How is AI early detection of skin cancer done?
Derm.AI
Researchers of Fraunhofer have developed a digital tool to speed-up skin cancer diagnosis significantly. The mobile application records skin lesions and delivers the same to hospitals’ dermatology departments. By doing image analysis and using AI, the software analyses the skin cancer risk of those people. Accordingly, dermatologists decide the way forward. Skin cancer is deceptive. During the early stages, it looks like a birthmark or mole and does not cause pain. However, the German Cancer Society released data mentioning that over 200000 people suffer from this form of cancer in Germany every year. 3764 patients even died from it in 2017 itself. Of all the skin cancers, malignant melanoma is the most serious. It is caused by extensive exposure to the UV rays of the sun. Thankfully, if diagnosed early, it can also be cured. The Fraunhofer Center for Assistive Information and Communication Solutions (AICOS) in Porto and Lisbon developed Derm.AI which combines pictures of the lesion with image-analysis software and AI. It swiftly assesses any dangerous changes on the skin’s surface. Once dermatologists access this information, they strategize the treatment plan for the patients. They would most definitely start the treatment for the patients who are diagnosed with skin cancer immediately. But they may treat people who are at more risk of getting it before analyzing people who are less susceptible if a choice has to be made. The Derm.AI aims to enhance the Teledermatology processes in the Portuguese National Health System as well. A Senior Scientist at Fraunhofer AICOS named Maria Vasconcelos says, “In recent years, GPs have been increasingly concerned with spotting skin cancer early. People who notice dark spots or other perceptible changes to their skin need clarity quickly. But in regions with few specialists, it often takes a long time to get an appointment for the initial assessment. Often patients also have to travel long distances for these appointments, too. This is where our Derm.AI solution comes in.”
AI-powered SPL
The scientists at MIT also developed an AI-powered SPL analysis system to assess the pigmented lesion accurately to detect any anomaly. It has achieved over 90% accuracy in differentiating suspicious lesions from non-suspicious ones.
AI Dermatologist
AI Dermatologist is another AI tool that also helps to diagnose skin cancer at its early stage. Developed by a team of scientists and doctors, it can assess and detect risks of more than 29 diseases including skin cancer, papillomavirus, benign formations, precancerous lesions, 6 types of acne, etc.
How does AI work to detect skin cancer?
Each of these AI tool has a similar way of working to determine if a certain lesion on the skin surface is a matter of concern or not. To get a diagnosis, a person has to take a photo on the phone and send it through the chosen AI app. While some apps do the diagnosis themselves and send the report almost immediately, other apps may send their analysis to dermatologists for further verification. Accordingly, the treatment can be started. Since these apps are being developed by researchers and scientists of different institutions in different countries, they may be available for use in those particular places only. You may suggest AI early detection of skin cancer if you know somebody who could be suffering or susceptible to skin cancer.
