Skip to main content
  • Research Article
  • Open access
  • Published:

Natural variables separate the endemic areas of Clonorchis sinensis and Opisthorchis viverrini along a continuous, straight zone in Southeast Asia



Clonorchiasis and opisthorchiasis, caused by the liver flukes Clonorchis sinensis and Opisthorchis viverrini respectively, represent significant neglected tropical diseases (NTDs) in Asia. The co-existence of these pathogens in overlapping regions complicates effective disease control strategies. This study aimed to clarify the distribution and interaction of these diseases within Southeast Asia.


We systematically collated occurrence records of human clonorchiasis (n = 1809) and opisthorchiasis (n = 731) across the Southeast Asia countries. Utilizing species distribution models incorporating environmental and climatic data, coupled machine learning algorithms with boosted regression trees, we predicted and distinguished endemic areas for each fluke species. Machine learning techniques, including geospatial analysis, were employed to delineate the boundaries between these flukes.


Our analysis revealed that the endemic range of C. sinensis and O. viverrini in Southeast Asia primarily spans across part of China, Vietnam, Thailand, Laos, and Cambodia. During the period from 2000 to 2018, we identified C. sinensis infections in 84 distinct locations, predominantly in southern China (Guangxi Zhuang Autonomous Region) and northern Vietnam. In a stark contrast, O. viverrini was more widely distributed, with infections documented in 721 locations across Thailand, Laos, Cambodia, and Vietnam. Critical environmental determinants were quantitatively analyzed, revealing annual mean temperatures ranging between 14 and 20 °C in clonorchiasis-endemic areas and 24–30 °C in opisthorchiasis regions (P < 0.05). The machine learning model effectively mapped a distinct demarcation zone, demonstrating a clear separation between the endemic areas of these two liver flukes with AUC from 0.9 to1. The study in Vietnam delineates the coexistence and geographical boundaries of C. sinensis and O. viverrini, revealing distinct endemic zones and a transitional area where both liver fluke species overlap.


Our findings highlight the critical role of specific climatic and environmental factors in influencing the geographical distribution of C. sinensis and O. viverrini. This spatial delineation offers valuable insights for integrated surveillance and control strategies, particularly in regions with sympatric transmission. The results underscore the need for tailored interventions, considering regional epidemiological variations. Future collaborations integrating eco-epidemiology, molecular epidemiology, and parasitology are essential to further elucidate the complex interplay of liver fluke distributions in Asia.


The prevalence and infection rates of liver fluke diseases are high across Asian regions, particularly notable in the Mekong River Basin. Those diseases caused by Clonorchis sinensis and Opisthorchis viverrini are highly prevalent food-borne trematodiasis (FBTs) [1, 2]. Extensive research indicates that clonorchiasis is widespread in parts of Russia, Republic of Korea, southern and northeastern China, extending its endemicity to the northern provinces of Vietnam, exhibiting localized epidemics [3]. In contrast, opisthorchiasis primarily affects the lower Mekong regions, including Thailand, Laos, Cambodia, and the central and southern provinces of Vietnam [4]. Notably, Vietnam is the only country where both types of human liver fluke infections co-exist [5, 6]. This co-endemicity presents unique public health challenges and necessitates targeted intervention strategies to effectively address the burden of these parasitic infections.

First documented in Vietnam in 1887, C. sinensis infection was followed by the discovery of O. viverrini transmission in central Vietnamese provinces in 1994 [7]. A 1992 epidemiological survey in Phu Yen Province in central Vietnam found C. sinensis prevalence ranging from 23.5 to 31.0% in northern Nam Dinh Province, with O. viverrini prevalence up to 43.5% in males and 29.4% in females, concentrated among 40–59-year-olds [8]. Subsequently, reports of both liver fluke infections have accumulated in Vietnam. As the only country endemic for both species, Vietnam provides an opportunity to elucidate the interface between their geographic distributions from an epidemiological perspective [9].

Whereas C. sinensis is distributed across parts of Russia, Republic of Korea, parts of China and northern Vietnam, O. viverrini is concentrated in the lower Mekong River Basin of Thailand, Laos, Cambodia, and central-southern Vietnam [9]. Co-endemicity of both fluke species has been uniquely observed in Vietnam, but early epidemiological data are limited by diagnostic constraints [10]. C. sinensis was initially reported from northern Vietnamese provinces only, while O. viverrini was restricted to central endemic foci of Vietnam. Although O. viverrini has been reported in cats in southern Vietnam, no human cases were documented [11]. Diagnosis, based on microscopic egg morphology in the 1970s–1980s likely resulted in substantial overestimates of national clonorchiasis prevalence, as other intestinal trematode eggs may have been misclassified as those of C. sinensis. Updated epidemiological assessments using accurate diagnostics are needed to delineate the endemic boundaries and disease burden posed by each liver fluke species across different regions, particularly in Vietnam.

The life cycles of C. sinensis and O. viverrini are similar, with humans and some other mammals (pigs, cats, dogs and rodents) as the definite hosts and freshwater snails and fish as two intermediate hosts following each other in that order [12]. Human infections of both flukes are acquired by consuming undercooked freshwater fish harboring infective metacercariae. However, the first intermediate hosts differ between the species. The spatial distribution of the two fluke infections is heavily influenced by the presence of populations of susceptible snail species and suitable environmental conditions [13, 14]. Elucidating the intermediate host profiles and environmental limits of each liver fluke species is imperative to understand their endemic boundaries and opportunities for sustained disease control.

