Skin Lesion Segmentation Method Research Articles

Skin melanoma detection and classification with deep learning

Because of its rapid growth and high mortality rate, melanoma skin cancer is among the most dangerous types of skin cancer. Consequently, melanoma treatment relies heavily on early detection. Based on the U-Net architecture with VGG-16 encoder and semantic segmentation, we present a skin lesion segmentation approach for dermoscopic pictures in this study. Diagnostic imaging systems can assess the characteristics of the segmented skin lesion and assign them a classification based on those aspects. Even on computers without powerful GPUs, the training accuracy is still high enough (over 95%) using the suggested strategy, which uses fewer resources. We use the ISIC dataset, which contains dermoscopy images, to train the model in our trials. We compare the suggested skin lesion segmentation method to others that use deep learning and analyze the Sorensen-Dice and Jaccard scores to determine how well it performs. In terms of skin lesion segmentation, the experimental findings demonstrated that the proposed method outperformed the alternatives.

EA-Net: Research on skin lesion segmentation method based on U-Net

Accurate segmentation of skin lesions is a challenging task because the task is highly influenced by factors such as location, shape and scale. In recent years, Convolutional Neural Networks (CNNs) have achieved advanced performance in automated medical image segmentation. However, existing CNNs have problems such as inability to highlight relevant features and preserve local features, which limit their application in clinical decision-making. This paper proposes a CNN with an added attention mechanism (EA-Net) for more accurate medical image segmentation.EA-Net is based on the U-Net network model framework. Specifically, we added a pixel-level attention module (PA) to the encoder section to preserve the local features of the image during downsampling, making the feature maps input to the decoder more relevant to the ground-truth. At the same time, we added a spatial multi-scale attention module (SA) after the decoding process to increase the spatial weight of the feature maps that are more relevant to the ground-truth, thereby reducing the gap between the output results and the ground-truth. We conducted extensive segmentation experiments on skin lesion images from the ISIC 2017 and ISIC 2018 datasets. The results demonstrate that, when compared to U-Net, our proposed EA-Net achieves an average Dice score improvement of 1.94% and 5.38% for skin lesion tissue segmentation on the ISIC 2017 and ISIC 2018 datasets, respectively. The IoU also increases by 2.69% and 8.31%, and the ASSD decreases by 0.3783 pix and 0.5432 pix, indicating superior segmentation performance. EA-Net can achieve better segmentation results when the original image of skin lesions has an obscure boundary and the segmentation area contains interference factors, which proves that the addition of attention mechanism in the encoder and the application of comprehensive attention mechanism can improve the performance of neural network in the field of skin lesions image segmentation.

MSREA-Net: An Efficient Skin Disease Segmentation Method Based on Multi-Level Resolution Receptive Field

Aiming at the low contrast of skin lesion image and inaccurate segmentation of lesion boundary, a skin lesion segmentation method based on multi-level split receptive field and attention is proposed. Firstly, the depth feature extraction module and multi-level splitting receptive field module are used to extract image feature information; secondly, the hybrid pooling module is used to build long-term and short-term dependencies and integrate global information and local information. Finally, the reverse residual external attention module is introduced to construct the decoding part, which can mine the potential relationship between data sets and improve the network segmentation ability. Experiments on ISBI2017 and ISIC2018 data sets show that the Dice similarity coefficient and Jaccard index reach 88.67% and 91.84%, 79.25% and 81.48%, respectively, and the accuracy reaches 93.89% and 96.16%. The segmentation method is superior to the existing algorithms as a whole. Simulation experiments show that the network has a good effect on skin lesion image segmentation and provides a new method for skin disease diagnosis.

Detection of Melanoma Using Distinct Features

Abstract: Melanoma is one of the most dangerous types of skin cancer because it grows fast and causes the majority of skin cancer fatalities. Hence, early detection is a very important task to treat melanoma. In this work, we suggest a method for dermoscopic images that segments skin lesions based on the U-Net architecture with VGG-16 encoder and the semantic segmentation. Base on the segmented skin lesion, diagnostic imaging systems can evaluate skin lesion features to classify them. The suggested approach requires fewer resources for training, and it is suitable for computing systems without powerful GPUs, but the training accuracy is still high enough (above 95 %). In the experiments, we train the model on the ISIC dataset–a common dermoscopic image dataset. To assess the performance of the proposed skin lesion segmentation method, we evaluate the Sorensen-Dice and the Jaccard scores and compare to other deep learning-based skin lesion segmentation methods. Experimental results showed that skin lesion segmentation qualities of the proposed method are better than ones of the compared methods.

