Search

We evaluated the midterm outcomes of lifting posterior mitral annuloplasty for enhancing leaflet coaptation in mitral valve repair.Between October 2007 and December 2012, 341 consecutive patients with significant mitral regurgitation underwent lifting posterior mitral annuloplasty using a specially designed fabric annuloplasty strip that lifts the middle portion of the posterior annulus. Associated procedures for mitral valve repairs, such as patch valvuloplasty for posterior leaflet prolapse (n=80), new chord placement for anterior leaflet prolapse (n=33), commissurotomy (n=29), and posterior leaflet extension (n=23), were performed in 141 patients (41.3%).Thirty-day mortality was 0.9%. Nine late deaths (2.6%) occurred. Mean overall survival at 5 years was 96.0%±1.1%. During the mean follow-up period of 38±17 months, six patients (1.8%) underwent valve-related reoperation (5-year freedom from valve-related reoperation, 98.1%±0.8%). At 5 years, mean freedom from recurrence of mitral regurgitation grade 3+ to 4+ (moderate to severe) was 95.1%±1.6%. The mean valve pressure gradient (PG) was 3.2±1.5 mmHg across all strip sizes at the time of follow-up.Lifting posterior mitral annuloplasty using an innovative annuloplasty strip in mitral valve repair has a low rate of recurrent regurgitation or valve-related reoperation with rare relevant complications.

In the semiconductor manufacturing process, large amounts of data are collected from various sensors of multiple facilities. The collected data from sensors have several different characteristics due to variables such as types of products, former processes and recipes. In general, Statistical Quality Control (SQC) methods assume the normality of the data to detect out-of-control states of processes. Although the collected data have different characteristics, using the data as inputs of SQC will increase variations of data, require wide control limits, and decrease performance to detect outof- control. Therefore, it is necessary to separate similar data groups from mixed data for more accurate process control. In the paper, we propose a regression tree using split algorithm based on Pearson distribution to handle non-normal distribution in parametric method. The regression tree finds similar properties of data from different variables. The experiments using real semiconductor manufacturing process data show improved performance in fault detecting ability.

In manufacturing industries, development of measurement leads to increase the number of monitoring variables and eventually the importance of multivariate control comes to the fore. Statistical process control (SPC) is one of the most widely used as multivariate control chart. Nevertheless, SPC is restricted to apply in processes because its assumption of data as following specific distribution. Unfortunately, process data are composed by the mixture of several processes and it is hard to estimate as one certain distribution. To alternative conventional SPC, therefore, nonparametric control chart come into the picture because of the strength of nonparametric control chart, the absence of parameter estimation. SVDD based control chart is one of the nonparametric control charts having the advantage of flexible control boundary. However,basic concept of SVDD has been an oversight to the important of data characteristic, density distribution. Therefore, we proposed DW-SVDD (Density Weighted SVDD) to cover up the weakness of conventional SVDD. DW-SVDD makes a new attempt to consider dense of data as introducing the notion of density Weight. We extend as control chart using new proposed SVDD and a simulation study of various distributional data is conducted to demonstrate the improvement of performance., {"references":["H. Hotelling, Multivariate Quality Control. McGraw-Hill, 1947.","D. C. Montgomery, Introduction to statistical quality control. New\nYork Singapore : John Wiley, 1985.","S. T. Bakir, \"Distribution-free quality control charts based on signedrank-\nlike statistics,\" Communications in Statistics - Theory and Methods,\nvol. 35, no. 4, pp. 743-757, 2006.","S. Bersimis, S. Psarakis, and J. Panaretos, \"Multivariate statistical process\ncontrol charts: An overview,\" Quality and Reliability Engineering\nInternational, vol. 23, no. 5, pp. 517-543, 2007.","D. M. J. Tax and R. P. W. Duin, \"Support vector domain description,\"\nPattern Recognition Letters, vol. 20, no. 11-13, pp. 1191-1199, 1999.","K. Lee, D. W. Kim, K. H. Lee, and D. Lee, \"Density-induced support\nvector data description,\" IEEE Transactions on Neural Networks, vol. 18,\nno. 1, pp. 284-289, 2007.","D. J. Tax and R. W. Duin, \"Support vector data description,\" Machine\nLearning, vol. 54, no. 1, pp. 45-66, 2004.","J. Shawe-Taylor and N. Cristianini, Kernel methods for pattern analysis.\nCambridge University Press, 2004.","M. M. Breunig, H. P. Kriegel, R. T. Ng, and J. Sander, \"Lof: Identifying\ndensity-based local outliers,\" Sigmod Record, vol. 29, no. 2, pp. 93-104,\n2000.\n[10] R. Sun and F. Tsung, \"A kernel-distance-based multivariate control\nchart using support vector methods,\" International Journal of Production\nResearch, vol. 41, no. 13, pp. 2975-2989, 2003.\n[11] S. Kumar, A. K. Choudhary, M. Kumar, R. Shankar, and M. K. Tiwari,\n\"Kernel distance-based robust support vector methods and its application\nin developing a robust k-chart,\" International Journal of Production\nResearch, vol. 44, no. 1, pp. 77-96, 2006.\n[12] T. Sukchotrat, S. B. Kim, and F. Tsung, \"One-class classification-based\ncontrol charts for multivariate process monitoring,\" IIE Transactions,\nvol. 42, no. 2, pp. 107-120, 2010.\n[13] B. Efron and R. J. Tibshirani, An Introduction to the Bootstrap. Chapman\nand Hall, New York, 1993.\n[14] J. Friedman, T. Hastie, and R. Tibshirani, The Elements of Statistical\nLearning : Data Mining, Inference, and Prediction, ser. Springer Series\nin Statistics. New York, NY: Springer-Verlag New York, 2009.\n[15] B. Liu, Y. Xiao, L. Cao, Z. Hao, and F. Deng, \"Svdd-based outlier\ndetection on uncertain data,\" Knowledge and Information Systems, pp.\n1-22, 2012."]}

