prashanth-gajula/maleria-cell-classification

In the current situation, human health is a major concern. Various diseases have a negative impact on human society and malaria is one of the deadliest diseases it has taken many lives, This infectious disease which started in Africa infected every continent except Antarctica, If the infection becomes severe it may cause kidney failure, seizures, mental confusion, coma, and death so it is important to diagnose malaria at an early stage to save lives. Our goal is to classify the given cell image into one of the two classes parasitized or un-infected. The dataset used for this project contains 27,562 cell images of two classes, and in this project, we used computer vision and deep learning techniques like CNN, and some pre-trained techniques like Resnet-50, VGG16, and Densenet-201. 70% of the total data i.e 19,292 images are used for training and 8266 images are used for Validation, to evaluate the performance Accuracy was used as a metric and achieved an validation accuracy of 0.90%,0.88%,0.74,0.68,0.92 for the above-mentioned models. In the current situation, human health is a major concern. Various diseases have a negative impact on human society and malaria is one of the deadliest diseases it has taken many lives, This infectious disease which started in Africa infected every continent except Antarctica, If the infection becomes severe it may cause kidney failure, seizures, mental confusion, coma, and death so it is important to diagnose malaria at an early stage to save lives. Our goal is to classify the given cell image into one of the two classes parasitized or un-infected. The dataset used for this project contains 27,562 cell images of two classes, and in this project, we used computer vision and deep learning techniques like CNN, and some pre-trained techniques like Resnet-50, VGG16, and Densenet-201. 70% of the total data i.e 19,292 images are used for training and 8266 images are used for Validation, to evaluate the performance Accuracy was used as a metric and achieved an validation accuracy of 0.90%,0.88%,0.74,0.68,0.92 for the above-mentioned models.