ULFAC-Net: Ultra-Lightweight Fully Asymmetric Convolutional Network for Skin Lesion Segmentation

Segmentation of skin lesions is a critical step in the process of skin lesion diagnosis. Such segmentation is challenging due to the irregular shape, fuzzy contours and severe noise interference in the skin lesion region. Existing deep learning-based skin lesion segmentation methods are usually computationally expensive, hindering their deployment in dermoscopic devices with poor computational power. To address these challenges, we propose an ultralightweight fully asymmetric convolutional network for skin lesion segmentation, called ULFAC-Net. we use a parallel asymmetric convolutional (PAC) module to extract features instead of the traditional square convolution, and innovatively propose a PAC module with dual attention (Att-PAC) to enhance the feature representation. Based on the PAC and Att-PAC modules, we further propose a lightweight textual information submodule. To balance the number of parameters and performance of the model, we also hand-design an asymmetric encoder-decoder architecture. In this paper, we validate the effectiveness and robustness of the proposed ULFAC-Net on four publicly available skin lesion segmentation datasets (ISIC2018, ISBI2017, ISIC2016 and PH2 datasets). The experimental results show that ULFAC-Net achieves competitive segmentation performance with only 0.842 million(0.842M) parameters and 3.71 gigabytes of floating point operations (GFLOPs) compared to other state-of-the-art methods.

JAAL-Net: a joint attention and adversarial learning network for skin lesion segmentation

Objective. Skin lesion segmentation plays an important role in the diagnosis and treatment of melanoma. Existing skin lesion segmentation methods have trouble distinguishing hairs, air bubbles, and blood vessels around lesions, which affects the segmentation performance. Approach. To clarify the lesion boundary and raise the accuracy of skin lesion segmentation, a joint attention and adversarial learning network (JAAL-Net) is proposed that consists of a generator and a discriminator. In the JAAL-Net, the generator is a local fusion network (LF-Net) utilizing the encoder-decoder structure. The encoder contains a convolutional block attention module to increase the weight of lesion information. The decoder involves a contour attention to obtain edge information and locate the lesion. To aid the LF-Net generate higher confidence predictions, a discriminant dual attention network is constructed with channel attention and position attention. Main results. The JAAL-Net is evaluated on three datasets ISBI2016, ISBI2017 and ISIC2018. The intersection over union of the JAAL-Net on the three datasets are 90.27%, 89.56% and 80.76%, respectively. Experimental results show that the JAAL-Net obtains rich lesion and boundary information, enhances the confidence of the predictions, and improves the accuracy of skin lesion segmentation. Significance. The proposed approach effectively improves the performance of the model for skin lesion segmentation, which can assist physicians in accurate diagnosis well.

A Collaborative Learning Model for Skin Lesion Segmentation and Classification.

The automatic segmentation and classification of skin lesions are two essential tasks in computer-aided skin cancer diagnosis. Segmentation aims to detect the location and boundary of the skin lesion area, while classification is used to evaluate the type of skin lesion. The location and contour information of lesions provided by segmentation is essential for the classification of skin lesions, while the skin disease classification helps generate target localization maps to assist the segmentation task. Although the segmentation and classification are studied independently in most cases, we find meaningful information can be explored using the correlation of dermatological segmentation and classification tasks, especially when the sample data are insufficient. In this paper, we propose a collaborative learning deep convolutional neural networks (CL-DCNN) model based on the teacher-student learning method for dermatological segmentation and classification. To generate high-quality pseudo-labels, we provide a self-training method. The segmentation network is selectively retrained through classification network screening pseudo-labels. Specially, we obtain high-quality pseudo-labels for the segmentation network by providing a reliability measure method. We also employ class activation maps to improve the location ability of the segmentation network. Furthermore, we provide the lesion contour information by using the lesion segmentation masks to improve the recognition ability of the classification network. Experiments are carried on the ISIC 2017 and ISIC Archive datasets. The CL-DCNN model achieved a Jaccard of 79.1% on the skin lesion segmentation task and an average AUC of 93.7% on the skin disease classification task, which is superior to the advanced skin lesion segmentation methods and classification methods.

