New AI Detects Chagas Parasite Directly from a Smartphone

A multidisciplinary team formed by Instituto de Salud Carlos III (ISCIII), CIBER-ISCIII, Mundo Sano Foundation, Universidad Politécnica de Madrid (UPM), Spotlab, and the Universidad Mayor de San Simón, has developed an artificial intelligence (AI) system to detect Trypanosoma cruzi —the parasite that causes Chagas disease or American trypanosomiasis— in real time. The system couples a mobile phone camera to a conventional optical microscope and runs the AI models locally, without requiring an internet connection or expensive equipment.

The algorithm has been created using human and animal model samples from Bolivia, and it achieves an accuracy near 86% in human samples and a sensitivity above 96% in pilot tests, while always maintaining the final supervision and validation of a specialist.

This technology could facilitate the diagnosis of Chagas disease in resource-limited areas compared to the complexity and specialization required by other methods. The AI model has been released as open source, so the door is open to adapt it to other tropical diseases.

The project is backed by the Bill & Melinda Gates Foundation, the European Union (H2020), and NextGenerationEU funds.

Artificial intelligence algorithm for real-time detection and counting of Trypanosoma cruzi parasites using smartphone microscopy