Related Experiment Video
Updated: Aug 8, 2026
13:58
Flat Mount Imaging of Mouse Skin and Its Application to the Analysis of Hair Follicle Patterning and Sensory Axon Morphology
Published on: June 25, 2014
Techniques for a structural analysis of dermatoscopic imagery
M G Fleming1, C Steger, J Zhang
1Department of Dermatology, Medical College of Wisconsin and Zablocki VA Hospital, Milwaukee, USA.
Summary
Automated techniques accurately detect and characterize skin structures in dermatoscopic images. This improves the analysis of pigmented lesions using advanced algorithms and image segmentation.
Area of Science:
- Dermatology
- Medical Imaging
- Computer Vision
Background:
- Accurate analysis of dermatoscopic images is crucial for diagnosing skin conditions.
- Automated methods can enhance the efficiency and consistency of dermatoscopic analysis.
Purpose of the Study:
- To develop and validate automated techniques for detecting and characterizing dermatoscopic structures.
- To enable global segmentation of pigmented lesions using novel image processing approaches.
Main Methods:
- Utilized algorithms for grayscale shape extraction (Steger).
- Incorporated a snake algorithm and modified region competition strategy (Zhu and Yuille).
- Developed a novel global segmentation approach based on stabilized inverse diffusion equations.
Main Results:
- Successfully developed automated detection and characterization of pigment network and brown globules.
- Reported procedures for detecting air bubbles and hairs in dermatoscopic images.
- Achieved global segmentation of pigmented lesions.
Conclusions:
- The developed techniques offer robust automated analysis of dermatoscopic images.
- These methods have the potential to improve diagnostic accuracy and efficiency in dermatology.
