Artificial Intelligence Skin Cancer Screening Study: Dyplens 2.0

• Research type

  Research Study

• Full title

  Building on the AISCSS Study; Evaluating Accuracy of a New Low-Cost Imaging Device in AI-Based Skin Cancer Detection. A Retrospective Study Utilising Prospectively Collected Data.

• IRAS ID

  372459

• Contact name

  Alexander Anderson

• Contact email

  alexanderanderson@nhs.net

• Sponsor organisation

  Royal Cornwall Hospitals NHS Trust

• Duration of Study in the UK

  2 years, 0 months, 1 days

• Research summary

  This study builds on the previous AISCSS study which investigated the accuracy of an AI algorithm (DermDx, MetaOptima, Canada) in diagnosing skin cancers from images taken with a magnified device (dermatoscope). The AI algorithm proved very useful as a screening tool, with a sensitivity of 98.1% (specificity 57.6%) for malignant or
  premalignant lesions.

  This linked study is designed to test the algorithm performance on images taken with a low cost clip on device (Dyplens). The cohort of patients will be the same as the initial AISCSS study (adults referred on the urgent skin cancer pathway and seen in a community lesion imaging clinic). An extra Dyplens image will be captured alongside
  usual images and metadata in our community lesion imaging clinics and retrospectively analysed with the DermDx algorithm alongside the standard dermoscopic images. This will allow us to calculate sensitivity and specificity of the algorithm on images taken with the low cost Dyplens device.

• REC name

  West Midlands - South Birmingham Research Ethics Committee

• REC reference

  26/WM/0114

• Date of REC Opinion

  8 Jun 2026

• REC opinion

  Further Information Favourable Opinion