Dataset Information

Identification of the high-risk area for schistosomiasis transmission in China based on information value and machine learning: a newly data-driven modeling attempt.

ABSTRACT:

Background

Schistosomiasis control is striving forward to transmission interruption and even elimination, evidence-lead control is of vital importance to eliminate the hidden dangers of schistosomiasis. This study attempts to identify high risk areas of schistosomiasis in China by using information value and machine learning.

Methods

The local case distribution from schistosomiasis surveillance data in China between 2005 and 2019 was assessed based on 19 variables including climate, geography, and social economy. Seven models were built in three categories including information value (IV), three machine learning models [logistic regression (LR), random forest (RF), generalized boosted model (GBM)], and three coupled models (IV + LR, IV + RF, IV + GBM). Accuracy, area under the curve (AUC), and F1-score were used to evaluate the prediction performance of the models. The optimal model was selected to predict the risk distribution for schistosomiasis.

Results

There is a more prone to schistosomiasis epidemic provided that paddy fields, grasslands, less than 2.5 km from the waterway, annual average temperature of 11.5-19.0 °C, annual average rainfall of 1000-1550 mm. IV + GBM had the highest prediction effect (accuracy = 0.878, AUC = 0.902, F1 = 0.920) compared with the other six models. The results of IV + GBM showed that the risk areas are mainly distributed in the coastal regions of the middle and lower reaches of the Yangtze River, the Poyang Lake region, and the Dongting Lake region. High-risk areas are primarily distributed in eastern Changde, western Yueyang, northeastern Yiyang, middle Changsha of Hunan province; southern Jiujiang, northern Nanchang, northeastern Shangrao, eastern Yichun in Jiangxi province; southern Jingzhou, southern Xiantao, middle Wuhan in Hubei province; southern Anqing, northwestern Guichi, eastern Wuhu in Anhui province; middle Meishan, northern Leshan, and the middle of Liangshan in Sichuan province.

Conclusions

The risk of schistosomiasis transmission in China still exists, with high-risk areas relatively concentrated in the coastal regions of the middle and lower reaches of the Yangtze River. Coupled models of IV and machine learning provide for effective analysis and prediction, forming a scientific basis for evidence-lead surveillance and control.

SUBMITTER: Gong YF

PROVIDER: S-EPMC8237418 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundSchistosomiasis is a neglected tropical disease endemic in 54 countries. A major Schistosoma species, Schistosoma mansoni, is sustained via a life cycle that includes both human and snail hosts. Mathematical models of S. mansoni transmission, used to elucidate the complexities of the transmission cycle and estimate the impact of intervention efforts, often focus primarily on the human host. However, S. mansoni incurs physiological costs in snails that vary with the age of the snail when first infected. Snail demography and the age of snail infection could thus affect the force of infection experienced by humans, which is frequently used to predict the impact of various control strategies.MethodsTo address how these snail host and parasite interactions influence model predictions, we developed deterministic models of schistosomiasis transmission that include varying complexity in the snail population age structure. Specifically, we examined how model outputs, such as schistosome prevalence in human and snail populations, respond to the inclusion of snail age structure.ResultsOur models suggest that snail population age structure modifies the force of infection experienced by humans and the relationship between snail infection prevalence and corresponding human infection prevalence. There are significant differences in estimated snail infection, cercarial density and mean worm burden between models without snail population dynamics and those with snail populations, and between models with a homogeneous snail population and those with age stratification. The variation between finely age-stratified snail populations and those grouped into only juvenile and adult life stages is, however, minimal.ConclusionsThese results indicate that including snails and snail age structure in a schistosomiasis transmission model alters the relationship between snail and human infection prevalence. This highlights the importance of accounting for a heterogeneous intermediate host population in models of schistosomiasis transmission where the impact of proposed control measures is being considered.

