An instrument using artificial intelligence (MI) is being developed for the early detection of melanoma with the help of researchers of the ELKH Institute for Computer Science & Control (SZTAKI). IToBoS is a research project funded by the European Union's (EU) Horizon 2020 research and innovation program to develop a new diagnostic tool for the early detection of melanoma that utilizes all available information about the patient. This holistic assessment tool seeks to understand the specific characteristics of each patient, paving the way for personalized, early detection of melanoma.
Skin cancer is the most common cancer in humans, with melanoma causing the highest number of deaths. According to the latest statistics, it is currently the sixth most common type of cancer in Europe, with more than 144,000 new cases diagnosed in 2018. When melanoma is detected early, more than 90 percent of patients are still alive after five years, but once the cancer cells start to spread — known as metastasis — the rate after five years drops to 23 percent or even lower. This means that rapid diagnosis is essential for treatment to occur before local and metastatic spread.
The goal of the iToBoS project is to create an artificial intelligence system capable of integrating information from a variety of sources, from dermoscopic images to complete medical records and genomic data. The iToBoS project is developing a new diagnostic tool to achieve this goal, as well as a cognitive assistant employing AI. With these tools, healthcare professionals may be able to recognize a higher rate of skin cancer and make a comprehensive, patient-specific diagnosis.
"In this large-scale international project, our researchers are able to utilize their experience in the fields of data management, online clouds and artificial intelligence," said Róbert Lovas, Deputy Director of SZTAKI.
The new diagnostic tool also uses the latest advances in artificial intelligence to facilitate the use of data already extracted with currently available technologies, such as dermoscopic images, and data obtained with the new hardware proposed in iToBoS.
The underlying algorithms will integrate additional patient information from different sources (such as the patient’s medical history, location of all moles, and genomics, age and gender) in order to provide a comprehensive, holistic assessment of each patient’s characteristics and their individual moles.
By systematically examining the patient on a regular basis, the system will also be able to determine any changes to each individual mole. This is one of the most informative things to look for when it comes to recognizing skin cancer. With the proposed new approach, physicians can diagnose skin diseases earlier and with greater accuracy, thereby increasing the efficiency and effectiveness of personalized clinical decision-making.
The consortium of 19 partner organizations is led by the University of Girona (Spain). The project has a duration of 48 months (April 1, 2021 – March 31, 2025) and a total budget of EUR 12 million (approximately HUF 4.31 billion).