Automated colour identification in melanocytic lesions

Colour information plays an important role in classifying skin lesion. However, colour identification by dermatologists can be very subjective, leading to cases of misdiagnosis. Therefore, a computer-assisted system for quantitative colour identification is highly desirable for dermatologists to use. Although numerous colour detection systems have been developed, few studies have focused on imitating the human visual perception of colours in melanoma application. In this paper we propose a new methodology based on QuadTree decomposition technique for automatic colour identification in dermoscopy images. Our approach mimics the human perception of lesion colours. The proposed method is trained on a set of 47 images from NIH dataset and applied to a test set of 190 skin lesions obtained from PH2 dataset. The results of our proposed method are compared with a recently reported colour identification method using the same dataset. The effectiveness of our method in detecting colours in dermoscopy images is vindicated by obtaining approximately 93% accuracy when the CIELab1 colour space is used.

[1]  Sumeet Dua,et al.  Region Quad-Tree Decomposition Based Edge Detection for Medical Images , 2010, The open medical informatics journal.

[2]  Shih-Fu Chang,et al.  Quad-tree segmentation for texture-based image query , 1994, MULTIMEDIA '94.

[3]  Gerald Schaefer,et al.  Lesion border detection in dermoscopy images , 2009, Comput. Medical Imaging Graph..

[4]  D. Mery,et al.  Color measurement in L ¿ a ¿ b ¿ units from RGB digital images , 2006 .

[5]  R. H. Moss,et al.  Skin lesion classification using relative color features , 2007, Skin research and technology : official journal of International Society for Bioengineering and the Skin (ISBS) [and] International Society for Digital Imaging of Skin (ISDIS) [and] International Society for Skin Imaging.

[6]  Randy H. Moss,et al.  The Median Split Algorithm for Detection of Critical Melanoma Color Features , 2013, VISAPP.

[7]  Amjad Rehman,et al.  Improved quadtree image segmentation approach to region information , 2014 .

[8]  E Epstein.  The use of the dermatoscope to identify early melanoma using the three‐colour test , 2002, The British journal of dermatology.

[9]  R. H. Moss,et al.  Colour histogram analysis for melanoma discrimination in clinical images , 2003, Skin research and technology : official journal of International Society for Bioengineering and the Skin (ISBS) [and] International Society for Digital Imaging of Skin (ISDIS) [and] International Society for Skin Imaging.

[10]  Mohammad Aldeen,et al.  Border detection in dermoscopy images using hybrid thresholding on optimized color channels , 2011, Comput. Medical Imaging Graph..

[11]  Ashraf A. Kassim,et al.  Hierarchical Segmentation-Based Image Coding Using Hybrid Quad-Binary Trees , 2009, IEEE Transactions on Image Processing.

[12]  B. Thiers,et al.  The CASH (color, architecture, symmetry, and homogeneity) algorithm for dermoscopy , 2008 .

[13]  S. Menzies,et al.  A sensitivity and specificity analysis of the surface microscopy features of invasive melanoma , 1996, Melanoma research.

[14]  A. M. Arymurthy,et al.  Color and Texture Feature for Remote Sensing - Image Retrieval System: A Comparative Study , 2011 .

[15]  Costantino Grana,et al.  Computer description of colours in dermoscopic melanocytic lesion images reproducing clinical assessment , 2003 .

[16]  H. Kittler,et al.  Epiluminescence microscopy-based classification of pigmented skin lesions using computerized image analysis and an artificial neural network , 1998, Melanoma research.

[17]  Jordi Vitrià,et al.  Using an ICA representation of high dimensional data for object recognition and classification , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[18]  Pedro M. Ferreira,et al.  PH2 - A dermoscopic image database for research and benchmarking , 2013, 2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[19]  Mário A. T. Figueiredo,et al.  Color identification in dermoscopy images using Gaussian mixture models , 2014, 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[20]  W. Stolz,et al.  The ABCD rule of dermatoscopy. High prospective value in the diagnosis of doubtful melanocytic skin lesions. , 1994, Journal of the American Academy of Dermatology.

[21]  James Bailey,et al.  Computer-Aided Diagnosis of Melanoma Using Border- and Wavelet-Based Texture Analysis , 2012, IEEE Transactions on Information Technology in Biomedicine.