Assessing the risk of dengue severity using demographic information and laboratory test results with machine learning

Machine learning-based models for prediction of in-hospital mortality in patients with dengue shock syndrome

BACKGROUND

Severe dengue children with critical complications have been attributed to high mortality rates, varying from approximately 1% to over 20%. To date, there is a lack of data on machine-learning-based algorithms for predicting the risk of in-hospital mortality in children with dengue shock syndrome (DSS).

AIM

To develop machine-learning models to estimate the risk of death in hospitalized children with DSS.

METHODS

This single-center retrospective study was conducted at tertiary Children’s Hospital No. 2 in Viet Nam, between 2013 and 2022. The primary outcome was the in-hospital mortality rate in children with DSS admitted to the pediatric intensive care unit (PICU). Nine significant features were predetermined for further analysis using machine learning models. An oversampling method was used to enhance the model performance. Supervised models, including logistic regression, Naïve Bayes, Random Forest (RF), K-nearest neighbors, Decision Tree and Extreme Gradient Boosting (XGBoost), were employed to develop predictive models. The Shapley Additive Explanation was used to determine the degree of contribution of the features.

RESULTS

In total, 1278 PICU-admitted children with complete data were included in the analysis. The median patient age was 8.1 years (interquartile range: 5.4-10.7). Thirty-nine patients (3%) died. The RF and XGboost models demonstrated the highest performance. The Shapley Addictive Explanations model revealed that the most important predictive features included younger age, female patients, presence of underlying diseases, severe transaminitis, severe bleeding, low platelet counts requiring platelet transfusion, elevated levels of international normalized ratio, blood lactate and serum creatinine, large volume of resuscitation fluid and a high vasoactive inotropic score (> 30).

CONCLUSION

We developed robust machine learning-based models to estimate the risk of death in hospitalized children with DSS. The study findings are applicable to the design of management schemes to enhance survival outcomes of patients with DSS.

Prognostic Models in Patients with Dengue: A Systematic Review

There is uncertainty regarding the usefulness of predictive models for dengue prognosis. We performed a systematic review to identify and evaluate prognostic models in patients with dengue. We conducted a literature search in PubMed, Embase, and Literatura Latinoamericana y del Caribe en Ciencias de la Salud (LILACS) up to May 24, 2023. We included case-control and cohort studies that developed or validated multivariable prognostic models related to severity, hospitalization, intensive care unit (ICU) admission, or mortality in patients of any age with a laboratory-based diagnosis of dengue. A narrative synthesis of the performance measures of the prognostic models evaluated in each study was performed. Of the 4,211 articles, a total of 35 studies reporting information on 43 prognostic models were included. Among these, 35 were developmental and 8 were for external validation. Most models were designed to predict severity (n = 30), followed by mortality (n = 10), hospitalization (n = 2), and ICU admission (n = 1). The reported C-statistic in the models ranged from 0.70 to 0.95 for severity, 0.83 to 0.99 for mortality, 0.87 for hospitalization, and 0.92 for ICU admission. Calibration measures were poorly reported in the vast majority of models. According to the Prediction Study Risk of Bias Assessment Tool, the risk of bias was considered high for all included models, and applicability was of low concern for most models. Our study identified multiple prognostic models, particularly for predicting severity and mortality in patients with dengue. Although most models demonstrated acceptable discriminative ability, calibration measures were poorly reported, and the overall methodological design was poor.

Machine learning for predicting severe dengue in Puerto Rico

BACKGROUND

Distinguishing between non-severe and severe dengue is crucial for timely intervention and reducing morbidity and mortality. World Health Organization (WHO)-recommended warning signs offer a practical approach for clinicians but have limited sensitivity and specificity. This study aims to evaluate machine learning (ML) model performance compared to WHO-recommended warning signs in predicting severe dengue among laboratory-confirmed cases in Puerto Rico.

METHODS

We analyzed data from Puerto Rico's Sentinel Enhanced Dengue Surveillance System (May 2012-August 2024), using 40 clinical, demographic, and laboratory variables. Nine ML models, including Decision Trees, K-Nearest Neighbors, Naïve Bayes, Support Vector Machines, Artificial Neural Networks, AdaBoost, CatBoost, LightGBM, and XGBoost, were trained using fivefold cross-validation and evaluated with area under the receiver operating characteristic curve (AUC-ROC), sensitivity, and specificity. A subanalysis excluded hemoconcentration and leukopenia to assess performance in resource-limited settings. An AUC-ROC value of 0.5 indicates no discriminative power, while values closer to 1.0 reflect better performance.

