A Robust Approach for Malaria Parasite Identification with CNN Based Feature Extraction and Classification using SVM
Kanak Ohdar, Akriti Nigam · 2023
Malaria is a fatal illness caused by the transmission of the Plasmodium parasite through Anopheles mosquitoes. It causes symptoms like fever, chill, headache, and fatigue. In severe cases, it also leads to organ failure and death. The conventional method of detecting malaria is through microscopy. The development of new tools and algorithms for malaria diagnosis and treatment is crucial. In this study, a CNN-based model has been proposed for the feature extraction from raw thin blood smear images. The extracted features are then classified into one of the four species types of namely Falciparum, Vivax, Ovale and Malariae using an SVM classifier. The proposed model achieves a significantly good accuracy of 98%, demonstrating the effectiveness of deep learning techniques in Malaria diagnosis and species classification. This study has significant implications for the development of new tools for Malaria diagnosis and treatment, which could help to speed up the process of detection and help the medical practitioner make an accurate analysis of infection, particularly in rural areas, where medical facilities are scarce.