Abstract:Background and objective Melanoma is a highly malignant skin tumor. Accurate segmentation of skin lesions from dermoscopy images is pivotal for computer-aided diagnosis of melanoma. However, blurred lesion boundaries, variable lesion shapes, and other interference factors pose a challenge in this regard. Methods This work proposes a novel framework called CFF-Net (Cross Feature Fusion Network) for supervised skin lesion segmentation. The encoder of the network includes dual branches, where the CNNs branch aims to extract rich local features while MLPs branch is used to establish both the global-spatial-dependencies and global-channel-dependencies for precise delineation of skin lesions. Besides, a feature-interaction module between two branches is designed for strengthening the feature representation by allowing dynamic exchange of spatial and channel information, so as to retain more spatial details and inhibit irrelevant noise. Moreover, an auxiliary prediction task is introduced to learn the global geometric information, highlighting the boundary of the skin lesion. Results Comprehensive experiments using four publicly available skin lesion datasets (i.e., ISIC 2018, ISIC 2017, ISIC 2016, and PH2) indicated that CFF-Net outperformed the state-of-the-art models. In particular, CFF-Net greatly increased the average Jaccard Index score from 79.71% to 81.86% in ISIC 2018, from 78.03% to 80.21% in ISIC 2017, from 82.58% to 85.38% in ISIC 2016, and from 84.18% to 89.71% in PH2 compared with U-Net. Ablation studies demonstrated the effectiveness of each proposed component. Cross-validation experiments in ISIC 2018 and PH2 datasets verified the generalizability of CFF-Net under different skin lesion data distributions. Finally, comparison experiments using three public datasets demonstrated the superior performance of our model. Conclusion The proposed CFF-Net performed well in four public skin lesion datasets, especially for challenging cases with blurred edges of skin lesions and low contrast between skin lesions and background. CFF-Net can be employed for other segmentation tasks with better prediction and more accurate delineation of boundaries.

Weakly supervised semantic segmentation for skin cancer via CNN superpixel region response

Attention-Guided Network with Densely Connected Convolution for Skin Lesion Segmentation

Combining unsupervised constraints on weakly supervised semantic segmentation of skin cancer

Deep Learning-Based Semantic Segmentation of Non-Melanocytic Skin Tumors in Whole-Slide Histopathological Images.

Semi- and Weakly Supervised Directional Bootstrapping Model for Automated Skin Lesion Segmentation.

Unsupervised sub-segmentation for pigmented skin lesions.

A novel SpaSA based hyper-parameter optimized FCEDN with adaptive CNN classification for skin cancer detection

Channel Attention Separable Convolution Network for Skin Lesion Segmentation

SSCD-Net: Semi-supervised Skin Cancer Diagnostical Network Combined with Curriculum Learning, Disease Relation and Clinical Information

Skin lesion image segmentation by using backchannel filling CNN and level sets

Portable Skin Lesion Segmentation System with Accurate Lesion Localization Based on Weakly Supervised Learning

Self-Supervised Skin Lesion Segmentation: An Annotation-Free Approach

Skin lesion classification using CNNs with grouping of multi-scale attention and class-specific loss weighting

Segmentation of skin lesion using superpixel guided generative adversarial network with dual-stream patch-based discriminators

Accurate Weakly-Supervised Deep Lesion Segmentation using Large-Scale Clinical Annotations: Slice-Propagated 3D Mask Generation from 2D RECIST

Efficient and real-time skin lesion image segmentation using spatial-frequency information and channel convolutional networks

Lesion Region Segmentation Via Weakly Supervised Learning

A Collaborative Learning Model for Skin Lesion Segmentation and Classification

Skin Lesion Segmentation through Generative Adversarial Networks with Global and Local Semantic Feature Awareness

Complementary Network with Adaptive Receptive Fields for Melanoma Segmentation

Dynamically Aggregating MLPs and CNNs for Skin Lesion Segmentation with Geometry Regularization