Lesion Boundary Segmentation With Artifacts Removal and Melanoma Detection in Skin Lesion Images

Melanoma is a severe form of skin cancer characterized by the rapid multiplication of pigment-producing cells. A problem on analysis of these images is interesting because of the existence of artifacts that produces noise such as hair, veins, water residue, illuminations, and light reflections. An important step in the diagnosis of melanoma is the removal and reduction of these artifacts that can inhibit the examination to accurately segment the skin lesion from the surrounding skin area. A simple method for artifacts removal for extracting skin lesion is implemented based on image enhancement and morphological operators. This is used for training together with some augmentation techniques on images for melanoma detection. The experimental results show that artifact removal and lesion segmentation in skin lesion images performed a true detection rate of 95.37% for melanoma skin lesion segmentation, and as high as 92.5% accuracy for melanoma detection using both GoogLeNet and Resnet50.

Download Full-text

Hybrid fully convolutional networks‐based skin lesion segmentation and melanoma detection using deep feature

International Journal of Imaging Systems and Technology ◽

10.1002/ima.22377 ◽

2019 ◽

Vol 30 (2) ◽

pp. 348-357 ◽

Cited By ~ 2

Author(s):

Kalyanakumar Jayapriya ◽

Israel Jeena Jacob

Keyword(s):

Skin Lesion ◽

Lesion Segmentation ◽

Convolutional Networks ◽

Melanoma Detection ◽

Fully Convolutional Networks ◽

Deep Feature

Download Full-text

Automatic Skin Lesion Segmentation and Melanoma Detection: Transfer Learning Approach with U-Net and DCNN-SVM

Proceedings of International Joint Conference on Computational Intelligence - Algorithms for Intelligent Systems ◽

10.1007/978-981-13-7564-4_32 ◽

2019 ◽

pp. 371-381 ◽

Cited By ~ 1

Author(s):

Zabir Al Nazi ◽

Tasnim Azad Abir

Keyword(s):

Skin Lesion ◽

Transfer Learning ◽

Learning Approach ◽

Lesion Segmentation ◽

Melanoma Detection

Download Full-text

Skin Lesion Segmentation in Dermoscopic Images with Combination of YOLO and GrabCut Algorithm

Diagnostics ◽

10.3390/diagnostics9030072 ◽

2019 ◽

Vol 9 (3) ◽

pp. 72 ◽

Cited By ~ 25

Author(s):

Halil Murat Ünver ◽

Enes Ayan

Keyword(s):

Skin Lesion ◽

Automatic Segmentation ◽

Critical Role ◽

Skin Lesions ◽

Lesion Segmentation ◽

Low Contrast ◽

Melanoma Detection ◽

Sensitivity Rate ◽

Dermoscopic Image ◽

Grabcut Algorithm

Skin lesion segmentation has a critical role in the early and accurate diagnosis of skin cancer by computerized systems. However, automatic segmentation of skin lesions in dermoscopic images is a challenging task owing to difficulties including artifacts (hairs, gel bubbles, ruler markers), indistinct boundaries, low contrast and varying sizes and shapes of the lesion images. This paper proposes a novel and effective pipeline for skin lesion segmentation in dermoscopic images combining a deep convolutional neural network named as You Only Look Once (YOLO) and the GrabCut algorithm. This method performs lesion segmentation using a dermoscopic image in four steps: 1. Removal of hairs on the lesion, 2. Detection of the lesion location, 3. Segmentation of the lesion area from the background, 4. Post-processing with morphological operators. The method was evaluated on two publicly well-known datasets, that is the PH2 and the ISBI 2017 (Skin Lesion Analysis Towards Melanoma Detection Challenge Dataset). The proposed pipeline model has achieved a 90% sensitivity rate on the ISBI 2017 dataset, outperforming other deep learning-based methods. The method also obtained close results according to the results obtained from other methods in the literature in terms of metrics of accuracy, specificity, Dice coefficient, and Jaccard index.

