The aim of the researchers is to make the diagnosis of the disease, which is endemic in 21 countries, cheaper and more accessible.
A group of Brazilian researchers developed a algorithm capable of detecting the Trypanosoma cruzi, the protozoan that causes Chagas disease, in images of blood samples obtained by cell phone and analyzed under an optical microscope. The technology was presented in an article published on May 27 in the scientific journal PeerJ.
The researchers’ goal is to automate and make the way to identify the disease cheaper. Currently, the diagnosis is made by trained microscopists who detect the parasites in blood samples using a professional microscope coupled with a high-resolution camera — which makes the method expensive and not very accessible to low-income people.
In addition, according to immunologist Helder Nakaya, coordinator of the study, if the algorithm works well for Chagas disease, it will be possible to adapt it for other purposes, such as analysis of stool samples, dermatology and colposcopy.
The research, which received support from the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP), brought together professionals from various areas, such as biology, mathematics and computing.
To enable the detection and counting of trypomastigotes (the form found in the blood of patients with the disease in the acute phase) of the Trypanosoma cruzi, the researchers developed a learning approach called random forest. Then, micrographs of blood smear samples recorded in images with a resolution of 12 megapixels were analyzed.
From this, characteristics such as shape, size, color and texture of 1,314 parasites were extracted. At this stage, scientists, together with machine learning and image processing specialists, “taught” the machine to recognize Trypanosoma.
Then, the samples were divided into training and test sets and, later, classified using the random forest algorithm. The precision and sensitivity values obtained by the survey were considered high: 87.6% and 90.5%, respectively.
With the algorithm, the group was able to automate the analysis of images acquired through the cell phone, obtaining an alternative to reduce costs and increase efficiency in the use of the optical microscope. The researchers’ proposal is to leave the algorithm open for the scientific community to contribute with other data and resources.
THE Chagas disease is an inflammation caused by the parasite Trypanosoma cruzi. found in the feces of insects, mainly barber. Transmission occurs from the bite and ingestion of raw food and contaminated with the feces of the transmitter, through blood transfusion and organ transplants contaminated with the disease and through direct contact with the parasite and/or other infected animals.
You symptoms of Chagas disease vary from the acute (more moderate or asymptomatic) to the chronic phase. In the first stage, the main signs are:
In the chronic phase, symptoms may include:
Difficulty swallowing;
Irregular heartbeats.
The disease is endemic in 21 countries in the Americas, affecting approximately six million people with an incidence of 30 thousand cases per year and leading, on average, to 14 thousand deaths per year. It is estimated that about 70 million people live in areas of exposure to barber and are at risk of contracting the disease.
Chagas disease has been classified by the World Health Organization (WHO) as one of the neglected tropical diseases (NTDs). Its prevention is linked to the control of the barber bug.