在現代醫學中，醫療影像已廣泛用於疾病的診斷與判定，而在應用電腦的輔助診斷過程前，器官的辨識與定位即是一項重要的前處理，其中，又以腹腔影像所包含的器官最多，也最為複雜，如：肝、腎、脾、胃、直腸、膀胱等；確定了器官在影像中的位置，後續如：自動疾病診斷、體積計算、三維重構等，便很容易可以達成，而對於解剖教學上，亦能提供幫助。隨著資訊科技的迅速發展，新一代的寬頻網路傳輸環境也已逐漸架構完成中，這使得整合傳輸大量影像資料與提供視訊會議功能的遠距醫療系統變得可行，藉由遠距醫療系統，不僅可以提供遠端醫師在看診時的意見諮詢服務，也彌補了醫療資源不均所帶來的差異。因此，本論文即是基於新一代的寬頻網路傳輸環境，對於建構互動式之遠距腹腔醫療會診系統做一研究。
本論文提出了一套腹腔CT (Computerized Tomography)影像之器官辨識方法，此方法結合了一multi-module contextual類神經網路用以對影像進行分割，同時，對於分割後的區域，我們應用了fuzzy觀念來描述這些區域的特徵，並且建構出fuzzy規則，在fuzzy推論過程中，鑑於RBF (Radial Basis Function) 類神經網路的快速收斂與區域模型(local model)的特性，同時與fuzzy inference在功能上的同等(functional equivalence)，我們提出了一robust RBF類神經網路，利用sigmoidal functions與 robust objective function以解決傳統類神經網路中高斯函數(Gaussian function)無法逼近常數值以及訓練樣本含有重大誤差的問題，並結合fuzzy規則以應用在器官辨識上，在實驗過程中，我們測試了超過四十組腹腔影像，而結果也顯示出本方法在器官辨識上確實有良好的效果。
在互動式的遠距醫療環境裡，內含1)醫療影像及傳輸標準(Digital Imaging and Communications in Medicine, DICOM)，使本系統可以直接與醫院影像儲存系統(Picture Archiving and Communication System, PACS)溝通，使用者可透過DICOM標準自醫院內部隨時調閱影像；2)多滑鼠顯示機制(telepointer)及命令啟動之同步控制機制，讓討論者可彈性操控畫面，並得以透過滑鼠指標對影像區域進行討論；3)提供視訊會議功能，使用者可以進行面對面溝通；4)內建醫學影像分析及處理工具，便於醫師做即時性影像分析；5)資料安全及憑證機制，以提供使用者認證及安全性之網路資料傳輸，以確保病人資料能受到保護。同時，為整合會診過程中的眾多資料傳輸，系統內建了資料流管理機制，針對不同資料的即時性、重要性、與封包遺失容忍度等特性，我們訂定了不同的資料傳輸優先順序，並結合不同的傳輸佇列以控制資料傳輸及流量，進一步提升系統整體效能。

Using medical image analysis for diagnosis and treatment of disease relies on the recognition and identification of the organs as one prerequisite step. Once the organ contour has been identified, several objectives can be fulfilled easily such as three dimensional treatment planning, organ volume computing, disease diagnosis, and so on. As such, recognition and identification of the organs is an important pre-processing during analysis. However, analysis of abdominal medical image is more difficult because there are several organs in abdominal images and the tissues usually show overlapping gray levels.
The dissertation proposes to identify abdominal organs from CT (Computerized Tomography) image series, by combining a multi-module contextual neural network, fuzzy rules, and fuzzy-inference-based RBF (Radial Basis Function) neural network. The multi-module contextual neural network is to segment each image slice through a divide-and-conquer concept embedded within multiple neural network modules, where the results obtained from each module are forwarded to other modules for integration, in which contextual constraints are enforced. With this approach, the difficulties arising from partial volume effects, gray level similarities of adjacent organs, and contrast media affect can be reduced to the extreme. To address the issue of high variations in organ position and shape, fuzzy inference is adopted. In this dissertation, a RBF network is implemented to perform the fuzzy inference for recognizing the organ of interest due to its properties of the local model and the functional equivalence between the fuzzy inference systems. This approach has been tested on more than forty sets of abdominal CT images, where each set consists of about 40 image slices, and the experimental results show the high promise of the proposed method.
