A bioespectroscopia como ferramenta de triagem para a fibromialgia: melhorando o diagnóstico para potencializar a funcionalidade
The American College of Rheumatology (ACR) presented in 2010 a consensus for the diagnosis of fibromyalgia (FM). However, they can use an assessment and diagnosis guide as there are many cases of underdiagnosis or false diagnosis. This causes a lack of chemical, immunological markers or tests for...
Na minha lista:
| Autor principal: | |
|---|---|
| Outros Autores: | |
| Formato: | Dissertação |
| Idioma: | pt_BR |
| Publicado em: |
Universidade Federal do Rio Grande do Norte
|
| Assuntos: | |
| Endereço do item: | https://repositorio.ufrn.br/handle/123456789/30382 |
| Tags: |
Adicionar Tag
Sem tags, seja o primeiro a adicionar uma tag!
|
| Resumo: | The American College of Rheumatology (ACR) presented in 2010 a consensus for the diagnosis
of fibromyalgia (FM). However, they can use an assessment and diagnosis guide as there are
many cases of underdiagnosis or false diagnosis. This causes a lack of chemical, immunological
markers or tests for FM detection. This project aims to use biospectroscopy (spectroscopy
without infrared) and multivariate classification techniques as new technologies for the
identification of FM, using only blood plasma as the material of analysis. This is a crosssectional analytical study with 126 members divided into the Fibromyalgia group and the
control group. For all subjects, a sociodemographic questionnaire was applied, collected
clinical data on the impact of FM, investigated with pain intensity, anxiety levels, quality of
life; as well as a collection of 10 ml of blood from each participant. Analysis of their blood
plasma using attenuated total reflection Fourier-transform infrared (ATR-FTIR) spectroscopy
in conjunction with chemometric techniques, hence, providing a low-cost, fast and accurate
diagnostic approach. Different chemometric algorithms were tested to classify the spectral data;
genetic algorithm with linear discriminant analysis (GA-LDA) achieved the best diagnostic
results with a sensitivity of 89.5% in an external test set. The GA-LDA model identified 24
spectral wavenumbers responsible for class separation; amongst these, the Amide II (1545 cm1
) and proteins (1425 cm-1
) were identified to be discriminant features. Clinical data showed
significant difference between groups in FIQ (p = 0.0001), anxiety (p = 0.001), pain (p =
0.0001) and quality of life (p = 0.0001). These results reinforce the potential of ATR-FTIR
spectroscopy with multivariate analysis as a new tool to screen and detect patients with
fibromyalgia in a fast, low-cost, non-destructive and minimally invasive fashion. |
|---|