Global technology consultancy DataArt announced the launch of “SkinCareAI” – a prototype app that analyses skin images to detect early signs of melanoma.
Using the latest advances in machine learning (ML) technology, SkinCareAI was developed by DataArt ML expert Andrey Sorokin for the ISIC (International Skin Imaging Collaboration) Challenge. Drawing from worldwide data, the app uses ML algorithms for early melanoma detection.
Andrey Sorokin said: “We can compress tens of thousands of hours of hard work by the best dermatologists from around the globe into a melanoma diagnosis tool accessible to anyone with a smartphone.”
“Imagine your suspicious-looking nevus being examined by thousands of doctors based on their collective experience and expertise from thousands of cases. This is exactly what our prototype smartphone app offers.
“We designed the app to demonstrate to the healthcare sector and the medical community the use of machine learning for analysis of medical imagery in an AI diagnostic process. The app does not replace a doctor’s diagnoses but rather supports and enhances the doctor’s work.”
Users upload photos of skin lesions or use the app to take them, and the images are analysed in seconds. After “comparing” the images to thousands of others, the app classifies the lesion, identifies its morphologic properties, and suggests the next steps.
“All functions are fast and user-friendly,” said Sorokin. “However, we do not recommend relying only on the app for skin cancer detection. Sometimes people leave things too long, and so the app encourages users to self-check and seek examination by a doctor if needed. The app also helps the medical community by bringing collective intelligence to individual practitioners.”
“We are extremely pleased with this app and very confident that when further developed with a medical partner, it will help save lives through early detection.”
Skin cancer is a life-threatening condition. With 5,000,000 new cases each year in the US alone, skin cancer is a major health threat. Melanoma is the deadliest type of skin cancer, but if detected early, the survival rate exceeds 95%.
By demonstrating SkinCareAI at upcoming industry events, DataArt’s Healthcare and Life Sciences practice is looking to raise awareness of the role that ML technology can play in melanoma detection and to find an industry partner for further developing the app and bringing it to market.