在本篇論文中，提出一套結合腕骨與指骨年齡判讀的全自動化流程，可以用來判讀發育中的兒童手掌X光影像的骨骼年齡，利用此骨骼年齡與實際年齡的比對，醫生可以提早發現某些生長發育上的異常。在前處理中首先利用圓弧掃描演算法切割出左手X光影像，再根據database中手掌型態學上的統計定義出腕骨的定位點，利用重心及moving average等方法找出腕骨中心，由中心與定位點切割出carpal bone region of interest(CROI)，最後是使用 Gabor Filter此具有方向性的方法smooth影像、Canny影像邊緣偵測法以及 Local Variance影像邊緣偵測法這兩種尋找影像邊緣的工具，將腕骨與指骨從背景及肉質等不需要的區域當中切割出來。由前處理的步驟完成後所得到的骨骼影像抽取特徵，以指骨第一個關節的水平投影圖做一階discrete cosine transform(DCT)和一階discrete wavelet transform(DWT)以及腕骨的生長面積比率為抽取特徵。為了更準確地判讀骨骼年齡，我們將腕骨與指骨年齡判讀系統各自分別運用模糊理論和類神經網路兩種不同的方法，並且分析腕骨、指骨與結合腕骨與指骨資訊而成的結合系統在各個年齡層時的錯誤量。實驗結果顯示，腕骨在年齡小於八歲半的時候能夠比指骨達到較好的準確度，且在結合腕骨與指骨的系統之後準確度會比只有腕骨或指骨做年齡判讀的系統更高，在誤差兩歲的容許範圍以內，結合腕骨指骨的年齡判讀系統正確率約略可以達到八成五左右。

In this article，it brings up an automatic procedure that combines the ways to assess the age of the carpal and the phalangeal. The procedure is made to assess the bone age of the growing children by analyzing their X-ray images of the palms. By comparing the bone age with the real age, doctors can find the problems of development early. In the preceding process, we segment the X-ray image of the left hand firstly, and according to the characteristic of morphology statistics in database, make an orientation in the carpal. Then, we find out the central point of the carpal, and segment carpal bone region of interest(CROI) from the intersection of the central point and the orientation. Lastly we use Gabor Filter smoothing image and two tools--- canny edge detector and local variance edge detector---which find the image’s edge to cut out the region of the carpal and the phalangeal from the needless parts such as the background.
Feature extracts from the bone image we obtain after finishing the preceding process. We use the horizontal projection of phalangeal’s first joint to make one dimension discrete cosine transform(DCT) and one dimension discrete wavelet transform(DWT). And the carpal area ratio are considered the carpal feature vector. In order to assess the bone age more exactly, we use two different kind of ways---fuzzy and neural network---on the carpal and the phalangeal age assessment systems individually, then analyze the amount of error in each age of these two assessment systems and an integral system that combines both. The result of the experiment shows that the carpal age assessment system would be able to get a better accuracy than the phalangeal age assessment system. Furthermore, the system combines both would be more correct than only one of the age assessment systems. In the odds allowed range within two years of age, the age assessment system combines the ways to assess the age of the carpal and the phalangeal can reach about 85% of the correctness.

[1] D. J. Michael and A. C. Nelson, “HANDX: A Model-Based System For Automatic Segmentation Of Bones From Digital Hand Radiographs” , IEEE Trans. Med. Imag., Vol.8, No.1, pp.64-69, March 1989.
[2] 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.
[3] E. Pietka, A. Gertych, S. Pospiech, F. Cao, H. K. Huang, “Computer-Assisted Bone Age Assessment: Image Preprocessing and Epiphyseal/Metaphyseal ROI Extraction”, IEEE Transactions On Medical Imaging, Vol.20, No.8, pp.715-729, August 2001.
[4] R. K. Cope and P. I. Rockett, “Efficacy of Gaussian smoothing in Canny edge detector”, Electronics Letters, Vol.36, No.19, pp.1615-1617, 14 September 2000.
[5] G. R. Milner, R. K. Levick and R. Kay, ”Assessment Of Bone Ages: A Comparison Of The Greulich and Pyle, and The Tanner and Whitehouse Methods”, Clin Radiol, Vol.37, pp.119-212, 1986
[6] E. Pietka, M. 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,pp.616-620, December 1991.
[7] E. Pietka, L. Kaabi, M. L. Kuo, and H. K. Huang, “Feature Extraction in Carpal-Bone Analysis”, IEEE Trans. Med. Imag., Vol.12, No.1,pp.44-49, March 1993.
[8] E. Pietka and H. K. Huang, “Epiphyseal Fusion Assessment Based On Wavelets Decomposition Analysis”, Computerized Medical Imaging and Graphics, Vol.19, No.6, pp.465-472, 1995.
