臺灣博碩士論文加值系統

對18歲以下的未成年追蹤、紀錄骨骼發育情況，能夠及早發現不正常的發育，並用藥改善。而一般醫師目視x光片對指骨做年齡的判斷，容易因為精神狀況及辨識的經驗不同，而有不同的判斷；尤有甚者，同一位醫師對同一張影像的判斷，也有可能在不同時間而有不同的結果。且因為判斷的過程必須聚精會神、耗神費力，同一位醫師每天所能做判斷的x光片張數也有限制。我們的動機就是希望能夠自動處理x光片，對年齡做出正確的辨識，而能提供給醫師做為判斷的參考；並且能夠大量提供具客觀性、可重覆的資料給醫生。
本篇論文呈現如何以電腦自動化做指骨年齡的辨識與估計。
在前處理，輸入影像的灰階直方圖首先被「正規化」，同時還包括擺正位置及角度轉正。其次定義我們指骨有興趣的區域，以食、中指及中指、無名指之間的節點為左右邊界，以中指尖為上界，以虎口處為下界，找到我們的有興趣區域。找到有興趣區域之後，切割為黑白影像；只有骨頭部份是白色，其餘都化為背景黑色。
切割完成後接著抽出特徵點，並且將特徵點送入類神經網路做辨識。我們的特徵點是以中指第一及第三關節為準。
本篇論文主要的貢獻，在於找出一組很簡單、卻極為有效的特徵。以這組特徵進行辨識的結果十分令人滿意。針對我們目前資料庫中的影像，我們可以得到非常高的正確率。

This paper presents the details of the design and implementation of a computerized approach to phalangeal age assessment. Histogram of a CR (Computed Radiography) hand image to be analyzed is standardized first in the pre-processing stage, together with its location and orientation. Then the PROI (Phalangeal Region Of Interests) is defined and found. After the PROI is located, the image is segmented to binary. Age assessment using neural network, following the feature extraction stage, completes our experiment. The morphological features of the epiphysis of proximal phalanx of the third digit and the epiphysis of the third metacarpal bones are extracted to classify the skeleton age. Finding out a simple but powerful feature vector for skeleton age assessment is the main contribution in this study. The simulation results using the features are satisfactory, and we can get high accuracy within one year error for the database we have so far.

Ⅰ．INTRODUCTION 1
Ⅱ．PHYSIOLOGY OF SKELETAL GROWTH 3
Ⅲ．PROCESSING PROCEDURES — IMAGE PROCESSING 13
A. Pre-Processing 13
A-1 Extract the left hand 13
A-2 Rotation 14
B. Finding the PROI 18
C. Enhancement 20
D. LUM filters 22
D-1 LUM smoother 23
D-2 LUM sharpener 23
D-3 LUM filters 24
D-4 Asymmetric LUM filter 26
E. Segmentation 27
Ⅳ．Age Assessment — Neural Network 30
A. Network Architecture 31
B. Feature Vector 36
C. Learning Rule 38
D. Training 40
Ⅴ．Experimental Results 41
Ⅵ．Conclusion and Discussion 46
Ⅶ．REFERENCES 47

