Malaria remains one of the most predominant infectious and lethal diseases in the tropical regions of the world. Hence, it requires accurate detection at early stages for efficacious treatment and containment. The aim of the research is to support the existing manual pathological examination and inference techniques, with automated processes that will eradicate the possibility of errors because of human intervention. An AI-driven technique that uses deep learning with a Multiclass CNN model is proposed, to classify just not the normal and malaria infected red cells, but also the gametocytes of the malaria parasites from blood smear images, which is a novel application in clinical practice especially in the field of pathology. Developing an AI-based platform to classify the images with a high level of accuracy, available on the internet will potentially bring this service in countries like India to geographically inaccessible regions at little cost.