Avaliação e otimização das leituras do teste imunocromatográfico Point-of-Care (POC-CCA) realizado em uma área de alta endemicidade no nordeste brasileiro para esquistossomose, mediante análise de imagens

• Ano de defesa: 2024
• Autor(a) principal: Morais, Francisca Janaina Damasceno
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Palavras-chave em Português: Cromatografia de Afinidade; Esquistossomose; Diagnóstico; Schistosoma mansoni; Chromatography, Affinity; Schistosomiasis; Diagnosis; CNPQ::CIENCIAS DA SAUDE::MEDICINA::ANATOMIA PATOLOGICA E PATOLOGIA CLINICA
• Link de acesso: http://repositorio.ufc.br/handle/riufc/79135

Schistosomiasis, caused by the parasite Schistosoma mansoni, is a Neglected Tropical Disease with a high impact in underdeveloped countries. Kato-Katz (KK) is the method recommended by the World Health Organization, however it has low sensitivity in areas of low endemicity. POC-CCA, an immunochromatographic test that allows the detection of parasitic antigens in urine, has been presented as a sensitive and rapid alternative. However, such a test is subject to inter-reader variations in the interpretation of the visual reading of the results, especially when the intensity of the response to the test is weak (trace), which may influence the determination of prevalence. Therefore, the objective of this study was to quantify the coloration of POC-CCA test results to reduce the subjectivity of the reading. For this, a database with 158 images from POC-CCA tests was used, carried out with material from an area of high endemicity for schistosomiasis. The image analysis computer program ImageJ® is a public Java-based image processing software available online that was used to quantify the color intensity pixels of the results. The POC-CCA reading strategy by pixel quantification showed good performance with satisfactory power to discriminate positive and negative results (area under the ROC curve equal to 0.859). The ideal cutoff point was 0.042 pixels, which corresponded to Youden's J maximum of 0.656. Under these conditions, the ideal sensitivity was 0.779 and the ideal specificity was 0.877. Compared to the results of Kato-Katz and other POC-CCA reading strategies, pixel quantification showed prevalences of positives and negatives similar to those of the Kato-Katz method and visual qualitative reading. However, the prevalence of positives was lower than that observed with the G Score reading strategy (G1 to G10). A weak positive correlation (r2 = 0.4812), but statistically significant (p< 0.0001), was identified between the OPG results of the Kato-Katz test and the pixel quantification of the POC-CCA test. A good agreement was observed between the reading strategy by pixel quantification with the Kato- test (Kappa value = 0.6572, p< 0.0001), as well as with the reading strategies stratified by G Score (Kappa value = 0.5498, p< 0.0001) and visual qualitative (Kappa value = 0.5383, p< 0.0001). Although the image processing technique has shown satisfactory performance, there is potential for improvements in usability and execution, including optimization of parameters and signal cutoff metrics. It is concluded that the visual interpretation of the POC-CCA test is subjective, however, pixel quantification offers an objective reading, eliminating subjectivity and allowing a more precise cut to determine the results.