While diagnosis of abdominal diseases usually relies on various areas of experts, in view of the fast development of telecommunication, this dissertation also design the abdominal organ identification into an interactive cooperative system where physicians could deliver their discussions on the diagnosis and identification results on the provided system. The interactive telemedicine system is built with CSCW (Computer-Supported Cooperative Work), DICOM (Digital Imaging and Communications in Medicine) standard, security functions, image processing/analysis tools, and streaming management. The built-in CSCW creates a collaborative consultation environment for synchronous interactive face-to-face discussion. The security functions provide the privacy and integrity in patient data transmission. The DICOM standard enables the medical image access to the PACS (Picture Archiving and Communication System) connecting with various imaging modalities. The image processing/analysis tools supported by CSCW functions provide useful tools for physicians to examine the images, and short-code messages are defined to transmit the image operation command for maintaining the system consistency between users. The streaming management module dynamically tunes the sending rate of the data according to its priority and the available bandwidth to accommodate the transmission to different network situations. These functions are tested on the NGN (Next Generation Network) transmission for its characteristics including transmission latency, jitter, data loss rate, and multicast performance. The experiments show that adopting the short-code message drastically reduces the bandwidth requirement and also the user waiting time, under which the basic bandwidth requirement of the system during consultation is about 160 Kbps.

[1] C. C. Lee, P. C. Chung, and H. M. Tsai, “Identifying Multiple Abdominal Organs from CT Image Series Using a Multimodule Contextual Neural Network and Spatial Fuzzy Rules,” accepted and to appeare in IEEE Transactions on Information Technology in Biomedicine.
[2] N. R. Pal and S. K. Pal, "A review on image segmentation techniques," Pattern Recognition, Vol. 26, No. 9, pp.1277-1294, 1993.
[3] M. Kobashi and L. G. Shapiro, “Knowledge-based organ identification from CT images,” Pattern Recognition, vol. 28, No.4, pp. 475-491, 1995.
[4] J. E. Koss, F. D. Newman, T. K. Johnson, and D. L. Kirch, “Abdominal organ segmentation using texture transform and a Hopfield neural network,” IEEE Trans. Med. Imag., vol. 18, no. 7, pp. 640-648, 1999.
[5] G. Bueno, M. Fisher, and R. Aldridge, “Segmentation of clinical structures for radiotherapy treatment planning: A comparison of two morphological approaches,” in Proc. of Seventh Australian and New Zealand Intelligent Information System Conference, pp. 15-18, 2001
[6] S. W. Yoo et al., “Organ segmentation by comparing of gray value portion on abdominal CT image,” in Proc. of 5th Int. Conf. on Signal Processing (WCCC-ICSP 2000), pp.1201-1208, 2000.
[7] B. Tsagaan et al., “Segmentation of kidney by using a deformable model,” in Proc. of Int. Conf. on Image Processing, vol. 3, pp. 1059-1062, 2001.
[8] H. M. Ladak et al., “Prostate segmentation from 2D ultrasound images,” in Proc. of the 22nd Annual EMBS International Conference, Chicago, pp. 3188-3191, 2000.
[9] F. S. Cohen, Z. Huang, and Z. Yang, "Invariant matching and identification of curves using B-splines curve representation," IEEE Trans. Image Processing, Vol. 4, No. 1, pp.1-10, 1995.
[10] R. N. Dave and T. Fu, "Robust shape detection using Fuzzy clustering: Practical applications," Fuzzy Sets and Systems 65, pp.161-185, 1994.
