Melanoma detection algorithm based on feature fusion.

A Computer Aided-Diagnosis (CAD) System for melanoma diagnosis usually makes use of different types of features to characterize the lesions. The features are often combined into a single vector that can belong to a high dimensional space (early fusion). However, it is not clear if this is the optimal strategy and works on other fields have shown that early fusion has some limitations. In this work, we address this issue and investigate which is the best approach to combine different features comparing early and late fusion. Experiments carried on the datasets PH2 (single source) and EDRA (multi source) show that late fusion performs better, leading to classification scores of Sensitivity = 98% and Specificity = 90% (PH(2)) and Sensitivity = 83% and Specificity = 76% (EDRA).

A clinically oriented system for melanoma diagnosis using a color representation.

Computer Aided-Diagnosis (CAD) systems have been proposed to help dermatologists diagnose melanomas. However, these systems fail to provide a medical explanation for the diagnosis. This makes the dermatologists unsure about their use, since they are not easy to understand. In this paper we propose a CAD system that extracts a clinically inspired color description of the lesion and then, uses this information to discriminate melanomas from benign lesions. The proposed system is also capable of showing the extracted color features, making the system and its decisions more comprehensible for practitioners. The development of this system is hampered by the lack of a database of detailed annotate dermoscopy images. Nonetheless, we are able to tackle this issue using an image annotation framework based on the Correspondence-LDA algorithm. This method is applied with success to the identification of relevant colors in dermoscopy images, obtaining an average Precision of 84.9% and a Recall of 85.5%. The proposed color representation is then used to classify skin lesions, resulting in a Sensitivity of 78.9% and Specificity of 76.7%, these values are promising and comparable with the state-of-the art.

Unified approach for lesion border detection based on mixture modeling and local entropy thresholding.

BACKGROUND/PURPOSE: Computer-aided design (CAD) methods are highly valuable for the analysis of skin lesions using digital dermoscopy due to low rate of diagnostic accuracy of expert dermatologist. In computerized diagnostic methods, automatic border detection is the first and crucial step. METHOD: In this study, a novel unified approach is proposed for automatic border detection (ABD). A preprocessing step is performed by normalized smoothing filter (NSF) to reduce background noise. Mixture models technique is then utilized to initially segment the lesion area roughly. Afterward, local entropy thresholding is performed to extract the lesion candidate pixels and the lesion border is smoothed using morphological reconstruction. RESULTS: The overall ABD system is tested on a set of 100 dermoscopy images with ground truth. A comparative study was conducted with the other three state-of-the-art methods using statistical metrics. This ABD technique has the minimal average error probability rate of 5%, true detection of 92.10% and false positive rate of 6.41%. CONCLUSION: Results demonstrate that the proposed method segments the lesion area accurately. Sample dataset and execute software are available online and can be downloaded from: http://cs.ntu.edu.pk/research.

Melanoma recognition framework based on expert definition of ABCD for dermoscopic images.

BACKGROUND/PURPOSE: Melanoma Recognition based on clinical ABCD rule is widely used for clinical diagnosis of pigmented skin lesions in dermoscopy images. However, the current computer-aided diagnostic (CAD) systems for classification between malignant and nevus lesions using the ABCD criteria are imperfect due to use of ineffective computerized techniques. METHODS: In this study, a novel melanoma recognition system (MRS) is presented by focusing more on extracting features from the lesions using ABCD criteria. The complete MRS system consists of the following six major steps: transformation to the CIEL*a*b* color space, preprocessing to enhance the tumor region, black-frame and hair artifacts removal, tumor-area segmentation, quantification of feature using ABCD criteria and normalization, and finally feature selection and classification. RESULTS: The MRS system for melanoma-nevus lesions is tested on a total of 120 dermoscopic images. To test the performance of the MRS diagnostic classifier, the area under the receiver operating characteristics curve (AUC) is utilized. The proposed classifier achieved a sensitivity of 88.2%, specificity of 91.3%, and AUC of 0.880. CONCLUSIONS: The experimental results show that the proposed MRS system can accurately distinguish between malignant and benign lesions. The MRS technique is fully automatic and can easily integrate to an existing CAD system. To increase the classification accuracy of MRS, the CASH pattern recognition technique, visual inspection of dermatologist, contextual information from the patients, and the histopathological tests can be included to investigate the impact with this system.

Lesion border detection in dermoscopy images using ensembles of thresholding methods.