[1] Marek Brejl and Milan Sonka, “Object Localization and Border Detection Criteria Design in Edge-Based Image Segmentation: Automated Learning from Examples.” IEEE Trans. Med. Imag., vol. 19, no. 10, October 2000.
[2] Smadar Shiffman, Geoffrey D. Rubin, and Sandy Napel, “Medical Image Segmentation Using Analysis of Isolable-Contour Maps.” IEEE Trans. Med. Imag., vol 19, no. 11, November 2000.
[3] Bir Bhanu and Jing Peng, “Adaptive Integrated Image Segmentation and Object Recognition.”, IEEE Trans. Syst., Man, Cybern, vol. 30, no. 4, November 2000.
[4] David J. Michael and Alan C. Nelson, “HANDX: A Model-Based System for Automatic Segmentation of Bones from Digital Hand Radiographs.” IEEE Trans. Med. Imag., vol. 8, no. 1, March 1989.
[5] Ewa Pietka, Michael F. McNitt-Gray, M. L. Kuo, and H. K. Huang, “Computer-Assisted Phalangeal Analysis in Skeletal Age Assessment.”, IEEE Trans. Med. Imag., vol. 10, no. 4, December 1991.
[6] Chair-Li Chang, Heang-Ping Chan, Loren T. Niklason, Jeffery Crabbe, John Mathews, Mark Cobby, and Ronald S. Adler, “Computer-Aided Detection and Characterization of Hyperparathyroidism in Digital Hand Radiographs.” IEEE Trans. 1992.
[7] Ewa Pietka, Lotfi Kaabi, M. L. Kuo, and H. K. Huang, “Feature Extraction in Carpal-Bone Analysis.”, IEEE Trans. Med. Imag., vol. 12, no. 1, March 1993.
[8] Nanning Zheng, Yuanliang Zhang, Wenming Li, and Mori Shinsaku, “A Neural Net Model for Unsupervised Pattern Classification and Its Application to Image Segmentation.” IEEE, 1995.
[9] Jianguo Zhang and H. K. Huang, “Automatic Background Recognition and Removal (ABRR) in Computed Radiography Images.”, IEEE Trans. Med. Imag., vol. 16, no. 6, December 1997.
[10] Mahbubul Islam Chowdhury and John A. Robinson, “Improving image segmentation using edge information.”, IEEE Trans., 2000.
[11] Sugata Ghosal and Rajiv Mehrotra, “Detection of Composite Edges.”, IEEE Trans. Med. Imag., vol. 3, no. 1, January 1994.
[12] Takahiro Sugiyama and Keiichi Abe, “Edge Feature Analysis by a Vectorized Feature Extractor and in Multiple Edges.” IEEE Proceedings of ICPR, 1996.
[13] Woobeom Lee and Wookhyun Kim, “Self-Organization Neural Network for Multiple Texture Image Segmentation.”, IEEE TENCON, 1999.
[14] Qin-sheng Chen, Michel Defrise, and F. Deconinck, “Symmetric Phase-Only Matched Filtering of Fourier-Mellin Transforms for Image Registration and Recognition.”, IEEE Trans. Pattern Anal. Machine Intell., vol. 16, no. 12, December 1994.
[15] Russell C. Hardie and Charles G. Boncelet, “LUM Filters: A Class of Rank-Order-Based Filters for Smoothing and Sharpening.”, IEEE Trans. Signal Processing, vol. 41, no. 3, March 1993.
[16] Russell C. Hardie and Charles G. Boncelet, “Gradient-Based Edge Detection Using Nonlinear Edge Enhancing Prefilters.”, IEEE Trans. Imag., vol. 4, no. 11, November 1995.
[17] The MATHWORKS® Inc., “Image Processing Toolbox —For Use with MATLAB®”, User’s Guide, Version 2.
[18] The MATHWORKS® Inc., “Neural Network Toolbox —For Use with MATLAB®”, User’s Guide, Version 3.0.
[19] Milner GR, Levick RK and Kay R., ”Assessment of bone ages: A comparison of the Greulich and Pyle, and the Tanner and Whitehouse methods.”, Clin Radiol 1986; 37:119-212.
[20] Frisch H, Riedl S and Waldhor T., “Computer-aided estimation of skeletal age and comparison with bone age evaluations by the method of Greulich-Pyle and Tanner-Whitehouse.” Pediatr Radiol 1996; 26:226-31.
[21] Ricketts IW, Cairns AY., “Classification of hand bones for bone age assessment.” Proc. ICECS ’96 1996; 2:1088-91.
[22] Wastl S, Dickhaus H., “Computerized classification of maturity stages of hand bones of children and juveniles.” 18th Annual International Conference of the IEEE Engineering in Medicine and Biology Society 1996; 3:1155-6.
[23] Mahmoodi S, Sharif BS, Chester EG, Owen JP, Lee REJ., “Automated vision system for skeletal age assessment using knowledge based techniques.” Proc IPA97 1997;2:809-13.]