Skin Lesion Segmentation Using Clustering Techniques

Cluster analysis has been widely used in various disciplines such as pattern recognition, computer vision, and data mining. In this work we investigate the applicability of two spatial clustering algorithms, namely DBSCAN and STING, to a new problem domain: Color segmentation of skin lesion (tumor) images. Automated segmentation is a key step in the computerized analysis of skin lesion images since the accuracy of the subsequent steps (feature extraction, classification, etc.) crucially depends on the accuracy of this very first step. In this paper, we develop two unsupervised methods for segmentation of skin lesion images: one based on DBSCAN clustering algorithm and the other based on STING clustering algorithm. Experiments on a database of over hundred skin lesion images show that DBSCAN-based segmentation algorithm performs significantly better than the STING-based one.

[1]  Theodore Johnson,et al.  Exploratory Data Mining and Data Cleaning , 2003 .

[2]  Hans-Peter Kriegel,et al.  A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise , 1996, KDD.

[3]  P. Schmid.  Segmentation of digitized dermatoscopic images by two-dimensional color clustering , 1999, IEEE Transactions on Medical Imaging.

[4]  Jacob Cohen.  A Coefficient of Agreement for Nominal Scales , 1960 .

[5]  Hans-Peter Kriegel,et al.  Density-Based Clustering in Spatial Databases: The Algorithm GDBSCAN and Its Applications , 1998, Data Mining and Knowledge Discovery.

[6]  T Lee,et al.  Dullrazor®: A software approach to hair removal from images , 1997, Comput. Biol. Medicine.

[7]  Hiroshi Motoda,et al.  Feature Extraction, Construction and Selection: A Data Mining Perspective , 1998 .

[8]  M S Woolfson,et al.  Application of region-based segmentation and neural network edge detection to skin lesions. , 2004, Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society.

[9]  W.V. Stoecker,et al.  Boundary detection in skin tumor images: An overall approach and a radial search algorithm , 1990, Pattern Recognit..

[10]  Anil K. Jain,et al.  Data clustering: a review , 1999, CSUR.

[11]  Emmanuel Ifeachor,et al.  Computer aided lesion border detection applied to macroscopic images , 2001 .

[12]  Jiong Yang,et al.  STING: A Statistical Information Grid Approach to Spatial Data Mining , 1997, VLDB.

[13]  Guillermo Sapiro,et al.  Segmenting skin lesions with partial-differential-equations-based image processing algorithms , 2000, IEEE Transactions on Medical Imaging.

[14]  Anil K. Jain,et al.  Algorithms for Clustering Data , 1988 .

[15]  Guillermo Sapiro,et al.  Segmenting skin lesions with partial-differential-equations-based image processing algorithms , 2000 .

[16]  Clement T. Yu,et al.  Segmentation of skin cancer images , 1999, Image Vis. Comput..