<![CDATA[Malaria is a mosquito-borne disease that has been responsible for numerous deaths in humans for decades, caused by a single bite from an Anopheles mosquito. For proper treatment, the affected person needs to undergo a series of tests. Currently, separating infected and uninfected cells is done manually, which is time-consuming overall. However, this article presents a developedexpert system called Plasmodium Detector (Plasmo-D), which leverages computer vision to detect malaria-infected cells using image recognition. Plasmo-D was built as an Android application, featuring an information menu, splash screen, and classification screen, along with an image recognition system that utilized computer vision. Data sourcing of 27,528 cell images was obtained from the Data Library of the United States National Library of Medicine. Training was conducted using Microsoft Azure, and the application was deployed to Android using Java programming language and an Android XML (Extensible Markup Language) user interface, along withTensorFlow Lite. Five iterations were conducted, and the parameters studied included cell images, backgrounds, visual style, size, type, lighting, and camera angle. High accuracy (up to 99.8%) in classifying parasitized, uninfected, and irrelevant images (‘not a cell’).]]>