SL-HarDNet: Skin lesion segmentation with HarDNet.

Automatic segmentation of skin lesions from dermoscopy is of great significance for the early diagnosis of skin cancer. However, due to the complexity and fuzzy boundary of skin lesions, automatic segmentation of skin lesions is a challenging task. In this paper, we present a novel skin lesion segmentation network based on HarDNet (SL-HarDNet). We adopt HarDNet as the backbone, which can learn more robust feature representation. Furthermore, we introduce three powerful modules, including: cascaded fusion module (CFM), spatial channel attention module (SCAM) and feature aggregation module (FAM). Among them, CFM combines the features of different levels and effectively aggregates the semantic and location information of skin lesions. SCAM realizes the capture of key spatial information. The cross-level features are effectively fused through FAM, and the obtained high-level semantic position information features are reintegrated with the features from CFM to improve the segmentation performance of the model. We apply the challenge dataset ISIC-2016&PH2 and ISIC-2018, and extensively evaluate and compare the state-of-the-art skin lesion segmentation methods. Experiments show that our SL-HarDNet performance is always superior to other segmentation methods and achieves the latest performance.

Applying an adaptive Otsu-based initialization algorithm to optimize active contour models for skin lesion segmentation.

Medical image processing has gained much attention in developing computer-aided diagnosis (CAD) of diseases. CAD systems require deep understanding of X-rays, MRIs, CT scans and other medical images. The segmentation of the region of interest (ROI) from those images is one of the most crucial tasks. Although active contour model (ACM) is a popular method to segment ROIs in medical images, the final segmentation results highly depend on the initial placement of the contour. In order to overcome this challenge, the objective of this study is to investigate feasibility of developing a fully automated initialization process that can be optimally used in ACM to more effectively segment ROIs. In this study, a fully automated initialization algorithm namely, an adaptive Otsu-based initialization (AOI) method is proposed. Using this proposed method, an initial contour is produced and further refined by the ACM to produce accurate segmentation. For evaluation of the proposed algorithm, the ISIC-2017 Skin Lesion dataset is used due to its challenging complexities. Four different supervised performance evaluation metrics are employed to measure the accuracy and robustness of the proposed algorithm. Using this AOI algorithm, the ACM significantly (p≤0.05) outperforms Otsu thresholding method with 0.88 Dice Score Coefficients (DSC) and 0.79 Jaccard Index (JI) and computational complexity of 0(mn). After comparing proposed method with other state-of-the-art methods, our study demonstrates that the proposed methods is superior to other skin lesion segmentation methods, and it requires no training time, which also makes the new method more efficient than other deep learning and machine learning methods.

Semi-Supervised Skin Lesion Segmentation via Iterative Mask Optimization

Deep learning-based skin lesion segmentation methods have achieved promising results in the community. However, they are usually based on fully supervised learning and require many high-quality ground truths. Labeling the ground truths takes a lot of labor, material, and financial resources. We propose a novel semi-supervised skin lesion segmentation method to solve this problem. First, a hierarchical image segmentation algorithm is used to generate optimal segmentation maps. Then, fully supervised training is performed on a small part of the images with ground truths. The resulting pseudo masks are generated to train the rest of the images. The optimal segmentation maps are utilized in this process to refine the pseudo masks. Experiments show that the proposed method can improve the performance of semi-supervised learning for skin lesion segmentation by reducing the gap with fully supervised learning methods. Moreover, it can reduce the workload of labeling the ground truths. Extensive experiments are conducted on the open dataset to validate the efficiency of the proposed method. The results show that our method is competitive in improving the quality of semi-supervised segmentation.

