Skin cancer can be self-detected using your phone
An app to detect skin cancer at early stages will be introduced in primary health care. But already now anyone can do a self-test with its help.
Some types of skin cancer, such as melanoma, are very aggressive. Oncologists recommend self-prophylaxis: regularly examine the body and monitor the appearance of moles, their growth and changes. Some doctors even recommend periodically photographing yourself and your moles to make it easier to see the changes. If they are really there, if the neoplasms behave restlessly (changes in size, shape, color of moles) - it is necessary to sound the alarm and go to a dermatologist to be examined. At the risk of the appearance of altered cells, the dermatologist sends the patient for consultation with an oncologist.
Early diagnosis of melanoma is crucial, and now there is an app that can help even a general practitioner or GP to handle the visual analysis of a patient's skin condition. This medical decision support service can be used by doctors in primary care.
The application is called "ProMoles" and uses a neural network that evaluates and draws a conclusion about neoplasms on the skin, comparable in accuracy to that of a dermatovenerologist or oncologist. This was confirmed by tests conducted on the basis of the N.N. Blokhin NMIC Oncology Center in Moscow. As a result, doctors during medical check-ups and regular appointments will be able to make an initial diagnosis of suspicious skin growths by examining the patient.
The prevalence of melanoma is increasing: our love of immoderate sunbathing and going to the solarium are to blame. "For example, in the Nizhny Novgorod region, the annual cost of treating patients with melanoma stage 3-4 is about 780 million rubles," says Nikolai Karyakin, rector of the Volga Institute of Traumatology and Orthopedics (PIMU). - However, if we introduce the ProRodinks medical decision support service, these costs, according to our calculations, can be reduced almost twofold due to early stage diagnostics".
"Over the past three years, more than 500,000 people have already used the ProMoles application - a non-medical information system for identifying the risks of developing skin cancer. Registration of our product as a medical device opens up opportunities for its use at the primary care stage and routing patients to qualified specialists in case of risks," Anton Melnikov, AIMED's development director, explained to Rossiyskaya Gazeta.
According to Irena Shlivko, Head of the Department of Skin and Venereal Diseases at PIMU and Project Manager, the mobile application is already being used by ordinary people, and it works: out of 1 million users, 14,000 have been diagnosed with suspected melanoma, and another 10,000 with other types of skin cancer.
The relevance of the application was also confirmed by Igor Sinelnikov, member of the board of the Association of Melanoma Specialists, chief oncologist of the Melanoma Unit Scientific Medical Center. "In the system of Russian medical education there are no separate courses to train specialists - general practitioners, other specialties to work specifically with malignant skin tumors. The application allows a specialist who has no special training to identify them quite effectively, both using artificial intelligence programs, photographing a mole, and with the help of specialists who control this artificial intelligence," the expert said.
According to the oncologist, the application is so easy to use that it will be useful for people who have no medical education at all, those who have no idea what moles can be dangerous. In other words, it can be used for self-examinations, in order to consult an oncologist if necessary.
"At the second stage, you should definitely be seen by a doctor, because any diagnosis, which is made by a machine, artificial intelligence or remote specialist, is preliminary and requires subsequent face-to-face consultation with a specialist and, possibly, the performance of some procedures, additional examination," - emphasized Igor Sinelnikov.