Support Vector Machine Classification Based on Correlation Prototypes Applied to Bone Age Assessment

Bone age assessment (BAA) on hand radiographs is a frequent and time-consuming task in radiology. We present a method for (semi)automatic BAA which is done in several steps: 1) extract 14 epiphyseal regions from the radiographs; 2) for each region, retain image features using the image retrieval in medical application framework; 3) use these features to build a classifier model (training phase); 4) evaluate performance on cross-validation schemes (testing phase); 5) classify unknown hand images (application phase). In this paper, we combine a support vector machine (SVM) with cross correlation to a prototype image for each class. These prototypes are obtained choosing one random hand per class. A systematic evaluation is presented comparing nominal- and real-valued SVM with k nearest neighbor classification on 1097 hand radiographs of 30 diagnostic classes (0-19 years). Mean error in age prediction is 1.0 and 0.83 years for 5-NN and SVM, respectively. Accuracy of nominal- and real-valued SVM based on six prominent regions (prototypes) is 91.57% and 96.16%, respectively, for accepting about two years age range.

[1]  Sven Kreiborg,et al.  The BoneXpert Method for Automated Determination of Skeletal Maturity , 2009, IEEE Transactions on Medical Imaging.

[2]  Tai-Lang Jong,et al.  A Fully Automatic Computerized Bone Age Assessment Procedure Based on Phalange Ossification Analysis , 2003 .

[3]  L. Morris.  Assessment of Skeletal Maturity and Prediction of Adult Height (TW3 Method) , 2003 .

[4]  Ma Preece,et al.  Growth and bone age , 1997 .

[5]  H. K. Huang,et al.  Computer-assisted bone age assessment: image preprocessing and epiphyseal/metaphyseal ROI extraction , 2001, IEEE Transactions on Medical Imaging.

[6]  Thomas Martin Deserno,et al.  A generic concept for the implementation of medical image retrieval systems , 2007, Int. J. Medical Informatics.

[7]  Whoi-Yul Kim,et al.  Computerized Bone Age Assessment Using DCT and LDA , 2007, MIRAGE.

[8]  H. K. Huang,et al.  Digital hand atlas for web-based bone age assessment: system design and implementation , 2000, Medical Imaging.

[9]  L. Bocchi,et al.  An artificial neural network architecture for skeletal age assessment , 2003, Proceedings 2003 International Conference on Image Processing (Cat. No.03CH37429).

[10]  I.W. Ricketts,et al.  Classification of hand bones for bone age assessment , 1996, Proceedings of Third International Conference on Electronics, Circuits, and Systems.

[11]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[12]  Thomas Martin Deserno,et al.  A Generic Concept for the Implementation of Medical Image Retrieval Systems , 2005, MIE.

[13]  Yi-Hong Chou,et al.  Computerized geometric features of carpal bone for bone age estimation. , 2007, Chinese medical journal.

[14]  Thomas Martin Deserno,et al.  Segmentation of medical images combining local, regional, global, and hierarchical distances into a bottom-up region merging scheme , 2005, SPIE Medical Imaging.

[15]  Chih-Jen Lin,et al.  LIBSVM: A library for support vector machines , 2011, TIST.

[16]  Carlos Alberola-López,et al.  A computational TW3 classifier for skeletal maturity assessment. A Computing with Words approach , 2004, J. Biomed. Informatics.

[17]  Thomas Martin Deserno,et al.  A classification framework for content-based extraction of biomedical objects from hierarchically decomposed images , 2006, SPIE Medical Imaging.

[18]  H. K. Huang,et al.  Bone age assessment of children using a digital hand atlas , 2007, Comput. Medical Imaging Graph..

[19]  Marcos Martín-Fernández,et al.  Automatic bone age assessment: a registration approach , 2003, SPIE Medical Imaging.

[20]  A. Oestreich.  Hand Bone Age: A Digital Atlas of Skeletal Maturity , 2005 .

[21]  Thomas M. Deserno,et al.  Scene analysis with structural prototypes for content-based image retrieval in medicine , 2008, SPIE Medical Imaging.

[22]  Chih-Jen Lin,et al.  A Practical Guide to Support Vector Classication , 2008 .

[23]  Thomas Martin Deserno,et al.  Web-based bone age assessment by content-based image retrieval for case-based reasoning , 2012, International Journal of Computer Assisted Radiology and Surgery.

[24]  Chih-Jen Lin,et al.  A comparison of methods for multiclass support vector machines , 2002, IEEE Trans. Neural Networks.

[25]  Thomas M. Deserno,et al.  Bone age assessment by content-based image retrieval and case-based reasoning , 2011, Medical Imaging.

[26]  Arnaldo de Albuquerque Araújo,et al.  MammoSys: A content-based image retrieval system using breast density patterns , 2010, Comput. Methods Programs Biomed..

[27]  Chih-Jen Lin,et al.  A Comparison of Methods for Multi-class Support Vector Machines , 2015 .

[28]  H. H. Thodberg,et al.  Validation and reference values of automated bone age determination for four ethnicities. , 2010, Academic radiology.

[29]  T M Lehmann,et al.  Content-based Image Retrieval in Medical Applications , 2004, Methods of Information in Medicine.

[30]  E Pietka,et al.  Digital hand atlas and web-based bone age assessment: system design and implementation. , 2000, Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society.