Implementation of personalized medicine in malignant melanoma patients aided by artificial intelligence.

This project aims to improve the diagnosis, treatment, quality of life and survival of patients with melanoma. Currently, all melanoma patients are divided into categories depending on the risk of having metastasis and potentially dying from the disease. To classify patients, a method (staging system) is used that includes information on a limited number of tumor characteristics and patient metastases. This information is used to predict the evolution and risk of relapse of the patient and to decide on the follow-up, diagnostic tests and treatment options for each patient. Unfortunately, this method is very inaccurate and many treatments and tests are not adequately indicated. Many patients with melanoma undergo a sentinel lymph node biopsy, an invasive surgical procedure for analyzing the lymph nodes. However, 80% of patients do not benefit from this intervention because they have normal lymph nodes. When the tumor is classified as "aggressive ", the patient is subjected to systemic therapies at high risk of dangerous toxicity. However, only half of the patients have a benefit from the treatments. Every day in the world, many patients are subjected to treatments that affect their quality of life. In addition, these treatments are extremely expensive. The healthcare cost in cancer is every year higher and, in the near future, many patients will not have access to the new drugs that scientists are discovering. A precise method for the diagnosis and prediction of the prognosis is essential to change this situation. However, the current method ignores a lot of information about the patient and the tumor. This information is essential to precisely establish the best treatment for each patient. Unfortunately, due to the huge amount of data, we cannot use this information without a complex mathematical integration. Today with the new technologies of artificial intelligence, mathematicians can, with the help of powerful computers, analyze many data and train automated systems that can help doctors to improve the diagnosis and treatment of the patient.

This project will be an ambitious collaboration of physicians, genetic and mathematics that will be working together for three years. It will include data from many patients from Catalan hospitals (more than 6000000 data entries of 14000 patients) about the tumor and patient, tests, medications, genetics and others, at the time of the diagnosis and during the follow-up visits. This information has been collected during the last decade in an unprecedented research effort in the world. But now we need to analyze and integrate it usefully with the help of artificial intelligence and build a method that will help physicians around the world to improve the care of patients with melanoma.