A team of researchers from SpotLab, the Universidad Politécnica de Madrid (UPM), the Instituto de Salud Carlos III (ISCIII), the Universidad Mayor de San Simón in Bolivia, the Biomedical Research Networking Centre (CIBER-ISCIII) and the Fundación Mundo Sano has developed an artificial intelligence (AI) system that detects parasites in real time using a mobile phone attached to a microscope.
Around 8 million people live with Chagas, a potentially fatal parasitic disease endemic to Latin America. Chagas disease, or American trypanosomiasis, is caused by the parasite Trypanosoma cruzi. An estimated 8 million people worldwide are infected, the disease causes more than 10,000 deaths every year, and over 100 million people are at risk of contracting it. It is endemic in 21 Latin American countries and is transmitted mainly through the faeces of triatomine bugs (known as vinchucas), but also through contaminated food, from mother to child, or via blood transfusions.
In its acute phase, when large numbers of parasites are circulating, the disease can be diagnosed by examining a blood sample under a microscope or, in some cases such as immunocompromised patients, a sample of cerebrospinal fluid. “But this examination requires specialised staff and more time to identify the parasites. In many affected areas, health facilities are low complexity and have a single laboratory worker with little or no experience, which makes it impossible to carry out a cross-check to identify the parasite and confirm the result. That complicates diagnosis,” explains Dr Mary Cruz Torrico, of the Universidad Mayor de San Simón in Cochabamba, Bolivia.
To overcome these limitations, researchers from the National Centre for Microbiology at the ISCIII, the AI company for diagnostics and biopharmaceutical research SpotLab, the UPM, the Bioengineering, Biomaterials and Nanomedicine Area (CIBER-BBN) of CIBER-ISCIII, the Universidad Mayor de San Simón (Cochabamba, Bolivia) and the Fundación Mundo Sano have developed a portable artificial intelligence (AI) system capable of detecting and counting Chagas parasites in real time. The system attaches a mobile phone camera to the eyepiece of a conventional optical microscope using a 3D-printed adapter, and runs the AI models on the phone itself, with no need for an internet connection or expensive equipment.
The results have just been published in the journal PLOS Neglected Tropical Diseases, and the AI models are openly available for the scientific community to download and test.
“Combining expertise in parasitology, microscopy, engineering and artificial intelligence has allowed us to reinvent the use of a tool as classic as the microscope. We now face another fundamental challenge: moving this development from scientific publication to use in real-world conditions,” says Dr María Flores-Chávez, principal investigator on the project at the ISCIII.
To train and validate the system, the research team used human blood and cerebrospinal fluid samples from Bolivia, along with samples from animal models. The main model achieved a precision of around 86% on human samples and analyses each image in roughly 350 milliseconds. In a laboratory pilot test, in which a person examined the samples using the phone, the system detected more than 96% of the parasites present, a very high sensitivity that makes it especially useful as a screening tool. The human specialist always has the final say and reviews the detections.
“The analysis is carried out directly on a mobile phone connected to a conventional microscope, with no need for an internet connection or expensive equipment. This brings diagnosis closer to resource-limited settings and makes it accessible anywhere in the world,” says Dr Lin Lin, principal investigator on the project for SpotLab.
Dr María Jesús Ledesma, principal investigator on the project for the Universidad Politécnica de Madrid, adds: “This innovation has enormous potential to improve the diagnosis and monitoring of Chagas disease, particularly in resource-limited settings. Its modular design also means it can be adapted to detect other parasites and diseases, making it a scalable solution for tackling neglected tropical diseases, in line with the goals set by the World Health Organization.”
The research was supported by the European Union’s H2020 programme, the Bill and Melinda Gates Foundation, the Comunidad de Madrid’s industrial doctorates programme, Spain’s Ministry of Science, Innovation and Universities, and NextGenerationEU.
Reference: Lin L, Solano AV, Gonzales F, Torrico MC, Illanes D, Díez N, Bermejo-Peláez D, Dacal E, Vallés-López R, Pastor L, Mancebo-Martín R, Ledesma-Carbayo MJ, Luengo-Oroz M, Rubio JM, Flores-Chavez M. (2026). Artificial intelligence algorithm for real-time detection and counting of Trypanosoma cruzi parasites using smartphone microscopy. PLoS Neglected Tropical Diseases, 20(5):e0012955. doi: 10.1371/journal.pntd.0012955
Article in PLOS NTDs: https://doi.org/10.1371/journal.pntd.0012955
AI model (HuggingFace): SpotLab/Chagas_ssd
Dataset (Zenodo): https://zenodo.org/records/15007339
The team’s previous work on filariasis: https://doi.org/10.1371/journal.pntd.0012117