本論文希望發展一套指骨ROI的自動化切割系統，進而在後續加入特徵抽取與分析，組成一套全自動的指骨骨齡判讀系統，幫助醫生進行自動化骨齡判讀。
在論文中主要研究灰階指骨影像切割，針對年齡比較小、指尖的指節X光影像，加入前處理調整灰階分佈後，後端切割效果會變得更好，然而年紀大的指骨則沒有改善甚至更糟。所以流程再加入年紀的預知識，以腕骨面積密度隨著年齡變大而增長的特性，在前處理前先利用腕骨大約判定年紀的大小，再考慮是否加入所提前處理，以便後續骨頭邊緣切割。
論文中提出了一套指節切割處理流程，一開始X光影像輸入進來，能夠成功的把九個指節切出(食指、中指、無名指各三個)，接著為了能準確找到這9個EMROI(epiphyseal/metaphyseal bone region of interest)的骨頭邊界，在前處理前，將X光影像的腕骨區域以canny和sobel operator找出待分析X光影像edge像素總數，以此訊息為依據把年紀分為兩類(8歲以上或以下)，判斷為年紀小的則將指骨影像經過前處理，調整影像灰階分布，再進行骨頭切割邊界；年紀大的則直接切割骨頭邊界。
在實驗過程中，採用了四種切割方法，分別是GVF snake、圓形除均、adaptive two-means clustering algorithm以及level set evolution。為了驗證加入預知識後切割骨頭邊界與未加預知識直接切割骨頭邊界的效果，我們採用六種的錯誤測量指標：ME (misclassification error) 、RFAE (relative foreground area error) 、NU (non-uniformity) 、MHD (modified Hausdorff distances)、EMM (edge mismatch)、Mean Errors，實驗結果證實加入預知識後切割骨頭邊界，錯誤指標的值普遍獲得改善，其中mean errors指標的值在四種切割方法中分別降低了28%到33%。

Abstract
The objective of this thesis is to develop an automatic and accurate phalangeal segmentation system, which can constitute a fully automatic phalanx bone age assessment system by adding feature extraction and analysis/classification stages in the future.
Study shows that generally the result of segmentation can be further improved by a proposed preprocessing of adjusting the gray level distribution of the knuckle images, i.e., EMROI (epiphyseal/metaphyseal bone region of interest), before entering the bone edge detection/segmentation stage. The improvements are more profound in the cases of younger children however, for children with age over 8 and elder the effects of the proposed preprocessing become less effective and even deteriorate the segmentation results. Therefore this thesis proposes an age-dependent preprocessing scheme before entering the bone edge detection/segmentation stage. Observing that the density of carpal area increases with age, we propose a method utilizing such growth characteristics of carpal to quickly determine the prior knowledge of age for the hand radiogram under segmentation.
A knuckle segmentation process is proposed. First, the input X-ray hand-image is processed and nine knuckle images of the index finger, middle finger, ring finger (i.e., 9 EMROI’s) are segmented. Then the prior knowledge of age is determined by using the carpal area density characteristics and the age-dependent preprocessing based on the prior knowledge of age is applied to the 9 EMROI’s. Finally, image segmentation is applied to detect the bone edges and/or contours in those knuckle images.
Four image segmentation methods, namely GVF snake, round-average deduction method, adaptive two-means clustering algorithm and the level set evolution are adopted in our experiments. Six error measures: ME (misclassification error), RFAE (relative foreground area error), NU (non-uniformity), MHD (modified Hausdorff distances), EMM (edge mismatch), and Mean Errors are used to assess the effectiveness of the proposed phalanx image segmentation process. The experimental results show that the error measures for the cases of incorporating prior knowledge of age and proposed age-dependent preprocessing are generally better than those cases without using prior knowledge of age and proposed preprocessing. For example, the values of mean errors are improved from 28% to 33% for the four different segmentation methods, respectively.

[1] W. W. Greulich, S. I. Pyle , Radiographic Atlas of Skeletal Development of the Hand and Wrist, CA: Stanford University Press, 1959.
[2] J. M. Tanner, R. H. Whitehouse, N. Cameron, H. Goldstein, M. J. R. Healy, W. A. Marshall, Assessment of Skeletal Maturity and Prediction of Adult Height(TW2 Method), Orlando, Fla.:Academic Press, 1984.
