Leveraging Supervised Machine Learning Techniques for Identification of Malaria Cells using Blood Smears

Plasmodium parasite is identified as amenable for spreading a disease named Malaria. Under the stodgy method, the blood splotch is first smeared on the slide, scrutinized under the microscope, and parasites (which can cause malaria) in blood cells are detected. For beneficial parasite detection, image processing proves to be very much dominant. The reason for this is accuracy in the results. This research presents Malaria detection in blood smear images using supervised learning methods. This proposed method starts with the preprocessing in which images are resized and converted into grayscale. The thresholding technique is implemented to identify blobs for segmentation. For feature extraction, GoogLeNet is maneuvered, and the results of the classification show that this method has an accuracy of 95.8% for detecting malaria in blood smear images.