RESULTS

Among the 1708 laboratory-confirmed dengue cases, 24.3% were classified as severe. Gradient boosting algorithms achieved the highest predictive performance, with an AUC-ROC of 97.1% (95% CI: 96.0-98.3%) for CatBoost using the full 40-variable feature set. Feature importance analysis identified hemoconcentration (≥ 20% increase during illness or ≥ 20% above baseline for age and sex), leukopenia (white blood cell count < 4000/mm3), and timing of presentation at 4-6 days post-symptom onset as key predictors. When excluding hemoconcentration and leukopenia, the CatBoost AUC-ROC was 96.7% (95% CI: 95.5-98.0%), demonstrating minimal reduction in performance. Individual warning signs like abdominal pain and restlessness had sensitivities of 79.0% and 64.6%, but lower specificities of 48.4% and 59.1%, respectively. Combining ≥ 3 warning signs improved specificity (80.9%) while maintaining moderate sensitivity (78.6%), resulting in an AUC-ROC of 74.0%.

CONCLUSIONS

ML models, especially gradient boosting algorithms, outperformed traditional warning signs in predicting severe dengue. Integrating these models into clinical decision-support tools could help clinicians better identify high-risk patients, guiding timely interventions like hospitalization, closer monitoring, or the administration of intravenous fluids. The subanalysis excluding hemoconcentration confirmed the models' applicability in resource-limited settings, where access to laboratory data may be limited.

A machine learning-based risk score for prediction of mechanical ventilation in children with dengue shock syndrome: A retrospective cohort study

BACKGROUND

Patients with severe dengue who develop severe respiratory failure requiring mechanical ventilation (MV) support have significantly increased mortality rates. This study aimed to develop a robust machine learning-based risk score to predict the need for MV in children with dengue shock syndrome (DSS) who developed acute respiratory failure.

METHODS

This single-institution retrospective study was conducted at a tertiary pediatric hospital in Vietnam between 2013 and 2022. The primary outcome was severe respiratory failure requiring MV in the children with DSS. Key covariables were predetermined by the LASSO method, literature review, and clinical expertise, including age (< 5 years), female patients, early onset day of DSS (≤ day 4), large cumulative fluid infusion, higher colloid-to-crystalloid fluid infusion ratio, severe bleeding, severe transaminitis, low platelet counts (< 20 x 109/L), elevated hematocrit, and high vasoactive-inotropic score. These covariables were analyzed using supervised models, including Logistic Regression (LR), Random Forest (RF), Support Vector Machine (SVM), k-Nearest Neighbor (KNN), and eXtreme Gradient Boosting (XGBoost). Shapley Additive Explanations (SHAP) analysis was used to assess feature contribution.

RESULTS

A total of 1,278 patients were included, with a median patient age of 8.1 years (IQR: 5.4-10.7). Among them, 170 patients (13.3%) with DSS required mechanical ventilation. A significantly higher fatality rate was observed in the MV group than that in the non-MV group (22.4% vs. 0.1%). The RF and SVM models showed the highest model discrimination. The SHAP model explained the significant predictors. Internal validation of the predictive model showed high consistency between the predicted and observed data, with a good slope calibration in training (test) sets 1.0 (0.934), and a low Brier score of 0.04. Complete-case analysis was used to construct the risk score.

CONCLUSIONS

We developed a robust machine learning-based risk score to estimate the need for MV in hospitalized children with DSS.

Machine-learning prediction models for any blood component transfusion in hospitalized dengue patients

BACKGROUND

Blood component transfusions are a common and often necessary medical practice during the epidemics of dengue. Transfusions are required for patients when they developed severe dengue fever or thrombocytopenia of 10×109/L or less. This study therefore investigated the risk factors, performance and effectiveness of eight different machine-learning algorithms to predict blood component transfusion requirements in confirmed dengue cases admitted to hospital. The objective was to study the risk factors that can help to predict blood component transfusion needs.

METHODS

Eight predictive models were developed based on retrospective data from a private group of hospitals in India. A python package SHAP (SHapley Additive exPlanations) was used to explain the output of the "XGBoost" model.

RESULTS

Sixteen vital variables were finally selected as having the most significant effects on blood component transfusion prediction. The XGBoost model presented significantly better predictive performance (area under the curve: 0.793; 95 % confidence interval: 0.699-0.795) than the other models.

CONCLUSION

Predictive modelling techniques can be utilized to streamline blood component preparation procedures and can help in the triage of high-risk patients and readiness of caregivers to provide blood component transfusions when required. This study demonstrates the potential of multilayer algorithms to reasonably predict any blood component transfusion needs which may help healthcare providers make more informed decisions regarding patient care.

Utilization of machine learning for dengue case screening

Dengue causes approximately 10.000 deaths and 100 million symptomatic infections annually worldwide, making it a significant public health concern. To address this, artificial intelligence tools like machine learning can play a crucial role in developing more effective strategies for control, diagnosis, and treatment. This study identifies relevant variables for the screening of dengue cases through machine learning models and evaluates the accuracy of the models. Data from reported dengue cases in the states of Rio de Janeiro and Minas Gerais for the years 2016 and 2019 were obtained through the National Notifiable Diseases Surveillance System (SINAN). The mutual information technique was used to assess which variables were most related to laboratory-confirmed dengue cases. Next, a random selection of 10,000 confirmed cases and 10,000 discarded cases was performed, and the dataset was divided into training (70%) and testing (30%). Machine learning models were then tested to classify the cases. It was found that the logistic regression model with 10 variables (gender, age, fever, myalgia, headache, vomiting, nausea, back pain, rash, retro-orbital pain) and the Decision Tree and Multilayer Perceptron (MLP) models achieved the best results in decision metrics, with an accuracy of 98%. Therefore, a tree-based model would be suitable for building an application and implementing it on smartphones. This resource would be available to healthcare professionals such as doctors and nurses.

Retrospective Analysis of Severe Dengue by Dengue Virus Serotypes in a Population with Social Security, Mexico 2023

BACKGROUND

Risk factors for severe dengue manifestations have been attributed to various factors, including specific serotypes, sex, and age. Mexico has seen the re-emergence of DENV-3, which has not circulated in a decade.

OBJECTIVE

To describe dengue serotypes by age, sex, and their association with disease severity in dengue-positive serum samples from epidemiological surveillance system units.

MATERIALS AND METHODS

A descriptive analysis was conducted to evaluate the frequency of dengue severity by sex, age, disease quarter, geographical location, and dengue virus serotypes. The study was conducted using laboratory samples from confirmed dengue cases through RT-qPCR from the epidemiological surveillance laboratory network of the Mexican Social Security Institute, Mexico. Simple frequencies and proportions were calculated using the z-test for proportional differences between groups. Bivariate analysis with adjusted Chi2 was performed, and binary logistic regression models were constructed using the forward Wald method considering the model's predictive capacity. The measure of association was the odds ratio, with 95% confidence intervals. Statistical significance was set to an alpha level of <0.05.

RESULTS

In 2023, 10,441 samples were processed for dengue RT-qPCR at the IMSS, with a predominance of serotype DENV-3 (64.4%). The samples were mostly from women (52.0%) and outpatient cases (63.3%). The distribution of dengue severity showed significant variations by age, with a lower proportion of severe cases in young children and a higher proportion in the 5- to 14-year-old group. Hospitalizations increased significantly with severity. Warm regions had more cases overall and severity. Cases were most frequent from July to September. While DENV-2 was associated with severity, DENV-4 was not. Binary regression identified higher risk in women, age extremes, and DENV-2, with an overall predictive model of 58.5%.

CONCLUSIONS

Women, age groups at the extremes of life, and the DENV-2 serotype presented severe risk of dengue in a population with social security in Mexico during 2023.

Using Computational Simulations Based on Fuzzy Cognitive Maps to Detect Dengue Complications

Dengue remains a globally prevalent and potentially fatal disease, affecting millions of people worldwide each year. Early and accurate detection of dengue complications is crucial to improving clinical outcomes and reducing the burden on healthcare systems. In this study, we explore the use of computational simulations based on fuzzy cognitive maps (FCMs) to improve the detection of dengue complications. We propose an innovative approach that integrates clinical data into a computational model that mimics the decision-making process of a medical expert. Our method uses FCMs to model complexity and uncertainty in dengue. The model was evaluated in simulated scenarios with each of the dengue classifications. These maps allow us to represent and process vague and fuzzy information effectively, capturing relationships that often go unnoticed in conventional approaches. The results of the simulations show the potential of our approach to detecting dengue complications. This innovative strategy has the potential to transform the way clinical management of dengue is approached. This research is a starting point for further development of complication detection approaches for events of public health concern, such as dengue.

Development of a machine learning model for early prediction of plasma leakage in suspected dengue patients

BACKGROUND

At least a third of dengue patients develop plasma leakage with increased risk of life-threatening complications. Predicting plasma leakage using laboratory parameters obtained in early infection as means of triaging patients for hospital admission is important for resource-limited settings.

METHODS

A Sri Lankan cohort including 4,768 instances of clinical data from N = 877 patients (60.3% patients with confirmed dengue infection) recorded in the first 96 hours of fever was considered. After excluding incomplete instances, the dataset was randomly split into a development and a test set with 374 (70%) and 172 (30%) patients, respectively. From the development set, five most informative features were selected using the minimum description length (MDL) algorithm. Random forest and light gradient boosting machine (LightGBM) were used to develop a classification model using the development set based on nested cross validation. An ensemble of the learners via average stacking was used as the final model to predict plasma leakage.

RESULTS

Lymphocyte count, haemoglobin, haematocrit, age, and aspartate aminotransferase were the most informative features to predict plasma leakage. The final model achieved the area under the receiver operating characteristics curve, AUC = 0.80 with positive predictive value, PPV = 76.9%, negative predictive value, NPV = 72.5%, specificity = 87.9%, and sensitivity = 54.8% on the test set.

CONCLUSION

The early predictors of plasma leakage identified in this study are similar to those identified in several prior studies that used non-machine learning based methods. However, our observations strengthen the evidence base for these predictors by showing their relevance even when individual data points, missing data and non-linear associations were considered. Testing the model on different populations using these low-cost observations would identify further strengths and limitations of the presented model.

Random forest model can predict the prognosis of hospital-acquired Klebsiella pneumoniae infection as well as traditional logistic regression model

OBJECTIVE

To explore if random forest (RF) model can predict the prognosis of hospital-acquired Klebsiella pneumoniae infection as well as traditional logistic regression(LR) model.

METHODS

A total of 254 cases of hospital-acquired Klebsiella pneumoniae infection in a tertiary hospital in Beijing from January 2016 to December 2020 were retrospectively collected. Appropriate influencing factors were selected by referring to relevant articles from the aspects of basic clinical information and contact history before infection, and divided into a training set and a test set. Both the RF and LR models were trained by the training set, and using testing set to compare these two models.

RESULTS

The prediction accuracy of the LR model was 87.0%, the true positive rate of the LR model was 94.7%; the false negative rate of the LR model was 5.3%; the false positive rate of the LR model was 35%; the true negative rate of the LR model was 65%; the sensitivity of the LR model for the prognosis prediction of hospital-acquired Klebsiella pneumoniae infection was 94.7%; and the specificity was 65%. The prediction accuracy of the RF model was 89.6%; the true positive rate of the RF model was 92.1%; the false negative rate of the RF model was 7.9%; the false positive rate of the RF model was 21.4%; the true negative rate of the RF model was 78.6%; the sensitivity of the RF model for the prognosis prediction of hospital-acquired Klebsiella pneumoniae infection was 92.1%; and the specificity was 78.6%. ROC curve shows that the area under curve(AUC) of the LR model was 0.91, and that of the RF model was 0.95.

CONCLUSION

The RF model has higher specificity, sensitivity, and accuracy for the prognostic prediction of hospital-acquired Klebsiella pneumoniae infection than the LR model and has greater clinical application prospects.