Project description:Schistosomiasis is a parasitic disease and its distribution, in space and time, can be influenced by environmental factors such as rivers, elevation, slope, land surface temperature, land use/cover and rainfall. The aim of this study is to identify the areas with suitable conditions for schistosomiasis transmission on the basis of physical and environmental factors derived from satellite imagery and spatial analysis for Akure North Local Government Area (LGA) of Ondo State. Nigeria. This was done through methodology multicriteria evaluation (MCE) using Saaty's analytical hierarchy process (AHP). AHP is a multi-criteria decision method that uses hierarchical structures to represent a problem and makes decisions based on priority scales. In this research AHP was used to obtain the mapping weight or importance of each individual schistosomiasis risk factor. For the purpose of identifying areas of schistosomiasis risk, this study focused on temperature, drainage, elevation, rainfall, slope and land use/land cover as the factors controlling schistosomiasis incidence in the study area. It is by reclassifying and overlaying these factors that areas vulnerable to schistosomiasis were identified. The weighted overlay analysis was done after each factor was given the appropriate weight derived through the analytical hierarchical process. The prevalence of urinary schistosomiasis in the study area was also determined by parasitological analysis of urine samples collected through random sampling. The results showed varying risk of schistosomiasis with a larger portion of the area (82%) falling under the high and very high risk category. The study also showed that one community (Oba Ile) had the lowest risk of schistosomiasis while the risk increased in the four remaining communities (Iju, Igoba, Ita Ogbolu and Ogbese). The predictions made by the model correlated strongly with observations from field study. The high risk zones corresponded to known endemic communities. This study revealed that environmental factors can be used in identifying and predicting the transmission of schistosomiasis as well as effective monitoring of disease risk in newly established rural and agricultural communities.

Project description:BackgroundSchistosomiasis is the most widespread water-based disease in sub-Saharan Africa. Transmission is governed by the spatial distribution of specific freshwater snails that act as intermediate hosts and human water contact patterns. Remote sensing data have been utilized for spatially explicit risk profiling of schistosomiasis. We investigated the potential of remote sensing to characterize habitat conditions of parasite and intermediate host snails and discuss the relevance for public health.MethodologyWe employed high-resolution remote sensing data, environmental field measurements, and ecological data to model environmental suitability for schistosomiasis-related parasite and snail species. The model was developed for Burkina Faso using a habitat suitability index (HSI). The plausibility of remote sensing habitat variables was validated using field measurements. The established model was transferred to different ecological settings in Côte d'Ivoire and validated against readily available survey data from school-aged children.Principal findingsEnvironmental suitability for schistosomiasis transmission was spatially delineated and quantified by seven habitat variables derived from remote sensing data. The strengths and weaknesses highlighted by the plausibility analysis showed that temporal dynamic water and vegetation measures were particularly useful to model parasite and snail habitat suitability, whereas the measurement of water surface temperature and topographic variables did not perform appropriately. The transferability of the model showed significant relations between the HSI and infection prevalence in study sites of Côte d'Ivoire.Conclusions/significanceA predictive map of environmental suitability for schistosomiasis transmission can support measures to gain and sustain control. This is particularly relevant as emphasis is shifting from morbidity control to interrupting transmission. Further validation of our mechanistic model needs to be complemented by field data of parasite- and snail-related fitness. Our model provides a useful tool to monitor the development of new hotspots of potential schistosomiasis transmission based on regularly updated remote sensing data.

Project description:BACKGROUND:Bulinus species are freshwater snails that transmit the parasitic trematode Schistosoma haematobium. Despite their importance, the diversity of these intermediate host snails and their evolutionary history is still unclear in Zimbabwe. Bulinus globosus and B. truncatus collected from a urogenital schistosomiasis endemic region in the Madziwa area of Zimbabwe were characterized using molecular methods. METHODS:Malacological survey sites were mapped and snails were collected from water contact sites in four communities in the Madziwa area, Shamva district for a period of one year, at three-month intervals. Schistosoma haematobium infections in snails were determined by cercarial shedding and the partial mitochondrial cytochrome c oxidase subunit 1 gene (cox1) was used to investigate the phylogeny and genetic variability of the Bulinus spp. collected. RESULTS:Among the 1570 Bulinus spp. snails collected, 30 (1.9%) B. globosus were shedding morphologically identified schistosomes. None of the B. truncatus snails were shedding. The mitochondrial cox1 data from 166 and 16 samples for B. globosus and B. truncatus, respectively, showed genetically diverse populations within the two species. Twelve cox1 haplotypes were found from the 166 B. globosus samples and three from the 16 B. truncatus samples with phylogenetic analysis showing that the haplotypes fall into well-supported clusters within their species groups. Both B. truncatus and B. globosus clustered into two distinct lineages. Overall, significant negative values for both Tajima's D statistic and the Fu's Fs statistic were observed for B. globosus and B. truncatus. CONCLUSIONS:The study provided new insights into the levels of genetic diversity within B. globosus and additional information on B. truncatus collected from a small geographical area in Zimbabwe. Low prevalence levels of infection observed in the snails may reflect the low transmission level of urogenital schistosomiasis in the area. Our results contribute towards the understanding of the distribution and population genetic structure of Bulinus spp. supporting the mapping of the transmission or risk of transmission of urogenital schistosomiasis, particularly in Zimbabwe.