Dataset Information

A machine learning model of microscopic agglutination test for diagnosis of leptospirosis

ABSTRACT: Leptospirosis is a zoonosis caused by the pathogenic bacterium Leptospira. The Microscopic Agglutination Test (MAT) is widely used as the gold standard for diagnosis of leptospirosis. In this method, diluted patient serum is mixed with serotype-determined Leptospires, and the presence or absence of aggregation is determined under a dark-field microscope to calculate the antibody titer. Problems of the current MAT method are 1) a requirement of examining many specimens per sample, and 2) a need of distinguishing contaminants from true aggregates to accurately identify positivity. Therefore, increasing efficiency and accuracy are the key to refine MAT. It is possible to achieve efficiency and standardize accuracy at the same time by automating the decision-making process. In this study, we built an automatic identification algorithm of MAT using a machine learning method to determine agglutination within microscopic images. The machine learned the features from 316 positive and 230 negative MAT images created with sera of Leptospira-infected (positive) and non-infected (negative) hamsters, respectively. In addition to the acquired original images, wavelet-transformed images were also considered as features. We utilized a support vector machine (SVM) as a proposed decision method. We validated the trained SVMs with 210 positive and 154 negative images. When the features were obtained from original or wavelet-transformed images, all negative images were misjudged as positive, and the classification performance was very low with sensitivity of 1 and specificity of 0. In contrast, when the histograms of wavelet coefficients were used as features, the performance was greatly improved with sensitivity of 0.99 and specificity of 0.99. We confirmed that the current algorithm judges the positive or negative of agglutinations in MAT images and gives the further possibility of automatizing MAT procedure.

SUBMITTER: Oyamada Y

PROVIDER: S-EPMC8594833 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundLeptospirosis is diagnosed on clinical grounds, and confirmed by microscopic agglutination test (MAT). IgM-ELISA (Serion-Virion) and immunochromatography test (Leptocheck-WB) are two immunodiagnostic assays for leptospirosis. Their sensitivity, specificity and applicability in Sri Lanka have not been systematically evaluated.MethodsClinically diagnosed leptospirosis patients (n = 919) were recruited from three hospitals in the Western Province of Sri Lanka, during June 2012 to December 2013. MAT, IgM-ELISA and Leptocheck-WB were performed on all patient sera. MAT titer of ? 400 in single sample, four-fold rise or seroconversion ? 100 in paired samples were considered as positive for MAT. For diagnostic confirmation, MAT was performed during both acute and convalescent phases. Anti-leptospiral IgM ? 20 IU/ml and appearance of a band in the test window were considered as positive for IgM-ELISA and Leptocheck-WB test respectively. Patients with an alternative diagnosis (n = 31) were excluded. Data analysis was performed using two methods, i) considering MAT as reference standard and ii) using Bayesian latent class model analysis (BLCM) which considers each test as imperfect.ResultsMAT, IgM-ELISA and Leptocheck-WB positivity were 39.8%, 45.8% and 38.7% respectively during the acute phase. Acute-phase MAT had specificity and sensitivity of 95.7% and 55.3% respectively, when compared to overall MAT positivity. IgM-ELISA and Leptocheck-WB had similar diagnostic sensitivity when compared with acute-phase MAT as the gold standard, although IgM-ELISA showed higher specificity (84.5%) than Leptocheck-WB (73.3%). BLCM analysis showed that IgM-ELISA and Leptocheck-WB had similar sensitivities (86.0% and 87.4%), while acute-phase MAT had the lowest sensitivity (77.4%). However, acute-phase MAT had high specificity (97.6%), while IgM-ELISA and Leptocheck-WB showed similar but lower specificity (84.5% and 82.9%).ConclusionsBoth IgM-ELISA and Leptocheck-WB shows similar sensitivities and specificities. IgM-ELISA may be superior to MAT during the acute phase and suitable for early diagnosis of leptospirosis. Leptocheck-WB is suitable as a rapid immunodiagnostic screening test for resource limited settings.

Project description:Visceral leishmaniasis (VL), the most severe form of leishmaniasis, is endemic in Europe with Mediterranean countries reporting endemic status alongside a worrying northward spread. Serological diagnosis, including immunochromatographic test based on the recombinant antigen rK39 (rK39-ICT) and a direct agglutination test (DAT) based on the whole parasite antigen, have been validated in regions with high VL burden, such as eastern Africa and the Indian subcontinent. To date, no studies using a large set of patients have performed an assessment of both methods within Europe.We selected a range of clinical serum samples from patients with confirmed VL (including HIV co-infection), Chagas disease, malaria, other parasitic infections and negative samples (n = 743; years 2009-2015) to test the performance of rK39-ICT rapid test (Kalazar Detect Rapid Test; InBios International, Inc., USA) and DAT (ITM-DAT/VLG; Institute of Tropical Medicine Antwerp, Belgium). An in-house immunofluorescence antibody test (IFAT), was included for comparison. Estimated sensitivities for rK39-ICT and DAT in HIV-negative VL patients were 83.1% [75.1-91.2] and 84.2% [76.3-92.1], respectively. Sensitivity was reduced to 67.3% [52.7-82.0] for rK39 and increased to 91.3% [82.1-100.0] for DAT in HIV/VL co-infected patients. The in-house IFAT was more sensitive in HIV-negative VL patients, 84.2% [76.3-92.1] than in HIV/VL patients, 79.4% [73.3-96.2]. DAT gave 32 false positives in sera from HIV-negative VL suspects, compared to 0 and 2 for rK39 and IFAT, respectively, but correctly detected more HIV/VL patients (42/46) than rK39 (31/46) and IFAT (39/46).Though rK39-ICT and DAT exhibited acceptable sensitivity and specificity a combination with other tests is required for highly sensitive diagnosis of VL cases in Spain. Important variation in the performance of the tests were seen in patients co-infected with HIV or with other parasitic infections. This study can help inform the choice of serological test to be used when screening or diagnosing VL in a European Mediterranean setting.

Dataset Information

A machine learning model of microscopic agglutination test for diagnosis of leptospirosis

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets