Utilization of machine learning for dengue case screening

Bohm, Bianca Conrad; Borges, Fernando Elias de Melo; Silva, Suellen Caroline Matos; Soares, Alessandra Talaska; Ferreira, Danton Diego; Belo, Vinícius Silva; Lignon, Julia Somavilla; Bruhn, Fábio Raphael Pascoti

doi:10.1186/s12889-024-19083-8

Cited by 2 publications

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

Machine learning for improved dengue diagnosis, Puerto Rico

Madewell,

Rodriguez,

Thayer

et al. 2024

Preprint

View full text Add to dashboard Cite

Background: Diagnosing dengue accurately, especially in resource-limited settings, remains challenging due to overlapping symptoms with other febrile illnesses and limitations of current diagnostic methods. This study aimed to develop machine learning (ML) models that leverage readily available clinical data to improve diagnostic accuracy for dengue, potentially offering a more accessible and rapid diagnostic tool for healthcare providers. Methods: We used data from the Sentinel Enhanced Dengue Surveillance System (SEDSS) in Puerto Rico (May 2012-June 2024). SEDSS primarily targets acute febrile illness but also includes cases with other symptoms during outbreaks (e.g., Zika and COVID-19). ML models (logistic regression, random forest, support vector machine, artificial neural network, adaptive boosting, light gradient boosting machine [LightGBM], and extreme gradient boosting [XGBoost]) were evaluated across different feature sets, including demographic, clinical, laboratory, and epidemiological variables. Model performance was assessed using the area under the receiver operating characteristic curve (AUC), where higher AUC values indicate better performance in distinguishing dengue cases from non-dengue cases. Results: Among 49,679 patients in SEDSS, 1,640 laboratory-confirmed dengue cases were identified. The XGBoost and LightGBM models achieved the highest diagnostic accuracy, with AUCs exceeding 90%, particularly with comprehensive feature sets. Incorporating predictors such as monthly dengue incidence, leukopenia, thrombocytopenia, rash, age, and absence of nasal discharge significantly enhanced model sensitivity and specificity for diagnosing dengue. Adding more relevant clinical and epidemiological features consistently improved the models' ability to correctly identify dengue cases. Conclusions: ML models, especially XGBoost and LightGBM, show promise for improving diagnostic accuracy for dengue using widely accessible clinical data, even in resource-limited settings. Future research should focus on developing user-friendly tools, such as mobile apps, web-based platforms, or clinical decision systems integrated into electronic health records, to implement these models in clinical practice and exploring their application for predicting dengue.

show abstract

Machine learning for improved dengue diagnosis, Puerto Rico

Madewell,

Rodriguez,

Thayer

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

An automated diagnosis & classification of dengue using advance artificial neural network

Hayat,

Anwar,

Aziz

2024

Comput. Artif. Intell.

View full text Add to dashboard Cite

In this research, an advanced artificial neural network (ANN)-based approach for prognosis and classification of dengue disease is presented. Dengue diagnosis usually relies on clinical assessment; subsequently, there might be a high probability of misdiagnoses due to the complex hodgepodge of symptoms of dengue with other vector-borne diseases. It is needed to develop a system that can help doctors to identify dengue disease much faster than the manual system, which takes longer time and more cost to detect the diseases. Such a system may help users to take an early action before it becomes serious. The study involved three phases: pre-processing, neural network processing, and post-processing. In the pre-processing phase, data were gathered from three high-severity dengue outbreak sites in Pakistan (Benazir Bhutto Hospital, CITI Lab Rawalpindi, and Meo Hospital Lahore) where the dengue outbreak severity was high during the year of 2011. After cleaning and normalizing, 768 samples were obtained, split into 560 for training and 208 for testing. Nineteen critical parameters were selected with input from physicians, medical staff, and prior research. This study presents a supervised feed-forward neural network (FFNN) with two hidden layers, trained using backpropagation and optimized with the Levenberg-Marquardt algorithm, achieving nearly 100% accuracy, minimal runtime, and a very low MSE (0.00000000000032521). The model reached 100% sensitivity, 99.8% precision, and 98.7% specificity, surpassing prior results in dengue diagnosis. The findings support improved diagnostic accuracy and confidence, providing a framework for physicians. Key factors in achieving optimal results include careful selection of architecture, data normalization, parameter selection, and critical evaluation.

show abstract

Utilization of machine learning for dengue case screening

Cited by 2 publications

References 33 publications

Machine learning for improved dengue diagnosis, Puerto Rico

Machine learning for improved dengue diagnosis, Puerto Rico

An automated diagnosis & classification of dengue using advance artificial neural network

Contact Info

Product

Resources

About