Association of dengue virus non-structural-1 protein with disease severity: a brief review

Dengue virus (DENV) was discovered by P. M. Ashburn and Charles F. Craig in 1907. Evidence of dengue-like illness was observed before 1907 and DENV epidemics have been reported from different parts of the world since then, with increased morbidity rates every year. DENV typically causes a febrile illness that ranges from mild asymptomatic infection to fatal dengue haemorrhagic fever (DHF) and/or dengue shock syndrome (DSS). Host mechanisms through which mild infection progresses to the fatal forms are still unknown. Few factors have been associated to aid severe disease acquisition, DENV non-structural 1 (NS1) protein being one of them. NS1 is a highly conserved glycoprotein among the Flavivirus and is often used as a biomarker for dengue diagnosis. This review focuses on assessing the role of NS1 in severe dengue. In this review, hospital-based studies on the association of dengue NS1 with severe dengue from all over the world have been assessed and analysed and the majority of the studies positively correlate high NS1 levels with DHF/DSS acquisition. The review also discusses a few experimental studies on NS1 that have shown it contributes to dengue pathogenesis. This review assesses the role of NS1 and disease severity from hospital-based studies and aims to provide better insights on the kinetics and dynamics of DENV infection with respect to NS1 for a better understanding of the role of NS1 in dengue.

Discovery and validation of circulating miRNAs for the clinical prognosis of severe dengue

Background

Early prognostic markers of severe dengue may improve case management and reduce dengue-related mortalities. This study aimed to identify circulating microRNAs (miRNAs) as biomarkers for predicting severe dengue.

Methodology

Serum samples from dengue-infected patients were collected on the first day of admission. Patients were followed up for 14 days after admission to determine the final diagnosis. Participants were divided into non-severe and severe dengue, as defined by WHO 2009 criteria. Circulating microtranscriptome analysis was performed using NanoString miRNA Expression Assay. The expression level of candidate miRNAs were then validated by quantitative reverse transcription-PCR method.

Principal findings

The discovery cohort (N = 19) lead to the identification of 37 differentially expressed miRNAs between the two groups. Six up-regulated candidate miRNAs were selected and further validated in the larger cohort (N = 135). MiR574-5p and miR1246 displayed the highest diagnostic performance in discriminating between severe from non-severe dengue (ROC-AUC = 0.83). Additionally, miR574-5p and miR1246 had high sensitivity and high negative predictive value for detecting severe dengue. Multivariate analysis suggested that serum miR574-5p was an independent predictor of severe dengue (odds ratio 3.30, 95% CI 1.81–6.04; p<0.001).

Conclusion

Our study indicated that circulating miRNAs, especially miR-574-5p and miR-1246, might be a promising diagnostic and prognostic biomarker for severe dengue upon hospital admission, especially when using these biomarkers on days 1 to 2 before the onset of severe dengue complications.

Dengue infection, a mosquito-borne disease, is an expanding global problem. It has a broad clinical spectrum that includes severe and non-severe clinical manifestations with a high risk of death. Identifying early prognostic markers of severe complications may improve case management and reduce dengue-related mortalities. The circulating microRNA (miRNA) profile has been widely used to identify potential biomarkers against viral infections. Our data revealed that the circulating miRNA expression pattern of severe dengue patients was significantly different from the non-severe group. In addition, circulating miRNAs, especially miR-574-5p and miR-1246, could be promising diagnostic and prognostic biomarkers for severe dengue. These data have implications for developing biomarkers for clinical use and could improve risk prediction in dengue patients.

An autonomous cycle of data analysis tasks for the clinical management of dengue

Dengue is the most widespread vector-borne disease worldwide. Timely diagnosis and treatment of dengue is the main objective of medical professionals to decrease mortality rates. In this paper, we propose an autonomous cycle that integrates data analysis tasks to support decision-making in the clinical management of dengue. Particularly, the autonomous cycle supports dengue diagnosis and treatment. The proposed system was built using machine learning techniques for classification tasks (artificial neural networks and support vector machines) and evolutionary techniques (a genetic algorithm) for prescription tasks (treatment). The system was quantitatively evaluated using dengue-patient datasets reported by healthcare institutions. Our system was compared with previous works using qualitative criteria. The proposed system has the ability to classify a patient's clinical picture and recommend the best treatment option. In particular, the classification of dengue was done with 98% accuracy and a genetic algorithm recommends treatment options for particular patients. Finally, our system is flexible and easily adaptable, which will allow the addition of new tasks for dengue analysis.

