Brazilian technology aims to revolutionize skin cancer diagnosis

By Roseli Andrion  |  FAPESP Innovative R&D – Accounting for more than 31% of all cases of malignant tumors in Brazil, skin cancer is the most common type of the disease in the country and in the world. Despite this high incidence, diagnosis is still difficult. This is because the main tool for evaluating lesions is the dermatoscope – a device similar to a magnifying glass – combined with the experience of the professional.

In order to detect skin cancer quickly, accurately and, above all, affordably, the Brazilian startup Unnawave, based in São José dos Campos, in the state of São Paulo, is developing a non-invasive, stand-alone device that uses infrared thermography. Fernando Malheiros, the mechanical engineer in charge of the project and a specialist in computer models of biological tissue, says that the device aims to detect skin cancer using mathematical processing of images captured by high-definition cameras.

The researcher, who developed computer models of biological tissue during his master’s and doctoral studies, has become one of the world’s leading authorities on the use of this technique. “I’ve published articles on everything from the mechanical properties of tissue to how it behaves in terms of energy and energy emission at temperature,” he explains. “Based on this, I began to create mathematical models of skin cancer in partnership with the Hospital de Amor in Barretos [in the interior of the state of São Paulo].”

At the beginning of his work, Malheiros’ goal was to determine whether it was possible to use these models to differentiate cancerous tissue from healthy tissue. “I wanted to know if, after creating a mathematical and physical model of how skin cancer behaves, I’d be able to differentiate it from healthy skin,” he explains.

The answer came from infrared thermography, a technique that uses a specific frequency band to analyze the heat emitted by tissue. The device picks up signals from the skin lesion and processes them mathematically to highlight the difference between healthy and cancerous tissue, the researcher explains. “The system then tells us if there is any indication of skin cancer.”

Clinical trials

In the initial phase of clinical trials, 101 lesions were evaluated. The participants are patients who have already been examined and referred for minor surgery. “They already have a lesion that’s suspicious for cancer and are evaluated with the device. The lesion is then removed and sent for biopsy. The result is then compared with the assessment made with the device.”

The results from the device are more accurate than the evaluations made by doctors using a dermatoscope. “With the new device, our accuracy is approximately 90%,” says Malheiros. “This is a significant improvement over the traditional method, which has an accuracy rate of 76%.”

The project is currently in the second phase of clinical trials and the research is focused primarily on melanoma, the most aggressive form of skin cancer. “Melanoma affects the melanocytes that produce melanin and is more aggressive and less common. The lethality of melanoma is unfortunately very high when it’s diagnosed late.”

What sets the technique developed by Malheiros apart is its ability to detect melanoma early by distinguishing it from a normal mole, a major challenge for dermatologists. “Melanoma and moles are identical in terms of their visual characteristics,” he says. “That’s why early detection is crucial, because this type infiltrates the tissue and metastasizes very quickly: 30 or 90 days is a very long time.”

The device created by Unnawave has important differentials. One of them is the ability to estimate the edge of the lesion: something crucial in the treatment of skin cancer. “If I remove enough skin, I solve the problem, but if I remove only part of the lesion, it becomes a problem. So providing a planar perspective of the lesion is very useful for the physician’s decision-making.”

Another major advantage is detection on dark skin. According to Malheiros, the dermatoscope uses visible light, which makes it difficult to diagnose patients with phototypes 5 and 6 because there isn’t enough contrast between the mole and the skin. “Our sensor works in a non-visible frequency band, so it has the same effectiveness and behavior for any skin type.”

Science with a purpose

According to the researcher, the project goes beyond the commercial aspect. “Cancer is humanity’s greatest fear,” he explains. “A 2019 survey by Datafolha [a research institute affiliated with the Folha de S.Paulo newspaper], which interviewed 2,074 people from across the country, showed that the greatest fear of 27% of Brazilians was having cancer. The runner-up in the survey was unemployment, at 14%. When I chose what to research, I thought about the medical field, so that it’d be something that would last longer than my lifetime.”

Technology can have a significant impact on public health. “Skin cancer goes unnoticed by many people because there are only 14,000 dermatologists in Brazil today. It isn’t possible to have a specialist in every health center,” says Malheiros. “What’s more, if you’ve already had a lesion, there’s a good chance you’ll have another one within five years. And skin cancer is cumulative, meaning that the longer an individual lives, the more likely they are to develop skin lesions. So the trend is for the numbers to go up. That’s why it’s vital to make new diagnostic tools more accessible, especially in areas where dermatologists are scarce.”

In this context, a protocol using an affordable, stand-alone device would be useful for early and mass diagnosis, especially in regions with less access to specialists. “The cost of treating skin cancer is many times greater than that of preventing it. If in a group of 50,000 people I find one melanoma, it’s already worth the cost of treating one recurrent metastasis. So treating it early is much cheaper and more effective.”

As the world’s population grows, the number of cancer diagnoses tends to increase. This makes the logic of prevention even more important. “Advances have created advanced cancer treatments, such as targeted molecular therapies, but they can cost hundreds of thousands of dollars per dose. I think even health insurance companies will take a closer look at prevention in the coming years,” reflects the researcher.

Malheiros’ project is supported by FAPESP’s Innovative Research in Small Businesses program (PIPE). He is continuing to improve the device and hopes to have a working prototype by the end of this cycle. In the future, the tool will be available in health centers across the country to help diagnose the disease early and thus increase patients’ chances of a complete cure.