Diagnóstico salivar do Enterovírus 71 por meio de espectroscopia ATR-FTIR acoplada a algoritmos de inteligência artificial

• Autor(a) principal: Alves, Deborah Cristina Teixeira
• Data de Publicação: 2024
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da UFU
• Texto Completo: https://repositorio.ufu.br/handle/123456789/41411; http://doi.org/10.14393/ufu.di.2024.5016
• topic: Enterovirus 71 humano; Espectroscopia Infravermelho por Transformada de Fourier; Saliva; Enterovirus 71, human; Fourier Transform Infrared Spectroscopy; CNPQ::CIENCIAS DA SAUDE::ODONTOLOGIA; Odontologia; ODS::ODS 9. Indústria, Inovação e infraestrutura - Construir infraestrutura resiliente, promover a industrialização inclusiva e sustentável, e fomentar a inovação.; ODS::ODS 3. Saúde e bem-estar - Assegurar uma vida saudável e promover o bem-estar para todos, em todas as idades.; ODS::ODS 12. Consumo e produção responsáveis - Assegurar padrões de produção e de consumo sustentáveis.

Enterovirus 71 (EV71) infection is associated with various diseases, with hand, foot, and mouth disease (HFMD) and herpangina being among the most common. These diseases are typically self-limiting and cause mild symptoms. However, due to the virus's neurotropism, severe complications can occur, such as aseptic meningitis, acute flaccid paralysis, and meningoencephalitis. EV71-related illnesses affect adults, adolescents, and children. HFMD outbreaks caused by EV71 often occur worldwide, especially in the Asia-Pacific region. Current diagnostic methods for EV71 viral detection include viral isolation, reverse transcription-polymerase chain reaction (RT-PCR), and neutralization testing. However, these techniques are time-consuming, involve toxic reagents, are costly, and require specific primers and specialized training. Despite the virus being present in the oropharynx, studies on using saliva as a diagnostic fluid for EV71 are limited. In this context, this study aimed to investigate the accuracy of a sustainable, rapid, and cost-effective technique based on Attenuated Total Reflectance Fourier Transform Infrared Spectroscopy (ATR-FTIR) to detect EV71 in different dilutions of human saliva. Saliva was collected from 10 healthy individuals, and each sample underwent tenfold serial dilution with EV71 starting from a concentration of 1x107 plaque-forming units per milliliter (PFU/mL). The dataset comprised 90 samples of infected or non-infected saliva. For infected saliva, 10 analyses were performed for each dilution: 1x107, 1x106, 1x105, 1x104, 1x103, 1x102, 10 e 1 PFU/mL, totaling 80 samples of infected saliva. A separate aliquot from each sample was set aside and used as a control, totaling 10 control saliva samples. ATR-FTIR coupled with four different artificial intelligence algorithms (Adaptive Boosting, Artificial Neural Networks, Random Forest, and Support Vector Machine) was used to detect EV71 diluted at eight different concentrations. The Adaptive Boosting (AdaBoost) algorithm achieved the best results with an accuracy of 70%, sensitivity of 70%, and specificity of up to 80% at a concentration of 1x107 PFU/mL. This proof-of-concept study demonstrated the potential of ATR-FTIR as a reagent-free, sustainable platform requiring an ultra-low saliva volume for detecting EV71 at concentrations greater than 1x107 PFU/mL.