Effective Performance of Bins Approach for Classification of Malaria Parasite using Machine Learning

Malaria is a severe infectious disease transferred through the bite of an infected mosquito caused by a blood parasite of the genus Plasmodium. In the past, conventional microscopy techniques have proven to be time-consuming and had observed a lack of differentiation due to poor accuracy and a few algorithms used. In this paper, our approach primarily focuses on image processing techniques to process and enhance stained thin blood smear images for feature extraction, as well as machine learning techniques for the final classification of feature space. Our emphasis is also to address the drawbacks, as mentioned earlier, by taking input cell images and performing classification using CNN. As an alternate approach, we have also computed color features using a novel Bins Approach Algorithm, statistical features using color moments, and texture features using GLCM, which also equally played a pivotal role in feature extraction for classification. Further, these images are classified into parasitized and uninfected cells by applying machine learning classifiers such as Linear SVM, Random Forest Algorithm (RNN), and KNN over feature space. The proposed algorithms have been experimented using the subset of Lister Hill National Center for Biomedical Communication (LHNCBC) dataset, a division of the National Library of Medicine (NLM). The performance of the algorithms is evaluated and compared using different parameters like accuracy, precision, recall, and F1-score. The proposed application of Bins Approach in malaria parasite detection has proved better in terms of all parameters as compared to the other existing algorithms.