Classification of Malaria Parasite Plasmodium Falciparum Based on Blood Smear Images Using Support Vector Machine Approach

Malaria remained a significant global health issue, particularly in tropical and subtropical regions. The disease resulted in a substantial number of clinical cases and deaths each year, with high-risk groups including infants, toddlers, and pregnant women. Accurate and prompt diagnosis was a key factor in managing the disease. To address this issue, the research aimed to develop an automated system for the classification of Plasmodium falciparum malaria parasites based on blood smear images. The methods employed included image feature selection using Principal Component Analysis (PCA) and the Support Vector Machine (SVM) approach for classification. The research findings indicated that in the image feature selection process, the category of normal malaria exhibited distinctive characteristics with PC1 and PC2 values that tended to be negative and dispersed, whereas the category of parasitic malaria displayed greater variability in both PC1 and PC2 components. Furthermore, the evaluation of the classification system's accuracy using SVM with three different kernel types showed promising results. The average accuracy through K-fold cross-validation for the polyinomial, linear, and radial basis function kernels was 96.7%, 98.9%, and 94.4%, respectively. These results highlighted the significant potential of SVM utilization in the classification of Plasmodium falciparum malaria parasites based on blood smear images.