Aplicação de espectroscopia ATR-FTIR suportada por algoritmos de inteligência artificial permite detecção de alterações moleculares no líquido sinovial em modelo animal de infecção pelo vírus Chikungunya

Abstract

Chikungunya virus (CHIKV) infection often manifests with symmetric arthralgia in peripheral joints, tenosynovitis, inflammation in joint ligaments, and approximately 60% of patients may develop chronic symptoms, directly impacting their quality of life. The correct diagnosis of CHIKV infection is complex and involves multi-step techniques with high-cost molecular biology reagents. Given the need to explore innovative approaches for CHIKV infection detection in synovial fluid, attenuated total reflectance Fourier-transform infrared (ATR-FTIR) spectroscopy without the use of reagents has emerged as a promising tool for identifying molecules related to the presence of viral particles in this fluid. Thus, this study aimed to identify alterations in functional groups of synovial fluid components using the ATR-FTIR platform supported by artificial intelligence algorithms to detect potential chondral and osteochondral lesions in an in vivo model of CHIKV infection. C57/BL6 mice (n=6) were inoculated with 20 µL (1 x 105 FFU of CHIKV), while control group mice (n=6) received 20 µL of placebo. Synovial fluid was collected 7 days after CHIKV infection, and the infrared spectra of the synovial fluid were analyzed supported by artificial neural network algorithm classification. The results showed a sensitivity of 66%, specificity of 100%, and accuracy of 83% in detecting alterations in the CHIKV infection model. Identifying vibrational modes at 1655 cm-1 , 1543 cm-1 , and 1060 cm-1 related to lipids could be used to discriminate synovial fluid from CHIKV-infected mice compared to controls. Thus, this study demonstrated the potential of the biophotonic platform to advance the application of ATR-FTIR spectroscopy in orthopedic diseases, pioneering in obtaining infrared spectra of synovial fluid. These findings highlight the potential of ATR-FTIR platforms supported by artificial neural network algorithms as a sustainable, reagent-free, non-invasive, and rapid tool for detecting alterations induced by CHIKV infection in synovial fluid