It remains unclear whether clear boundaries exist in the endemic distributions of the two parasites, and which factors that have resulted in their segregated geographic patterns[5]. Predicted distribution maps for clonorchiasis and opisthorchiasis exist, but a proper, integrated analysis investigating the geographic boundaries between the two trematodiases has yet to be conducted. Defining the specific ecologic limits and spatial overlap of these two helminths is imperative to devise integrated control strategies that account for their sympatry across certain endemic regions. A comprehensive approach combining predictive mapping and delineation of the niche boundaries would provide novel insights into their distinct epidemiology in Asia.

The application of machine learning methods for disease distribution prediction represents a major focus in this field. Such approaches primarily address the classification problem of delineating disease ranges based on input variable patterns using algorithms that identify relationships within data to categorize outcome or input variables[15]. Machine learning techniques comprise supervised learning, in which models are trained on known input–output variable pairs to predict outcomes for new inputs (e.g. logistic regression), and unsupervised learning, which uncovers inherent structure within existing data to inform clustering[16]. Given compiled databases of geolocated C. sinensis and O. viverrini human infection records across the Mekong River Basin, we applied machine learning models to elucidate the niche boundaries between the two liver flukes and characterize key determinants of their divergent endemic patterns. Since algorithms trained on ecological and social data can provide novel insights unavailable from traditional statistical approaches, we applied such techniques for comparative predictive mapping of clonorchiasis and opisthorchiasis distributions. By integrating epidemiological perspectives from the literature with computational modelling techniques, we aimed to elucidate the complex and divergent ecology of these two liver fluke species. The primary aim is to compile location-specific cases of infections to input into ecological niche modelling by machine learning techniques, aiming to map transmission patterns rather than quantify infection prevalence rates. Then secondary aim is to document infection risk factors, identifying potential influences on the models to understand the relationship between environmental variables and geo-locations. This comprehensive approach allows for a detailed comparison of environmental conditions that support the sustenance of these parasites in endemic areas.


Study design

We conducted an integrative modelling study using a mix of data sources to map the niche boundaries and model the divergent epidemiology of clonorchiasis and opisthorchiasis in Asia. The integration of multiple data modalities aimed to provide novel insights into the differing ecology and transmission dynamics of these two liver fluke infections. The modelling framework incorporated three main components: (1) a systematic literature review of prior epidemiological studies reporting prevalence and risk factors; (2) compilation of environmental, socioeconomic, and disease burden data from regional databases; and (3) implementation of species distribution modelling algorithms to delineate environmental niches. The models enabled predictive mapping of each disease's ecological niche across Asia based on inferred associations between disease occurrence and environmental conditions.

Data sources

Literature review

We systematically searched major biomedical, regional, and grey literature databases, including PubMed, Scopus, EBSCOhost, Web of Science, Cochrane Library, Cairn, OpenGrey, and Scielo from database inception through December 2018. The search strategy utilized Medical Subject Headings (MeSH) terms including ["Clonorchis sinensis", "Clonorchis sinenses", "Clonorchiasis", "Opisthorchis sinensis", "Opisthorchis sinenses","Liver fluke"], AND ["Asia", "Epidemiology", "Prevalence"], and relevant variants in [All Fields]. Equivalent subject headings and keywords were used for searches in other databases.

Inclusion criteria were cross-sectional surveys, cohort studies, or case–control studies reporting primary prevalence data or risk factors for clonorchiasis and/or opisthorchiasis in Asia. Studies were required to have laboratory diagnostic testing for infection.

Exclusion criteria were case reports, reviews, opinion pieces, policy documents, animal studies, and studies without primary prevalence data. Two independent reviewers screened all titles, abstracts, and full texts for eligibility. Data on prevalence, diagnostics, location, sample size, demographics, and risk factors were extracted from included studies into a standardized form using Zotero version: 6.0.31 (The Roy Rosenzweig Center for History and New Media, Fairfax, USA). Any discrepancy was resolved by consensus.

This comprehensive literature search aimed to compile all relevant epidemiological data on clonorchiasis and opisthorchiasis prevalence and risk factors needed to inform model development.


For C. sinensis infection data in the Mekong River region, we supplemented the literature review by compiling primary data on population infection rates from databases in Vietnam and Guangxi Zhuang Autonomous Region of China. Cross-sectional surveys, conducted between 2000 and 2018 in Vietnam, were systematically searched to extract geolocated presence/absence data based on faecal egg detection at the survey point and the regional level. Infection status was coded as Yes (positive) or No (negative) for C. sinensi in the databases.

For Guangxi Zhuang Autonomous Region, China, population-level data on C. sinensis infections were obtained from the 3rd National Survey on Key Parasitic Diseases conducted between 2014 and 2016, which covered 31 provinces (municipalities, autonomous regions) in rural and urban areas of China. Apart from C. sinensis, testing included tapeworms, intestinal protozoa and other key parasitic infections via faecal examination in the sampled population. A stratified cluster sampling method was used, classifying China into 5 endemic zones for C. sinensis and sampling within each zone. All individuals in the selected clusters underwent testing. Stool specimens were examined by the Kato-Katz thick smear technique using two smears per specimen to detect intestinal helminth eggs.

