ABSTRACT
Malaria is a severe and contagious disease worldwide. The disease is transmitted through Anopheles mosquitoes, specifically females. The World Health Organization's (WHO) report unveiled that there were 0.228 billion cases and around 567,000 deaths due to malaria in 2018. In 2022, cases rose to 0.249 billion, deaths rose to around 608,000. Detection of malaria parasites at preliminary stages can reduce the mortality rates. The two supremely used traditional methods of malaria testing are RDTs and light microscopy. However, the traditional methods are tedious, prone to errors, rely on manual labor, and require specialized pathologists. Automated systems using image processing, deep learning, and machine learning can be devised for the rapid and efficient diagnosis of the disease. This work describes the comprehensive study of several already existing computer-aided diagnostic approaches and models in this area aiming to pave the way for enhanced and accelerated malaria identification in the upcoming years.
