Texture Analysis for Dermoscopic Image Processing (original) (raw)
Related papers
IOP Conference Series: Materials Science and Engineering
One of the damaging diseases among people in the world is skin cancer. Skin cancer leftovers an important scientific, clinical and public task. Swarm intelligence techniques (SITs) are new, improved and modern methods for optimization algorithms. Failure of detection in skin cancer images can be seen in SITs. This work proposes an efficient image and examines for some samples in this disease. The study presents a simple technique for a pre-processing and an automatic detection of SITs to make the needed analysis. This paper estimated all these various models using the PH 2 , Dermis, ISIC (2016 ,2017, 2018) segmentation challenge dataset. The input images are improved for better processing than, the lesion sampling is segmented from the improved image by using Otsu thresholding and median filter operations. In the earlier studies, skin cancer is analyzed by means of several optimization algorithms. Now, the outcomes of the above algorithms were compared with the dice coefficient and it was demonstrated that the value of 97.35% which is nearer to manual segmentation. The accuracy the value of 98.58% when used for solving the same problem. To this end, a somewhat comprehensive analysis was showed to compare the effectiveness of many parameters' combinations.
2020
In recent years, the incidence of skin lesions has been one of the most rapidly increasing of all commonly occurring cancers. This deadliest form of melanoma must be detected early to be effectively treated. Because of the trouble and objectivity of human clarification, a significant research field has developed around the computerized examination of dermoscopy images. One reason to apply swarm intelligence systems is that an optimal solution can be advanced with a sensible computational application. This work introduces an artificial bee colony technique (ABC), distinctions, and applications. The planned ABC is a more suitable algorithm and one that requires smaller amounts of factors that need to be adjusted in comparison to other modern artificial swarm intelligence techniques (MASITs) for distinguishing unhealthy in skin tumor lesions. In these swarm's intelligence optimization algorithms have been positively executed for melanoma problems and provided extraordinary results ...
International Journal of Electrical and Computer Engineering (IJECE), 2020
Methods of image processing can recognize the images of melanoma lesions border in addition to the disease compared to a skilled dermatologist. New swarm intelligence technique depends on meta-heuristic that is industrialized to resolve composite real problems which are problematic to explain by the available deterministic approaches. For an accurate detection of all segmentation and classification of skin lesions, some dealings should be measured which contain, contrast broadening, irregularity quantity, choice of most optimal features, and so into the world. The price essential for the action of progressive disease cases is identical high and the survival percentage is low. Many electronic dermoscopy classifications are advanced depend on the grouping of form, surface and dye features to facilitate premature analysis of malignance. To overcome this problematic, an effective prototypical for accurate boundary detection and arrangement is obtainable. The projected classical recovers...
Discrimination between Healthy and Unhealthy Mole Lesions using Artificial Swarm Intelligence
IOP Conference Series: Materials Science and Engineering
In recent years, occurrence rates of skin melanoma have shown a rapid increase, resulting in enhancements to death rates. Based on the difficulty and subjectivity of human clarification, computer examination of dermoscopy images has thus developed into a significant research field in this area. One the reasons for applying heuristic methods is that good solutions can be developed with only reasonable computational exertion. This paper thus presents an artificial swarm intelligence method with variations and suggestions. The proposed artificial bee colony (ABC) is a more suitable algorithm in comparison to other algorithms for detecting melanoma in the skin tumour lesions, being flexible, fast, and simple, and requiring fewer adjustments. These is characteristics are recognized assisting dermatologists to detect malignant melanoma (MM) at the lowest time and effort cost. Automatic classification of skin cancers by using segmenting the lesion's regions and selecting of the ABC technique for the values of the characteristic principles allows. Information to be fed into several well-known algorithms to obtain skin cancer categorization: in terms of whether the lesion is suspicious, malignant, benign (healthy and unhealthy nevi). This segmentation approach can further be utilized to develop handling and preventive approaches, thus decreasing the danger of skin cancer lesions. One of the most significant stages in dermoscopy image examination is the segmentation of the melanoma. Here, various PH2 dataset image were utilized along with their masks to estimate the accuracy, sensitivity, and specificity of various segmentation techniques. The results show that a modified automatic based on ABC images have the highest accuracy and specificity compares with the other algorithms. The results show that a modified automatic based on ABC images displayed the highest accuracy and specificity in such testing.
Diagnosis of Skin Lesions Based on Dermoscopic Images Using Image Processing Techniques
International Journal of Signal Processing, Image Processing and Pattern Recognition, 2016
Great effort has been put into the development of diagnosis methods for the most dangerous type of skin diseases-melanoma. This paper aims to develop a prototype capable of segment and classify skin lesions in dermoscopy images based on ABCD rule. The proposed work is divided into four distinct stages: (1) preprocessing, consists of filtering and contrast enhancing techniques, (2) segmentation, thresholding, and statistical properties are computed to localize the lesion, (3) features extraction, asymmetry is calculated by averaging the calculated results of the two methods: entropy and bi-fold. Border irregularity is calculated by accumulate the statistical scores of the eight segments of the segmented lesion. Color feature is calculated among the existence of six candidate colors: white, black, red, light-brown, dark-brown, and blue-gray. Diameter is measured by the conversion operation from the total number of pixels in the greatest diameter into millimeter (mm), and (4) classification, the summation of the four extracted feature scores multiplied by their weights to yield a total dermoscopy score (TDS); hence, the lesion is classified into benign, suspicious, or malignant. The prototype is implemented in MATLAB and the dataset used consists of 200 dermoscopic images from Hospital Pedro Hispano, Matosinhos. The achieved results show an acceptable performance rates, an accuracy 90%, sensitivity 85%, and specificity 92.22%.
