AI-supported Point-of-Care Microscopy
Imported infectious diseases such as malaria and other parasitic infections remain a diagnostic challenge due to their often non-specific clinical presentation and the limited availability of specialized microscopy expertise. Rapid and reliable diagnostic pathways are essential for the timely initiation of appropriate treatment and for improving clinical decision-making and patient outcomes.
This is why GHIDA is collaborating with SpotLab to explore the clinical application of an artificial intelligence-supported point-of-care microscopy platform. The technology combines conventional microscopy with smartphone-based digitalization, AI-assisted image analysis, and telemedicine-based expert consultation (https://spotlab.ai/).
As an initial area of collaboration, the technology will be implemented in Asella, Jimma and Düsseldorf to evaluate remote malaria diagnostics under different epidemiological and infrastructural conditions. Digitized blood-smear images can be securely shared with experienced microscopists and parasitologists, allowing diagnostic assessments to be performed remotely when sufficient expertise is not available locally.
The collaboration will examine the feasibility and diagnostic performance of this approach, as well as its integration into existing clinical workflows. Particular attention will be given to image quality, the reliability and timeliness of remote diagnostic assessments, user acceptance, and the training requirements of healthcare professionals.
In parallel, GHIDA and SpotLab are exploring further opportunities to apply and evaluate the technology within broader clinical and research initiatives, including projects aimed at improving the diagnosis and management of imported infectious diseases in Germany.
By combining digital microscopy, artificial intelligence, telemedicine, and capacity building, the collaboration aims to strengthen diagnostic expertise across sites and facilitate a bidirectional exchange of knowledge between Ethiopia and Germany. Ultimately, this approach may contribute to more accessible, timely, and reliable diagnostics for malaria and other parasitic infectious diseases.