[11] E. Persoon and K. S. Fu, "Shape discrimination using Fourier descriptors," IEEE Trans. Patt. Anal. Machine Intell., Vol. PAMI-8. No. 3, pp.388-397, 1986.
[12] A. H. Mir, M. Hanmandlu, and S. N. Tondon, “Description of shapes in CT images: the usefulness of time-series modeling techniques for identifying organs,” IEEE Engineering in Medicine and Biology, pp. 79-84, 1999.
[13] H. Li et al., “Object recognition in brain CT-scans: Knowledge-based fusion of data from multiple feature extractors,” IEEE Trans. Med. Imag., vol. 14, no. 2, pp. 212-229, 1995.
[14] C. Li et al., “Knowledge based classfication and tissue labeling of MR images of human brain,” IEEE Trans. Med. Imag., vol. 12, no. 4, pp. 740-749, 1993.
[15] C. W. Chang et al., “A fuzzy rule-based system for labeling the structures in 3D human brain magnetic resonance images,” in Proc. of IEEE Int. Fuzzy Systems., pp. 1978-1982, 1996.
[16] S. P. Raya, “Low-level segmentation of 3D magnetic resonance brain images: A rule-based system,” IEEE Trans. Med. Imag., vol. 9, no. 3, pp. 327-337, 1990.
[17] X. Li, S. Bhide, M. R. Kabuka, “Labeling of MR brain images using boolean neural network,” IEEE Trans. Med. Imag. vol. 15, no. 5, pp. 628-629, 1996.
[18] C. Nikuo, G. Bueno, F. Heitz, and J. P. Armspach, “A joint physics-based statistical deformable model for multimodal brain image analysis,” IEEE Trans. Med. Imag., vol. 20, no. 10, pp. 1026-1037, 2001.
[19] J. S. R. Jang and C. T. Sun, “Functional equivalence between radial basis function networks and fuzzy inference systems,” IEEE Trans. Neural Networks, vol. 4, pp. 156-158, Jan. 1993.
[20] K. J. Hunt, R. Hass, and R. Murray-Smith, “Extending the functional equivalence of radial basis function networks and fuzzy inference systems,” IEEE Trans. Neural Networks, vol. 7, no. 3, pp. 776-781, May. 1996.
[21] M. Chen and D. A. Linkens, “A Fast Fuzzy Modelling Approach Using Clustering Neural Networks,” Proc. IEEE World Congress on Computational Intelligence 1998, Anchorage, Alaska 4-9, pp.1088-1092, May 1998.
[22] C. C. Lee, P. C. Chung, J. R. Tsai and C. I Chang, “Robust Radial Basis Function Neural Networks,” IEEE Trans. Syst. Man, Cybern., Part B, Vol. 29, No. 6, pp.674-685, Dec. 1999.
[23] H. K. Huang, “Teleradiology technologies and some service models,” Computerized Medical Imaging and Graphics, vol. 20, pp. 59-68, 1996.
[24] J. Zhang, J. N. Stahl, H. K. Huang, X. Zhou, S. L. Lou, and K. S. Song, “Real-Time Teleconsultation with High-Resolution and Large-Volume Medical Images for Collaborative Healthcare,” IEEE Trans. Inform. Technol. Biomed., vol. 4, no. 2, pp. 178-184, 2000.
[25] E. J. Gomez, F. del Pozo, E. J. Ortiz, N. Malpica, and H. Rahms, “A broadband multimedia collaborative system for advanced teleradiology and medical imaging diagnosis,” IEEE Trans. Inform. Technol. Biomed., vol. 2, no. 3, pp. 146-155, 1998.
[26] R. Sacile, “Telemedicine Systems for Collaborative Diagnosis over the Internet Towards Virtual Collaboratories,” Proc. 2000 Academia/Industry Working Conference on Research Challenges, pp. 191-196, 2000.