By compiling primary epidemiological records from these standardized national surveys in China and Vietnam, we obtained geolocated C. sinensis infection data needed to parameterize niche modelling and epidemiological comparisons between the two liver flukes. The original data of O. viverrini infection in endemic countries of Southeast Asia is extracted from the 113 studies and combined with the reported data from WHO (Department of Neglected tropical diseases of WHO Western Pacific), and details of system review screening were shown in Additional file 1.

Environmental data

We compiled 26 natural climatic, and socio-cultural predictor variables, including distance to water bodies, elevation, slope, normalized difference vegetation index (NDVI), land cover, 19 bioclimatic variables (Bio1–Bio19), human influence index (HII), human footprint index (HFP), based on the variables used in Zhao’s study [17], and Zheng’s study[18] for modelling with liver fluke and snails. We additionally included local habit of raw fish consumption-eating as a predictor variable. Our approach was to use a comprehensive set of environmental and socio-economic factors to capture these fine-scale differences. Factors may similarly increase overall risk, but specific values pinpoint geographic boundaries. The machine learning framework integrated with ecological data successfully learned these distinct signatures, enabling accurate discrimination for mapping. All databases used for these 26 predictor variables are shown in Table 1.

Table 1 Environmental and climatic variables influencing liver fluke infection

Topographic variables, such as water distance, elevation, slope, NDVI, and land cover were extracted from the Shuttle Radar Topography Mission (SRTM) at 5 km-resolution ( Water distance calculates the Euclidean distance from each grid cell to the nearest wetland, including lakes, wetlands, and river floodplains, representing proximity to water bodies (in meters). Elevation denotes altitude above the mean sea level (in meters). Slope describes the rate of change in elevation. NDVI is an index of green vegetation density ranging from -1 to 1, with values below 0 indicating water, cloud, snow; near 0 barren land; and above 0 vegetation cover increasing with density. Land cover was defined using the Moderate Resolution Imaging Spectroradiometer (MODIS) MCD12Q1 product (, aggregated and reprojected to match the 15-class University of Maryland scheme.

Geospatial data layers were extracted for each liver fluke survey location to assess environmental factors associated with C. sinensis and O. viverrini transmission. Univariate comparisons were conducted between survey points for each variable using Mann–Whitney U tests.

Climatic data were obtained from WorldClim v.1.4 at 5 km-resolution (, interpolated from global weather station data from 1955 to 2000 for China. The 19 bioclimatic variables represent annual trends, seasonality, and limiting factors calculated from monthly temperature and rainfall. These are more biologically meaningful than temperature/rainfall alone.

To represent anthropogenic effects on the environment, we extracted two human influence indices: HII quantifies direct human pressures on ecosystems using population density, built environments, transportation networks, land use/land cover, and nightlights ( HII ranges 0–64, with higher values indicating greater human environmental impacts. HFP shows relative human pressure, with red indicating more intense activity.

Eating habits of raw fish

The study considered the use of data on the consumption of raw fish because eating raw or undercooked fish is a well-known risk factor for infections with C. sinensis and O. viverrini. These liver flukes can infect humans who consume freshwater fish containing the larval stages of the parasites. This dietary habit directly relates to the transmission dynamics of these parasites, making it a critical factor to examine in understanding the geographical distribution and risk of infection.

By understanding where and how often people consume raw fish, we define the eating habits with reference to raw fish were recorded for mapping sections by provinces and municipalities in all countries based on literature review, coded as 1 if present the eating habits with raw fish or 0 if absent the eating habits with raw fish. Also, we collecting data from affected populations through direct questioning about their dietary habits through the help of local disease control centres and experts from each country as we have consulted with, specifically the dietary habits for consumption of raw or undercooked freshwater fish with specific municipalities areas. Finally, provincial polygons were rasterized to assign presence across each province.

Assessment and extraction of variable data

The compiled databases were separated into two groups based on human infection with C. sinensis and O. viverrini for comparative analysis. We statistically summarized and mapped the locations of the two parasite infections. For normally distributed continuous variables, means and standard deviations were calculated, with t-tests used to compare groups. As land cover comprised 15 categorical classes, non-normal variables were summarized using median and interquartile range (IQR) and compared between groups with non-parametric tests.

All data processing and analyses were conducted in R V.4.0.2 (Lucent Technologies, Jasmine Mountain, USA). Variables were assessed for collinearity and eliminated if the variance inflation factor (VIF) exceeded 5. We then used random forest (RF) models to rank predictor importance based on mean decrease in accuracy when excluded from the models. The top 10 most important variables for each parasite were retained for further niche modelling. This process filtered the database variable data to retain only relevant non-redundant predictors characterizing the fundamental and realized niches of the two liver flukes under study. It also allowed statistical comparisons to identify similarities and differences in their ecological and environmental constraints. These ‘curated’ database variable values provide the inputs for ensuing distribution modelling and mapping of the transmission risk.

