ML-Driven Approach forMalaria Infection Prediction from Blood Smear Image Analysis

Authors

  • K. Sharmila, P. Shailaha Author

DOI:

https://doi.org/10.48047/

Keywords:

Public health, Malaria disease, Machine learning, Portable diagnostic devices, Naïve Bayes model, Ensemble learning model.

Abstract

Many countries still struggle with malaria, a deadly disease spread by Plasmodium parasites in infected mosquitoes. Malaria therapy and control require early and precise detection. Healthcare workers can perform speedy and accurate malaria tests in distant or resource-limited situations using portable diagnostic instruments using the automated malaria detection system. The approach can help epidemiologists and health organizations track malaria prevalence and transmission, improving resource allocation. Traditional malaria detection involves experienced personnel manually examining blood smears under a microscope. Reliable, yet time-consuming, laborious, and dependent on microscopist expertise. Regression-based blood smear analysis can yield false-negative or false-positive results. Recently, machine learning-based methods have showed promise in automating malaria parasite detection in blood samples. Therefore,this research provides a high-accuracy, efficient machine learning-based malaria infection detection approach from blood smear image analysis. The proposed system employs two machine learning models such as naïve bayes, and ensemble learning model for performance assessment. Obtained simulation results demonstrate the ensemble learning model outperforms the naïve bayes classifier with improved prediction accuracy.

Downloads

Download data is not yet available.

Downloads

Published

2022-11-16

Issue

Section

Articles

How to Cite

K. Sharmila, P. Shailaha. (2022). ML-Driven Approach forMalaria Infection Prediction from Blood Smear Image Analysis. History of Medicine, 8(2). https://doi.org/10.48047/