Intelligent Engineering Systems through Artificial Neural Networks, Volume 20, 2010
In this paper, an unsupervised approach based on Evolving Vector Quantization (EVQ) is presented for enhancing dermatology images for skin lesion segmentation. Vector Quantization (VQ) as a famous compression technique has been widely used in image signal compression and speech signal compression. The EVQ algorithm extends the Linde, Buzo, and Gray (LBG) Vector Quantization method with Particle Swarm Optimization to cluster the pixels inside the image based on merging similar gray value pixels. The proposed enhancement technique is evaluated using 100 dermoscopy skin lesion images for skin lesion segmentation. The EVQ algorithm is applied to the individual color planes red, green, and blue, respectively. Segmentation results using these three planes are compared and scored based on manual borders obtained from three dermatologists. In addition, Differential Equation-based Particle Swarm Optimization is implemented and their results are compared with the standard PSO.
Detection of Pigment Networks in Dermoscopy Images
Journal of Physics: Conference Series
One of the most important structures in dermoscopy images is the pigment network, which is also one of the most challenging and fundamental task for dermatologists in early detection of melanoma. This paper presents an automatic system to detect pigment network from dermoscopy images. The design of the proposed algorithm consists of four stages. First, a preprocessing algorithm is carried out in order to remove the noise and improve the quality of the image. Second, a bank of directional filters and morphological connected component analysis are applied to detect the pigment networks. Third, features are extracted from the detected image, which can be used in the subsequent stage. Fourth, the classification process is performed by applying feed-forward neural network, in order to classify the region as either normal or abnormal skin. The method was tested on a dataset of 200 dermoscopy images from Hospital Pedro Hispano (Matosinhos), and better results were produced compared to previous studies.
Skin Cancer Detection in Dermoscopy Images Using Sub-Region Features
Lecture Notes in Computer Science, 2017
In the medical field, the identification of skin cancer (Malignant Melanoma) in dermoscopy images is still a challenging task for radiologists and researchers. Due to its rapid increase, the need for decision support systems to assist the radiologists to detect it in early stages becomes essential and necessary. Computer Aided Diagnosis (CAD) systems have significant potential to increase the accuracy of its early detection. Typically, CAD systems use various types of features to characterize skin lesions. The features are often concatenated into one vector (early fusion) to represent the image. In this paper, we present a novel method for melanoma detection from images. First the lesions are segmented by combining Particle Swarm Optimization and Markov Random Field methods. Then the K-means is applied on the segmented lesions to separate them into homogeneous clusters, from which important features are extracted. Finally, an Artificial Neural Network with Radial Basis Function is applied for the detection of melanoma. The method was tested on 200 dermoscopy images. The experimental results show that the proposed method achieved higher accuracy in terms of melanoma detection, compared to alternative methods.
Automatic Detection of Melanoma Skin Cancer using Texture Analysis
International Journal of Computer Applications, 2012
Melanoma is considered the most dangerous type of skin cancer. Early and accurate diagnosis depends mainly on important issues, accuracy of feature extracted and efficiency of classifier method. This paper presents an automated method for melanoma diagnosis applied on a set of dermoscopy images. Features extracted are based on gray level Co-occurrence matrix (GLCM) and Using Multilayer perceptron classifier (MLP) to classify between Melanocytic Nevi and Malignant melanoma. MLP classifier was proposed with two different techniques in training and testing process: Automatic MLP and Traditional MLP. Results indicated that texture analysis is a useful method for discrimination of melanocytic skin tumors with high accuracy. The first technique, Automatic iteration counter is faster but the second one, Default iteration counter gives a better accuracy, which is 100 % for the training set and 92 % for the test set.
An efficient of estimation stages for segmentation skin lesions based optimization algorithm
International Journal of Electrical and Computer Engineering (IJECE), 2021
Modern dermatology distinguishes premature diagnosis for example an important part in reducing the death percentage and promising less aggressive treatment for patients. The classifications comprise various stages that must be selected suitably using the characteristics of the filter pointing to get a dependable analysis. The dermoscopic images hold challenges to be faced and overcome to enhance the automatic diagnosis of hazardous lesions. It is calculated to survey a different metaheuristic and evolutionary computing working for filter design systems. Approximately general computing techniques are observed to improve features of infect design method. Nevertheless, the median filter (MF) is normally multimodal with respect to the filter factors and so, reliable approaches that can provide optimal solutions are required. The design of MF depends on modern artificial swarm intelligence technique (MASIT) optimization algorithm which has proven to be more effective than other population-based algorithms to improve of estimation stages for segmentation skin lesions. A controlled artificial bee colony (ABC) algorithm is advanced for solving factors optimization problems and, also the physical-programming-depend on ABC way is applied to proposal median filter, and the outcomes are compared to another approaches.