Machine learning-based in-hospital mortality prediction of HIV/AIDS patients with Talaromyces marneffei infection in Guangxi, China

Objective

Talaromycosis is a serious regional disease endemic in Southeast Asia. In China, Talaromyces marneffei (T. marneffei) infections is mainly concentrated in the southern region, especially in Guangxi, and cause considerable in-hospital mortality in HIV-infected individuals. Currently, the factors that influence in-hospital death of HIV/AIDS patients with T. marneffei infection are not completely clear. Existing machine learning techniques can be used to develop a predictive model to identify relevant prognostic factors to predict death and appears to be essential to reducing in-hospital mortality.

Methods

We prospectively enrolled HIV/AIDS patients with talaromycosis in the Fourth People’s Hospital of Nanning, Guangxi, from January 2012 to June 2019. Clinical features were selected and used to train four different machine learning models (logistic regression, XGBoost, KNN, and SVM) to predict the treatment outcome of hospitalized patients, and 30% internal validation was used to evaluate the performance of models. Machine learning model performance was assessed according to a range of learning metrics, including area under the receiver operating characteristic curve (AUC). The SHapley Additive exPlanations (SHAP) tool was used to explain the model.

Results

A total of 1927 HIV/AIDS patients with T. marneffei infection were included. The average in-hospital mortality rate was 13.3% (256/1927) from 2012 to 2019. The most common complications/coinfections were pneumonia (68.9%), followed by oral candida (47.5%), and tuberculosis (40.6%). Deceased patients showed higher CD4/CD8 ratios, aspartate aminotransferase (AST) levels, creatinine levels, urea levels, uric acid (UA) levels, lactate dehydrogenase (LDH) levels, total bilirubin levels, creatine kinase levels, white blood-cell counts (WBC) counts, neutrophil counts, procaicltonin levels and C-reactive protein (CRP) levels and lower CD3⁺ T-cell count, CD8⁺ T-cell count, and lymphocyte counts, platelet (PLT), high-density lipoprotein cholesterol (HDL), hemoglobin (Hb) levels than those of surviving patients. The predictive XGBoost model exhibited 0.71 sensitivity, 0.99 specificity, and 0.97 AUC in the training dataset, and our outcome prediction model provided robust discrimination in the testing dataset, showing an AUC of 0.90 with 0.69 sensitivity and 0.96 specificity. The other three models were ruled out due to poor performance. Septic shock and respiratory failure were the most important predictive features, followed by uric acid, urea, platelets, and the AST/ALT ratios.

Conclusion

The XGBoost machine learning model is a good predictor in the hospitalization outcome of HIV/AIDS patients with T. marneffei infection. The model may have potential application in mortality prediction and high-risk factor identification in the talaromycosis population.

Talaromyces marneffei can cause a fatal deeply disseminated fungal infection- talaromycosis. It is widely distributed in Southeast Asia and spreading globally, the disease is insidious and responsible for significant deaths. Clinicians need easy-to-use tools to make decisions on which patients are at a higher risk of dying after infecting T. marneffei. In this study, conducted in Southern China, we have evolved XGBoost machine learning model. 15 clinical indicators and laboratory measures were used to estimate a patient’s risk of dying in the hospital due to the T. marneffei infection. The study showed that the machine learning model has good predictive ability when tested in an internal testing population of patients. We expect that the model could help clinicians assess a patient’s risk of death in just the time of admission to help decide on early treatment timing of high-risk patients who are likely to die.

An 8-gene machine learning model improves clinical prediction of severe dengue progression

BACKGROUND

Each year 3-6 million people develop life-threatening severe dengue (SD). Clinical warning signs for SD manifest late in the disease course and are nonspecific, leading to missed cases and excess hospital burden. Better SD prognostics are urgently needed.