Learning from dermoscopic images in association with clinical metadata for skin lesion segmentation and classification

Automatic segmentation and classification of lesions are two clinically significant tasks in the computer-aided diagnosis of skin diseases. Both tasks are challenging due to the nonnegligible lesion differences in dermoscopic images from different patients. In this paper, we propose a novel pipeline to efficiently implement skin lesions’ segmentation and classification tasks, which consists of a segmentation network and a classification network. To improve the performance of the segmentation network, we propose a novel module of Multi-Scale Holistic Feature Exploration (MSH) to thoroughly exploit perceptual clues latent among multi-scale feature maps as synthesized by the decoder. The MSH module enables holistic exploration of features across multiple scales to more effectively support downstream image analysis tasks. To boost the performance of the classification network, we propose a novel module of Cross-Modality Collaborative Feature Exploration (CMC) to discover latent discriminative features by collaboratively exploiting potential relationships between cross-modal features of dermoscopic images and clinical metadata. The CMC module enables dynamically capturing versatile interaction effects among cross-modal features during the model’s representation learning procedure by discriminatively and adaptively learning the interaction weight associated with each crossmodality feature pair. In addition, to effectively reduce background noise and boost the lesion discrimination ability of the classification network, we crop the images based on lesion masks generated by the best segmentation model. We evaluate the proposed pipeline on the four public skin lesion datasets, where the ISIC 2018 and PH2 are for segmentation, and the ISIC 2019 and ISIC 2020 are combined into a new dataset, ISIC 2019&2020, for classification. It achieves a Jaccard index of 83.31% and 90.14% in skin lesion segmentation, an AUC of 97.98% and an Accuracy of 92.63% in skin lesion classification, which is superior to the performance of representative state-of-the-art skin lesion segmentation and classification methods. Last but not least, the new model for segmentation utilizes much fewer model parameters (3.3 M) than its peer approaches, leading to a greatly reduced number of labeled samples required for model training, which obtains substantially stronger robustness than its peers.

Machine learning based skin lesion segmentation method with novel borders and hair removal techniques.

The effective segmentation of lesion(s) from dermoscopic skin images assists the Computer-Aided Diagnosis (CAD) systems in improving the diagnosing rate of skin cancer. The results of the existing skin lesion segmentation techniques are not up to the mark for dermoscopic images with artifacts like varying size corner borders with color similar to lesion(s) and/or hairs having low contrast with surrounding background. To improve the results of the existing skin lesion segmentation techniques for such kinds of dermoscopic images, an effective skin lesion segmentation method is proposed in this research work. The proposed method searches for the presence of corner borders in the given dermoscopc image and removes them if found otherwise it starts searching for the presence of hairs on it and eliminate them if present. Next, it enhances the resultant image using state-of-the-art image enhancement method and segments lesion from it using machine learning technique namely, GrabCut method. The proposed method was tested on PH2 and ISIC 2018 datasets containing 200 images each and its accuracy was measured with two evaluation metrics, i.e., Jaccard index, and Dice index. The evaluation results show that our proposed skin lesion segmentation method obtained Jaccard Index of 0.77, 0.80 and Dice index of 0.87, 0.82 values on PH2, and ISIC2018 datasets, respectively, which are better than state-of-the-art skin lesion segmentation techniques.

Weakly supervised skin lesion segmentation based on spot‐seeds guided optimal regions

Automatic skin lesion segmentation is the most critical and relevant task in computer-aided skin cancer diagnosis. Methods based on convolutional neural networks (CNNs) are mainly used in current skin lesion segmentation. The requirement of huge pixel-level labels is a significant obstacle to achieve semantic segmentation of skin lesion by CNNs. In this paper, a novel weakly supervised framework for skin lesion segmentation is presented, which generates high-quality pixel-level annotations and optimizes the segmentation network. A hierarchical image segmentation algorithm can predict a boundary map for training images. Then, the optimal regions of candidate hierarchical levels are selected. Afterward, Superpixels-CRF built on the optimal regions is guided by spot seeds to propagate information from spot seeds to unlabeled regions, resulting in high-quality pixel-level annotations. Using these high-quality pixel-level annotations, a segmentation network can be trained and segmentation masks can be predicted. To iteratively optimize the segmentation network, the predicted segmentation masks are refined and the segmentation network are retrained. Comparative experiments demonstrate that the proposed segmentation framework reduces the gap between weakly and fully supervised skin lesion segmentation methods, and achieves state-of-the-art performance while reducing human labeling efforts.

Hyper-fusion network for semi-automatic segmentation of skin lesions.

