臨床上，椎弓足螺絲失效主要為斷裂及鬆脫，故設計擁有強抗彎強度及抗拉強度的螺絲為一個重要的課題。本研究利用田口品質法及神經遺傳法搜尋最佳的螺絲設計並探討螺絲參數的影響程度。建構L25的直交表，並用有限元素法模擬螺絲強度，並分別找到其最佳的設計。本研究以變異度分析計算各螺絲參數的貢獻度為標準，探討兩種以神經遺傳法計算參數貢獻度的方法(即:profile method 和 connection weight)之準確度。結果顯示，神經遺傳法擁有比田口方法找到較佳的螺絲設計能力;並利用connection weights 計算出與變異度分析相近的解。

Improper design of pedicle screw may seriously affect the bending strength which resists breakage and pullout strength which resists loosening. With the increasing importance of optimization method which has been sprung up these decades in engineering field, many methods were evolved. One of conventional approaches to find optimum is Taguchi method which is applied widely and successfully. Another approach utilized in this study is Neuro-Genetic which is regarded as the powerful method of optimization. Two capabilities of both methods are considered in this study. One is the capability of searching optimal solution. Another one is the capability of calculating contribution of each factor.
The study simulates the performances including bending strength and pullout strength of pedicle screws by finite element analysis (FEA). The orthogonal array L25 is constructed to study optimization by Taguchi method and calculate contribution of each factor by analysis of variation (ANOVA). Then, 6 artificial neural networks (ANNs) trained by these 25 data are inserted into the genetic algorithm (GA) individually as the fitness function. According to the predicted performances (calculated from ANN), the GA will research the optimal design from generation to generation. On the aspect of contribution, the result of ANOVA is considered as the standard. Based on the consideration, the Profile Method and the Connection Weights Approach are compared which one’s contribution is more similar to the standard by Student’s T-Test.
The results show that the bending strength of optimal design found by Neuro-Genetic method is 7.65% lower than Taguchi method; and is 3.23% higher for pullout strength. For the parameter of IP in bending case, the contribution of Connection Weights Approach is more similar to the standard than the Profile method significantly. But there are no differences for other parameters. In conclusions, the study can optimize the design by the Neuro-Genetic method, beyond doubt. In addition, the contribution can be calculated by the Connection Weights Approach in place of the ANOVA.

[1] C.S. Chen, W.J. Chen, C.K. Cheng et al., "Failure Analysis of Broken pedicle screws on spinal instrumentation," Medical Engineering & Physics, vol. 27, pp. 487-496, 2005
[2] L.L. Stambough, F.E. Khatib, A.M. Genaidy, R.L. Huston, "Strength and Fatigue Resistance of Thoracolumbar Spine Implants: an Experimental Study of Selected Clinical Devices," Journal of Spinal Disorders & Techniques, vol. 12, pp. 410-414, 1999
[3] S.D. Cook, S.L. Salkeld, T. Stanley et al., "Biomechanical Study of Pedicle Screw Fixation in Severely Osteoporotic Bone," The Spine Journal, vol. 4, pp. 402-408, 2004
[4] B.W. Cunningham, J.C. Sefter, Y. Shono, P.C. McAfee, "Static and Cyclical Biomechanical Analysis of Pedicle Screw Spinal Constructs," Spine, vol. 25, pp. 1-12, 2000
[5] P.Q. Chen, S.J. Lin, S.S. Wu, H. So, "Mechanical Performance of the New Posterior Spinal Implant: Effect of Materials, Connecting Plant, and Pedicle Screw Design," Spine, vol. 28, pp. 881-886, 2003
[6] M.R. Mikles, F.A. Asghar, E.P. Frankenburg et al., "Biomechanical Study of Lumbar Pedicle Scres in a Corpectomy Model Assessing Significance of Screw Height," Journal of Spinal Disorders & Techniques, vol. 17, pp. 272-276, 2004
[7] C.C. Hsu, C.K. Chao, J.L. Wang, S.M. Hou, Y.T. Tsai, and J. Lin, “Increase of Pullout Strength of Spinal Pedicle Screws with Conical Core: Biomechanical Tests and Finite Element Analyses,” Journal of Orthopedic Research, vol.23, pp. 788-794, 2005
[8] E. Barrado, M. Vega, R. Paedo et al., "Optimization of a Purification Method for Metal-containing Wastewater by Use of a Taguchi Experimental Design," Water Research, vol.30 pp. 2309-2314, 1996
[9] A.R. Khoei, I. Masers, D.T. Gethin, "Design Optimization of Aluminium Recycling Processes Using Taguchi Technique," Journal of Materials Processing Technology, vol. 127, pp. 96-106, 2002
[10] J.Z. Zhang, J.C. Chen, E.D. Kirby, "Surface Roughness Optimization in an End-milling Operation Using the Taguchi Design Method," Journal of Materials Processing Technology, vol. 184, pp. 233-239, 2007
[11] A. Adnani, M. Basri, E.A. Malek et al., "Optimization of Lipase- catalyzed Synthesis of Xylitol Ester by Taguchi Robust Design Method," Industrial Crops and Products, vol. 31, pp. 350-356, 2010
[12] C.T. Su and T.L. Chiang, “Optimizing the IC Wire Bonding Process Using a Neural Networks/Genetic Algorithms Approach,” Journal of Intelligent Manufacturing, vol. 14, pp. 229-238, 2003
