Deep Learning Based Approach For Malaria Detection in Blood Cell Images
Amogh Manoj Joshi, Ananta Kumar Das, Subhasish Dhal
Abstract
Malaria, a life-threatening disease, develops due to the bite of female Anopheles mosquito. It spreads the plasmodium parasites in human blood, killing hundreds of millions of people every year. Modern scientific advancements play a pivotal role to combat the disease, along with biomedical research by the medical experts to possibly eradicate this disease from all parts of the world. With the significant development in deep learning research, faultless identification of medical imaging has become an important factor in medical diagnosis and decision-making. To this end, we present a deep learning based approach using a convolutional neural network for detecting malaria from microscopic cell images using image classification. The proposed CNN model implemented using 5-fold cross validation approach outperforms all the existing methods in terms of accuracy and other evaluation metrics, thus achieving the best results till date in malaria detection using deep learning.