AI-mediated assessment and treatment of skin disease - version 1

  • Research type

    Research Study

  • Full title

    Artificial Intelligence-Mediated Assessment and Treatment of Skin disease

  • IRAS ID

    299722

  • Contact name

    Eugene Healy

  • Contact email

    ehealy@soton.ac.uk

  • Sponsor organisation

    University of Southampton

  • Duration of Study in the UK

    1 years, 6 months, days

  • Research summary

    Dermatologists use various approaches to treat skin disease, including medical, surgical and physical therapies. There is a need to develop more accurate assessments of the severity of certain skin diseases, for example psoriasis and vitiligo, because this would aid decisions on choosing appropriate treatments and monitoring improvement of disease during treatment. Ultraviolet light (UV) therapy is one of the main physical treatments for psoriasis and vitiligo, however, a UV-phototherapy approach that delivers UV only to skin lesions while avoiding normal skin would offer significant benefits over currently available UV therapies.

    We wish to develop an artificial intelligence-mediated approach to (i) the assessment of the severity of psoriasis and vitiligo and (ii) future UV-phototherapy for these skin diseases so that UV is directed to the skin lesions while avoiding uninvolved skin. This process will involve training and testing a neural network for automated recognition and scoring of severity of psoriasis lesions and of vitiligo lesions. Photographs of skin disease will initially be used to train and test the neural network. Subsequently, the ability of the neural network to direct treatment to lesional skin will be examined using a laser diode that emits light within the visible light spectrum, and subsequently using UV, to demonstrate that the light from the laser diode and from the UV is directed only to areas of skin disease in photographs of the skin condition and that the normal skin in the photographs remains unirradiated.

    We also wish to obtain patient and public input into the use of artificial intelligence in phototherapy for skin conditions, including their views and thoughts on the concept and development of this approach for UV treatment of skin disease, and on the design of future clinical studies for psoriasis and vitiligo to assess the use of an artificial intelligence-mediated UV-phototherapy unit.

  • REC name

    West of Scotland REC 4

  • REC reference

    22/WS/0015

  • Date of REC Opinion

    1 Apr 2022

  • REC opinion

    Further Information Favourable Opinion