Smart Malaria Classification: A Novel Machine Learning Algorithms for Early Malaria Monitoring and Detecting Using IoT-Based Healthcare Environment

Malaria, caused by the Plasmodium parasite and transmitted by female Anopheles mosquitoes, poses a significant risk to nearly half of the global population, with sub-Saharan Africa being the most affected. A rapid and accurate detection method is crucial due to its high mortality rate and swift transmission. This study proposes a real-time malaria monitoring and detection system using an Internet of Things (IoT) framework. The system collects real-time symptom data via wearable sensors, employs edge computing for processing, utilizes cloud infrastructure for data storage, and applies machine learning models for data analysis. The five key components of the framework are wearable sensor-based symptom data collection and uploading, edge (fog) computing, cloud infrastructure, machine learning models for data analysis, and doctors (physicians). The study compares four machine learning techniques: Support Vector Machine (SVM), Artificial Neural Network (ANN), K-Nearest Neighbor (KNN), and Na & iuml;ve Bayes. SVM outperformed the other algorithms, achieving 98% training accuracy, 96% test accuracy, and a 95% AUC score. Based on the findings, we anticipate that real-time symptom data would enable the proposed system can effectively and accurately diagnose malaria, classifying cases as either Parasitized or Normal.