Computer aided measurement of melanoma depth of invasion in microscopic images

Mojgan Mokhtari, Mahdie Rezaeian, Shahriar Gharibzadeh, Vahid Malekian*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

21 Citations (Scopus)

Abstract

This paper presents a novel computer aided technique for measurement of melanoma depth of invasion. Melanoma is the deadliest form of skin cancer with worldwide increasing incidences. For a conclusive diagnosis of melanoma, skin biopsies should be examined under a microscope. Visual inspection of microscopic samples is often subjective, time-consuming, cumbersome and prone to human errors. This fact demonstrates the necessity of developing an automated method which assists pathologists in evaluating histopathological samples more accurately in the busy clinical environment. To the best of our knowledge, this is the first time that a computer-assisted diagnosis algorithm has been applied in measurement of melanoma invasion depth. The proposed method uses a clustering algorithm for granular layer extraction and a pre-trained SVM classifier for detection of malignant melanocytes. The experimental results with average error of 3.9. μm demonstrate that the proposed method is reliable and effective.

Original languageEnglish
Pages (from-to)40-48
Number of pages9
JournalMicron
Volume61
DOIs
Publication statusPublished - Jun 2014
Externally publishedYes

Keywords

  • Computer-assisted diagnosis
  • Depth of invasion
  • Melanoma
  • SVM

Fingerprint

Dive into the research topics of 'Computer aided measurement of melanoma depth of invasion in microscopic images'. Together they form a unique fingerprint.

Cite this