Classifying real-world macroscopic images in the primary-secondary care interface using transfer learning: implications for development of artificial intelligence solutions using nondermoscopic images.

Journal: Clinical and experimental dermatology
PMID: 37990943

Abstract

BACKGROUND: The application of deep learning (DL) to diagnostic dermatology has been the subject of numerous studies, with some reporting skin lesion classification performance on curated datasets comparable to that of experienced dermatologists. Most skin disease images encountered in clinical settings are macroscopic, without dermoscopic information, and exhibit considerable variability. Further research is necessary to determine the generalizability of DL algorithms across populations and acquisition settings.

Authors

  • Jacob Carse
    CVIP (Computer Vision and Image Processing), School of Science and Engineering, University of Dundee, Dundee, UK.
  • Tamas Suveges
    CVIP, School of Science and Engineering, University of Dundee, Dundee DD1 4HN, UK.
  • Gillian Chin
    Department of Dermatology, Ninewells Hospital and Medical School, Dundee, UK.
  • Shareen Muthiah
    Department of Dermatology, Forth Valley Dermatology Centre, Stirling, UK.
  • Colin Morton
    Department of Dermatology, Forth Valley Dermatology Centre, Stirling, UK.
  • Charlotte Proby
    Department of Dermatology, Ninewells Hospital and Medical School, Dundee, UK.
  • Emanuele Trucco
  • Colin Fleming
    Ninewells Hospital and Medical School, Dundee, UK.
  • Stephen McKenna
    CVIP (Computer Vision and Image Processing), School of Science and Engineering, University of Dundee, Dundee, UK.

Keywords

External Resources

Popular Topics
Recent Journals