[3] C. M. Tiu, T. C. Liu, C. W. Hsieh, D. M. Niu, J. D. Chen, T. L. Jong, “Turner syndrome phalangeal screening based on a two-stage linear regression concept,” Pediatrics International, Vol. 51, No. 4, pp. 453-459, 2009.
[4] T. W. Todd, Atlas of skeletal maturation, St. Louis, The C.V. Mosby Company, 1937.
[5] J. M. Tanner, M. J. R. Healy, “Assessment of skeletal maturity and prediction of adult height (TW3) method,” The Bone & Joint Journal, Vol. 84, No. 2, pp. 310-311, 2002.
[6] P. K. Bull, “Bone age assessment: a large scale comparison of the Greulich and Pyle, and Tanner and Whitehouse (TW2) methods,” Arch Dis Child, Vol. 81, pp. 172-173, 1999.
[7] S. Bolboaca, L. Jäntschi, “PHP and MySQL Medical Application Based on Tanner Whitehouse Standard,” Proceeding of the UNITECH'03 International Scientific Conference, pp. 304-308, 2003.
[8] M. Law, H. Huang, L. Chan, X. Zhang, J. Sun, “The Development and Implementation of a Prototype DICOM-based Radiotherapy Information System,” Scientific poster, 2004.
[9] G. Luo, J. J. Kaufman, “COMPUTATIONAL METHODS FOR ULTRASONIC BONE ASSESSMENT,” Ultrasound in Med. & Biol., Vol. 25, No. 5, pp. 823-832, 1999.
[10] D. J. Michael, A. C. Nelson, “HANDX: A Model-Based System for Automatic Segmentation of Bones from Digital Hand Radiographs,” IEEE TRANSACTIOSS ON hlEDlCAL IMAGING, Vol. 8, No. 1, 1989.
[11] E. Pietka, M. F. McNitt-Gray, M. L. Kuo, H. K. Huang, “Computer-Assisted Phalangeal Analysis in Skeletal Age Assessment,” IEEE Trans. Med. Imag., Vol. 10, No. 4, 1991.
[12] E. Pietka, H. K. Huang, “Epiphyseal Fusion Assessment Based On Wavelets Decomposition Analysis,” Computerized Medical Imaging and Graphics, Vol. 19, No. 6, pp. 465-472, 1995.
[13] E. Pietka, “Computer automated approach to the extraction of epiphyseal regions in hand radiographs,” Journal of Digital Imaging, Vol. 14, No. 4, 2001.
[14] H. H. Lin, “The Study of Computerized Bone Age Cluster Assessment System Based on Hand Radiographs,” PHD dissertation of NCHU, 2010.
[15] H. H. Lin, “Alpha-gamma equalization-enhanced hand radiographic image segmentation scheme,” Optical Engineering, Vol. 48, No. 10, 2009.
[16] C. W. Hsieh, T. L. Jong, C. H. Chang, C. H. Chen, C. M. Tiu, “The phalanges-based morphological features for skeleton age assessments,” 15th conference on CVGIP20, pp. 729-733, 2002.
[17] C. H. Chang, “An Investigation of Computerized Automatic Bone Age Assessment System Based on Phalangeal Image Features,” master thesis of NTHU, 2003.
[18] B. C. Fan, C. W. Hsieh, T. L. Jong, C. M. Tiu, “Automatic bone age estimation based on carpal-bone image--a preliminary report,” Europe PMC, Vol. 64, pp. 203-208, 2001.
[19] 張博超, “Cloud Bone Age Assessment Service Based on Tanner and Whitehouse Method,” Master Thesis of NTHU, 2012.
[20] 王甫安, “A Web-Based TW Scoring System for Bone Age Assessment,” Master Thesis of NTHU, 2011.
[21] “BoneXpert[Online],” Visiana, Available: http://www.bonexpert.com/.
[22] H. H. Thodberg, S. Kreiborg, A. Juul, K. D. Pedersen, “The BoneXpert method for automated determination of skeletal maturity.,” IEEE Trans Med Imaging., Vol. 28, No. 1, pp. 52-66, 2009.
