Design of Ensemble Learning Model to Diagnose Malaria Disease Using Convolutional Neural Network

Malaria is one of the most fatal infectious diseases caused by the Plasmodium parasite. It often goes undiagnosed due to human error. Machine learning and many other computational algorithms of soft computing can be used as a novelty approach which can be effectively applied on diagnosing malaria. In this paper, a novel deep neural network model is identified as optical microscopy. An ensemble model emerged the concept of convolutional neural network as LeNet, AlexNet and ResNet to obtain the accurate result. The accuracy of input modified ResNet comes 90.50% after training for four epochs; accuracy of modified AlexNet with input size of 64 * 64 * 3 is 93.50% with just three epochs, and finally, we applied modified LeNet which gives accuracy of 95.50% with 10 epochs from UCI dataset. The obtained result shows that deep neural network models have potential effect in healthcare domain especially in malaria diagnosis.