Border Detection in Skin Lesion Images Using an Improved Clustering Algorithm

2020 ◽  
Vol 16 (4) ◽  
pp. 15-29
Author(s):  
Jayalakshmi D. ◽  
Dheeba J.
Keyword(s):  
Skin Cancer ◽  
Skin Lesion ◽  
Clustering Algorithm ◽  
Median Filter ◽  
Second Phase ◽  
Border Detection ◽  
Early Computer ◽  
Segmentation Methods ◽  
Two Phases ◽  
Skin Cancer Detection

The incidence of skin cancer has been increasing in recent years and it can become dangerous if not detected early. Computer-aided diagnosis systems can help the dermatologists in assisting with skin cancer detection by examining the features more critically. In this article, a detailed review of pre-processing and segmentation methods is done on skin lesion images by investigating existing and prevalent segmentation methods for the diagnosis of skin cancer. The pre-processing stage is divided into two phases, in the first phase, a median filter is used to remove the artifact; and in the second phase, an improved K-means clustering with outlier removal (KMOR) algorithm is suggested. The proposed method was tested in a publicly available Danderm database. The improved cluster-based algorithm gives an accuracy of 92.8% with a sensitivity of 93% and specificity of 90% with an AUC value of 0.90435. From the experimental results, it is evident that the clustering algorithm has performed well in detecting the border of the lesion and is suitable for pre-processing dermoscopic images.

scholarly journals Self-Adaptive K-Means Based on a Covering Algorithm

Complexity ◽  
2018 ◽  
Vol 2018 ◽  
pp. 1-16 ◽  
Author(s):  
Yiwen Zhang ◽  
Yuanyuan Zhou ◽  
Xing Guo ◽  
Jintao Wu ◽  
Qiang He ◽  
...  
Keyword(s):  
Large Scale ◽  
Clustering Algorithm ◽  
Real Data ◽  
Second Phase ◽  
Data Sets ◽  
Number Of Clusters ◽  
Large Scale Data ◽  
Long Time ◽  
Two Phases ◽  
Selection Of

The K-means algorithm is one of the ten classic algorithms in the area of data mining and has been studied by researchers in numerous fields for a long time. However, the value of the clustering number k in the K-means algorithm is not always easy to be determined, and the selection of the initial centers is vulnerable to outliers. This paper proposes an improved K-means clustering algorithm called the covering K-means algorithm (C-K-means). The C-K-means algorithm can not only acquire efficient and accurate clustering results but also self-adaptively provide a reasonable numbers of clusters based on the data features. It includes two phases: the initialization of the covering algorithm (CA) and the Lloyd iteration of the K-means. The first phase executes the CA. CA self-organizes and recognizes the number of clusters k based on the similarities in the data, and it requires neither the number of clusters to be prespecified nor the initial centers to be manually selected. Therefore, it has a “blind” feature, that is, k is not preselected. The second phase performs the Lloyd iteration based on the results of the first phase. The C-K-means algorithm combines the advantages of CA and K-means. Experiments are carried out on the Spark platform, and the results verify the good scalability of the C-K-means algorithm. This algorithm can effectively solve the problem of large-scale data clustering. Extensive experiments on real data sets show that the accuracy and efficiency of the C-K-means algorithm outperforms the existing algorithms under both sequential and parallel conditions.

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

2021 ◽  
Vol 45 (1) ◽  
pp. 122-129
Author(s):  
Dang N.H. Thanh ◽  
Nguyen Hoang Hai ◽  
Le Minh Hieu ◽  
Prayag Tiwari ◽  
V.B. Surya Prasath
Keyword(s):  
Skin Cancer ◽  
Skin Lesion ◽  
Semantic Segmentation ◽  
Segmentation Method ◽  
Lesion Segmentation ◽  
Computing Systems ◽  
Segmentation Methods ◽  
Dermoscopic Image ◽  
Better Than

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.

