Related Experiment Video
Updated: Jun 8, 2026
07:15
Transient Optical Clearing Using Absorbing Molecules for Ex Vivo and In Vivo Imaging
Published on: July 11, 2025
Border detection in dermoscopy images using hybrid thresholding on optimized color channels
Rahil Garnavi1, Mohammad Aldeen, M Emre Celebi
1Department of Electrical and Electronic Engineering, NICTA Victoria Research Laboratory, Universty of Melbourne, Parkville, Melbourne, Victoria 3010, Australia. r.garnavi@ee.unimelb.edu.au
Summary
This study introduces an automated method for detecting skin lesion borders in dermoscopy images by optimizing color channels and using hybrid thresholding. The approach achieves high accuracy and efficiency, rivaling expert dermatologists.
Area of Science:
- Medical Imaging
- Computer Vision
- Dermatology
Background:
- Automated border detection is crucial for dermoscopy image analysis.
- Optimal color channel selection for border detection remains under-explored.
- Existing methods lack focus on specificity and sensitivity optimization.
Purpose of the Study:
- To propose an automated dermoscopy border detection method.
- To identify optimal color channels for enhanced lesion border detection.
- To develop a hybrid thresholding technique for improved accuracy.
Main Methods:
- Color channel optimization tested on 30 dermoscopy images with dermatologist ground truth.
- Hybrid thresholding method applied to 85 dermoscopy images.
- Two-stage approach prioritizing specificity then sensitivity.
Main Results:
- The method demonstrates high accuracy, precision, sensitivity, and specificity.
- Achieves competitive performance against state-of-the-art methods.
- Outperforms or matches a dermatology registrar's effectiveness.
Conclusions:
- The proposed automated method effectively detects lesion borders in dermoscopy images.
- Optimal color channel selection and hybrid thresholding enhance detection accuracy.
- This technique offers a potentially faster and more effective alternative to current methods.
