Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/19735
標題: 掌骨X光影像電腦化骨齡群集估測系統之研究
The Study of Computerized Bone Age Cluster Assessment System Based on Hand Radiographs
作者: 林秀霞
Lin, Hsiu-Hsia
關鍵字: 骨骺/幹骺特徵區域;feature extraction;骨齡群集估測;模糊類神經網路;骨骼發展階段;先驗知識;epiphyseal/metaphyseal regions of interest;bone age cluster assessment;fuzzy neural network;a priori knowledge;skeletal development stages
出版社: 資訊科學與工程學系所
引用: [1] D. B Darling, Radiography of Infants and Children, 1st ed. Springfield, IL: Charles C Thomas, Ch.6, 1979, pp.370-372. [2] F.E. Johnston and S.B. Jahina, The contribution of the carpal bones to the assessment of skeletal age Amer. J. Phys. Anthrop. Vol. 23, No. 4, 1965, pp.349-354. [3] D.R. Kirks, Practical Pediatric Imaging. Diagnostic Radiology of Infants and Children, 1st ed. Boston, MA: Little, Brown & Co., 1984. [4] K.S. Pospiech, A. Gertych, E. Pietka, F. Cao and H.K. Huang, Wavelet decomposition based features in description of epiphyseal fusion, Analysis of Biomedical Signals and Images, Proceedings of 15th Euro conference Biosignal, Brn ,2000, pp.246-248. [5] K.S. Prasanna and A. Gurdial, Image thresholding using two-dimensional Tsallis-Havrda-Charvát entropy, Pattern recogn. lett., Vol.27, No. 6, 2006, pp.520-528. [6] W.W. Greulich and S.I. Pyle, Radiographic Atlas of Skeletal Development of Hand Wrist (ed2), Stanford, CA: Standford University Press ,1971. [7] D.G. King, D.M. Steventon, M.P. O'Sullivan, A.M. Cook, V.P. Hornsby and I.G.. Jefferson, Reproducibility of bone ages when performed by radiology registrars: An audit of Tanner and Whitehouse II versus Greulich and Pyle methods, Br. J. Radiol. Vol .67, No. 8, 1994, pp.848-851. [8] J.M. Tanner and R.H. Whitehouse, Assessment of skeletal maturityand prediction of adult height (TW2 Method) London: Academic Press.,1975. [9] A.F. Roch, C.G. Rochman and G.H.Davila, Effect of training of replicability of assessment of skeletal maturity (Greulich-Pyle), Amer. J. Roentgenol. Vol.108 , 1970, pp.511-515. [10] H. Kwon, S.Z. Der and N.M.Nasrabadi, An adaptive segmentation algorithm using iterative local feature extraction for hyperspectral imagery, IEEE Int. Conf. on Image Processing Vol.1, 2001, pp.74-77. [11] E. Pietka, S.P. Kurkowska, A. Gertych and F. Cao, Integration of computer assisted bone age assessment with clinical PACS, Comput. Med. Imag. Grap. Vo.27, 2003, pp. 217-228. [12] E. Pietka, A. Gertych, S. Pospiech, H.K. Huang and F. Cao, Computer assisted bone age assessment: image pre-processing and ROI extraction, IEEE Trans. Medical on Imaging, Vol.20, No.8, 2001, pp.715-729. [13] D.J. Michael and A.C. Nelson, HANDX, A model-based system for automatic segmentation of bones from digital hand radiographs, IEEE Transactions on Medical Imaging, Vol.8, No.1 ,1989, pp.64-69. [14] G..K. Manos, A.Y. Cairns and I.W. Rickets, D. Sinclair, Segmenting radiographs of the hand and wrist, Computer Methods and Programs in Biomedicine, Vol.43, No. 3, 1994, pp.227-237. [15] S. Mahmoodi, B.S. Sharif, E.G. Chester, J.P. Owen and R.E.J. Lee, Automated vision system for skeletal age assessment using knowledge based techniques, IEEE conference publication, ISSN 0537-9989, issue 443, 1997, pp.809-813. [16] F. Vogelsang, M. Kohnen, H. Schneider, F. Weiler, M.W. Kilbinger, B.B. Wein and R.W. G¨unther, Skeletal maturity determination from hand radiograph by model based analysis, Proceedings SPIE, Vol.3979, 2000, pp.294-305. [17] Y. L.Hu, W. Wang and B.C. Yin, A research of bone maturation evalu tion system based on active shape models, Journal of Image and Graphics, Vol.8(A), No.1, 2003, pp. 33-50. [18] T.F. Cootes, A. Hill, C.J. Taylor and J. Haslam, The use of active shape models for locating structures