Parasites & Vectors (Aug 2024)
Laboratory validation of the automated diagnosis of intestinal parasites via fecal sample processing for the recovery of intestinal parasites through the dissolved air flotation technique
Abstract
Abstract Background Techniques for diagnosing intestinal parasites need technological advancements in the preanalytical (collection/processing) and analytical (detection) stages. The dissolved air flotation (DAF) technique effectively recovers parasites from processed feces for routine diagnosis. Artificial intelligence (AI) is a practical and affordable alternative to modernize the analysis stage of microscopy images and generates high efficiency in the parasitological examination of feces. Methods The objective of this study was to standardize a laboratory protocol for stool processing using the DAF technique in conjunction with an automated diagnosis of intestinal parasites (DAPI) system. A total of 400 samples were obtained to perform the tests with the use of DAF to verify the recovery of the parasites as a function of the chemical reagent (polymer and surfactant), the volume of the flotation tube, and standardization of smear assembly on a microscopy slide, with automated analysis by DAPI. The DAF protocol that obtained the most satisfactory results in terms of parasite recovery (P 0.05). The surfactants showed a range of parasite recoveries between 41.9% and 91.2% in the float supernatant. We obtained a maximum positivity of 73% of the assembled slides when we applied DAF processing with 7% CTAB surfactant and 57% positivity with the modified TF-Test technique. Regarding diagnostic performance, the TF-Test-modified and DAF techniques used in fecal processing for subsequent computerized analysis by AI presented sensitivities of 86% and 94%, with kappa agreements of 0.62 and 0.80 (substantial), respectively. Conclusions The DAF protocol defined in this study and the DAPI system are innovative processes for parasite recovery and fecal debris elimination that are favorable for effectively detecting pathogenic structures in laboratory diagnosis. Graphical Abstract
Keywords