Segmentation of skin lesions is an important step for imaging-based clinical decision support systems. Automatic skin lesion segmentation methods based on fully convolutional networks (FCNs) are regarded as the state-of-the-art for accuracy. When there are, however, insufficient training data to cover all the variations in skin lesions, where lesions from different patients may have major differences in size/shape/texture, these methods failed to segment the lesions that have image characteristics, which are less common in the training datasets. FCN-based semi-automatic segmentation methods, which fuse user-inputs with high-level semantic image features derived from FCNs offer an ideal complement to overcome limitations of automatic segmentation methods. These semi-automatic methods rely on the automated state-of-the-art FCNs coupled with user-inputs for refinements, and therefore being able to tackle challenging skin lesions. However, there are a limited number of FCN-based semi-automatic segmentation methods and all these methods focused on 'early-fusion', where the first few convolutional layers are used to fuse image features and user-inputs and then derive fused image features for segmentation. For early-fusion based methods, because the user-input information can be lost after the first few convolutional layers, consequently, the user-input information will have limited guidance and constraint in segmenting the challenging skin lesions with inhomogeneous textures and fuzzy boundaries. Hence, in this work, we introduce a hyper-fusion network (HFN) to fuse the extracted user-inputs and image features over multiple stages. We separately extract complementary features which then allows for an iterative use of user-inputs along all the fusion stages to refine the segmentation. We evaluated our HFN on three well-established public benchmark datasets - ISBI Skin Lesion Challenge 2017, 2016 and PH2 - and our results show that the HFN is more accurate and generalizable than the state-of-the-art methods, in particular with challenging skin lesions.

Machine Learning and Deep Learning Methods for Skin Lesion Classification and Diagnosis: A Systematic Review.

Computer-aided systems for skin lesion diagnosis is a growing area of research. Recently, researchers have shown an increasing interest in developing computer-aided diagnosis systems. This paper aims to review, synthesize and evaluate the quality of evidence for the diagnostic accuracy of computer-aided systems. This study discusses the papers published in the last five years in ScienceDirect, IEEE, and SpringerLink databases. It includes 53 articles using traditional machine learning methods and 49 articles using deep learning methods. The studies are compared based on their contributions, the methods used and the achieved results. The work identified the main challenges of evaluating skin lesion segmentation and classification methods such as small datasets, ad hoc image selection and racial bias.

SLSNet: Skin lesion segmentation using a lightweight generative adversarial network

The determination of precise skin lesion boundaries in dermoscopic images using automated methods faces many challenges, most importantly, the presence of hair, inconspicuous lesion edges and low contrast in dermoscopic images, and variability in the color, texture and shapes of skin lesions. Existing deep learning-based skin lesion segmentation algorithms are expensive in terms of computational time and memory. Consequently, running such segmentation algorithms requires a powerful GPU and high bandwidth memory, which are not available in dermoscopy devices. Thus, this article aims to achieve precise skin lesion segmentation with minimum resources: a lightweight, efficient generative adversarial network (GAN) model called SLSNet, which combines 1-D kernel factorized networks, position and channel attention, and multiscale aggregation mechanisms with a GAN model. The 1-D kernel factorized network reduces the computational cost of 2D filtering. The position and channel attention modules enhance the discriminative ability between the lesion and non-lesion feature representations in spatial and channel dimensions, respectively. A multiscale block is also used to aggregate the coarse-to-fine features of input skin images and reduce the effect of the artifacts. SLSNet is evaluated on two publicly available datasets: ISBI 2017 and the ISIC 2018. Although SLSNet has only 2.35 million parameters, the experimental results demonstrate that it achieves segmentation results on a par with the state-of-the-art skin lesion segmentation methods with an accuracy of 97.61%, and Dice and Jaccard similarity coefficients of 90.63% and 81.98%, respectively. SLSNet can run at more than 110 frames per second (FPS) in a single GTX1080Ti GPU, which is faster than well-known deep learning-based image segmentation models, such as FCN. Therefore, SLSNet can be used for practical dermoscopic applications.

