Name: BRUNA FERRO BRUN
Publication date: 07/10/2024
Examining board:
Name |
Role |
|---|---|
| GABRIELY SILVEIRA FOLLI | Examinador Interno |
| LEANDRO DOS SANTOS | Examinador Externo |
| LUIS FELIPE DAS CHAGAS E SILVA DE CARVALHO | Examinador Externo |
| MARA REJANE BARROSO BARCELOS | Examinador Interno |
| MARCIA HELENA CASSAGO NASCIMENTO | Presidente |
Summary: Fourier transform infrared spectroscopy (FTIR) has been an emerging technology in the medical field for the detection, screening and discrimination of different diseases. The concentration of D-dimer in plasma, a product of fibrin degradation during fibrinolysis, is an important marker in clinical evaluation, especially in hypercoagulable conditions. However, currently available tests for D-dimer analysis do not present satisfactory evaluation parameters. To date, the use of FTIR has not been investigated for its potential for the analysis of plasma D-dimer concentrations. Therefore, this study proposes a new methodological alternative, practical and effective, for the evaluation of patients according to D-dimer concentrations. Objective: Evaluate the ability to distinguish and quantify human plasma samples with different concentrations of D-dimer using FTIR associated with multivariate data analysis. Methods: One hundred plasma samples from patients admitted to a hospital intensive care unit were analyzed. After the standardization stage of sample preparation and spectral acquisition, the data were organized for statistical and multivariate pattern recognition analyses. For qualitative analysis, principal component analysis (PCA) was performed as an unsupervised method; linear discriminant analysis with genetic algorithm (GA-LDA) and partial least squares with linear discriminant analysis (PLS-DA) were performed as supervised methods. For quantitative analysis, the partial least squares (PLS) regression method was applied. For the analyses, a cutout was made in the fingerprint region of the spectra (1800 - ~1000 cm-1). Results: The GA-LDA method effectively classified patients according to the reference value of D-dimer (0.5 g/mL and >0.5 g/mL) with 87.5% specificity and 100% sensitivity in the training set, and 85.7% specificity and 95.6% sensitivity in the test set. However, the PLS method did not show overwhelming parameters of D-dimer concentration quantification. Conclusion: FTIR spectroscopy has been shown as a potential tool for patient classification according to D-dimer reference values. FTIR spectral analyses associated with clinical evidence may contribute to faster and more accurate medical diagnosis, reduce patient morbidity, and save resources and demands on professionals.
