Background/aims: Epiluminescence microscopy (ELM), also known as dermoscopy or dermatoscopy, is a non-invasive, in vivo technique, that permits visualization of features of pigmented melanocytic neoplasms that are not discernable by examination with the naked eye. ELM offers a completely new range of visual features. One such feature is the solid pigment, also called the blotchy pigment or dark structureless area. Our goal was to automatically detect this feature and determine whether its presence is useful in distinguishing benign from malignant pigmented lesions.
Methods: Here, a texture-based algorithm is developed for the detection of solid pigment. The factors d and a used in calculating neighboring gray level dependence matrix (NGLDM) numbers were chosen as optimum by experimentation. The algorithms are tested on a set of 37 images. A new index is presented for separation of benign and malignant lesions, based on the presence of solid pigment in the periphery.
Results: The NGLDM large number emphasis N2 was satisfactory for the detection of the solid pigment. Nine lesions had solid pigment detected, and among our 37 lesions, no melanoma lacked solid pigment. The index for separation of benign and malignant lesions was applied to the nine lesions. We were able to separate the benign lesions with solid pigment from the malignant lesions with the exception of only one lesion, a Spitz nevus that mimicked a malignant melanoma.
Conclusion: Texture methods may be useful in detecting important dermatoscopy features in digitized images and a new index may be useful in separating benign from malignant lesions. Testing on a larger set of lesions is needed before further conclusions can be made.
M. Anantha et al., "Detection of Solid Pigment in Dermatoscopy Images using Texture Analysis," Skin Research and Technology, vol. 6, no. 4, pp. 193-198, John Wiley & Sons, Nov 2000.
The definitive version is available at http://dx.doi.org/10.1034/j.1600-0846.2000.006004193.x
Electrical and Computer Engineering
National Institutes of Health (U.S.). Small Business Innovation Research Program
Keywords and Phrases
Pigment; Algorithm; Article; Benign Tumor; Clinical Feature; Controlled Study; Data Analysis; Diagnostic Accuracy; Diagnostic Procedure; Epiluminescence Microscopy; Human; Human Tissue; Image Analysis; Imaging System; Juvenile Melanoma; Mathematical Computing; Melanocyte; Melanoma; Skin Cancer; Skin Defect; Skin Examination; Skin Pigmentation; Skin Surface; Skin Tumor; Surface Property; Tumor Classification; Visual Discrimination; Dermatoscopy; Solid Pigment; Texture
International Standard Serial Number (ISSN)
Article - Journal
© 2000 John Wiley & Sons, All rights reserved.