skip to main content

Title: Identification of Snails and Schistosoma of Medical Importance via Convolutional Neural Networks: A Proof-of-Concept Application for Human Schistosomiasis
In recent decades, computer vision has proven remarkably effective in addressing diverse issues in public health, from determining the diagnosis, prognosis, and treatment of diseases in humans to predicting infectious disease outbreaks. Here, we investigate whether convolutional neural networks (CNNs) can also demonstrate effectiveness in classifying the environmental stages of parasites of public health importance and their invertebrate hosts. We used schistosomiasis as a reference model. Schistosomiasis is a debilitating parasitic disease transmitted to humans via snail intermediate hosts. The parasite affects more than 200 million people in tropical and subtropical regions. We trained our CNN, a feed-forward neural network, on a limited dataset of 5,500 images of snails and 5,100 images of cercariae obtained from schistosomiasis transmission sites in the Senegal River Basin, a region in western Africa that is hyper-endemic for the disease. The image set included both images of two snail genera that are relevant to schistosomiasis transmission – that is, Bulinus spp. and Biomphalaria pfeifferi – as well as snail images that are non-component hosts for human schistosomiasis. Cercariae shed from Bi. pfeifferi and Bulinus spp. snails were classified into 11 categories, of which only two, S. haematobium and S. mansoni , are major etiological agents more » of human schistosomiasis. The algorithms, trained on 80% of the snail and parasite dataset, achieved 99% and 91% accuracy for snail and parasite classification, respectively, when used on the hold-out validation dataset – a performance comparable to that of experienced parasitologists. The promising results of this proof-of-concept study suggests that this CNN model, and potentially similar replicable models, have the potential to support the classification of snails and parasite of medical importance. In remote field settings where machine learning algorithms can be deployed on cost-effective and widely used mobile devices, such as smartphones, these models can be a valuable complement to laboratory identification by trained technicians. Future efforts must be dedicated to increasing dataset sizes for model training and validation, as well as testing these algorithms in diverse transmission settings and geographies. « less
Authors:
; ; ; ; ; ; ; ; ; ; ; ;
Award ID(s):
2024385 2024383 2011179
Publication Date:
NSF-PAR ID:
10285676
Journal Name:
Frontiers in Public Health
Volume:
9
ISSN:
2296-2565
Sponsoring Org:
National Science Foundation
More Like this
  1. Temperature constrains the transmission of many pathogens. Interventions that target temperature-sensitive life stages, such as vector control measures that kill intermediate hosts, could shift the thermal optimum of transmission, thereby altering seasonal disease dynamics and rendering interventions less effective at certain times of the year and with global climate change. To test these hypotheses, we integrated an epidemiological model of schistosomiasis with empirically determined temperature-dependent traits of the human parasiteSchistosoma mansoniand its intermediate snail host (Biomphalariaspp.). We show that transmission risk peaks at 21.7 °C (Topt), and simulated interventions targeting snails and free-living parasite larvae increasedToptby up to 1.3 °Cmore »because intervention-related mortality overrode thermal constraints on transmission. ThisToptshift suggests that snail control is more effective at lower temperatures, and global climate change will increase schistosomiasis risk in regions that move closer toTopt. Considering regional transmission phenologies and timing of interventions when local conditions approachToptwill maximize human health outcomes.

    « less
  2. Secor, W. Evan (Ed.)
    Schistosome parasites infect more than 200 million people annually, mostly in sub-Saharan Africa, where people may be co-infected with more than one species of the parasite. Infection risk for any single species is determined, in part, by the distribution of its obligate intermediate host snail. As the World Health Organization reprioritizes snail control to reduce the global burden of schistosomiasis, there is renewed importance in knowing when and where to target those efforts, which could vary by schistosome species. This study estimates factors associated with schistosomiasis risk in 16 villages located in the Senegal River Basin, a region hyperendemic formore »Schistosoma haematobium and S . mansoni . We first analyzed the spatial distributions of the two schistosomes’ intermediate host snails ( Bulinus spp. and Biomphalaria pfeifferi , respectively) at village water access sites. Then, we separately evaluated the relationships between human S . haematobium and S . mansoni infections and (i) the area of remotely-sensed snail habitat across spatial extents ranging from 1 to 120 m from shorelines, and (ii) water access site size and shape characteristics. We compared the influence of snail habitat across spatial extents because, while snail sampling is traditionally done near shorelines, we hypothesized that snails further from shore also contribute to infection risk. We found that, controlling for demographic variables, human risk for S . haematobium infection was positively correlated with snail habitat when snail habitat was measured over a much greater radius from shore (45 m to 120 m) than usual. S . haematobium risk was also associated with large, open water access sites. However, S . mansoni infection risk was associated with small, sheltered water access sites, and was not positively correlated with snail habitat at any spatial sampling radius. Our findings highlight the need to consider different ecological and environmental factors driving the transmission of each schistosome species in co-endemic landscapes.« less
