Grayscale medical image annotation using local relational features

Image annotation and classification are important areas where pattern recognition algorithms can be applied. In this article we report the insights that we have gained during our participation in the ImageCLEF medical annotation task during the years 2006 and 2007. Grayscale radiograph images taken from clinical routine had to be classified into one of the 116 base classes or labeled with attributes which described various properties of the radiograph. We present an algorithm based on local relational features which is robust with respect to illumination changes. It incorporates the geometric constellation of the feature points during the matching process and thus obtains superior performance. Furthermore, a hierarchical classification scheme is presented which reduces the computational complexity of the classifier.

[1]  Jing Liu,et al.  Medical Image Annotation and Retrieval Using Visual Features , 2006, CLEF.

[2]  Nicu Sebe,et al.  Wavelet-based salient points for image retrieval , 2000, Proceedings 2000 International Conference on Image Processing (Cat. No.00CH37101).

[3]  Hans Burkhardt,et al.  Image classification using cluster cooccurrence matrices of local relational features , 2006, MIR '06.

[4]  Hermann Ney,et al.  Deformations, patches, and discriminative models for automatic annotation of medical radiographs , 2008, Pattern Recognit. Lett..

[5]  Michael Kohnen,et al.  The IRMA code for unique classification of medical images , 2003, SPIE Medical Imaging.

[6]  David G. Lowe,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004, International Journal of Computer Vision.

[7]  Matti Pietikäinen,et al.  Multiresolution Gray-Scale and Rotation Invariant Texture Classification with Local Binary Patterns , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[8]  Robert M. Haralick,et al.  Textural Features for Image Classification , 1973, IEEE Trans. Syst. Man Cybern..

[9]  Gabriela Csurka,et al.  Visual categorization with bags of keypoints , 2002, eccv 2004.

[10]  Pietro Perona,et al.  Object class recognition by unsupervised scale-invariant learning , 2003, 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings..

[11]  Francesca Odone,et al.  Image Kernels , 2002, SVM.

[12]  Barbara Caputo,et al.  Discriminative cue integration for medical image annotation , 2008, Pattern Recognit. Lett..

[13]  Patrick Haffner,et al.  Support vector machines for histogram-based image classification , 1999, IEEE Trans. Neural Networks.

[14]  Eugene Kim,et al.  Overview of the ImageCLEFmed 2006 Medical Retrieval and Medical Annotation Tasks , 2006, CLEF.

[15]  Cordelia Schmid,et al.  A Performance Evaluation of Local Descriptors , 2005, IEEE Trans. Pattern Anal. Mach. Intell..

[16]  Hans Burkhardt,et al.  Learning Taxonomies in Large Image Databases , 2007 .

[17]  Stéfan Jacques Darmoni,et al.  MedIC/CISMeF at ImageCLEF 2006: Image Annotation and Retrieval Tasks , 2006, CLEF.

[18]  Michael Kohnen,et al.  Quality of DICOM header information for image categorization , 2002, SPIE Medical Imaging.

[19]  Nicu Sebe,et al.  Comparing salient point detectors , 2001, IEEE International Conference on Multimedia and Expo, 2001. ICME 2001..

[20]  Hermann Ney,et al.  Sparse Patch-Histograms for Object Classification in Cluttered Images , 2006, DAGM-Symposium.

[21]  Nicu Sebe,et al.  Fast spatial pattern discovery integrating boosting with constellations of contextual descriptors , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[22]  Thomas Martin Deserno,et al.  Automatic medical image annotation in ImageCLEF 2007: Overview, results, and discussion , 2008, Pattern Recognit. Lett..