An improved Internet-based melanoma screening system with dermatologist-like tumor area extraction algorithm

Hitoshi Iyatomi, Hiroshi Oka, M. Emre Celebi, Masahiro Hashimoto, Masafumi Hagiwara, Masaru Tanaka, Koichi Ogawa

Research output: Contribution to journalArticle

169 Citations (Scopus)


In this paper, we present an Internet-based melanoma screening system. Our web server is accessible from all over the world and performs the following procedures when a remote user uploads a dermoscopy image: separates the tumor area from the surrounding skin using highly accurate dermatologist-like tumor area extraction algorithm, calculates a total of 428 features for the characterization of the tumor, classifies the tumor as melanoma or nevus using a neural network classifier, and presents the diagnosis. Our system achieves a sensitivity of 85.9% and a specificity of 86.0% on a set of 1258 dermoscopy images using cross-validation.

Original languageEnglish
Pages (from-to)566-579
Number of pages14
JournalComputerized Medical Imaging and Graphics
Issue number7
Publication statusPublished - 2008 Oct



  • Computer-aided diagnosis (CAD)
  • Dermoscopy
  • Melanoma
  • Neural network
  • Tumor extraction

ASJC Scopus subject areas

  • Computer Science Applications
  • Radiology Nuclear Medicine and imaging

Cite this