Loading...

Multi-class segmentation of skin lesions via joint dictionary learning

Moradi, N ; Sharif University of Technology | 2021

618 Viewed
  1. Type of Document: Article
  2. DOI: 10.1016/j.bspc.2021.102787
  3. Publisher: Elsevier Ltd , 2021
  4. Abstract:
  5. Melanoma is the deadliest type of human skin cancer. However, it is curable if diagnosed in an early stage. Recently, computer aided diagnosis (CAD) systems have drawn much interests. Segmentation is a crucial step of a CAD system. There are different types of skin lesions having high similarities in terms of color, shape, size and appearance. Most available works focus on a binary segmentation. Due to the huge variety of skin lesions and high similarities between different types of lesions, multi-class segmentation is still a challenging task. Here, we propose a method based on joint dictionary learning for multi-class segmentation of dermoscopic images. The key idea is based on combining data from different feature spaces to build a more informative structure. We consider training data from two different spaces. Then, two dictionaries are jointly learned using the K-SVD algorithm. The final segmentation is accomplished by a graph-cut method based on both the topological information of lesions and the learned dictionaries. We evaluate our proposed method on the ISIC 2107 dataset to segment three classes of lesions. Our method achieves better results, specially for challenging skin lesions, compared to the only available method for multi-class segmentation of dermoscopic images. We also evaluate the performance of our method for binary segmentation and lesion diagnosis and compared the results with the other state-of-the-art methods. Experimental results show the efficiency and effectiveness of the proposed method in producing results that are more reliable for clinical applications, even using limited amount of training data. © 2021
  6. Keywords:
  7. Computer aided diagnosis ; Dermatology ; Image segmentation ; Binary segmentation ; Computer aided diagnosis systems ; Dermoscopic images ; Dictionary learning ; Graph-cut ; Joint dictionary learning ; Multi-class segmentations ; Skin lesion ; Sparse representation ; Training data ; Graphic methods ; Algorithm ; Classifier ; Clinical practice ; Computer assisted diagnosis ; Controlled study ; Diagnostic accuracy ; Epiluminescence microscopy ; fcn 16 segmentation algorithm ; fcn 32 segmentation algorithm ; fcn 8 segmentation algorithm ; fcn alexnet segmentation algorithm ; Feature extraction ; Histogram ; Human ; Image analysis ; Iterative region growing algorithm ; Learning algorithm ; Luminance ; Major clinical study ; Priority journal ; Region growing (imaging) ; Seborrheic keratosis ; Segmentation algorithm ; Sensitivity and specificity ; Singular value decomposition ; Skin defect ; Spitz nevus ; Transfer of learning
  8. Source: Biomedical Signal Processing and Control ; Volume 68 , 2021 ; 17468094 (ISSN)
  9. URL: https://www.sciencedirect.com/science/article/abs/pii/S1746809421003840