Noninvasive malaria detection beyond blood sampling.

Abstract

Malaria remains a major global health challenge, particularly in sub-Saharan Africa, despite being one of the oldest documented human diseases. Although blood-based detection remains the mainstream diagnostic modality, its use is constrained by the need for trained personnel, sterile procedures, patient discomfort, and reduced sensitivity for low-parasitemia and asymptomatic infections. Recent advances in sensing and diagnostics, increasingly integrated with machine learning and artificial intelligence, have created new opportunities for noninvasive malaria detection using accessible biological samples and physiological signals. This critical review synthesizes recent progress in noninvasive malaria detection across nonblood sampling matrices, biophysical detection mechanisms, and data-driven analytical approaches. We examine diverse sampling matrices, including urine, saliva, exhaled breath, and skin-emitted volatile organic compounds with attention to their associated biomarkers, biological relevance, and operational feasibility. We further categorize detection technologies into three principal domains: biological sample-based platforms, including immunoassays, molecular amplification, biosensors, and mass spectrometry; acoustic, photoacoustic, and optical systems, including ultrasound, photoacoustics, optical imaging, and spectroscopy; and digital and connected health platforms, including wearable monitoring, smartphone imaging, and machine learning-enabled analysis. We then compare these modalities in terms of analytical performance, operational suitability, and translational potential for integration into healthcare systems. Finally, we present an implementation-oriented roadmap that highlights translational, equity, artificial intelligence, scalability, and regulatory considerations, identifying practical paths toward field-ready, affordable, and programmatically relevant noninvasive malaria detection.