Over the past few years, the corona has created havoc across the globe but still, malaria is holding theposition of disease with the highest mortality rate in few parts of the world. Malaria is caused by thebite of female mosquitoes - Anopheles. According to WHO, in 2020 the total number of malaria casesreportedwasmorethan240millionandabout627,000deathswerereported[5].Itcanbeexaminedontime, so now the major concern is to identify if a person is affected by malaria or not. There are manytraditionalwaystotestformalariabuteithertheyrequirehighlycompetentdoctorsormaygiveresults in high time. Scaling of this old technique is very difficult and not having doctors with properexpertise in rural areas is also a problem. So, in this paper, we have used a Convolutional NeuralNetwork (CNN) to classify the blood images as infected or not and get the results faster. Threedifferent deep learning models were compared to find out the most accurate model which willautomatetheprocessandcanbeusedbydoctorsinremoteareastogetfasterresults.

ConvolutionalNeuralNetwork;DeepLearning;MalariaDetection;VGG-16,RESNET-50;Inceptionv3;layers;transferlearning