in medical images, Proc. First British machine Vision Conference, Vol.245, 1990, pp.407-412. [19] C.C. Han, C.H. Lee and W.L. Peng, Hand radiograph image segmentation using a coarse-to-fine strategy, Pattern Recognition, Vol.40, No. 1, 2007, pp.2994-3004. [20] D. Giordano, R. Leonardi, F. Maiorana, G. Scarciofalo and C. Spampinato, Epiphysis and metaphysis extraction and classification by adaptive thresholding and DoG filtering for automated skeletal bone age analysis, in proceedings of 29th Annual International Conference of the I3-10 Engineering in Medicine and Biology Society, 23-26, Auguset, 2007, pp. 6551-6556. [21] C.W. Hsieh, T.L. Jong, Y.H. Chou and C.M. Tiu, Computerized geometric features of carpal bone for bone age estimation, Chinese Medical Journal; Vol.120, No.9, 2007, pp. 767-770. [22] T. Pappas, An adaptive segmentation algorithm for image segmentation, IEEE Trans. on Signal Processing. Vol.40, 1992, pp. 901-914. [23] E. Pietka, A. Gertych, K.S. Pospiech, F. Cao, H.K. Huang and V. Gilzanz, Computer-assisted bone age assessment: graphical user interface for image processing and comparison, J. Digit. Imaging. Vol.17, No. 3, 2004, pp.175-188. [24] E. Pietka, A. Gertych asnd B.J. Liu, Segmentation of regions of interest and post-segmentation edge location improvement in computer-aided bone age assessment, Pattern Anal. Appl. Vol.10, No. 2, 2007, pp.115-123. [25] E. Pietka, L. Kaabi, M. L. Kuo and H. K. Huang, Feature extraction in carpal-bone analysis, IEEE transaction on medical imaging, Vol. 12, No.1, 1993, pp. 616-620. [26] C. W. Hsieh, T. L. Jong and C. M. Tiu, The phalangeal morphological characteristics for bone age recognition, 13th IEEE-NPSS Real time conference 2003, Montreal, Canada, May, 2003. [27] S. Aja-Fernandez, R.D. Luis-Garcia, M.A. Martin-Fernandez and C. Alberola-Lopez, A computational TW3 classifier for skeletal maturity assessment. A computing with words approach, J Biomed Inform, Vol. 37, No.2, 2004, pp. 99-107. [28] J. S. Lee, Refined filtering of image noise using local statistics, Comput. graph. image process, Vol,15, 1981, pp.380-389. [29] S.K. Pakin, R.S. Gaborski, L.L. Barski and K.J. Parker, Clustering approach to bone and soft tissue segmentation of digital radiographic images of extremities, Journal of Electronic Imaging, Vol.12, No. 1, 2003, pp.40-49. [30] N. Otsu, A threshold selection method from gray level histogram, IEEE Trans. Syst., Man, Cybern SMC-9, 1979, pp.62-66. [31] J.F. Canny, A computational approach to edge detection, I3-10 Transactions on Pattern Analysis and Machine Intelligence, Vol.8, No.6, 1986, pp.679-698. [32] J.B.T.M. Roerdink and A. Meijster, The watershed transform, definitions, algorithms and parallelization strategies, Fundamenta Informaticae Vol.41, 2000, pp.187-228. [33] C. Xu and J.L. Prince, Snakes, shapes, and gradient vector flow, IEEE Trans. Image Process., Vol. 7, No.3, 1998, pp.359-369. [34] A. Witkin, M. Kass and D. Terzopoulos, Snake: active contour model, Int. J. Comput., Vol.l, 1987, pp.321-331. [35] M. Friedman and A. Kandel, Introduction to Pattern Recognition: Statistical, Structural, neural, and Fuzzy Logic Approaches, 1999, World Scientific, Singapore. [36] J.T. Tou and R.C. Gonzalez, Pattern Recognition Principles Reading, 1974, Addison-Wesley. [37] J. Besag, Spatial interaction and the statistical analysis of lattice system, J. Roy. Stat. Soc. Ser. B 26, 1974, pp.192-236. [38] M.D. Levine and A.M. Nazif, Dynamic measurement of computer generated image segmentations, IEEE Trans. Pattern Anal. Machine Intell. Vol.7, No. 2, 1985, pp.155-164. [39] M. Sezgin and B. Sankur, Survey over image thresholding techniques and quantitative performance evaluation, J. Electron. Imaging