Performance Evaluation of the Schistoscope 5.0 for (Semi-)automated Digital Detection and Quantification of Schistosoma haematobium Eggs in Urine: A Field-based Study in Nigeria

Abstract

Conventional microscopy is the standard procedure for the diagnosis of schistosomiasis, despite its limited sensitivity, reliance on skilled personnel, and the fact that it is error prone. Here, we report the performance of the innovative (semi-)automated Schistoscope 5.0 for optical digital detection and quantification of Schistosoma haematobium eggs in urine, using conventional microscopy as the reference standard. At baseline, 487 participants in a rural setting in Nigeria were assessed, of which 166 (34.1%) tested S. haematobium positive by conventional microscopy. Captured images from the Schistoscope 5.0 were analyzed manually (semiautomation) and by an artificial intelligence (AI) algorithm (full automation). Semi- and fully automated digital microscopy showed comparable sensitivities of 80.1% (95% confidence interval [CI]: 73.2–86.0) and 87.3% (95% CI: 81.3–92.0), but a significant difference in specificity of 95.3% (95% CI: 92.4–97.4) and 48.9% (95% CI: 43.3–55.0), respectively. Overall, estimated egg counts of semi- and fully automated digital microscopy correlated significantly with the egg counts of conventional microscopy (r = 0.90 and r = 0.80, respectively, P < 0.001), although the fully automated procedure generally underestimated the higher egg counts. In 38 egg positive cases, an additional urine sample was examined 10 days after praziquantel treatment, showing a similar cure rate and egg reduction rate when comparing conventional microscopy with semiautomated digital microscopy. In this first extensive field evaluation, we found the semiautomated Schistoscope 5.0 to be a promising tool for the detection and monitoring of S. haematobium infection, although further improvement of the AI algorithm for full automation is required.

Diagnosis of Schistosomiasis without a Microscope: Evaluating Circulating Antigen (CCA, CAA) and DNA Detection Methods on Banked Samples of a Community-Based Survey from DR Congo

“…This suggests a possible co-infection of S. haematobium with S. mansoni , although these urine samples might also have been contaminated with stool and, therefore, could represent an S. mansoni infection only. Alternatively, ectopic egg elimination cannot be excluded, as S. mansoni eggs have been occasionally observed in urine as well [ 42 , 43 , 44 ]. In such cases, microscopy could have provided additional information concerning the Schistosoma species, provided that the infection intensity was sufficiently high.…”

Two-stage automated diagnosis framework for urogenital schistosomiasis in microscopy images from low-resource settings

“…In our previous work, we developed a low-cost automated digital microscope (Schistoscope V5.0) with AI for the detection S. haematobium eggs, 10 and we reported the results from a field validation study in Nigeria. 11 A U-Net model 25 trained with the S. haematobium dataset consisting of 5198 images captured from both clinical and spiked urine samples was used for automated egg detection. Although we achieved a high diagnostic sensitivity of 87.3%, the diagnostic specificity was low (48.9%).…”

“…The images were captured from urine samples collected in a rural area in central Nigeria with the Schistoscope V5.0. 11 The size of the captured images is 1520 × 2028 pixels. The details of the sample collection and preparation process are described in our previous works.…”

“…In this work, we address these challenges by first increasing the size of the S. haematobium (SH) dataset in our previous work 10 from 5198 to 12,051 images of clinical samples. 11 We carry out detection and counting of S. haematobium eggs present in each image by proposing a two-stage framework consisting of a DeepLabv3 with MobilenetV3 backbone deep convolutional neural network 12 trained on the SH dataset using a transfer learning approach. The second stage of our proposed framework is a refined segmentation and egg counting procedure, which adapts the region-based fitting of overlapping ellipses 13 to efficiently separate the boundaries of overlapping eggs in the image.…”

Reproducibility matters: intra- and inter-sample variation of the point-of-care circulating cathodic antigen test in two Schistosoma mansoni endemic areas in Uganda

“…Sensitivity can be improved by increasing the number of smears per stool and/ or stools sampled, but this increases logistical, temporal and financial costs. Artificial intelligence algorithms for automated or semiautomated identification of eggs enable faster reading, but the sensitivity does not yet outperform humans [10]. Mathematical models have informed easy-to-use tools to estimate true S. mansoni prevalence from observed Kato-Katz-based prevalence to improve interpretation of suboptimal sensitivity [11].…”