A multidisciplinary team formed by the Microbiology Service of the Vall d'Hebron University Hospital, the Microbiology Research Group of the Vall Hebron Research Institute (VHIR), the Universitat Politècnica de Catalunya - Barcelona Tech (UPC) and the Probitas Foundation, has presented a new diagnostic method for malaria.

The research has been developed largely in the Microbiology laboratory of the Vall d'Hebron Drassanes International Health Centre, with the support of the research groups in Computational Biology and Complex Systems (BIOCOM-UPC), Image and Video Processing (GPI) and Database Technologies and Information Management (DTMI) of the UPC. It is an artificial intelligence-based system that combines a mobile application with a low-cost robotic microscope. The design is designed to be a useful and effective method in resource-poor countries, where the disease is endemic. The results of the first prototype of iMAGING have been published in the journal Frontiers in Microbiology. The device has demonstrated a reliability of more than 90% in the laboratory; the next step will be to test it in the field.

Malaria is an infectious disease transmitted by mosquito bites and caused by parasites of the genus Plasmodium. The World Health Organisation estimates that there were 249 million cases worldwide in 2022, 93% of which were located in the African region, with 95% of deaths. The same report also warned that climate change and globalisation are causing the mosquito to spread to new areas that have little capacity and resources to cope. Currently, the gold standard for malaria diagnosis is the visualisation of parasites by an expert under an optical microscope using blood samples. This is a manual, time-consuming and repetitive procedure, which, coupled with a lack of laboratory technicians and instruments, results in a large under-diagnosis. Until now, any attempt to automate the process increased the cost exponentially, making it prohibitively expensive in countries with limited health resources.

An automatic microscope controlled via Bluetooth

The team's proposed solution is iMAGING, a mobile app that uses artificial intelligence to process digital images of blood samples to determine whether infection is present. If positive, it also determines the density and stage of the parasite infection. To capture the images, a robotic microscope has been designed from a normal optical microscope with parts created using 3D printing, which has made it relatively inexpensive to produce.

The prototype has been trained with more than 2500 images and has achieved a reliability of more than 96% for high density samples and 94% for low density samples. False positives and negatives have not reached 5% in any case. However, Dr. Joan Joseph i Munné, principal investigator of this project from VHIR's Microbiology group, explains that the "key test will be its performance in the field, however, if it is successful, it could open the door to adapt it to other Neglected Tropical Diseases".

The project is framed within the work of science and technology for human development promoted by the Centre for Development Cooperation (CCD) of the UPC. Daniel López Codina, researcher at BIOCOM-UPC, explains that "we need to continue working to develop quality, low-cost tools to improve the health of people living in countries with low or very low human development indices. We are very happy with the results achieved so far, and are confident that we will be able to adapt these tools to other neglected tropical diseases”.

The plan now is to continue training the artificial intelligence to make improvements in other aspects, for example, so that it can differentiate between the five different species of parasites that cause the disease. This will allow a much more personalised treatment, improving its effectiveness. This project is supported by the World Health Organisation as part of its initiative to diagnose haemoparasites through digital imaging in low- and middle-income countries.