Model development

We developed predictive models to classify and discriminate human infections of O. viverrini versus C. sinensis based on environmental variables, following the framework Y = f(x). The binary response variable Y would indicate O. viverrini (Y = 0) or C. sinensis (Y = 1) infection, the aim was delineating the potential transition zone where the probability for both species shifts between 0 and 1, indicating possible co-endemicity. The predictor variables (X) comprised 26 environmental, climatic and socio-cultural factors. To account for class imbalance, we used the SMOTE algorithm from the DMwR package to synthesize additional minority class examples. The models were constructed and evaluated using the Caret package in R. To enable consistent comparison across algorithms and assessment of variable importance, we selected six commonly used machine-learning classification methods to model environmental suitability for O. viverrini and C. sinensis transmission: linear regression (LM), decision trees (DT), neural networks (NNET), RF, gradient boosting machines (GBM) and extreme gradient boosting (XGBOOST). Details on each algorithm can be found at the Caret documentation ( All models were trained using tenfold cross-validation repeated 5 times, with hyperparameter tuning to optimize model performance. Model fitting performance, prediction accuracy, variable contributions, marginal response plots, and projected distribution maps were analysed and evaluated for each approach.

The fitted machine learning models were applied to an independent testing dataset to evaluate generalizability. Liver fluke presence/absence predictions were generated for each testing location and compared to observed outcomes to assess model discrimination. Testing performance was quantified using AUC, accuracy, Kappa value, sensitivity, and specificity metrics.

We evaluated and compared models based on the area under the receiver operating characteristic curve (AUC) by sensitivity, specificity, and Cohen's Kappa statistic. The optimal model was selected based on having the highest cross-validated AUC. This model was then finalized by refitting on the full dataset to generate the final prediction equation.

Model development aimed to maximize discrimination accuracy in predicting O. viverrini versus C. sinensis infections based on ecological and environmental factors relevant to their transmission dynamics and geographic distributions. The resulting model could then be applied to mapping transmission risk and predicting changes under climate change scenarios. Model development and validation followed a rigorous workflow for tuning, testing and application. The compiled database of values for the two infections was randomly split into a training set (70% of the data) for model calibration and a testing set (30% of the data) for independent evaluation. The training data underwent fivefold cross-validation, whereby the data were divided into 5 equal partitions. In each fold, models were fitted on 4 partitions and predictions generated for the held-out fold. This process was repeated, holding out each partition in turn to identify the optimal hyperparameters that minimized the cross-validation error. This was done as cross-validation prevents model overfitting and provides a realistic estimate of performance on new data.

Following cross-validation-based tuning, the final models were refit on the full training set using optimal hyperparameters. Model skill was quantified on the training set using the AUC as mentioned above. The tuned models were then applied to the previously held-out testing set to evaluate performance on new data. Variable importance was calculated by excluding each predictor and quantifying loss in testing AUC. Marginal effects of key predictors were generated from the finalized models to quantify variable-outcome relationships. Model predictions were mapped across the study region based on environmental inputs to predict risk areas for each species. Finally, an ensemble approach was taken by integrating predictions across algorithms to leverage model strengths.

Model assessment and prediction

Model calibration was assessed using calibration plots to evaluate agreement between predicted and observed outcomes. Classification metrics including AUC, accuracy, Kappa value, specificity, and sensitivity were calculated at the optimal probability threshold to quantify model discrimination ability. Variable importance was determined using the varImp function in the Caret package, which quantifies the decrease in model AUC with variable exclusion. This approach includes all predictors and ranks importance based on change in performance. Marginal effects of key variables were visualized using partial dependence plots (PDPs) from the pdp package. PDPs show the functional relationship between a predictor and the outcome while accounting for effects of other variables. To reduce computation time, PDPs were generated for the top three important variables.

The finalized models were applied to predict the probability of C. sinensis infection across gridded environmental data in China's Guangxi Zhuang Autonomous Region and the south-eastern Laos, Thailand, Cambodia, and Vietnam Regions. Predictions were mapped to visualize the geographic distribution of estimated risk. Any predictions of C. sinensis in Guangxi Zhuang Autonomous Region of China were considered erroneous given known distributions. To delineate species boundaries, we focused on areas of Vietnam and Laos where both species are endemic. Grid cells with a predicted probability of C. sinensis of 100% were classified as high risk for that species. Areas with intermediate probabilities of 0–1 were considered potential hybrid zones with sympatric transmission.


Geographic distribution

This study found the Southeast Asian range of O. viverrini and C. sinensis infections to be cantered in China, Vietnam, Thailand, Laos, and Cambodia (Prediction map was shown Figure1.png in In the study period (2000–2018), C. sinensis infections were identified in 84 places extracted from data in 15 studies, with a concentration in southern China (Guangxi Zhuang Autonomous Region) and northern Vietnam. In contrast, O. viverrini infections were seen in 721 places, extracted from 113 studies, with clusters appearing across Thailand, Laos, Cambodia, and Vietnam. These spatial patterns align with known endemic zones based on past national surveys and reported cases. However, our database compiled a larger number of geocoded epidemiological studies that provided mapping with higher resolution of local prevalence.

Environmental associations

Univariate comparisons conducted between survey points for elevation variable showed median elevation to be significantly lower in C. sinensis locations (median = 35.44 m, IQR: 12.2–98.7) compared to O. viverrini ones (median = 159.5 m, IQR: 45.6–212.4; P < 0.0001). No significant differences were found for slope, Bio10, Bio8, or Bio16 (See Table 2).

Table 2 Comparative ecological analysis of Clonorchis sinensis and Opisthorchis viverrini infections of environmental, climatic, and socio-economic variables

Model fitting with training data

Based on the compiled training dataset, the six algorithms learned associations between environmental predictors and liver fluke presence/absence to develop fitted models. All models achieved excellent fitting performance on the training data with AUC, accuracy, Kappa value, sensitivity, and specificity approaching 1 (NNET range: 0.981–0.998). (Table 3).