[9] A. M. M. D. Silva, S. D. Olabarriaga, C. A. Dietrich and C. A. A. Schmitz, “On Determining A Signature For Skeletal Maturity”, Computer Graphics and Image Processing, 2001 Proceedings of XIV Brazilian Symposium, pp.246-251, 15-18 Oct. 2001.
[10] M. Sonka, V. Hlavac and R. Boyle, Image Processing, Analysis, and Machine Vision, 2nd edition, PWS, 1999
[11] L. Hong, Y. Wan and A. Jain, “Fingerprint Image Enhancement: Algorithm and Performance Evaluation”, IEEE Transactions On Pattern Analysis and Machine Intelligence, Vol.20, No. 8, pp.777-789, August 1998.
[12] N. Mittal, D. P. Mital and K. L. Chan, “ Features For Texture Segmentation Using Gabor Filter”, Image Processing and Its Applications, 1999, Seventh International Conference On (Conf, Publ. No.465), Volume:1, Vol.1, pp.353-357, 13-15 July 1999.
[13] B. S. Sharif, S. A. Zaroug, E. G. Chester, J. P. Owen, E. J. Lee, “Bone Edge Detection In Hand Radiographic Images”, Engineering In Medicine and Biology Society, 1994. Engineering Advances: New Opportunities for Biomedical Engineers. Proceedings Of The 16th Annual International Conference Of The IEEE, Vol.1, pp.514-515, 3-6 Nov. 1994.
[14] J.H. Chen, “A Computerized Skeleton Age Assessment Based On Phalangeal Image and Neural Network Approach”, Master Thesis, EE Nthu, 2002.
[15] S. Haykin, Neural Networks, 2nd edition, Prentice Hall, 1999
[16] F. Cao, H. K. Huang, E. Pietka and V. Gilsanz, “Digital Hand Atlas and Web-Based Bone Age Assessment: System Design and Implementation”, Computerized Medical Imaging and Graphics 24 (2000), pp. 297-307.
[17] K. Bowyer, C. Kranenburg and S. Dougherty, “Edge Detector Evaluation Using Empirical ROC Curves”, Computer Vision and Pattern Recognition, 1999. IEEE Computer Society Conference on. , Volume: 1, Vol.1, pp.354-359, 23-25 June 1999.
[18] T. S. Yoo, “Multiscale Statistical Image Invariants”, 7th International Conference on Computer Vision
[19] D. Grennhill and E. R. Davies, “A New Approach To The Determination Of Unbiased Thresholds For Image Segmentation”, Image Processing and Its Applications, 1995., Fifth International Conference on,pp.519-523, 4-6 Jul 1995
[20] C. W. Hsieh, T. L. Jong, C. H. Chang, C. H. Chen and C. M. Tiu, “The Phalanges-Based Morphological Features For Skeleton Age Assessments”, 15th Conference on CVGIP02, Hsinchu, Aug. 2002.
[21] L. A. Zadeh, “Fuzzy sets” Information and Control, vol. 8, pp. 338-353,1965.
[22] C. W. Hsieh, T. L. Jong, Chi-Hsing Chang, “An Investigation of Computerized Automatic Bone Age Assessment System Based on Phalangeal Image Features”, 16th Conference on CVGIP, Hsinchu, Aug. 2003.
[23] I. W. Ricketts, and A. Y. Cairns, “Classification of Hand Bones for Bone Age Assessment”, ICECS ’96, pp. 1088-1091, vol.2, Rodos Palace Hotel and Convention Center, Rodos Freece, Oct. 13-16, 1996.
[24] S. Wastl, H. Dickhaus, “Computerized Classification of Maturity Stage of Hand Bones of Children and Juveniles”, 18th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, pp. 1155-1156, vol.3, Amsterdam The Netherlands, Oct. 31-Nov. 3, 1996.
[25] S. Mahmoodi, B. S. Sharif, E. G. Chester, J. P. Owen, R. E. J. Lee, “Automated Vision System for Skeletal Age Assessment using Knowledge Based Techniques”, IPA97, pp. 809-813, vol.2 , Trinity Colledge, Dublin, Ireland, July 15-17, 1997.
[26] Dinesh M. S, Bhanuprakash, Dr. Ashok Rao, “Vision system for bone measurement from digital hand radiograph”, Proceedings RC-IEEE-EMBS & 14th BMESI, pp. SPC9-SP10, New Delhi, Feb. 15-18, 1995.
[27] Greulich WW, Pyle SI “Radiographic atlas of skeletal development of the hand and wrist”, Stanford University Press, Stanford, California 1959.