Download Full-text

HAIR ARTIFACT REMOVAL AND SKIN LESION SEGMENTATION OF DERMOSCOPY IMAGES

Asian Journal of Pharmaceutical and Clinical Research ◽

10.22159/ajpcr.2018.v11s3.30025 ◽

2018 ◽

Vol 11 (15) ◽

pp. 36 ◽

Cited By ~ 5

Author(s):

Julie Ann Salido ◽

Conrado Ruiz Jr

Keyword(s):

Skin Lesion ◽

Detection Rate ◽

Median Filter ◽

Research Work ◽

Color Space ◽

Reference Image ◽

Skin Area ◽

Lesion Segmentation ◽

Artifact Removal ◽

Morphological Opening

Objective: The objective of this research is to perform automatic hair artifact removal and skin lesion segmentation on dermoscopy images.Methods: Dermoscopy images are images from the examination of the skin lesion using a dermatoscope. There are different types of skin lesion artifacts, structures, or objects that are present in dermoscopy images. This is a pertinent problem that can inhibit the proper examination and accurately segment the skin lesion from the surrounding skin area. Artifacts, such as hair strands, introduce additional features that can also cause problems during classification. Our process starts with hair removal using a median filter on each color space of RGB, a bottom hat filter, a binary conversion, a dilation and morphological opening, and then the removal of small connected pixels. The detected hair regions are then filled up using harmonic inpainting. Then, skin lesion segmentation is performed using a binary conversion, a dilation, a perimeter detection and morphological opening, and then the removal of small connected pixels.Results: Experiments were carried out on the PH2 dermoscopy images. The border of the lesion was quantified for evaluation by four statistical metrics with the lesions identified by the PH2 as the reference image, resulting with a true detection rate (TDR) of 82.31 and a false detection rate of 5.69.Conclusions: The results obtained in the research work on hair artifacts removal and skin lesion segmentation provides acceptable results in terms of TDR and low false-positive rates.

Download Full-text

Skin lesion segmentation using object scale-oriented fully convolutional neural networks

Signal Image and Video Processing ◽

10.1007/s11760-018-01410-3 ◽

2019 ◽

Vol 13 (3) ◽

pp. 431-438 ◽

Cited By ~ 9

Author(s):

Lin Huang ◽

Yi-gong Zhao ◽

Tie-jun Yang

Keyword(s):

Neural Networks ◽

Skin Lesion ◽

Convolutional Neural Networks ◽

Lesion Segmentation ◽

Fully Convolutional Neural Networks

Download Full-text

FAC-Net: Feedback Attention Network Based on Context Encoder Network for Skin Lesion Segmentation

Sensors ◽

10.3390/s21155172 ◽

2021 ◽

Vol 21 (15) ◽

pp. 5172

Author(s):

Yuying Dong ◽

Liejun Wang ◽

Shuli Cheng ◽

Yongming Li

Keyword(s):

Skin Lesion ◽

Real Life ◽

Good Deal ◽

Experimental Tests ◽

Skin Lesions ◽

Lesion Segmentation ◽

Attention Network ◽

Effective Feedback ◽

Public Datasets ◽

Learning Architectures

Considerable research and surveys indicate that skin lesions are an early symptom of skin cancer. Segmentation of skin lesions is still a hot research topic. Dermatological datasets in skin lesion segmentation tasks generated a large number of parameters when data augmented, limiting the application of smart assisted medicine in real life. Hence, this paper proposes an effective feedback attention network (FAC-Net). The network is equipped with the feedback fusion block (FFB) and the attention mechanism block (AMB), through the combination of these two modules, we can obtain richer and more specific feature mapping without data enhancement. Numerous experimental tests were given by us on public datasets (ISIC2018, ISBI2017, ISBI2016), and a good deal of metrics like the Jaccard index (JA) and Dice coefficient (DC) were used to evaluate the results of segmentation. On the ISIC2018 dataset, we obtained results for DC equal to 91.19% and JA equal to 83.99%, compared with the based network. The results of these two main metrics were improved by more than 1%. In addition, the metrics were also improved in the other two datasets. It can be demonstrated through experiments that without any enhancements of the datasets, our lightweight model can achieve better segmentation performance than most deep learning architectures.

Download Full-text