Table 3 Training model fit metrics for the machine learning approaches

Model predictions on independent testing data

All models achieved excellent prediction accuracy on the testing data with AUC, accuracy, kappa value, sensitivity, and specificity approaching 1 (NNET range: 0.991–0.996) (Table 4).

Table 4 Parameters of model performance in the testing set

Key environmental drivers

The machine-learning models identified Bio4 and Bio3 as consistently influential predictors of liver fluke presence across techniques (Fig. 1), agreeing with their known role in snail habitat suitability.

Fig. 1
figure 1

Feature Importance Analysis in Clonorchis sinensis and Opisthorchis viverrini infections of six predictive models (The models displayed are Linear Model (LM), Random Forest (RF), Gradient Boosting Machine (GBM), Decision Tree (DT), Neural Network (NNET), and eXtreme Gradient Boosting (XGBOOST). Each bar represents the importance score of ecological and environmental features, such as Bioclimatic variables (Bio1-Bio19), elevation, slope, land cover, NDVI (Normalized Difference Vegetation Index), HII (Human Influence Index), HFP (Human Footprint), and distance to water bodies)

Marginal variable effects

Based on the fitting and prediction results, all models except NNET showed good performance. Using the LM model, the variables Bio8, Bio1, and Bio18 were selected for partial dependence plots based on contribution over 75%. For the RF model, Bio4 and Bio3 were selected. The plots visualize the dependence between variables and predicted probability of liver fluke of C. sinensis presence (Y = 1). From Fig. 2, results showed Bio8 had a predicted probability of 0.987 at 22 °C, decreasing as Bio8 increased. Probability of presence increased with Bio1, with values below 0.1 when Bio1 = 22.4 °C and reaching 1 when Bio1 > 27 °C. For Bio18, predicted probability was above 0.87, increasing to over 0.9 when Bio18 exceeded 750 mm. Bio3 exhibited increasing predicted probability with values, with plateau creation at 0.93 when Bio3 > 50. In contrast, as Bio4 increased, predicted probability declined. These patterns align with known biological requirements of snail intermediate hosts and transmission potential. The marginal response plots aid interpretation of complex model dynamics.

Fig. 2
figure 2

Partial dependence plots (PDPs) for Clonorchis sinensis (Y = 1) infections of RF model predictive (The figure consists of PDPs that depict the relationship between selected Bioclimatic variables (Bio1, Bio3, Bio4, Bio8, Bio18) and the probability of Clonorchis sinensis infection as predicted by RF model. Each plot shows how changes in a specific Bioclimatic variable impact the model's predicted probability of infection, holding all other features constant)

Model predictions

The fitted machine-learning models were used to estimate the geographic distribution of liver fluke infection risk across the Mekong River basin and Guangxi Zhuang Autonomous Region of China (Prediction of liver fluke infection risk map was shown Figure 2.png in Each model generated a predicted probability surface for liver fluke presence at each location based on the input variables. All models successfully delineated areas of higher predicted risk for C. sinensis versus O. viverrini. However, the NNET and RF models projected some C. sinensis presence in northern Vietnam and southern China (Guangxi Zhuang Autonomous Region). The DT, GBM, and XGBOOST models limited O. viverrini suitability to western Guangxi Zhuang Autonomous Region of China, while the LM model predicted only C. sinensis across the all Guangxi Zhuang Autonomous Region of China.

Delineation of boundaries

To further investigate the geographic boundaries between C. sinensis and O. viverrini, the study area was narrowed to Vietnam and Laos. Of the model predictions, only the RF results met all simulated conditions for delineating species risk. The RF model generated a predicted risk map for C. sinensis and O. viverrini in Vietnam and Laos (Prediction map was shown Figure 3.png in Three types of regions have been classified in Figure 3.png, including: Region A (in red) indicated a higher predicted risk for O. viverrini transmission, with a focus in central and southern Vietnam plus few northern provinces of Vietnam bordering Laos, as well as whole areas of Laos. Region B (in blue) depicted higher C. sinensis transmission risk in northestern Vietnam, adjoining the border to Guangxi Zhuang Autonomous Region of China. Region C (in pink) represented a transitional zone of mixed species potential, where the RF model predicted O. viverrini transmission probability < 1. This spanned 4 northwestern provinces and 2 northcentral provinces of Vietnam, as well as small areas in northern Laos bordering Vietnam.


Of the six machine-learning models developed in this study, all except the NNET model, accurately classified and predicted C. sinensis and O. viverrini infections during model fitting and projection across the Mekong River Basin and Guangxi Zhuang Automomous Region of China. Despite known absence of O.viverrini in China, the RF model predicted potential presence in small areas of northwestern Guangxi Zhuang Autonomous Region of China, either indicating inaccurate classification or undiscovered transmission. The LM model predicted no any O. viverrini transmission risk in China, but modelled suitable areas across central and southern Vietnam, aligning with national surveys [19, 20]. Reportedly endemic areas across 21 northern provinces of Vietnam for C. sinensis infections, concentrated near the Red River Basin [14]. O. viverrini persists across 11 central provinces of Vietnam while control efforts have eliminated southern foci there [21]. However, complex co-endemic areas remain understudied, with no reports from north-western Vietnam [5]. Sithithaworn et al. previously delimited a diagonal boundary from Lai Chau Province northwest to Quang Binh Province central-east of Vietnam, designating upper and lower zones for C. sinensis and O. viverrini, respectively [20, 22]. We identified a transition zone of mixed transmission risk in Vietnam, with suitable environments for both flukes, spanning four northwestern provinces and two northcentral provinces. Visualizing the projected ranges advances an understanding of potential overlapping. As these infections depend on human culinary practices, mixed zones likely reflect localized food habits including raw fish dishes.

