Exploring mesh-based features for enhanced dermoscopic lesion classification
Authors:
Open Access
- 29 September 2026
- journal article
- research article
- Published by Springer Nature in Research on Biomedical Engineering
- Vol. 42 (4) , 54
Abstract
Purpose: Computational methods in digital image analysis have made significant progress in facilitating early diagnosis across various medical specialties, especially in dermoscopy. This study introduces a mesh-based classification approach that uses geometric and color information extracted from pigmented skin lesions to aid in the description and prediction of melanoma. Methods: The proposed method initially constructs Delaunay meshes to model segmented lesions under specific settings. Subsequently, mesh-derived and graph-based features are extracted and selected to feed machine learning algorithms. Results: An experimental evaluation of our proposal on 274 public dermoscopies with melanoma and nevus abnormalities indicates that our approach achieved competitive performance against deep and shallow learning approaches. In particular, by using fewer training images than the literature methods, the best proposal configurations yielded an average accuracy of 86% and reduced the number of features down to 24. Notably, mesh configurations with a variable number of vertices around the lesion border and a fixed number within the lesion demonstrated promising results. Conclusion: The mesh-based classification method proposed in this work shows promise in dermoscopy and presents potential for application in other medical image classification domains.Keywords
Funding Information
- Universidade Estadual do Oeste do Paraná
This publication has 65 references indexed in Scilit:
- Prospective Evidence on Artificial Intelligence−Assisted Melanoma DiagnosticsJAMA Dermatology, 2026
- Comparison of the impacts of dermoscopy image augmentation methods on skin cancer classification and a new augmentation method with wavelet packetsInternational Journal of Imaging Systems and Technology, 2023
- Automatic skin cancer detection in dermoscopy images by combining convolutional neural networks and texture featuresInternational Journal of Imaging Systems and Technology, 2020
- Performance Evaluation of Classification Algorithms on Diagnosis of Breast Cancer and Skin Disease2020
- The Role of Color and Morphologic Characteristics in Dermoscopic DiagnosisJAMA Dermatology, 2016
- Nonparametric Statistics for Non‐Statisticians2009
- Improved Dataset Characterisation for Meta-learning2002
- The Morphologic Criteria of the Pseudopod in Surface MicroscopyArchives of Dermatology, 1995
- Estimating attributes: Analysis and extensions of RELIEF1994
- Constrained delaunay triangulationsAlgorithmica, 1989