Malaria Diagnostics: A Scoping Review from Traditional to Artificial Intelligence–Powered Methods

Abstract

This review article provides a comprehensive analysis of malaria diagnostic technologies, ranging from conventional methods such as microscopy, rapid diagnostic tests (RDTs), and polymerase chain reaction (PCR) to the latest AI-powered automated diagnostic solutions. A total of 3,640 publications identified from three databases—PubMed, Scopus, and IEEE Xplore—were systematically screened, with 72 studies ultimately included in the review. The article offers an in-depth examination of the transformative potential of AI in malaria diagnostics, as well as the key challenges associated with its implementation in clinical practice.

Key Highlights

• Identified as a commercially available AI-based malaria diagnostic platform for automated parasite detection and species differentiation
• Achieved 100% sensitivity and 100% specificity in an evaluation conducted by a U.S. reference laboratory
• Multicenter clinical study conducted in Ethiopia and Ghana (n = 2,201) demonstrated:
• P. falciparum: 97.4% sensitivity, 98.8% specificity
• P. vivax: 95.9% sensitivity, 97.8% specificity
• Demonstrated significantly higher sensitivity and specificity compared with routine microscopy performed at primary healthcare facilities

Keywords