Convolutional Neural Network Based Malaria Parasite Infection Detection Using Thin Microscopic Blood Smear Samples
Sidharth S Prakash, Binsu C. Kovoor, K. Visakha · 2020 Second International Conference on Inventive Research in Computing Applications (ICIRCA) · 2020
Malaria stands as a major health issue across the globe even though it is both preventable and treatable. Malaria is usually diagnosed by an expert microbiologist through the examination of a microscopic blood smear sample. Malaria can be easily cured if diagnosed at an early stage and followed up with proper medical treatment. Computer-aided diagnosis is gaining popularity nowadays as it can be effectively used as a primary screening test in the absence of an expert microbiologist. Deep learning is an artificial intelligence technique where the machine is trained to mimic the thought process of a human brain. This paper focuses on building a deep convolutional neural network that can predict the malaria parasite infection from thin blood smear samples. The malaria dataset is obtained from the National Library of Medicine hosted by Lister Hill National Center for Biomedical Communications, USA. The convolutional neural network model is optimized against overfitting and attains an F1 score of over 94%.