  3. Background: Schistosomiasis is an emerging disease associated with changes to the environment that have increased human contact rates with disease-causing parasites, flatworms that are released from freshwater snails. For example, schistosomiasis remains a major public health problem in Northern Senegal, where prevalence in schoolchildren often reaches 90%. Aim: This study focuses on the impact of seasonality on the risk of human exposure (RHE) to Schistosoma mansoni, defined as the total number of cercariae (the free-living life stage that infects humans) shed from all Biomphalaria pfeifferi snails collected at a site using standardized methods. We focus on RHE because it ismore »rarely quantified and a recent study demonstrated that snails stop shedding cercariae when snail densities increase and thus per capita snail resources become limited [2], suggesting that densities of snails might not be directly proportional to RHE to schistosomes. Method: We sampled four water access points in three villages every other week during the early (Dry1) and later dry seasons (Dry2) and the rainy season, quantifying the abundance of infected and non-infected snail intermediate hosts, cercariae released per infected snail, and water chemistry. We used simple and multiple linear regressions to assess how seasonality and environmental parameters affect non-infected and infected snail abundance and RHE. Results: Although RHE was found across all seasons, the abundance of infected and non-infected snail intermediate hosts and cercariae, as well as prevalence (23.71%), were all highest in the rainy season. In the rainy season, RHE was positively associated with the density of snail hosts and their periphyton food resource. Conclusion: Although previous studies have examined the influence of seasonality on snail densities, few studies have explored the effects of seasonality on cercarial densities, which is the primary source of infection to humans. Our study demonstrates that cercarial densities are greater in the rainy season than in the early or late dry seasons. Given that cercarial densities directly pose risk of infection to humans, unlike non-infected or infected snails, these finding should help to inform decision making and schistosomiasis control efforts in West Africa.« less
  4. Schistosomiasis is a debilitating parasitic disease of poverty that affects more than 200 million people worldwide, mostly in sub-Saharan Africa, and is clearly associated with the construction of dams and water resource management infrastructure in tropical and subtropical areas. Changes to hydrology and salinity linked to water infrastructure development may create conditions favorable to the aquatic vegetation that is suitable habitat for the intermediate snail hosts of schistosome parasites. With thousands of small and large water reservoirs, irrigation canals, and dams developed or under construction in Africa, it is crucial to accurately assess the spatial distribution of high-risk environments thatmore »are habitat for freshwater snail intermediate hosts of schistosomiasis in rapidly changing ecosystems. Yet, standard techniques for monitoring snails are labor-intensive, time-consuming, and provide information limited to the small areas that can be manually sampled. Consequently, in low-income countries where schistosomiasis control is most needed, there are formidable challenges to identifying potential transmission hotspots for targeted medical and environmental interventions. In this study, we developed a new framework to map the spatial distribution of suitable snail habitat across large spatial scales in the Senegal River Basin by integrating satellite data, high-definition, low-cost drone imagery, and an artificial intelligence (AI)-powered computer vision technique called semantic segmentation. A deep learning model (U-Net) was built to automatically analyze high-resolution satellite imagery to produce segmentation maps of aquatic vegetation, with a fast and robust generalized prediction that proved more accurate than a more commonly used random forest approach. Accurate and up-to-date knowledge of areas at highest risk for disease transmission can increase the effectiveness of control interventions by targeting habitat of disease-carrying snails. With the deployment of this new framework, local governments or health actors might better target environmental interventions to where and when they are most needed in an integrated effort to reach the goal of schistosomiasis elimination.« less
  5. Helminths are parasites that cause disease at considerable cost to public health and present a risk for emergence as novel human infections. Although recent research has elucidated characteristics conferring a propensity to emergence in other parasite groups (e.g. viruses), the understanding of factors associated with zoonotic potential in helminths remains poor. We applied an investigator-directed learning algorithm to a global dataset of mammal helminth traits to identify factors contributing to spillover of helminths from wild animal hosts into humans. We characterized parasite traits that distinguish between zoonotic and non-zoonotic species with 91% accuracy. Results suggest that helminth traits relating tomore »transmission (e.g. definitive and intermediate hosts) and geography (e.g. distribution) are more important to discriminating zoonotic from non-zoonotic species than morphological or epidemiological traits. Whether or not a helminth causes infection in companion animals (cats and dogs) is the most important predictor of propensity to cause human infection. Finally, we identified helminth species with high modelled propensity to cause zoonosis (over 70%) that have not previously been considered to be of risk. This work highlights the importance of prioritizing studies on the transmission of helminths that infect pets and points to the risks incurred by close associations with these animals. This article is part of the theme issue ‘Infectious disease macroecology: parasite diversity and dynamics across the globe’.« less