Vol.13, No. 1, 2004, pp.146-165. [40] Y.J. Zhang, A Survey on evaluation methods for image segmentation, Pattern Recognition Vol.29, No. 8, 1974, pp.1335-1346. [41] L.A. Zadeh, Fuzzy Logic, IEEE Computer, Vol. 21, No. 4, 1988, pp. 83-93. [42] R. Jain and A. Abraham, A Comparative study of fuzzy classifiers on breast cancer data, 7th International Work Conference on Artificial and Natural Neural Networks, Lecture Notes in Computer Science, Jose Mira and Jose R. Alverez (Eds.), Springer Verlag, Germany, Vol. 2687, 2003, pp. 512-519. [43] H. Ishibuchi and T. Nakashima, A study on generating classification rules using histogram, Second International Conference on Knowledge-Based Intelligent Electronic Systems, 21-23 April 1998, pp. 132-140. [44] O.D. Richard, E.H. Peter and G.S. David, Pattern Classification, John Wiley & Sons, 2nd edition, 2001. [45] H. Simon, Neural Networks: A Comprehensive Foundation, Prentice-Hall, 1994. [46] B. Widrow and M.A. Lehr, 30 Years of adaptive neural networks: perceptron, madaline, and backpropagation, Proc. IEEE, Vol. 78, No. 9, 1990, pp. 1415-1442. [47] S. Haykin, Neural Networks: A Comprehensive Foundation 2e, Prentice Hall, 1998. [48] R.L. Kennedy, Y. Lee, B.V. Roy, C.D. Reed, and R.P. Lippmann, Solving Data Mining Problems Through Pattern Recognition, pp 10.23-10.32, Prentice Hall, 1998. [49] C.T. Lin, and C.S.G. Lee, Neural Fuzzy Systems, Prentice Hall, 1999. [50] J.S. Lim, Two Dimensional Signal & Image Processing Prentice-Hall, Englewood Cliffs, NJ, 1990, pp.536-540. [51] M. Niemeijer, B. V. Ginneken, C. A. Maas, F. J. A. Beek, and M. A. Viergever, Assessing the skeletal age from a hand radiograph: automating the Tanner-Whitehouse method. SPIE Medical Imaging. Editor(s): M. Sonka, J.M. Fitzpatrick, SPIE 5032, 2003, pp. 1197-1205. [52] R.C. Gonzalez and R.E. Woods. Digital Image Processing. Addison-Wesley Longman Publishing Co., Inc., Boston, MA, USA 2002. [53] C.H. Chang, C.W. Hsieh, T.L. Jong and C.M. Tiu, A fully automatic computerized bone age assessment procedure based on phalange ossification analysis. In 16th IPPR conference on CVGIP, Kinmen 2003, pp. 463-68. [54] H.H. Lin, S.G. Shu, S.W. Kuo, C.H. Wang, Y.P. Chan and S.S. Yu, Alpha-gamma equalization-enhanced hand radiographic image segmentation scheme. Opt. Eng.,Vol. 48, No. 10, 2009. [55] S.G. Shu, H.H. Lin, S.W. Kuo and S.S. Yu, Excluding background initial segmentation for radiographic image segmentation, International Journal of Innovative Computing, Information and Control, Vol. 5, No. 11, 2009, pp. 3849-3860. [56] D. Nauck, Neuro-Fuzzy Systems: An Overview, Fuzzy systems in Computer Science, Artificial Intelligence. Wiesbaden; 1994b. [57] D. Nauck and R. Kruse, Choosing Appropriate Neuro-Fuzzy Models. In Proc. 2nd European Congress on Fuzzy and Intelligent Technologies (EUFIT94); 1994a. [58] Y. Yuan and S. Suarga, On the Integration of Neural Networks and Fuzzy Logic Systems. International Conference on Systems, Man and Cybernetics, Canada; 1995. [59] S. Haykin, Neural Networks: A Comprehensive Foundation. Macmillan College Publishing Company; 1994. [60] W. Ng, Application of neural networks to adaptive control of nonlinear systems. John Wiley and Sons, Inc; 1997. [61] S.E. Fahlman and C. Lebiere, The cascade-correlation learning architecture. In Advances in Neural Information Processing Systems 2, San Mateo, CA: Morgan Kaufmann 1990; pp. 524-532. [62] C. Zang and M. Imregun, Combined neural network and reduced FRF techniques for slight damage detection using measured response data. Arch Appl Mech, Vol. 71, No. 8, 2001, pp. 525-536. [63] P.A. Ioannou and J. Sun, Robust Adaptive Control. Prentice-Hall, Upper Saddle River, NJ, USA; 1995.
