Skip to content



Why did we create Autoderm?

Because 15% of all telehealth consultations and 20% of a GP visit are skin disease queries.

What is autoderm?

Autoderm is a research backed, CE marked (class 1), dermatology search engine. We use artificial intelligence (AI) for faster and more accurate answers on dermatology concerns. Our current AI model can screen for 44 different skin disease types. This includes inflammatory skin diseases, skin tumors and genital skin concerns. The model can be accessed via our “API as a Service”. From just a smartphone photo, the model generates the top five ranked skin diseases in order of probability.


We use deep learning models and state-of-the-art convolutional neural networks built on top of PyTorch. Our current model has been trained usingon over 110,000 smartphone images of 44 different skin disease classes. This, and it represents 90% of what the general public are concerned about.

Our models can be accessed from an API that can be integrated into any internet connected platform. ntegrating our API takes on average 4 hours. Our customers are free to design the end user UX/UI interface and connect any links to the end output (answers) that describe the disease.

Autoderm in the wild!

  • Using artificial intelligence on dermatology conditions in Uganda: A case for diversity in training data sets for machine learning. Louis Henry Kamulegeya et al. bioRxiv 2019
  • Evaluation of the diagnostic accuracy of an online artificial intelligence app for skin disease diagnosis. Alexander Larson, Gothenburg, Sweden 2018
  • First Derm Skin Image Search
  • WeChat
  • Telegram

Where to go next?

Check examples in different programming languages using our API. [Work In Progress]

If you want to check our Swagger docs, go to /docs. Or if you prefer redoc, got to /redoc.