Segmentation of skin lesion images using discrete wavelet transform

Skin lesion segmentation is the essential step in automated computer-aided diagnosis of malignant melanoma. However, this task might be challenging when there are variations in appearances of lesion across different patients and it turns more complex when acquired dermoscopic images include some artifacts like dark corner, gel, low contrast and dense skin hair. In this paper, we propose a skin lesion segmentation method using discrete wavelet transform to tackle the complexities present in the dermoscopic images. The proposed method is based on the analysis of different colour components from various colour spaces such as YCbCr, HSV subjected to discrete wavelet decomposition to remove unwanted data in the form of marker ink, colour chart etc. Otsu and Histogram based thresholding is used to separate skin lesion region from the background. The proposed method is tested using publicly available PH2 and Kaggle's Skin Lesion Segmentation dataset. The results are compared with different state-of-the-art methods using evaluation metrics. The results reveal the effectiveness of the proposed method in comparison with other approaches, accordance with quantitative results and visual effects.

Skin lesion segmentation from dermoscopic images by using Mask R-CNN, Retina-Deeplab, and graph-based methods

Background and objectiveTimely diagnosis of skin cancer which is one of the most common cancers can greatly prevent death. Automatic skin lesion segmentation is an important part of an automatic skin cancer diagnosis system. Due to the wide variety in color, location, size, shape, and boundary contrast of lesions, the lesion segmentation is still a challenging problem. MethodsIn this study, we present a two-stage automatic skin lesion segmentation method. In the first stage, a detection-based segmentation structure, Retina-Deeplab, is proposed to be combined with the Mask R-CNN, which inherently detects and segments objects simultaneously. To combine the results of these two segmentation methods, two geodesic-based and graph-based combination approaches are proposed. ResultsThe proposed method is evaluated using three well-known skin image datasets (ISBI 2017, DermQuest, and PH2). Through the proposed two-step graph-based combination strategy, the Jaccard value of the overall lesion segmentation method reached 80.04%, which is 3.54% higher than the winner of the ISBI 2017 lesion segmentation challenge. ConclusionsThe proposed Retina-Deeplab segmentation method reached about 1% of the Jaccard value higher than the Mask R-CNN. Our overall segmentation method considered both overall characteristics of lesions in all images (by using CNN-based methods in the first stage) and image-specific features of lesions (by using geodesic-based/graph-based combination approaches in the second stage). The proposed two-step geodesic-based and graph-based combination approaches performed better than earlier combination strategies. Experiments demonstrated that the overall proposed lesion segmentation methods outperformed other state-of-the-art methods on well-known datasets.

Skin lesion segmentation method for dermoscopic images with convolutional neural networks and semantic segmentation

Melanoma skin cancer is one of the most dangerous forms of skin cancer because it grows fast and causes most of the skin cancer deaths. Hence, early detection is a very important task to treat melanoma. In this article, we propose a skin lesion segmentation method for dermoscopic images based on the U-Net architecture with VGG-16 encoder and the semantic segmentation. Base on the segmented skin lesion, diagnostic imaging systems can evaluate skin lesion features to classify them. The proposed method requires fewer resources for training, and it is suitable for computing systems without powerful GPUs, but the training accuracy is still high enough (above 95 %). In the experiments, we train the model on the ISIC dataset – a common dermoscopic image dataset. To assess the performance of the proposed skin lesion segmentation method, we evaluate the Sorensen-Dice and the Jaccard scores and compare to other deep learning-based skin lesion segmentation methods. Experimental results showed that skin lesion segmentation quality of the proposed method are better than ones of the compared methods.

Simple and Accurate Border Detection Algorithm for Melanoma Computer Aided Diagnosis.

The interest of the scientific community for computer aided skin lesion analysis and characterization has been increased during the last years for the growing incidence of melanoma among cancerous pathologies. The detection of melanoma in its early stage is essential for prognosis improvement and for guaranteeing a high five-year relative survival rate of patients. The clinical diagnosis of skin lesions is challenging and not trivial since it depends on human vision and physician experience and expertise. Therefore, a computer method that makes an accurate extraction of important details of skin lesion image can assist dermatologists in cancer detection. In particular, the border detection is a critical computer vision issue owing to the wide range of lesion shapes, sizes, colours and skin texture types. In this paper, an automatic and effective pigmented skin lesion segmentation method in dermoscopic image is presented. The proposed procedure is adopted to extract a mask of the lesion region without the adoption of other signal processing procedures for image improvement. A quantitative experimental evaluation has been performed on a publicly available database. The achieved results show the method validity and its high robustness towards irregular boundaries, smooth transition between lesion and skin, noise and artifact presence.