摘要: 
本論文提出了基於掌骨X光影像中骨骺/幹骺切割技術以及其對電腦化骨齡群集估測之相關研究。主要貢獻如下︰(1)基於X光片中複雜及不規則的背景及不論手掌在X光片中是否定位,提出一個全自動且有效的骨骺/幹骺特徵區域擷取方法。(2)提出一個以空間統計為基礎的distance approach adaptive two-means影像切割方法,其骨骺/幹骺區域的切割結果之準確度遠高於基於機率分佈的adaptive two-mean方法,同時亦大幅地降低了整個演算法的迭代次數及運算時間。(3)進一步地,我們考量不論是何種adaptive two-means的方法,其最終分類結果往往取決於初始分類的好壞,而初始分類又受到不同指節的特徵區域影像特性之影響。因此,為了更精進整個演算法對於的骨骺/幹骺區域切割的精準度及效率,我們針對不同指節提出兩個影像切割方法。針對遠端及中間指節,提出一個不考慮背景的影像切割方法,克服了在大部分影像切割分法如Sobel, two-means, Canny edge-detection 及 watershed中,因含背景導致所得到初始分類之精確度將嚴重影響最終切割結果的問題。針對近端指節,提出一個基於α-gamma equalization enhanced的影像切割方法,解決了在近端指節特徵區域影像中,因骨組織及軟組織之灰度階對比低而影響初始分類的問題。實驗結果顯示,上述我們提出的兩個方法,其最終得到之骨骺/幹骺區域切割的準確度均遠高於影像切割常用到之GVF的方法。(4)依據上述方法所得到的切割結果,萃取出各指節骨骺/幹骺區域之直徑比值作為特徵,提出一個基於模糊類神經網路系統之掌骨X光影像骨齡群集之估測系統,並重新定義出不同於傳統骨骼發展階段(前期及後期)的四個階段,以及建議各階段所對應的ROI處理及特徵選取方法。本系統最主要解決了以往骨齡估測必須依賴先驗知識(a priori knowledge)預先決定骨骼發展階段,據以作為選擇適當特徵項目及分類器的問題。實驗結果證明,我們提出的系統除了可以說明上述四個新定義骨骼發展階段之合理性之外,並可準確地估測掌骨X光影像之骨齡群集,使得整個電腦化骨齡估測系統更具彈性與可靠度。

This dissertation is focused on the study of segmentation for epiphyseal/metaphyseal regions of interest (EMROI) and bone age cluster assessment system using fuzzy neural network (FNN) based on phalangeal image segmentation. The contributions are illustrated as follows. (1)An automated ROI extraction with normalization is proposed to automatically select ROIs no matter whether the hand position is in upright or not. (2)A distance approach adaptive two-means clustering scheme is proposed to get a more precise segmentation on ROI and greatly reduce the processing time by comparing the probability approach adaptive two-means clustering method. (3) Furthermore, the performance of any above-mentioned segmentation techniques relies on the precisian of a given initial segmentation. In general, the initial segmentation is given by random or manual choices. Thus, two modification methods for optimum reliability, robustness and computational efficiency according to the properties of phalangeal ROIs are presented. For the distal and middle phalange, an excluding background initial segmentation method is proposed to overcome the initial segmentation problem for the bony structure segmentation. Experimental results verify that using this method can provide a better initial segmentation than other methods, such as Sobel, two-means, Canny edge-detection and watershed. An α-gamma equalization enhanced image segmentation scheme for proximal phalange is given to solve the problem of poor intensity contrast between soft tissues and bony tissues. The comparative experimental results show that these two methods really promote the performance and achieve a more accurate and robust epiphyseal/metaphyseal segmentation against GVF snake. (4) A bone age cluster assessment system using fuzzy neural network (FNN) based on phalangeal image segmentation is presented. This could avoid taking an a priori knowledge for bone age stage and provide a more flexible and reliable BAA system. Experimental results reveal that the presented FNN system provides a very well ability to assign the hand radiograph to an appropriate bone age cluster. The rationality of our four new defined stages which are different from the traditional skeletal development stages (the early and later stage). Furthermore, the related feature clustering analysis for various stages is discussed to obtain an accurate quantitative evaluation of specific features for the final BAA. Finally, it can provide a more flexible and reliable BAA system.
URI: http://hdl.handle.net/11455/19735
其他識別: U0005-2904201016081800
Appears in Collections:資訊科學與工程學系所

Show full item record
 

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.