Multi-Class Skin Cancer Detection and Classification Using Hybrid Features Extraction Techniques

2020 ◽  
Vol 10 (10) ◽  
pp. 2466-2472
Author(s):  
Mahnoor Masood ◽  
Khalid Iqbal ◽  
Qasim Khan ◽  
Ali Saeed Alowayr ◽  
Khalid Mahmood Awan ◽  
...  
Keyword(s):  
Skin Cancer ◽  
Skin Lesion ◽  
Early Stage ◽  
Noise Removal ◽  
Support Vector ◽  
Hybrid Features ◽  
Cancer Types ◽  
Biopsy Methods ◽  
Skin Cancer Detection ◽  
Aided Diagnosis

Skin cancer is measured as one of the fatal types of cancer diseases in humans, among numerous kinds of malignancy. Current diagnostic classifications are lacking in finding an effective treatment. The effective and early stage treatment of skin disease can increase the survival rate of patients. Substantial investigative work has been developed to improve computer aided diagnosis system to detect cancer at early stage. However, early detection of skin cancer still requires better accuracy through experiment on digital skin lesion images as a multiclass classification, rather than using biopsy methods. This paper presents an intelligent framework to detect and classify four types of skin cancer. Before classification, noise removal from skin lesion is performed by gaussian filter. Textural and colour features are extracted from skin lesion to detect and classify cancer into four types. Support vector Machine is trained to classify Melanoma, Nevus, Basal and Squamous skin cancer types. Extensive experiments are performed on standard benchmark skin cancer images dataset with an improvement in accuracy of 92.41% after comparison with the well-known methods.

scholarly journals Skin Lesion Border Detection Based on Best Statistical Model Using Optimal Colour Channel

2020 ◽  
Vol 3 (1) ◽  
pp. 18
Author(s):  
Alaa Ahmed Abbas Al-abayechi ◽  
Fadeheela Sabri Abu-Almash
Keyword(s):  
Skin Lesion ◽  
Median Filter ◽  
Bilateral Filter ◽  
Polynomial Model ◽  
Morphological Operations ◽  
Edge Detector ◽  
Border Detection ◽  
Canny Edge Detector ◽  
Edge Detectors ◽  
Canny Edge

This paper proposes an effective way to segment melanoma skin lesion in colour dermoscopic images, using an edge-based approach. The proposed method, different methods were combined to improve the segmentation performance. These methods are morphological operations, bilateral filter, spline, polynomial model and canny edge detector. Different methods were tested to select the best method that was produced the best outcome. These testing methods, bilateral filter provided the highest PSNR amongst other filters such as median filter, Gaussian and average filter. Two statistical models were implemented polynomial model and linear regression and selected the best performance as polynomial model. Four edge detectors were applied to detect the edge of skin lesion and select the best segmentation accuracy.  Manual border selection was used as the benchmark to evaluation the accuracy of the automatic border. The proposed method was able to achieve a good average accuracy of 96.69% based on canny edge detector. Our dataset consists of (70) dermoscopic images that includes melanoma and nevus.

DUNEScan: A Web Application for Uncertainty Estimation in Skin Cancer Detection with Deep Neural Networks

2021 ◽  
Author(s):  
Bogdan Mazoure ◽  
Alexander Mazoure ◽  
Jocelyn Bédard ◽  
Vladimir Makarenkov
Keyword(s):  
Neural Networks ◽  
Skin Cancer ◽  
Skin Lesion ◽  
Cancer Detection ◽  
Web Application ◽  
Depth Analysis ◽  
Crucial Information ◽  
The Mean ◽  
Variance Estimates ◽  
Skin Cancer Detection

Abstract Recent years have seen a steep rise in the number of skin cancer detection applications. While modern advances in deep learning made possible reaching new heights in terms of classification accuracy, no publicly available skin cancer detection software provide confidence estimates for these predictions. We present DUNEScan (Deep Uncertainty Estimation for Skin Cancer), a web application that performs an intuitive in-depth analysis of uncertainty in commonly used skin cancer classification models based on convolutional neural networks (CNNs). DUNEScan allows users to upload a skin lesion image, and quickly compares the mean and the variance estimates provided by a number of new and traditional CNN models. Moreover, our web application uses the Grad-CAM and UMAP algorithms to visualize the classification manifold for the user’s input, hence providing crucial information about its closeness to skin lesion images taken from the popular ISIC database. DUNEScan is freely available at: https://www.dunescan.org.

scholarly journals DUNEScan: a web server for uncertainty estimation in skin cancer detection with deep neural networks

