Related Experiment Video
Updated: Dec 17, 2025
09:52
Detection and Isolation of Circulating Melanoma Cells using Photoacoustic Flowmetry
Published on: November 25, 2011
16.2K
Simple and Accurate Border Detection Algorithm for Melanoma Computer Aided Diagnosis
Cataldo Guaragnella1, Maria Rizzi1
1DEI-Department of Electrics and Information Engineering, Politecnico di Bari, 70125 Bari, Italy.
Diagnostics (Basel, Switzerland)
|June 26, 2020
Summary
Early melanoma detection is crucial for patient survival. This study presents an effective computer-aided method for accurate skin lesion segmentation in dermoscopic images, aiding dermatologists in cancer diagnosis.
Area of Science:
- Dermatology
- Computer Vision
- Medical Imaging
Background:
- Increasing incidence of melanoma necessitates improved diagnostic tools.
- Clinical diagnosis of skin lesions is subjective and relies on physician expertise.
- Accurate segmentation of skin lesions is vital for computer-aided analysis.
Purpose of the Study:
- To present an automatic and effective method for pigmented skin lesion segmentation in dermoscopic images.
- To develop a computer vision approach for precise border detection of skin lesions.
- To assist dermatologists in early melanoma detection through image analysis.
Main Methods:
- Developed an automatic pigmented skin lesion segmentation method.
- The procedure extracts a lesion region mask directly from dermoscopic images.
- No additional image pre-processing steps were required.
Main Results:
- The proposed method demonstrated validity and high robustness.
- Effective segmentation was achieved despite irregular boundaries and smooth transitions.
- The method showed resilience to noise and artifacts in dermoscopic images.
Conclusions:
- The developed computer-aided method shows promise for accurate skin lesion segmentation.
- This technique can support dermatologists in the early detection of melanoma.
- The robustness of the method is suitable for real-world clinical applications.