[27] M. Akay, I. Marsic, A. Medl, and G. Bu, “A system for Medical Consultation and Education using Multimodal Human/Machine Communication,” IEEE Trans. Inform. Technol. Biomed., vol. 2, no. 4, pp. 282-291, 1998.
[28] J. N. Stahl, J. Zhang, C. Zellner, E. V. Pomerantsev, T. M. Chou, and H. K. Huang, “Teleconferencing with Dynamic Medical Images,” IEEE Trans. Inform. Technol. Biomed., vol. 4, no. 2, pp. 88-96, 2000.
[29] E. I. Karavatselou, G. P. K. Economou, C. A. Chassomeris, V. Danelli-Mylonas, and D. K. Lymberopoulos, “OTE-TS - a new value-added telematics service for telemedicine applications,” IEEE Trans. Inform. Technol. Biomed., vol. 5, no. 3, pp. 210-224, 2001.
[30] Y. Bouillon, F. Wendling, and F. Bartolomei, “Computer-supported collaborative work (CSCW) in biomedical signal visualization and processing,” IEEE Trans. Inform. Technol. Biomed., vol. 3, no. 1, pp. 28-31, 1999.
[31] L. Makris, I. Kamilatos, E. V. Kopsacheilis, and M. G. Strintzis, “Teleworks: A CSCW Application for Remote Medical Diagnosis Support and Teleconsultation,” IEEE Trans. Inform. Technol. Biomed., vol. 2, no. 2, pp. 62-73, 1998.
[32] L. Kleinholtz, R. Virchow, and M. Ohly, “Supporting cooperative medicine: The Bermed project,” IEEE Multimedia Mag., pp. 44-53, winter, 1995.
[33] F. Malamateniou, G. Vassilacopoulos, and P. Tsanakas, “A Workflow-Based Apporach to Virtual patient Record Security,” IEEE Trans. Inform. Technol. Biomed., vol. 2, no. 3, pp. 139-145, 1998.
[34] S. Tachakra, M. Loane, and C. U. Uche, “A follow-up study of remote trauma teleconsultations,” J. Telemed. Telecare, vol. 6, no. 6, pp. 330-334, 2000.
[35] E. D. Lemaire, Y. Boudrias, and G. Greene, “Low-bandwidth Internet-based videoconferencing for physical rehabilitation consultations,” J. Telemed. Telecare, vol. 7, no. 2, pp. 82-89, 2001.
[36] O. Cuzzani, M. Bromwich, E. Penno, and H. Gimbel, “The effect of transmission bandwidth on the quality of ophthalmological still and video images,” J. Telemed. Telecare, vol. 6, no. 4, pp. 11-13, 2000.
[37] C. C. Lee and P. C. Chung, “A Telemedicine System for Collaborative Teleconsultation based on Data Stream Management,“ in Proc. Of 2nd European Medical & Biological Engineering Conference, EMBEC'02, Vienna, Austria, Dec. 04-08, 2002.
[38] A. Rosenfeld, R.A. Hummel and S.W. Zucker, “Scene labeling by relaxation operations,” IEEE Trans. Syst. Man Cybern., vol. 6, pp.420-433, 1976.
[39] W. C. Lin, E. C. K. Tsao and C. T. Chen, "Constraint satisfaction neural networks for image segmentation," Pattern Recognition, Vol. 25, No. 7, pp.679-693, 1992.
[40] S. Haring, M. A. Viergever and J. N. Kok, "Kohonen networks for multiscale image segmentation," Image and Vision Computing, Vol. 12, No. 6, pp.339-344, 1994.
[41] R. Krishnapuram, J. M. Keller and Y. Ma, "Quantitative analysis of properties and spatial relations of Fuzzy image regions," IEEE Trans. Fuzzy Systems, Vol. 1, No. 3, pp.222-233, 1993.
[42] K. Miyajima and A. Ralescu, "Spatial organization in 2D segmentated images: Representation and recognition of primitive spatial relations," Fuzzy Sets and Systems 65, pp.225-236, 1994.