METHODS

We integrated 11 public datasets profiling the blood transcriptome of 365 dengue patients of all ages and from seven countries, encompassing biological, clinical, and technical heterogeneity. We performed an iterative multi-cohort analysis to identify differentially expressed genes (DEGs) between non-severe patients and SD progressors. Using only these DEGs, we trained an XGBoost machine learning model on public data to predict progression to SD. All model parameters were "locked" prior to validation in an independent, prospectively enrolled cohort of 377 dengue patients in Colombia. We measured expression of the DEGs in whole blood samples collected upon presentation, prior to SD progression. We then compared the accuracy of the locked XGBoost model and clinical warning signs in predicting SD.

RESULTS

We identified eight SD-associated DEGs in the public datasets and built an 8-gene XGBoost model that accurately predicted SD progression in the independent validation cohort with 86.4% (95% CI 68.2-100) sensitivity and 79.7% (95% CI 75.5-83.9) specificity. Given the 5.8% proportion of SD cases in this cohort, the 8-gene model had a positive and negative predictive value (PPV and NPV) of 20.9% (95% CI 16.7-25.6) and 99.0% (95% CI 97.7-100.0), respectively. Compared to clinical warning signs at presentation, which had 77.3% (95% CI 58.3-94.1) sensitivity and 39.7% (95% CI 34.7-44.9) specificity, the 8-gene model led to an 80% reduction in the number needed to predict (NNP) from 25.4 to 5.0. Importantly, the 8-gene model accurately predicted subsequent SD in the first three days post-fever onset and up to three days prior to SD progression.

CONCLUSIONS

The 8-gene XGBoost model, trained on heterogeneous public datasets, accurately predicted progression to SD in a large, independent, prospective cohort, including during the early febrile stage when SD prediction remains clinically difficult. The model has potential to be translated to a point-of-care prognostic assay to reduce dengue morbidity and mortality without overwhelming limited healthcare resources.

Assessment of the Risk of Severe Dengue Using Intrahost Viral Population in Dengue Virus Serotype 2 Patients via Machine Learning

Dengue virus, a positive-sense single-stranded RNA virus, continuously threatens human health. Although several criteria for evaluation of severe dengue have been recently established, the ability to prognose the risk of severe outcomes for dengue patients remains limited. Mutant spectra of RNA viruses, including single nucleotide variants (SNVs) and defective virus genomes (DVGs), contribute to viral virulence and growth. Here, we determine the potency of intrahost viral population in dengue patients with primary infection that progresses into severe dengue. A total of 65 dengue virus serotype 2 infected patients in primary infection including 17 severe cases were enrolled. We utilized deep sequencing to directly define the frequency of SNVs and detection times of DVGs in sera of dengue patients and analyzed their associations with severe dengue. Among the detected SNVs and DVGs, the frequencies of 9 SNVs and the detection time of 1 DVG exhibited statistically significant differences between patients with dengue fever and those with severe dengue. By utilizing the detected frequencies/times of the selected SNVs/DVG as features, the machine learning model showed high average with a value of area under the receiver operating characteristic curve (AUROC, 0.966 ± 0.064). The elevation of the frequency of SNVs at E (nucleotide position 995 and 2216), NS2A (nucleotide position 4105), NS3 (nucleotide position 4536, 4606), and NS5 protein (nucleotide position 7643 and 10067) and the detection times of the selected DVG that had a deletion junction in the E protein region (nucleotide positions of the junction: between 969 and 1022) increased the possibility of dengue patients for severe dengue. In summary, we demonstrated the detected frequencies/times of SNVs/DVG in dengue patients associated with severe disease and successfully utilized them to discriminate severe patients using machine learning algorithm. The identified SNVs and DVGs that are associated with severe dengue will expand our understanding of intrahost viral population in dengue pathogenesis.

Applied machine learning for the risk-stratification and clinical decision support of hospitalised patients with dengue in Vietnam

BACKGROUND

Identifying patients at risk of dengue shock syndrome (DSS) is vital for effective healthcare delivery. This can be challenging in endemic settings because of high caseloads and limited resources. Machine learning models trained using clinical data could support decision-making in this context.

METHODS

We developed supervised machine learning prediction models using pooled data from adult and paediatric patients hospitalised with dengue. Individuals from 5 prospective clinical studies in Ho Chi Minh City, Vietnam conducted between 12th April 2001 and 30th January 2018 were included. The outcome was onset of dengue shock syndrome during hospitalisation. Data underwent random stratified splitting at 80:20 ratio with the former used only for model development. Ten-fold cross-validation was used for hyperparameter optimisation and confidence intervals derived from percentile bootstrapping. Optimised models were evaluated against the hold-out set.