[23] 楊嘉棟, “The effect of phalanx contour location via alpha-gamma equalization,” Master Thesis of NTHU, 2012.
[24] C. Xu, J. L. Prince, “Gradient Vector Flow: A New External Force for Snakes,” IEEE Proc. Conf. on Comp. Vis. Patt. Recog., 1997.
[25] C. Xu, J. L. Prince, “Snakes, Shapes, and Gradient Vector Flow,” IEEE TRANSACTIONS ON IMAGE PROCESSING, Vol. 7, No. 3, 1998.
[26] C. Li, C. Xu, C. Gui, M. D. Fox, “Distance Regularized Level Set Evolution and Its,” IEEE TRANSACTIONS ON IMAGE PROCESSING, Vol. 19, No. 12, pp. 3243-3254, 2010.
[27] C. Li, C. Xu, C. Gui, M. D. Fox, “Level Set Evolution Without Re-initialization: A New Variational Formulation,” IEEE Computer Society Conference, 2005.
[28] 黃柏鈞, “The Extraction and Process of ROI of Phalangeal Bones Radiographs,” Master Thesis of NCHU, 2008.
[29] 黃耀仟, “Design and Implementation of an Automatic Phalangeal Segmentation System,” Master Thesis of NTHU, 2010.
[30] C. C. Liu, P. C. Chung, “Objects Extraction Algorithm of color image using adaptive forecasting filters created automatically,” International Journal of Innovative Computing, Vol. 10, No. 7, 2011.
[31] X. T. JIANG, Y. LV, “Improved Approach to Image Matching Based on Weighted Hausdorff Distance,” APPLICATION RESEARCH OF COMPUTERS, Vol. 24, No. 4, 2007.
[32] C. Li, “A Level Set Method for Image Segmentation in the Presence of Intensity Inhomogeneities with Application to MRI,” IEEE Trans. Image Processing, Vol. 20, No. 7, pp. 2007-2016, 2011.
[33] R. C. Gonzalez, R. E. Woods, Digital Image Processing., Boston, MA: Addison-Wesley Longman Publishing Co., 2002.
[34] K. Wagsta, C. Cardie, S. Rogers, S. Schroedl, “Constrained K-means Clustering with Background Knowledge,” Proceedings of the Eighteenth International Conference on Machine Learning, pp. 577-584, 2001.
[35] M. kass, “Snakes:Active contour models.,” Int. Journal of Computer Vision, Vol. 1, No. 4, pp. 321-331, 1987.
[36] K. Sato, “An accurate method of predicting mandibular growth potential based on bone maturity,” American Journal of Orthodontics and Dentofacial Orthopedics, Vol. 120, No. 3, 2001.
[37] S. Mahmoodi, B. S. Sharif, E. G. Chester, J. P. Owen, R. Lee, “Skeletal Growth Estimation Using Radiographic,” IEEE TRANSACTIONS ON INFORMATION TECHNOLOGY IN BIOMEDICINE, Vol. 4, No. 4, 2000.
[38] E. Pietka, “Feature extraction in carpal-bone analysis,” IEEE Transactions Medical Imaging, Vol. 12, No. 1, 1993.
[39] H. H. Lin, “Bone age cluster assessment and feature clustering analysis based on phalangeal image rough segmentation,” Pattern Recognition, Vol. 45, No. 1, 2012.
[40] H. TAMURA, M. Shunji, Y. Takashi, “Textural Features Corresponding to Visual Perception,” IEEE Transactions on, Vol. 8, No. 6, pp. 460-473, 1978.
[41] 李振銘, “The Study of Fully Automatic Bone Age Development Using the Area Ratio of Carpals,” Master Thesis of NTHU, 2010.
[42] C. W. Hsieh, T. L. Jong, Y. H. Chou, C. M. Tiu, “Computerized Carpal Bone Geometric Feature for Bone Age Estimation,” Chinese Medical Journal, Vol. 120, No. 9, pp. 767-770, 2007.
[43] E. Pietka, “Computer-Assisted Bone Age Assessment Based on Features Automatically Extracted From A Hand Radiograph,” Computerized Medical Imaging and Graphics, Vol. 19, No. 3, pp. 251-259, 1995.