[13] Y.M. Chiang, H.M. Chiang, Y.W. Cheng, "Wire Bonding Parameters Optimization and Yield Enhancement for Silicon Rectification Diode," The Dissertation from Department of Industrial Engineering and Management, 2006
[14] U. Esme, "Application of Taguchi Method for the Optimization of Resistance Spot Welding Process," The Arabian Journal for Science and Engineering, vol. 34, pp. 519-528, 2009
[15] W.G. Baxt, "Use of an Artificial Neural Network for the Diagnosis of Myocardial Infarction," Annals of Internal Medicine, vol. 115, pp. 843-848, 1991
[16] M. Binder, H. Kittler, A. Seeber et al., "Epiluminescence Microscopy-based Classification of Pigmented Skin Lesions Using Computerized Image Analysis and an Artificial Neural Network," Melanoma Research, vol. 8, pp. 261-266, 1998
[17] K.Y. Lee, Y.T. Cha, J. H. Park, "Short-term Load Forecasting Using an Artificial Neural Network," IEEE Transactions on Power Systems, vol. 7, pp. 124-132, 2002
[18] M.L.D. Wong, A.K. Nandi, "Automatic Digital Modulation Recognition Using Artificial Neural Network and Genetic Algorithm," Signal Processing, vol. 84, pp. 351-365, 2004
[19] B. Ozcelik, H. Oktem, H. Kurtaran, "Optimum Surface Roughness in End Milling Inconel 718 by Coupling Neural Network Model and Genetic Algorithm," The International Journal of Advanced Manufacturing Technology, vol. 27, pp. 234-241, 2005
[20] K. Panneerselvam, S. Aravindan, A.N. Haq, "Hybrid of ANN with Genetic Algorithm for Optimization of Frictional Vibration Joining Process of Plastics," The International Journal of Advanced Manufacturing Technology, vol. 42, pp. 669-677, 2009
[21] J.A. Ghani, I.A. Choudhury, H.H. Hassan, “Application of Taguchi Mthod in the Optimization of end milling parameters,” Journal of Materials Processing Technology, vol. 145, pp. 84-92, 2004
[22] N.N. Luong, “Design Optimization Study of Orthopedic Screw by Taguchi Method and Neurogenetic Method, ” in Department of Mechanical Engineering, NTUST, pp. 106, 2011
[23] M.H. Pope and M. Panjabi, “Biomechanical definitions of spinal instability,” Spine, vol. 10, pp. 255-256, 1985
[24] C. Shen, L. Wang, and Q. Li, “Optimization of Injection Molding Process Parameters Using Combination of Artificial Neural Network and Genetic Algorithm Method,” Journal of Materials Processing Technology, vol.183, pp.412-418, 2007
[25] D. Venkatesan, K. Kannan, R. Saravanan, “A Genetic Algorithm-Based Artificial Neural Network Model for the Optimization of Machining Process,” Neural Computing & Applications, vol. 18, pp. 135-140, 2009
[26] M. Negnevitsky, “Artificial Intelligence: A Guide to Intelligent Systems,” 2nd Edition, Pearson Education, Inc., 2010
[27] M. Gevrey, “Review and Comparison of Methods to Study the Contribution of Variables in artificial neural network models,” Ecological Modeling, vol. 160, pp. 249-264, 2003
[28] S. Lek, A. Belaud, P. Baran, I. Dimopoulos, and M. Delacoste, “Role of Some environmental Variables in Trout Abundance Models using Neural Networks,” Aquatic Living Resources, vol. 9, pp. 23-29, 1996
[29] J.D. Olden, M.K. Joy, and R.G. Death, “An Accurate Comparison of Mthods for Quantifying Variable Importance in Artificial Neural Networks using Simulated Data,” Ecological Modeling, vol. 178, pp. 389-397, 2004
[30] J.D. Olden and D.A. Jackson, “Illuminating the “Black Box”: a Randomization Approach for Understanding Variable Contributions in Artificial Neural Networks,” Ecological Modeling, vol. 154, pp. 135-150, 2002
[31] B. Ozcelik and T. Erzurumlu, “Comparison of the Warpage Optimization in the Plastic Injection Molding using ANOVA, Neural Network Model and Genetic Algorithm,” Journal of Materials Processing Technology, vol. 171, pp. 437-445, 2006
[32] J.J. Sheu and C.Y. Chen, “An Integration Method of Artificial Neural Network and Genetic Algorithm for Structure Design of a Scooter,” Intelligent Computing, vol. 4113, pp. 655-662, 2006
[33] M. Aitan, “Reducing Shrinkage in Injection Modings via the Taguchi, ANOVA and Neural Network Methods,” Materials & Design, vol. 31, pp. 599-604, 2010
[34] X.R. Cui, M.F. Abbod, Q. Liu, J.S. Shieh, T.Y. Chao, C.Y. Hsieh, Y.C. Yang, “Ensembled Artificial Neural Networks to Predict the Fitness Score for Body Composition Analysis,” The Journal of Nutrition, Health & Aging, vol. 15, pp. 341-348, 2011
[35] 張傑富，「腰椎椎弓足螺絲之多目標最佳化設計：有限元素分析與生物力學測試」，碩士論文，國立台灣科技大學，台北，2010
[36] C.C. Hsu, C.K. Chao, J.L. Wong, and J. Lin, “Multiobjective Optimization of Tibial Locking Screw Design using a Genetic Algorithm: Evaluation of Mechanical Performance,” Journal of Orthopaedic Research, vol.24, pp.908-916, 2006
[37] G.R. Fogel, C.A. Reitman, W. Liu, and S.I. Esses, “Physical Characteristics of Polyaxial-Headed Pedicle Screws and Biomechanical Comparison of Load with Their Failure,” Spine, vol.28, pp.470-473, 2003
[38] T.A. Tawfik, W.S. Reginald, “Employing the Taguchi Method in Optimizing the Scaffold Production Process for Artificial Bone Grafts,” Journal of Materials Processing Technology, vol. 209, pp. 1521-1532, 2009