Application of artificial intelligence in identifying pigmented skin diseases. DermaMarker – system for analysis and early detection of skin tumors.

The scientific novelty of DermaMarker lies in the fact that we have created a DermaMarker platform that scans pigmented skin abnormalities on the body and analyzes their danger using a neural network. We managed to train this neural network to recognize and distinguish skin neoplasms by comparing them with existing samples. The created platform defines 7 different diagnoses: Actinic keratoses and intraepithelial carcinoma / Bowen's disease (akiec), basal cell carcinoma (bcc), benign keratosis-like lesions (solar lentigines / seborrheic keratoses and lichen-planus like keratosis, bkl), dermatofibroma (df), melanoma (mel), melanocytic nevi (nv) and vascular lesions (angiomas, angiokeratomas, pyogenic granulomas and hemorrhage, vasc).

The platform-website offers mole diagnostics from a photograph. The testing procedure includes the following:

• Uploading a photo of a mole or pigmented skin area will be uploaded to the platform using the "Select File" button.
• Identification of the uploaded mole photo with the available samples. After successfully uploading a photo, its icon will be displayed on the screen, thereby the neural network will find similar results with Latin names. The whole procedure will be performed within 10 seconds when loading a high-quality image taken with good lighting and maximum magnification.
• The chatbot integrated in the website will give the result of comparison with samples: percentage probability of exposure to a certain skin deviation, and will also provide further recommendations to prevent the development of deviations.