[43] L. Shen, R. M. Rangayyan, and J. E. Leo Desautels, "Application of shape analysis to mammographic calcifications," IEEE Trans. Medical Imaging, Vol. 13, No. 2, pp.263-274, 1994.
[44] M. Sonka, V. Hlavac and R. Boyle, Image Processing, Analysis and Machine Vision, Chapman & Hall Publishing Company, 1993.
[45] J. Weir et al., An Imaging Atlas of Human Anatomy, Wolfe Publishing Ltd, 1992.
[46] O. H. Wegener, Whole Body Computerized Tomography, Schering AG West Germany, 1983.
[47] T. Takagi and M. Sugeno, “Fuzzy identification of systems and its application to modeling and control,” IEEE Trans. Syst. Man, Cybern., vol. 15, pp. 116-132, 1985.
[48] K. Liano, “A robust approach to supervised learning in neural network," ICNN'94, 1, pp. 513-516, 1994.
[49] F. Mokhtarian and A. Mackworth, "Scale-based description and representation of planar curves and Two-dimensional shapes," IEEE Trans. Patt. Anal. Machine Intell., Vol. PAMI-8. No. 1, pp.34-43, 1986.
[50] N. Ansari and E. J. Delp, "On detecting dominant points," Pattern Recognition, Vol. 24, No. 5, pp.441-451, 1991.
[51] Microsoft Co., Inter Locator Service, MSDN Library, Microsoft Co., Oct. 2000.
[52] Microsoft Co., IP Telephony with TAPI 3.0 White Paper, Microsoft Co., 1999.
[53] Microsoft Co., Telephony Application Programming Interface, MSDN Library, Microsoft Co., Oct. 2000.
[54] S. Greenberg, C. Gutwin, M. Roseman, “Semantic telepointers for groupware,” Proc. 16th Computer-Human Interaction Conference, Australian, pp. 54-61, 1996.
[55] H. Balakrishnan, H. S. Rahul, and S. Seshan, “An Integrated Congestion Management Architecture for Internet Hosts,” in Proc. ACM SIGCOMM, p.p. 175-187, Sep 1999.
[56] N. Laoutaris and I. Stavrakakis, “Intrastream synchronization for continuous media stream: a survey of playout schedulers,” IEEE Network, vol. 16, no. 3, pp. 30-40, May-Jun 2002.
[57] A. Zhang, Y. Song, and M. Mieike, “NetMedia: streaming multimedia presentations in distributed environments,” IEEE Multimedia, vol. 9, no. 1, Jan-Mar 2002.
[58] Microsoft Co., Microsoft DirectX, Available: http://microsoft.com/windows/directx/.
[59] M. Miller and M. Copper, “Security considerations for present and future medical databases,” Int. J. Bio-Med. Comput., vol. 41, pp. 39-46. 1996.
[60] S. Bengtsson and B. B. Solheim, “Enforcement of data protection privacy and security in medical informatics,” Proc. MEDINFO’92, Geneva, Switzerland: North-Holland, pp. 1561-1565, 1992.
[61] L. Makris, N. Argiriou, and M. G. Strintzis, “Network access and data security design for telemedicine applications,” Med. Informat. J., vol. 22, no. 2, pp. 133-142, 1997.
[62] NBS, “Data encryption standard, Nat. Bureau Standards,” U.S. Dept. Commerce, Washington, DC, NBS FIPS PUB 46, 1977.
[63] R. L. Rivest, A. Shamir, and L. M. Adleman, “A method for obtaining digital signatures and public-key cryptosystems,” Commun. ACM., vol. 21, no. 2, pp. 120-126, 1978.
[64] G. Borgefors, “An improved version of the chamfer matching algorithm,” Proc. Int. Conf. Pattern Recognition, 1984.
[65] L. Allen et al., “GBS IP multicast tunneling,” Proc. 21st Century Military Communications Conference, vol. 2, pp. 699-703, 2000.