It is important to predict yield in semiconductor test process in order to increase yield. In this study, yield prediction means finding out defective die, wafer or lot effectively. Semiconductor test process consists of some test steps and each test includes various test items. In other world, test data has a big and complicated characteristic. It also is disproportionably distributed as the number of data belonging to FAIL class is extremely low. For yield prediction, general data mining techniques have a limitation without any data preprocessing due to eigen properties of test data. Therefore, this study proposes an under-sampling method using support vector machine (SVM) to eliminate an imbalanced characteristic. For evaluating a performance, randomly under-sampling method is compared with the proposed method using actual semiconductor test data. As a result, sampling method using SVM is effective in generating robust model for yield prediction., {"references":["K. Kim, C.-G. Hwang, and J. G. Lee, \"Dram technology perspective for\ngigabit era,\" Electron Devices, IEEE Transactions on, vol. 45, no. 3, pp.\n598 -608, mar 1998.","P. Burggraaf, \"New roadmap unveiled-2000 begins with a revised\nindustry roadmap-a revised 1999 international technology roadmapfor\nsemiconductors identifies future challenges to the evolution of semiconductor,\"\nSolid State Technology, vol. 43, no. 1, pp. 31-48, 2000.","B. Ciciani and G. Iazeolla, \"A markov chain-based yield formula for\nvlsi fault-tolerant chips,\" Computer-Aided Design of Integrated Circuits\nand Systems, IEEE Transactions on, vol. 10, no. 2, pp. 252-259, 1991.","D. An, H. Ko, T. Gulambar, J. Kim, J. Baek, and S. Kim, \"A semiconductor\nyields prediction using stepwise support vector machine,\" in\nAssembly and Manufacturing, 2009. ISAM 2009. IEEE International\nSymposium on. IEEE, 2009, pp. 130-136.","G. Verdier and A. Ferreira, \"Adaptive mahalanobis distance and¡ formula\nformulatype=,\" Semiconductor Manufacturing, IEEE Transactions on,\nvol. 24, no. 1, pp. 59-68, 2011.","K. Kerdprasop and N. Kerdprasop, \"A data mining approach to automate\nfault detection model development in the semiconductor manufacturing\nprocess.\"","N. Chawla, K. Bowyer, L. Hall, and W. Kegelmeyer, \"Smote: synthetic\nminority over-sampling technique,\" arXiv preprint arXiv:1106.1813,\n2011.","X. Liu, J. Wu, and Z. Zhou, \"Exploratory undersampling for classimbalance\nlearning,\" Systems, Man, and Cybernetics, Part B: Cybernetics,\nIEEE Transactions on, vol. 39, no. 2, pp. 539-550, 2009.","V. Vapnik, The nature of statistical learning theory. springer, 1999.\n[10] E. Osuna, R. Freund, and F. Girosit, \"Training support vector machines:\nan application to face detection,\" in Computer Vision and Pattern Recognition,\n1997. Proceedings., 1997 IEEE Computer Society Conference on.\nIEEE, 1997, pp. 130-136.\n[11] S. Dumais, J. Platt, D. Heckerman, and M. Sahami, \"Inductive learning\nalgorithms and representations for text categorization,\" in Proceedings\nof the seventh international conference on Information and knowledge\nmanagement. ACM, 1998, pp. 148-155.\n[12] T. Joachims, \"Text categorization with support vector machines: Learning\nwith many relevant features,\" Machine learning: ECML-98, pp. 137-\n142, 1998.\n[13] B. Schlkopf and A. Smola, \"Learning with kernels: Support vector\nmachines, regularization, optimization, and beyond,\" 2002.\n[14] R. Barandela, J. S'anchez, V. Garc'ıa, and E. Rangel, \"Strategies for\nlearning in class imbalance problems,\" Pattern Recognition, vol. 36,\nno. 3, pp. 849-851, 2003."]}