Noninvasive diagnosis of melanoma with tensor decomposition-based feature extraction from clinical color image

We propose a method for feature extraction from clinical color images, with application in classification of skin lesions. Proposed feature extraction method is based on tensor decomposition of the clinical color image of skin lesion. Since color image is naturally represented as a three- way tensor, it is reasonable to use multi-way techniques to capture the underlying information contained in the image. Extracted features are elements of the core tensor in the corresponding multi-way decomposition, and represent spatial- spectral profile of the lesion. In contrast to common methods that exploit either texture or spectral diversity of the tumor only, the proposed approach simultaneously captures spatial and spectral characteristics. The procedure is tested on a problem of noninvasive diagnosis of melanoma from the clinical color images of skin lesions, with overall sensitivity 82.1% and specificity 86.9%. Our method compares favorably with the state of the art results reported in the literature and provides an interesting alternative to the existing approaches.

[1]  Tian Han,et al.  Investigation of Nonlinearity in Hyperspectral Imagery Using Surrogate Data Methods , 2008, IEEE Transactions on Geoscience and Remote Sensing.

[2]  Aapo Hyvärinen,et al.  Validating the independent components of neuroimaging time series via clustering and visualization , 2004, NeuroImage.

[3]  P Altmeyer,et al.  Diagnostic and neural analysis of skin cancer (DANAOS). A multicentre study for collection and computer‐aided analysis of data from pigmented skin lesions using digital dermoscopy , 2003, The British journal of dermatology.

[4]  S. Kong,et al.  Hyperspectral Image Analysis for Skin Tumor Detection , 2009 .

[5]  Jason Weston,et al.  Gene Selection for Cancer Classification using Support Vector Machines , 2002, Machine Learning.

[6]  Thomas L. Diepgen,et al.  Dermatology online Atlas , 1996 .

[7]  Motoaki Kawanabe,et al.  Kernel-Based Nonlinear Blind Source Separation , 2003, Neural Computation.

[8]  Chris H. Q. Ding,et al.  Are Tensor Decomposition Solutions Unique? On the Global Convergence HOSVD and ParaFac Algorithms , 2009, PAKDD.

[9]  H. Peter Soyer,et al.  Scanning for melanoma , 2010 .

[10]  Bernhard Schölkopf,et al.  Learning with kernels , 2001 .

[11]  J. Kruskal Three-way arrays: rank and uniqueness of trilinear decompositions, with application to arithmetic complexity and statistics , 1977 .

[12]  José M. Bioucas-Dias,et al.  Does independent component analysis play a role in unmixing hyperspectral data? , 2005, IEEE Trans. Geosci. Remote. Sens..

[13]  Rafael García,et al.  Computerized analysis of pigmented skin lesions: A review , 2012, Artif. Intell. Medicine.

[14]  Tom Minka,et al.  Automatic Choice of Dimensionality for PCA , 2000, NIPS.

[15]  Tapani Ristaniemi,et al.  2011 Ieee International Workshop on Machine Learning for Signal Processing Determining the Number of Sources in High-density Eeg Recordings of Event-related Potentials by Model Order Selection , 2022 .

[16]  Andrzej Cichocki,et al.  Efficient method for Tucker3 model selection , 2009 .

[17]  Josef Smolle,et al.  EARLY DIAGNOSIS OF MALIGNANT MELANOMA BY SURFACE MICROSCOPY , 1987, The Lancet.

[18]  Andrzej Cichocki,et al.  Blind multispectral image decomposition by 3D nonnegative tensor factorization. , 2009, Optics letters.

[19]  Edmund R. Malinowski,et al.  Determination of the number of factors and the experimental error in a data matrix , 1977 .

[20]  J. Rissanen,et al.  Modeling By Shortest Data Description* , 1978, Autom..

[21]  Ralph Braun,et al.  The performance of SolarScan: an automated dermoscopy image analysis instrument for the diagnosis of primary melanoma. , 2005, Archives of dermatology.

[22]  J. Cavanaugh A large-sample model selection criterion based on Kullback's symmetric divergence , 1999 .

[23]  H P Soyer,et al.  Dermoscopy of pigmented skin lesions--a valuable tool for early diagnosis of melanoma. , 2001, The Lancet. Oncology.

[24]  Bernhard Schölkopf,et al.  Nonlinear Component Analysis as a Kernel Eigenvalue Problem , 1998, Neural Computation.

[25]  Edmund R. Malinowski,et al.  Theory of error in factor analysis , 1977 .

