WHITE-BOX MACHINE LEARNING APPROACH TO PLASMODIUM FALCIPARUM MALARIA DETECTION USING MULTI-CHANNEL COLOUR SCHEME

Abstract

Malaria, a severe global health threat, remains a challenge despite ongoing efforts. This paper reviews various studies, highlighting disease severity and combat strategies. The substantial public health burden of malaria is evident, affecting millions globally. In 2019, 409,000 deaths were reported, 67% being children, necessitating preventive measures. Various strategies, such as laboratory networks, computing-based methods, and Geographic Information Systems, target malaria's impact. Despite progress, malaria persists in West Africa, requiring significant support and research. The paper's focus on white box algorithms for malaria diagnosis aligns with transparent, accurate healthcare solutions. With F1-score of 0.95 for "Infected," 0.95 for "Uninfected," and overall accuracy of 95.42%, the model detects Plasmodium falciparum. The study underscores comprehensive strategies to combat malaria's global burden.