FINDINGS

The final dataset included 4,131 patients (477 adults and 3,654 children). DSS was experienced by 222 (5.4%) of individuals. Predictors were age, sex, weight, day of illness at hospitalisation, indices of haematocrit and platelets over first 48 hours of admission and before the onset of DSS. An artificial neural network model (ANN) model had best performance with an area under receiver operator curve (AUROC) of 0.83 (95% confidence interval [CI], 0.76-0.85) in predicting DSS. When evaluated against the independent hold-out set this calibrated model exhibited an AUROC of 0.82, specificity of 0.84, sensitivity of 0.66, positive predictive value of 0.18 and negative predictive value of 0.98.

INTERPRETATION

The study demonstrates additional insights can be obtained from basic healthcare data, when applied through a machine learning framework. The high negative predictive value could support interventions such as early discharge or ambulatory patient management in this population. Work is underway to incorporate these findings into an electronic clinical decision support system to guide individual patient management.

Dengue models based on machine learning techniques: A systematic literature review

BACKGROUND

Dengue modeling is a research topic that has increased in recent years. Early prediction and decision-making are key factors to control dengue. This Systematic Literature Review (SLR) analyzes three modeling approaches of dengue: diagnostic, epidemic, intervention. These approaches require models of prediction, prescription and optimization. This SLR establishes the state-of-the-art in dengue modeling, using machine learning, in the last years.

METHODS

Several databases were selected to search the articles. The selection was made based on Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodology. Sixty-four articles were obtained and analyzed to describe their strengths and limitations. Finally, challenges and opportunities for research on machine-learning for dengue modeling were identified.

RESULTS

Logistic regression was the most used modeling approach for the diagnosis of dengue (59.1%). The analysis of the epidemic approach showed that linear regression (17.4%) is the most used technique within the spatial analysis. Finally, the most used intervention modeling is General Linear Model with 70%.

CONCLUSIONS

We conclude that cause-effect models may improve diagnosis and understanding of dengue. Models that manage uncertainty can also be helpful, because of low data-quality in healthcare. Finally, decentralization of data, using federated learning, may decrease computational costs and allow model building without compromising data security.

Different Profiles of Cytokines, Chemokines and Coagulation Mediators Associated with Severity in Brazilian Patients Infected with Dengue Virus

The incidence of dengue in Latin America has increased dramatically during the last decade. Understanding the pathogenic mechanisms in dengue is crucial for the identification of biomarkers for the triage of patients. We aimed to characterize the profile of cytokines (IFN-γ, TNF-α, IL-1β, IL-6, IL-18 and IL-10), chemokines (CXCL8/IL-8, CCL2/MCP-1 and CXCL10/IP-10) and coagulation mediators (Fibrinogen, D-dimer, Tissue factor-TF, Tissue factor pathway inhibitor-TFPI and Thrombomodulin) during the dengue-4 epidemic in Brazil. Laboratory-confirmed dengue cases had higher levels of TNF-α (p < 0.001), IL-6 (p = 0.005), IL-10 (p < 0.001), IL-18 (p = 0.001), CXCL8/IL-8 (p < 0.001), CCL2/MCP-1 (p < 0.001), CXCL10/IP-10 (p = 0.001), fibrinogen (p = 0.037), D-dimer (p = 0.01) and TFPI (p = 0.042) and lower levels of TF (p = 0.042) compared to healthy controls. A principal component analysis (PCA) distinguished between two profiles of mediators of inflammation and coagulation: protective (TNF-α, IL-1β and CXCL8/IL-8) and pathological (IL-6, TF and TFPI). Lastly, multivariate logistic regression analysis identified high aspartate aminotransferase-to-platelet ratio index (APRI) as independent risk factors associated with severity (adjusted OR: 1.33; 95% CI 1.03–1.71; p = 0.027), the area under the receiver operating characteristics curve (AUC) was 0.775 (95% CI 0.681–0.869) and an optimal cutoff value was 1.4 (sensitivity: 76%; specificity: 79%), so it could be a useful marker for the triage of patients attending primary care centers.