[26]  Aixia Guo,et al.  Gene Selection for Cancer Classification using Support Vector Machines , 2014 .

[27]  H. Akaike A new look at the statistical model identification , 1974 .

[28]  L. K. Hansen,et al.  Automatic relevance determination for multi‐way models , 2009 .

[29]  Joos Vandewalle,et al.  On the Best Rank-1 and Rank-(R1 , R2, ... , RN) Approximation of Higher-Order Tensors , 2000, SIAM J. Matrix Anal. Appl..

[30]  A. Cichocki,et al.  Tensor decompositions for feature extraction and classification of high dimensional datasets , 2010 .

[31]  Thomas M. Cover,et al.  Geometrical and Statistical Properties of Systems of Linear Inequalities with Applications in Pattern Recognition , 1965, IEEE Trans. Electron. Comput..

[32]  S. Menzies,et al.  Accuracy of computer diagnosis of melanoma: a quantitative meta-analysis. , 2003, Archives of dermatology.

[33]  G. Baudat,et al.  Kernel-based methods and function approximation , 2001, IJCNN'01. International Joint Conference on Neural Networks. Proceedings (Cat. No.01CH37222).

[34]  L. Tucker,et al.  Some mathematical notes on three-mode factor analysis , 1966, Psychometrika.

[35]  D. Chakrabarti,et al.  A fast fixed - point algorithm for independent component analysis , 1997 .

[36]  Aapo Hyvärinen,et al.  A Fast Fixed-Point Algorithm for Independent Component Analysis , 1997, Neural Computation.

[37]  A. Kopf,et al.  Thickness of malignant melanoma: global analysis of related factors. , 1987, The Journal of dermatologic surgery and oncology.

[38]  Joos Vandewalle,et al.  A Multilinear Singular Value Decomposition , 2000, SIAM J. Matrix Anal. Appl..

[39]  Anil K. Jain,et al.  Statistical Pattern Recognition: A Review , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[40]  M. Mihm,et al.  Automatic differentiation of melanoma from melanocytic nevi with multispectral digital dermoscopy: a feasibility study. , 2001, Journal of the American Academy of Dermatology.

[41]  Berkant Savas,et al.  A Newton-Grassmann Method for Computing the Best Multilinear Rank-(r1, r2, r3) Approximation of a Tensor , 2009, SIAM J. Matrix Anal. Appl..

[42]  A. Esteki,et al.  Independent Component Analysis as an Effective Tool for Automated Diagnosis of Melanoma , 2008, 2008 Cairo International Biomedical Engineering Conference.

[43]  Masafumi Hagiwara,et al.  An improved Internet-based melanoma screening system with dermatologist-like tumor area extraction algorithm , 2008, Comput. Medical Imaging Graph..

[44]  Yixin Wang,et al.  Melanoma-specific marker expression in skin biopsy tissues as a tool to facilitate melanoma diagnosis. , 2010, The Journal of investigative dermatology.

[45]  Tamara G. Kolda,et al.  Tensor Decompositions and Applications , 2009, SIAM Rev..

[46]  José M. Bioucas-Dias,et al.  Does independent component analysis play a role in unmixing hyperspectral data? , 2003, IEEE Transactions on Geoscience and Remote Sensing.

[47]  Y. Selen,et al.  Model-order selection: a review of information criterion rules , 2004, IEEE Signal Processing Magazine.

[48]  Harald Ganster,et al.  Automated Melanoma Recognition , 2001, IEEE Trans. Medical Imaging.

[49]  Kyuwan Choi,et al.  Detecting the Number of Clusters in n-Way Probabilistic Clustering , 2010, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[50]  Robert Tibshirani,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd Edition , 2001, Springer Series in Statistics.

[51]  A. Green,et al.  Computer image analysis of pigmented skin lesions , 1991, Melanoma research.

[52]  H. Kiers Towards a standardized notation and terminology in multiway analysis , 2000 .

[53]  Berkant Savas,et al.  Quasi-Newton Methods on Grassmannians and Multilinear Approximations of Tensors , 2009, SIAM J. Sci. Comput..

[54]  Tapani Ristaniemi,et al.  Fast and effective model order selection method to determine the number of sources in a linear transformation model , 2012, 2012 Proceedings of the 20th European Signal Processing Conference (EUSIPCO).

[55]  Andrzej Cichocki,et al.  Nonnegative Matrix and Tensor Factorization T , 2007 .

[56]  T. Schindewolf,et al.  Classification of melanocytic lesions with color and texture analysis using digital image processing. , 1993, Analytical and quantitative cytology and histology.