Dengue is the most widespread vector-borne disease globally. Its timely diagnosis and treatment can decrease mortality rates. Hoyos W. et al., in their recent study, proposed an autonomous cycle that integrated data analysis tasks to back decision-making in managing dengue. Notably, the autonomous cycle supported dengue diagnosis and treatment. They built this system using machine learning techniques for classification tasks (artificial neural networks and support vector machines) and evolutionary techniques (a genetic algorithm) for prescription tasks (treatment). They quantitatively evaluated the system using dengue-patient datasets reported by healthcare institutions and compared the system with previous works using qualitative criteria. Their proposed system classified the patients′ clinical picture and recommended the best treatment option. In particular, it classified dengue with 98% accuracy, and a genetic algorithm recommended treatment options for specific patients. Their system is flexible and easily adaptable, allowing new tasks to be added for dengue analysis.

This is the first study that utilized an autonomic approach to support the clinical management of dengue. Additionally, it is also the first work to propose a prescriptive model for the clinical management of this disease.

Future work is warranted to improve the models by implementing routine laboratory tests such as white blood cell counts and liver enzymes and cytokines levels. Additionally, including comorbidities such as diabetes and arterial hypertension will improve the performance of the models due to the influence of these diseases on the severity of dengue. Finally, creating available datasets with prescriptive or treatment variables will help to validate the results of prescriptive models.

Hoyos W, Aguilar J, Toro M. An autonomous cycle of data analysis tasks for the clinical management of dengue. Heliyon. 2022;8(10):e10846. https://doi.org/10.1016/j.heliyon.2022.e10846.