Deep Malaria Parasite Detection in Thin Blood Smear Microscopic Images

Malaria is a disease activated by type of microscopic parasite transmitted from infected female mosquito bites to humans. fatal that endemic in many regions the world. Quick diagnosis this will be very valuable for patients, as traditional methods require tedious work its detection. Recently, some automated have been proposed exploit hand-crafted feature extraction techniques however, their accuracies are not reliable. Deep learning approaches modernize world with superior performance. Convolutional Neural Networks (CNN) vastly scalable image classification tasks extract features through hidden layers model without any handcrafting. The detection malaria-infected red blood cells segmented images using convolutional neural networks can assist quick diagnosis, and useful fewer healthcare experts. contributions paper two-fold. First, we evaluate performance different existing deep models efficient malaria Second, propose customized CNN outperforms all observed models. It exploits bilateral filtering augmentation highlighting before training model. Due techniques, generalized avoids over-fitting. All experimental evaluations performed on benchmark NIH Dataset, results reveal algorithm 96.82% accurate detecting smears.