2022 ◽  
Vol 12 (1) ◽  
Author(s):  
Bogdan Mazoure ◽  
Alexander Mazoure ◽  
Jocelyn Bédard ◽  
Vladimir Makarenkov
Keyword(s):  
Neural Networks ◽  
Skin Cancer ◽  
Skin Lesion ◽  
Cancer Detection ◽  
Web Server ◽  
Uncertainty Estimation ◽  
Deep Uncertainty ◽  
Depth Analysis ◽  
Variance Estimates ◽  
Skin Cancer Detection

AbstractRecent years have seen a steep rise in the number of skin cancer detection applications. While modern advances in deep learning made possible reaching new heights in terms of classification accuracy, no publicly available skin cancer detection software provide confidence estimates for these predictions. We present DUNEScan (Deep Uncertainty Estimation for Skin Cancer), a web server that performs an intuitive in-depth analysis of uncertainty in commonly used skin cancer classification models based on convolutional neural networks (CNNs). DUNEScan allows users to upload a skin lesion image, and quickly compares the mean and the variance estimates provided by a number of new and traditional CNN models. Moreover, our web server uses the Grad-CAM and UMAP algorithms to visualize the classification manifold for the user’s input, hence providing crucial information about its closeness to skin lesion images  from the popular ISIC database. DUNEScan is freely available at: https://www.dunescan.org.

scholarly journals Skin Lesion Detection using Texture Based Segmentation and Classification by Convolutional Neural Networks (CNN)

2019 ◽  
Vol 9 (2) ◽  
pp. 2117-2120
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Skin Cancer ◽  
Skin Lesion ◽  
Lesion Detection ◽  
Technology Advancement ◽  
Classification Of Images ◽  
Occurrence Matrix ◽  
Skin Cancer Detection

Skin cancer is one of the dangerous cancers like breast cancer, brain tumour, and lung cancer. The detection of a skin lesion is melanoma or nonmelanoma is a very crucial issue. The earlier detection of melanoma is one of the best solutions for this issue. There is a various technique for detecting the skin lesion. Because of the technology advancement earlier detection of the skin lesion is possible. Malignant melanoma is a very harmful melanoma it is the cancerous cell that will lead to growth and that can be a mole in different colours red, black and brown. Skin lesion segmentation from dermoscopic images is a very challenging task nowadays because of the contrast of those images. there are various techniques for detecting the skin cancer base on the characteristics of the images shape, colour, textures. We proposed a system for skin cancer detection using texture-based segmentation and classification using Convolutional Neural Network. GLCM (Gray Level Co-occurrence Matrix) matrix is exacting the features from an image. And used Neural network tool for checking the accuracy of training network. Nowadays Deep Learning technique is very popular for classification of images. CNN is one of the techniques of Deep Learning. The proposed work will help in classification of skin lesion. Model will helpful for dermatologists for classifying melanomas.

scholarly journals Real-Time Mobile-Phone-Aided Melanoma Skin Lesion Detection Using Triangulation Technique

2020 ◽  
Vol 11 (3) ◽  
pp. 9-31
Author(s):  
Kumud Tiwari ◽  
Sachin Kumar ◽  
R. K. Tiwari
Keyword(s):  
Skin Cancer ◽  
Skin Lesion ◽  
Cancer Detection ◽  
Mobile Application ◽  
Laboratory Testing ◽  
Skin Lesions ◽  
Lesion Detection ◽  
Detection Accuracy ◽  
Skin Cancer Detection ◽  
Melanoma Skin

Melanoma is a harmful disease among all types of skin cancer. Genetic factors and the exposure of UV rays causes melanoma skin lesions. Early diagnosis is important to identify malignant melanomas to improve the patient prognosis. A biopsy is a traditional method which is painful and invasive when used for skin cancer detection. This method requires laboratory testing which is not very efficient and time-consuming to detect skin lesions. To solve the above issue, a computer aided diagnosis (CAD) for skin lesion detection is needed. In this article, we have developed a mobile application with the capabilities to segment skin lesions in dermoscopy images using a triangulation method and categorize them into malignant or bengin lesions through a supervised method which is convolution neural network (CNN). This mobile application will make the skin cancer detection non-invasive which does not require any laboratory testing, making the detection less time consuming and inexpensive with a detection accuracy of 81%.

Sign in / Sign up

Export Citation Format

Share Document