Classification of Malaria in Red Blood Cell Microscopic Images Using Deep Learning with EfficientNet Architecture and SVM

Malaria remains a global health concern, caused by the Plasmodium parasite's invasion of human red blood cells primarily through female Anopheles mosquito bites. Early detection is vital for timely treatment and preventing severe illness. However, current malaria detection methods demand high precision and expertise, particularly when handling large sample sizes. This study introduces a malaria detection model employing Deep Learning CNN with the EfficientNet architecture for feature extraction, and Support Vector Machine (SVM) classification. The dataset comprised 2,101 blood cell images categorized into malaria-infected and uninfected classes. The most robust performance was achieved with the EfficientNet-SVM architecture, yielding 96% accuracy, precision, recall, and F1-score. This research strives to enhance the efficiency and accuracy of malaria detection, crucial for timely intervention and disease prevention. The proposed model presents promising results in malaria detection and contributes to the ongoing fight against this perilous disease.