Segmentation of proximal femur in digital radiographic image using principal component model

In this paper we propose a segmentation of femur bone based on principal component model (PCM). The PCM is an analysis of principal component along with shape and appearance model of an object. Principal Component Analysis (PCA) is mostly used as a tool in exploratory data analysis and for making predictive models. An active shape model segmentation scheme is presented that is steered by optimal local features in the original formulation. A nonlinear NN-classifier is used, instead of the linear Mahalanobis distance, to find optimal displacements for landmarks. For each of the landmarks that describe the shape, at each resolution level taken into account during the segmentation procedure, a distinct set of optimal features is determined. The selection of features is automatic, using the training images and sequential feature forward and backward selection. The PCM based femur bone segmentation approach was tested on right proximal femur digital X-ray images obtained from 50 [n=50, age ± SD= 50.12 ± 13.7 years] Indian women and has produced 74% and 70% of sensitivity and specificity respectively.

[1]  Alan L. Yuille,et al.  Feature extraction from faces using deformable templates , 2004, International Journal of Computer Vision.

[2]  Guido Gerig,et al.  Elastic model-based segmentation of 3-D neuroradiological data sets , 1999, IEEE Transactions on Medical Imaging.

[3]  Timothy F. Cootes,et al.  Active Appearance Models , 1998, ECCV.

[4]  Milan Sonka,et al.  Object localization and border detection criteria design in edge-based image segmentation: automated learning from examples , 2000, IEEE Transactions on Medical Imaging.

[5]  Marko Subasic,et al.  Level Set Methods and Fast Marching Methods , 2003 .

[6]  James Graham,et al.  Improving the detection of osteoporosis from dental radiographs using Active Appearance Models , 2010, 2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro.

[7]  Marios S. Pattichis,et al.  Pre-classification of chest radiographs for improved active shape model segmentation of ribs , 2002, Proceedings Fifth IEEE Southwest Symposium on Image Analysis and Interpretation.

[8]  Anil K. Jain,et al.  Object Matching Using Deformable Templates , 1996, IEEE Trans. Pattern Anal. Mach. Intell..

[9]  Timothy F. Cootes,et al.  Active Shape Models-Their Training and Application , 1995, Comput. Vis. Image Underst..

[10]  Paul A. Yushkevich,et al.  Segmentation, registration, and measurement of shape variation via image object shape , 1999, IEEE Transactions on Medical Imaging.

[11]  Demetri Terzopoulos,et al.  Deformable models in medical image analysis: a survey , 1996, Medical Image Anal..

[12]  Stiliyan Kalitzin,et al.  Computer Assisted Human Follicle Analysis for Fertility Prospects with 3D Ultrasound , 1999, IPMI.

[13]  Horst Bischof,et al.  Fast Active Appearance Model Search Using Canonical Correlation Analysis , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[14]  Demetri Terzopoulos,et al.  Snakes: Active contour models , 2004, International Journal of Computer Vision.

[15]  Alejandro F. Frangi,et al.  Active shape model segmentation with optimal features , 2002, IEEE Transactions on Medical Imaging.

[16]  Rémi Ronfard,et al.  Region-based strategies for active contour models , 1994, International Journal of Computer Vision.

[17]  James A. Sethian,et al.  Level Set Methods and Fast Marching Methods , 1999 .