We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Utilizing convolutional neural networks to detect skin cancer: a review of initial trials.
- Authors
Ladda, Matthew; Champagne, Trevor
- Abstract
A convolutional neural network (CNN) is a type of deep, feed-forward artificial neural network able to detect, recognize, and classify visual imagery. Given the visual nature of dermatology, recent research has explored the ability of trained CNNs to detect skin cancers. The ability of CNNs to correctly categorize biopsy-confirmed images of skin lesions has been compared to that of dermatologists in several studies discussed herein. This article will review several studies that evaluated the ability of CNNs to detect skin cancer. Given the diagnostic accuracy that CNNs have demonstrated in studies to date, and the ease of which they could be incorporated into smartphones and digital dermatoscopes, CNNs have the potential to significantly improve the detection of skin cancer and potentially other dermatological diseases. Further research is needed regarding how CNNs could be effectively integrated into clinical practice.
- Subjects
SKIN disease diagnosis; SKIN tumors; DIGITAL image processing; MICROSCOPY; ARTIFICIAL neural networks; SMARTPHONES; MOBILE apps; DIAGNOSIS
- Publication
University of Toronto Medical Journal, 2019, Vol 96, Issue 1, p22
- ISSN
0833-2207
- Publication type
Article