The distinct geographic distributions of C. sinensis and O. viverrini motivated an analysis of geographic, climatic, and anthropogenic predictors, revealing divergence between the two flukes. For example, the former occurred mostly in low latitudes while the latter predominated in higher latitudes. In Thailand, O. viverrini is concentrated in the Northeast with a similar high latitude pattern in Laos [23]. Most georeferenced O. viverrini occurrences are from Thailand. Reports from Vietnam noted C. sinensis as concentrated around the Red River Basin in lower latitudes [24]. Climate also differed between endemic areas, with mean annual precipitation of 772 mm for C. sinensis versus 367 mm for O. viverrini. These factors likely influenced fluke distributions indirectly by impacting snail intermediate hosts. A study in Thailand found O. viverrini sensitive to rainfall and minimum temperature, with consistent prevalence from 41–356 mm monthly rainfall but a drop above 23 °C [25]. The most influential predictors varied among models constructed here. Considering variables contributing over 75% for O. viverrini, key factors were annual mean temperature (Bio1), temperature seasonality (Bio4), warmth index (Bio8), rainfall in warmest quarter (Bio18) and annual temperature range (Bio3). The LM dependency plot showed O. viverrini probability increasing with temperature seasonality. Infection likelihood peaked around 300% variance in seasonal temperatures and with over 750 mm precipitation in the warmest quarter. As annual mean temperature rose above 27 °C, O. viverrini probability approached 100%. These results highlight climatic factors, especially temperature and rainfall, as important delineators between O. viverrini and C. sinensis distributions. In differentiating the distribution of O. viverrini and C. sinensis, our study identified several critical influencing factors. For C. sinensis, factors such as higher temperatures and urbanized environments showed greater association, whereas O. viverrini distribution was more influenced by wetland ecosystems and certain agricultural practices, and lower elevation ranges for the Bithynia spp. snail hosts of O. viverrine [26]. The distinction in habitat preferences, intermediate host snail species, and human behaviours, including dietary differences across regions, were significant in delineating the distribution of these flukes. These variables were crucial in our machine learning models to predict geospatial distribution with greater specificity.

The results demonstrated environmental and climatic variables shape distributions of C. sinensis and O. viverrini. Divergence across factors enables classification, with models accurately categorizing infections in test data. However, as Max Kuhn notes, machine-learning risks finding spurious relationships if predictors closely parallel outcomes, producing apparent 100% accuracy for uninformative variables. While dividing flukes using single factors proved successful here, disease emergence involves complex interactions among environmental, climatic, and social determinants [27]. Despite ideal performance during training, GBM leaning solely on annual mean temperature range and temperature seasonality generated some geographically discordant projections. This highlights the need to validate models against real-world data, not just internal fit, when applying predictions. While these spatial models’ further knowledge of potential C. sinensis and O. viverrini distributions, multifaceted drivers and potential sampling biases warrant caution for public health planning until localized surveys confirm patterns. Elimination efforts require understanding mixed-disease contexts through on-the-ground investigation. Our models successfully distinguished northern endemic areas for C. sinensis from southern O. viverrini foci but also delimited a transitional zone of overlapping potential spanning 6 north-western Vietnamese provinces where both liver flukes may persist. This coexistence complicates control efforts designed for single infection. The policy recommendation should be with a multi-pronged approach in these zones of sympatry, incorporating coordinated interventions tailored to each species while integrating education and policy to maximize efficacy. This includes dual drug administration with praziquantel and tribendimidine to target both flukes, augmented diagnostics to distinguish infections, ecological modifications limiting snail intermediate hosts, and sociocultural promotion of cooked fish consumption given dietary habits underlying transmission. Robust surveillance is vital to monitor efforts [28].

Although this study collected georeferenced human infections to classify liver flukes using environmental predictors and machine-learning, limitations include reliance on reported occurrence points from prior literature, lack of animal infection data, exclusion of intermediate snail/fish host distributions, assumptions that all suitable environments have active transmission, and generalizability constraints of machine learning algorithms. Additionally, while we identified sympatric zones, molecular evidence would further confirm co-endemicity. Future work should incorporate such data to delineate ranges. Our ecological approach provided initial delineation in Vietnam, but molecular epidemiology is needed to confirm potential boundaries or zones of sympatry. Further studies should identify underlying drivers, be it intermediate host compatibility or fluke biology. Questions remain whether a sharp boundary exists and what drives separation. To address these complexities, a multidisciplinary collaboration that synthesizes eco-epidemiology, molecular epidemiology, malacology, and parasitology is essential. This would allow for a deeper understanding of fluke ecology, inform targeted control programs, and support the global effort to combat these neglected tropical diseases within the framework of the One Health concept [29].