BACKGROUND: Dermoscopy is one of the major imaging modalities used in the diagnosis of melanoma and other pigmented skin lesions. Due to the difficulty and subjectivity of human interpretation, automated analysis of dermoscopy images has become an important research area. Border detection is often the first step in this analysis. In many cases, the lesion can be roughly separated from the background skin using a thresholding method applied to the blue channel. However, no single thresholding method appears to be robust enough to successfully handle the wide variety of dermoscopy images encountered in clinical practice. METHODS: In this article, we present an automated method for detecting lesion borders in dermoscopy images using ensembles of thres holding methods. CONCLUSION: Experiments on a difficult set of 90 images demonstrate that the proposed method is robust, fast, and accurate when compared to nine state-of-the-art methods.

A perceptually oriented method for contrast enhancement and segmentation of dermoscopy images.

BACKGROUND/PURPOSE: Dermoscopy images often suffer from low contrast caused by different light conditions, which reduces the accuracy of lesion border detection. Accordingly for lesion recognition, automatic melanoma border detection (MBD) is an initial as well as crucial task. METHOD: In this article, a novel perceptually oriented approach for MBD is presented by combing region and edge-based segmentation techniques. The MBD system for color contrast and segmentation improvement consists of four main steps: first, the RGB dermoscopy image is transformed to CIE L*a*b* color space, lesion contrast is then enhanced by adjusting and mapping the intensity values of the lesion pixels in the specified range using the three channels of CIE L*a*b*, a hill-climbing algorithm is used later to detect region-of-interest (ROI) map in a perceptually oriented color space using color channels (L*,a*,b*) and finally, an adaptive thresholding is applied to determine the optimal lesion border. Manually drawn borders obtained from an experienced dermatologist are utilized as a ground truth for performance evaluation. RESULTS: The proposed MBD method is tested on a total of 100 dermoscopy images. A comparative study with three state-of-the-art color and texture-based segmentation techniques (JSeg, dermatologists-like tumor area extraction: DTEA and region-based active contours: RAC), is also conducted to show the effectiveness of our MBD method using measures of true positive rate (TPR), false positive rate (FPR), and error probability (EP). Among different algorithms, our MBD algorithm achieved TPR of 94.25%, FPR of 3.56%, and EP of 4%. CONCLUSIONS: The proposed MBD approach is highly accurate to detect the lesion border area. The MBD software and sample of dermoscopy images can be downloaded at http://cs.ntu.edu.pk/research.php.

A feature-preserving hair removal algorithm for dermoscopy images.

BACKGROUND/PURPOSE: Accurate segmentation and repair of hair-occluded information from dermoscopy images are challenging tasks for computer-aided detection (CAD) of melanoma. Currently, many hair-restoration algorithms have been developed, but most of these fail to identify hairs accurately and their removal technique is slow and disturbs the lesion's pattern. METHODS: In this article, a novel hair-restoration algorithm is presented, which has a capability to preserve the skin lesion features such as color and texture and able to segment both dark and light hairs. Our algorithm is based on three major steps: the rough hairs are segmented using a matched filtering with first derivative of gaussian (MF-FDOG) with thresholding that generate strong responses for both dark and light hairs, refinement of hairs by morphological edge-based techniques, which are repaired through a fast marching inpainting method. Diagnostic accuracy (DA) and texture-quality measure (TQM) metrics are utilized based on dermatologist-drawn manual hair masks that were used as a ground truth to evaluate the performance of the system. RESULTS: The hair-restoration algorithm is tested on 100 dermoscopy images. The comparisons have been done among (i) linear interpolation, inpainting by (ii) non-linear partial differential equation (PDE), and (iii) exemplar-based repairing techniques. Among different hair detection and removal techniques, our proposed algorithm obtained the highest value of DA: 93.3% and TQM: 90%. CONCLUSION: The experimental results indicate that the proposed algorithm is highly accurate, robust and able to restore hair pixels without damaging the lesion texture. This method is fully automatic and can be easily integrated into a CAD system.

Unsupervised border detection in dermoscopy images.

BACKGROUND: As a result of the advances in skin imaging technology and the development of suitable image processing techniques, during the last decade, there has been a significant increase of interest in the computer-aided diagnosis of skin cancer. Automated border detection is one of the most important steps in this procedure as the accuracy of the subsequent steps crucially depends on the accuracy of this step. METHODS: In this article, we present an unsupervised approach to border detection in dermoscopy skin lesion images based on a modified version of the JSEG algorithm. RESULTS: The method is tested on a set of 100 dermoscopy images. The border detection error is quantified by a metric that uses manually determined borders from a dermatologist as the ground truth. The results are compared with three other automated methods and manually determined borders by a second dermatologist. CONCLUSION: The results demonstrate that the presented method achieves both fast and accurate border detection in dermoscopy images.