Evaluation of the AiDx Assist device for automated detection of Schistosoma eggs in stool and urine samples in Nigeria
Brice Meulah (Leiden University Medical Center, CERMEL)
Pytsje T. Hoekstra (Leiden University Medical Center)
Samuel Popoola (Aidx Medical BV)
Satyajith Jujjavarapu (Aidx Medical BV)
Moses Aderogba (The Ending Neglected Diseases (END) Fund)
Joseph O. Fadare (Ekiti State University)
John A. Omotayo (Ekiti State University)
David Bell (Independent Consultant)
Cornelis H. Hokke (Leiden University Medical Center)
Lisette van Lieshout (Leiden University Medical Center)
G.V. Vdovin (TU Delft - Team Raf Van de Plas)
Jan Carel Diehl (TU Delft - Design for Sustainability)
Temitope Agbana (Aidx Medical BV)
Louise Makau-Barasa (The Ending Neglected Diseases (END) Fund)
Jacob Solomon (Federal Ministry of Health)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
Introduction: Schistosomiasis is a public health concern and there is a need for reliable field-compatible diagnostic methods in endemic settings. The AiDx Assist, an artificial intelligence (AI)-based automated microscope, has shown promising results for the detection of Schistosoma haematobium eggs in urine. It has been further developed to detect Schistosoma mansoni eggs in stool.
Methods: In this study, we evaluated the performance of the AiDx Assist for the detection of S. mansoni eggs in stool samples and further validated the performance of the AiDx Assist for the detection of S. haematobium eggs in urine samples. Additionally, the potential of the AiDx Assist for the detection of other helminths in stool samples was explored. In total, 405 participants from an area endemic for both S. mansoni and S. haematobium provided stool and urine samples which were subjected to AiDx Assist (semi- and fully automated), while conventional microscopy was used as the diagnostic reference.
Results: Only samples with complete test results were included in the final analysis, resulting in 375 stool and 398 urine samples, of which 38.4% and 65.3% showed Schistosoma eggs by conventional microscopy. The collected images of the stool samples were retrospectively examined for other helminth eggs via manual analysis. For the detection of S. mansoni eggs, the sensitivity of the semi-automated AiDx Assist (86.8%) was significantly higher compared to the fully automated AiDx Assist (56.9%) while the specificity was comparable, with 81.4% and 86.8%, respectively. Retrospectively, eggs of Ascaris lumbricoides and Trichuris trichiura were visualized. For the examination of urine samples, a comparable sensitivity in the detection of S. haematobium eggs was found between the semi-and the fully automated modes of the AiDx Assist, showing 94.6% and 91.9%, respectively. Furthermore, the specificity was comparable, with 90.6%and 91.3% respectively.
Discussion: The AiDx Assist met the World Health Organization Target Product Profile criteria in terms of diagnostic accuracy for the detection of S. haematobium eggs in urine samples and performed modestly in the detection of S. mansoni eggs in stool samples. With some further improvements, it has the potential to become a valuable diagnostic tool for screening multiple helminth parasites in stool and urine samples.
help_outline