This study delineated the boundary between C. sinensis and O. viverrini in the Mekong River Basin, identifying sympatric transmission in Vietnam concentrated in northwestern and northcentral provinces, and part of northern Laos. Environmental, climatic, and sociocultural factors diverged between the endemic areas, with rainfall in the warmest quarter, precipitation in the wettest month and annual mean temperature influencing distributions most. Machine-learning models effectively classified the endemic areas of liver flukes, demonstrating utility for mapping boundaries. Elimination of these neglected tropical diseases requires understanding the mosaic of species and targeting control and surveillance to local transmission patterns. Further molecular epidemiological studies can confirm the potential boundary and drivers shaping this divergence across Southeast Asia.

Availability of data and materials

The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.



Food-borne trematodiases


Neglected tropical diseases


Area under the curve


Random forest


Gradient boosting machines


eXtreme gradient boosting


Neural networks


Linear model


Decision trees


Medical subject headings


Normalized difference vegetation index


Human influence index


Human footprint index


Shuttle radar topography mission


Moderate resolution imaging spectroradiometer


Variance inflation factor


Synthetic minority over-sampling technique


  1. Na BK, Pak JH, Hong SJ. Clonorchis sinensis and clonorchiasis. Acta Trop. 2020;203: 105309.

    Article  CAS  PubMed  Google Scholar 

  2. Liu K, Tan J, Xiao L, Pan RT, Yao XY, Shi FY, Spatio-temporal disparities of Clonorchis sinensis infection in animal hosts in China: a systematic review and meta-analysis. Infect Dis Poverty. 2023;12(1):97. 

    Article  PubMed  PubMed Central  Google Scholar 

  3. Qian MB, Utzinger J, Keiser J, Zhou XN. Clonorchiasis. Lancet. 2016;387:800–10.

    Article  PubMed  Google Scholar 

  4. Andrews RH, Sithithaworn P, Petney TN. Opisthorchis viverrini: an underestimated parasite in world health. Trends Parasitol. 2008;24:497–501.

    Article  PubMed  PubMed Central  Google Scholar 

  5. Hung NM, Dung DT, Lan Anh NT, Van PT, Thanh BN, Van Ha N, et al. Current status of fish-borne zoonotic trematode infections in Gia Vien district, Ninh Binh province. Vietnam Parasit Vectors. 2015;8:21.

    Article  PubMed  Google Scholar 

  6. Doanh PN, Nawa Y. Clonorchis sinensis and Opisthorchis spp in Vietnam: current status and prospects. Trans R Soc Trop Med Hyg. 2016;110:13–20.

    Article  PubMed  Google Scholar 

  7. Humans IWG on the E of CR to. Opisthorchis viverrini and Clonorchis sinensis. Biological Agents. Int Agency Res Cancer. 2012. Accessed 2023 Nov 15.

  8. Dao TTH, Abatih EN, Nguyen TTG, Tran HTL, Gabriël S, Smit S, et al. Prevalence of Opisthorchis viverrini-Like fluke infection in ducks in Binh Dinh Province. Central Vietnam Korean J Parasitol. 2016;54:357–61.

    Article  PubMed  Google Scholar 

  9. Suwannatrai A, Saichua P, Haswell M. Epidemiology of Opisthorchis viverrini Infection. Adv Parasitol. 2018;101:41–67.

    Article  PubMed  Google Scholar 

  10. Echaubard P, Sripa B, Mallory FF, Wilcox BA. The role of evolutionary biology in research and control of liver flukes in Southeast Asia. Infect Genet Evol. 2016;43:381–97.

    Article  PubMed  PubMed Central  Google Scholar 

  11. Cam DTT, Yajima A, Viet KN, Montresor A. Prevalence, intensity and risk factor of Clonorchiasis and possible use of questionnaire to detect individuals at risk in northern Vietnam. Trans R Soc Trop Med Hyg. 2008;102:1263–8.

    Article  Google Scholar 

  12. Pakharukova MY, Lishai EA, Zaparina O, Baginskaya NV, Hong S-J, Sripa B, et al. Opisthorchis viverrini, Clonorchis sinensis and Opisthorchis felineus liver flukes affect mammalian host microbiome in a species-specific manner. PLoS Negl Trop Dis. 2023;17: e0011111.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  13. Tran AKT, Doan HT, Do AN, Nguyen VT, Hoang SX, Le HTT, et al. Prevalence, species distribution, and related factors of fish-borne trematode infection in Ninh Binh Province, Vietnam. Biomed Res Int. 2019;2019:8581379.

    Article  PubMed  PubMed Central  Google Scholar 

  14. Trung Dung D, Van De N, Waikagul J, Dalsgaard A, Chai J-Y, Sohn W-M, et al. Fishborne zoonotic intestinal trematodes, Vietnam. Emerg Infect Dis. 2007;13:1828–33.

    Article  PubMed  Google Scholar 

  15. Albaradei S, Thafar M, Alsaedi A, Van Neste C, Gojobori T, Essack M, et al. Machine learning and deep learning methods that use omics data for metastasis prediction. Comput Struct Biotechnol J. 2021;19:5008–18.

    Article  PubMed  PubMed Central  Google Scholar 

  16. Bi Q, Goodman KE, Kaminsky J, Lessler J. What is machine learning? A primer for the epidemiologist. Am J Epidemiol. 2019;188:2222–39.

    PubMed  Google Scholar 

  17. Zhao TT, Feng YJ, Doanh PN, Sayasone S, Khieu V, Nithikathkul C, et al. Model-based spatial-temporal mapping of opisthorchiasis in endemic countries of Southeast Asia. Elife. 2021;10: e59755.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. Zheng JX, Xia S, Lv S, Zhang Y, Bergquist R, Zhou XN. Infestation risk of the intermediate snail host of Schistosoma japonicum in the Yangtze River Basin: improved results by spatial reassessment and a random forest approach. Infect Dis Poverty. 2021;10(1):74.

