito Santo ( email )

Paula Vassallo

Federal University of Minas Gerais (UFMG) ( email )

BELO HORIZONTE
Brazil

Wanderson Romão

Federal University of Espirito Santo ( email )

José Geraldo Mill

Federal University of Espirito Santo - Department of Physiological Sciences ( email )

Francis L. Martin

Biocel UK Ltd ( email )

Valerio Garrone Barauna (Contact Author)

Federal University of Espírito Santo - Department of Physiological Science ( email )

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
49
Abstract Views
286
PlumX Metrics