本論文的研究動機是來自於一組肺高壓嚴重分級的診斷問題，因為偵測肺高壓的黃金標準是需要侵入性的檢查，如果能夠運用替代且非侵入性的診斷工具，尋找最佳切點來進行分群，將可嘉惠患者。因此，本研究提出針對多組的連續型分群資料的統計方法，在已知有序連續型資料且已知可分成若干群之下，我們發展在不同的母體分配假設下，運用最大化尤登指數的概念來尋找最佳切點。接著，本研究也利用蒙地卡羅模擬，來探討不同分布假設下切點估計的表現。最後，為了讓容易應用在實際臨床診斷上，設計出一個shiny 互動式介面，利用肺高壓嚴重分級當成例子說明提出方法的應用。

Many important clinical problems more than two category outcomes, for example, investigating the severity of pulmonary hypertension (PH). Luo and Xiong (2013) developed the Youden index to assess diagnostic accuracy when there are three ordinal diagnostic groups and presented the parametric and nonparametric methods to estimate the optimal Youden index, as well as the underlying optimal cut-points and the associated confidence intervals. This thesis extends the concept of Luo and Xiong. We purpose the statistical method to estimate Youden index and calculate the optimal cut-points based on the pre-specified distribution, normal, gamma, lognormal assumption, for multiple groups. Moreover, a Shiny interface is designed, taking the severity of PH as an example to illustrate the application of the proposed method.

Chang, W., Cheng, J., Allaire, J., Xie, Y., McPherson, J., et al. Shiny: web application framework for r. R Package Version 1, 5 (2017), 2017. [https://CRAN.Rproject.org/package=shiny].
Chemla, D., Castelain, V., Herve, P., Lecarpentier, Y., and Brimioulle, S. Haemodynamic evaluation of pulmonary hypertension. Eur Respir J 20, 5 (2002), 1314–1331.
Galiè, N., McLaughlin, V. V., Rubin, L. J., et al. An overview of the 6th world symposium on pulmonary hypertension. Eur Respir J, 53 (2019). 1802148 [https://doi.org/10.1183/13993003.021482018].
Hanley, J. A., and McNeil, B. J. The meaning and use of the area under a receiver operating characteristic (roc) curve. Radiology 143, 1 (1982), 29–36.
Harrell, F. E., Califf, R. M., Pryor, D. B., Lee, K. L., and Rosati, R. A. Evaluating the yield of medical tests. JAMA 247, 18 (1982), 2543–2546.
Luo, J., and Xiong, C. Youden index and associated cutpoints for three ordinal diagnostic groups. Communications in StatisticsSimulation and Computation 42, 6 (2013), 1213–1234.
Nakas, C. T., DalrympleAlford, J. C., Anderson, T. J., and Alonzo, T. A. Generalization of youden index for multipleclass classification problems applied to the assessment of externally validated cognition in parkinson disease screening. Statistics in Medicine 32, 6 (2013), 995–1003.
Nakas, C. T., and Yiannoutsos, C. T. Ordered multipleclass roc analysis with continuous measurements. Statistics in Medicine 23, 22 (2004), 3437–3449.
Pandya, R., and Pandya, J. C5. 0 algorithm to improved decision tree with feature selection and reduced error pruning. International Journal of Computer Applications 117, 16 (2015), 18–21.
Pepe, M. S., et al. The statistical evaluation of medical tests for classification and prediction. (2003). Oxford: Oxford University Press.
Perkins, N. J., and Schisterman, E. F. The inconsistency of“optimal＂cutpoints obtained using two criteria based on the receiver operating characteristic curve. American journal of epidemiology 163, 7 (2006), 670–675.
Scurfield, B. K. Multipleevent forcedchoice tasks in the theory of signal detectability. Journal of Mathematical Psychology 40, 3 (1996), 253–269.
Van Calster, B., Van Belle, V., Vergouwe, Y., and Steyerberg, E. W. Discrimination ability of prediction models for ordinal outcomes: relationships between existing measures and a new measure. Biometrical Journal 54, 5 (2012), 674–685.
Youden, W. J. Index for rating diagnostic tests. Cancer 3, 1 (1950), 32–35.