    Article  PubMed  PubMed Central  Google Scholar 

  19. De NV, Murrell KD, Cong LD, Cam PD, Chau LV, Toan ND, et al. The food-borne trematode zoonoses of Vietnam, Southeast Asian. J Trop Med Public Health. 2003;34(Suppl 1):12–34.

    Google Scholar 

  20. Sithithaworn P, Andrews RH, Nguyen VD, Wongsaroj T, Sinuon M, Odermatt P, et al. The current status of opisthorchiasis and clonorchiasis in the Mekong Basin. Parasitol Int. 2012;61:10–6.

    Article  PubMed  Google Scholar 

  21. Vo The Dung, Waikagul J, Thanh BN, Vo DT, Nguyen DN, Murrell KD. Endemicity of Opisthorchis viverrini liver flukes, Vietnam, 2011–2012. Emerg Infect Dis. 2014;20:152–4.

  22. Crellen T, Sithithaworn P, et al. Towards Evidence-based Control of Opisthorchis viverrini. Trends parasitol. 2021;37(5):370–80.

    Article  CAS  PubMed  Google Scholar 

  23. Kim T-S, Pak JH, Kim J-B, Bahk YY. Clonorchis sinensis, an oriental liver fluke, as a human biological agent of cholangiocarcinoma: a brief review. BMB Rep. 2016;49:590–7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  24. Suwannatrai A, Pratumchart K, Suwannatrai K, Thinkhamrop K, Chaiyos J, Kim CS, et al. Modeling impacts of climate change on the potential distribution of the carcinogenic liver fluke, Opisthorchis viverrini, in Thailand. Parasitol Res. 2017;116:243–50.

    Article  CAS  PubMed  Google Scholar 

  25. Petney TN, Andrews RH, Saijuntha W, Wenz-Mücke A, Sithithaworn P. The zoonotic, fish-borne liver flukes Clonorchis sinensis, Opisthorchis felineus and Opisthorchis viverrini. Int J Parasitol. 2013;43:1031–46.

    Article  PubMed  Google Scholar 

  26. Kiatsopit N, Sithithaworn P, Saijuntha W, Petney TN, Andrews RH. Opisthorchis viverrini: Implications of the systematics of first intermediate hosts, Bithynia snail species in Thailand and Lao PDR. Infect Genet Evol. 2013;14:313–9.

    Article  PubMed  Google Scholar 

  27. Zhang XX, Liu JS, Han LF, Simm G, Guo XK, Zhou XN. One health: new evaluation framework launched. Nature. 2022;604(7907):625.

    Article  CAS  PubMed  Google Scholar 

  28. Duan C, Li C, Ren R, Bai W, Zhou L. An overview of avian influenza surveillance strategies and modes. Science in One Health. 2023;2: 100043.

    Article  Google Scholar 

  29. Guo ZY, Zheng JX, Li SZ, Zhou XN. Orientation of one health development: think globally and act locally. Science in One Health. 2023;2: 100042.

    Article  Google Scholar 

Download references


We are grateful to experts from the National Institute of Parasitic Diseases at the Chinese Center for Disease Control and Prevention for their valuable insights and resources. Additionally, we would like to acknowledge the Western Pacific Office of the World Health Organization, particularly the NTD department, for their support and guidance throughout this project. The collaboration and assistance provided by all these individuals and institutions have been instrumental in the successful completion of our research.


This study was supported by the National Key Research and Development Program of China (No. 2021YFC2300800, 2021YFC2300804) and the International Joint Laboratory on Tropical Diseases Control in Greater Mekong Subregion (No. 21410750200), and The International Development Research Centre (IDRC), Canada (No. 108100-001). The funders had no role in study design, data collection, analysis, and interpretation, or in writing the report and the decision to submit the article for publication.

Author information

Authors and Affiliations



JXZ conceived the study, participated in its design, and was instrumental in drafting the manuscript. HHZ was responsible for extensive data collection and analyses. SX conducted the statistical analysis. MBQ was involved in study design and manuscript preparation. MNH, BS, and SS contributed to the study's design and coordination, providing vital regional insights, and assisting in drafting the manuscript. VK focused on data and insights from Cambodia and participated in manuscript drafting. RB played a key role in coordinating the study and significantly contributed to drafting and revising the manuscript. As the corresponding author, XNZ oversaw the entire project, ensuring accuracy and integrity, and was crucial in finalizing the manuscript. All authors have read and approved the final manuscript.

Corresponding author

Correspondence to Xiao-Nong Zhou.

Ethics declarations

Ethics approval and consent to participate.

Not applicable.

Consent for publication

Not applicable.

Competing interests

Xiao-Nong Zhou is an Editor-in-Chief of Infectious Diseases of Poverty. He was not involved in the peer-review or handling of the manuscript. The authors have no other competing interests to disclose.

Supplementary Information

Additional file 1.

The process of extracting data from public database.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zheng, JX., Zhu, HH., Xia, S. et al. Natural variables separate the endemic areas of Clonorchis sinensis and Opisthorchis viverrini along a continuous, straight zone in Southeast Asia. Infect Dis Poverty 13, 24 (2024).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: