Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Image analysis and pattern recognition for computer supported skin tumor diagnosis.

H Handels1, T Ross, J Kreusch

  • 1Institute for Medical Informatics, Medical University of Lübeck, Germany. Handels@medinf.mu-luebeck.de

Studies in Health Technology and Informatics
|June 29, 1999
PubMed
Summary

This study introduces a novel computer-aided method for identifying melanoma and naevocytic nevi using high-resolution skin surface profiles. The nearest neighbor classifier achieved the best diagnostic accuracy with an error rate of 2.3%.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

[Dermoscopy in special locations : Nails, acral skin, face, and mucosa].

Der Hautarzt; Zeitschrift fur Dermatologie, Venerologie, und verwandte Gebiete·2019
Same author

Fabrication of self-assembled spherical Gold Particles by pulsed UV Laser Treatment.

Scientific reports·2018
Same author

[Dermoscopy for malignant and benign skin tumors : Indication and standardized terminology].

Der Hautarzt; Zeitschrift fur Dermatologie, Venerologie, und verwandte Gebiete·2017
Same author

Medical Image Processing: New Perspectives in Computer Supported Diagnostics, Computer Aided Surgery and Medical Education and Training.

Yearbook of medical informatics·2016
Same author

Relationship between basophil chemotaxis and histamine releasability to immunological stimulation.

Inflammation research : official journal of the European Histamine Research Society ... [et al.]·2016
Same author

Quantitative surface topography as a tool in the differential diagnosis between melanoma and naevus.

Skin research and technology : official journal of International Society for Bioengineering and the Skin (ISBS) [and] International Society for Digital Imaging of Skin (ISDIS) [and] International Society for Skin Imaging (ISSI)·2016

Area of Science:

  • Dermatology
  • Computer Science
  • Medical Imaging

Background:

  • Melanoma and naevocytic nevi require accurate early detection for effective treatment.
  • Traditional diagnostic methods can be subjective and time-consuming.
  • Computer-aided diagnosis (CAD) systems offer potential for objective and efficient skin lesion analysis.

Purpose of the Study:

  • To develop and evaluate a novel computer-supported approach for recognizing melanoma and naevocytic nevi.
  • To assess the efficacy of high-resolution skin surface profiles and advanced image analysis techniques for skin lesion classification.
  • To compare the performance of different machine learning classifiers in differentiating between benign and malignant skin lesions.

Main Methods:

  • Generating high-resolution skin surface profiles using a laser profilometer (4x4 mm area, 125 points/mm resolution, 0.1 micron vertical resolution).

Related Experiment Videos

  • Extracting image features including Haralick's texture parameters, Fourier features, and fractal analysis features.
  • Employing genetic algorithms for optimal feature subset selection.
  • Classifying lesions using a feed-forward back-propagation neural network and a nearest neighbor classifier, with performance optimized through various network configurations and pruning techniques.

    • Main Results:

    • The nearest neighbor classifier achieved the highest classification performance with an error rate of 2.3%.
    • The best-performing neural network classifier, optimized via pruning, yielded an error rate of 4.5%.
    • Genetic algorithms effectively identified feature subsets that improved classification accuracy.

    Conclusions:

    • High-resolution skin surface profiling combined with advanced image analysis and machine learning provides a promising approach for computer-aided diagnosis of skin lesions.
    • The nearest neighbor classifier demonstrated superior performance in distinguishing between melanoma and naevocytic nevi in this study.
    • Further research and validation are warranted to integrate this technology into clinical practice for improved dermatological diagnostics.