Study of machine learning techniques for outcome assessment of leptospirosis patients
2024

Machine Learning for Predicting Leptospirosis Outcomes

Sample size: 1319 publication 10 minutes Evidence: high

Author Information

Author(s): Silva Andreia Ferreira da, Figueiredo Karla, Falcão Igor W. S., Costa Fernando A. R., da Rocha Seruffo Marcos César, de Moraes Carla Cristina Guimarães

Primary Institution: Federal University of Para

Hypothesis

Can machine learning algorithms accurately predict the outcomes of leptospirosis patients?

Conclusion

The study confirms that the Decision Tree model is the best choice for predicting outcomes in leptospirosis cases.

Supporting Evidence

  • The Random Forest model achieved an accuracy of 90.81% on the training dataset.
  • The Decision Tree model was identified as the best for future predictions.
  • Data was collected from 1,319 leptospirosis cases over a 10-year period.
  • Machine learning techniques can improve the diagnosis and treatment of leptospirosis.

Takeaway

This study used computers to help doctors figure out if people with a sickness called leptospirosis will get better or not.

Methodology

The study analyzed data from the National System of Aggressions and Notification (SINAN) using three machine learning algorithms: Decision Tree, Random Forest, and Adaboost.

Potential Biases

Potential biases due to the reliance on government health records and the socio-economic context of the data.

Limitations

The study relied on historical data, which may have missing values and underreporting of cases.

Participant Demographics

Patients from the state of Pará, Brazil, with leptospirosis from 2007 to 2017.

Digital Object Identifier (DOI)

10.1038/s41598-024-62254-1

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication