臺灣博碩士論文加值系統

隨著醫療的進步，許多疾病可以透過儀器或是潛在風險因子，提前知道其發生的可能性，像是糖尿病、高血壓、年紀或是其他疾病，皆為醫生參考的依據，而台灣實施全民健保已來，是少數保有完整就醫資料的國家，記錄著全國人民的就醫狀況。
本論文試圖透過全民健保資料大數據，利用機器學習方法，以貝氏logistic迴歸模型統計分析健保資料庫，將心房纖維顫動導致中風及老年失智症的風險因子，利用stan這個開放工具，使用logistic回歸模型進行統計及分析，算出每個bata常數的值，並計算出其勝算，因而知道每個風險因子導致此疾病的機率，除了探討單因子，還討論當同時擁有多項疾病時，導致中風或老年失智症的機率。
利用此方法進行推論，可以提供在醫療上，給予醫生確切的量化權重，相較於傳統的得分權重，有很大的不同。

As medical advances, many diseases can be through the instrument or potential risk factors, know in advance the likelihood of its occurrence, such as diabetes, high blood pressure, age, or other diseases, doctors are all the reference for the implementation of the National Health Insurance Taiwan It has come, a few countries to maintain complete medical treatment information, a record of medical treatment status of the people.
This paper attempts by the National Health Insurance data big data using machine learning methods to Bayesian logistic regression statistical analysis of health care database, will lead to a stroke and atrial fibrillation in elderly dementia risk factors, the use of open stan this tool, using logistic statistical analysis and regression model to calculate the value of each bata constant, and calculate the odds, and thus know the probability of each risk factor leading to this disease, in addition to explore a single factor, but also has a number of discussions when both diseases, leading to stroke or senile dementia chances.
With this method of inference, may be provided in the medical, to provide doctors with the exact quantization weighting score is compared to traditional heavy weight, they are very different.

論文審定書 i
誌謝 ii
摘要 iii
Abstract iv
目錄 v
圖次 vii
表次 viii
第一章序論 1
1.1 研究動機 1
1.2 研究目的 1
1.3 論文架構 3
第二章 Regression 4
2.1 Regression 4
2.1.1 線性迴歸 4
2.1.2 logistic regression模型 4
2.1.3分組資料的變數設定 7
2.1.4 logistic regression係數解釋 9
2.2 Bayesian logistic regression 14
2.3 Bayesian與傳統logistic regression之比較 15
第三章 研究背景 16
3.1 貝氏統計 16
3.1.1貝氏推論 16
3.1.2貝氏網路 17
3.1.3貝氏統計與傳統統計之比較 22
3.2 馬可夫鏈蒙地卡羅法(Markov chain Monte Carlo) 23
3.3 健保資料庫介紹 25
3.3.1 心房性心律不整及其風險因子 26
3.3.2 老年失智症及其風險因子 26
第四章 STAN 27
4.1 STAN 27
4.1.1 Hamiltonian Monte Carlo 27
4.1.2 no-U-turn抽樣 28
4.1.3 RSTAN 29
第五章 實驗方法 33
5.1 實驗Data 33
5.1.1 AF病患篩選結果 33
5.1.2 老年失智症病患篩選結果 35
5.2 利用rstan分析方法 37
5.2.1 AF病患rstan分析 37
5.2.2 老年失智症病患rstan分析 40
第六章 實驗結果分析與討論 44
6.1 RSTAN分析結果 44
6.1.1 AF與風險因子分析結果 44
6.1.2 老年失智症與風險因子分析結果 50
第七章 結論與未來展望 63
參考文獻 64

[1] 汪群超,The Markov Chain Monte Carlo Simulations, 2007.
[2] Markov chain Monte Carlo https://en.wikipedia.org/wiki/Markov_chain_Monte_Carlo
[3]陳定立, 馬可夫鏈蒙地卡羅法（Markov chain Monte Carlo）
中央研究院週報第1320期, 2011.
[4]Lam, Patrick. "MCMC Methods: Gibbs Sampling and the Metropolis-Hastings Algorithm." Harvard University.
[5]蔡佳泓,貝氏估計與WinBUGS在社會科學的應用,政大選舉研究中心
[6]Bradley, P. Carlin. Intermediate Bayesian Data Analysis Using WinBUGS and BRugs, 2007.
[7]詹皇鼎, 貝氏結構方程模式, 2011.
[8]G. Jay, Kerns. Introduction to Probability and Statistics Using R, 2014.
[9]Vincent,Zoonekynd. Statistics with R, pages 876-917, 2007.
[10]Chia-hung, Tsai. Bayesian Inference in Binomial Logistic Regression: A Case Study of the 2002 Taipei Mayoral Election, 2005.
[11]Lawrence, Joseph. Bayesian Inference for Linear and Logistic Regression Parameters, 2016.
[12]Stan Modeling Language User’s Guide and Reference Manual, 2015.
[13]Stan: Package “rstan”, 2016.
[14]Bob Carpenter, Andrew Gelman, Matt Hoffman, Daniel Lee, Ben Goodrich,
Michael Betancourt, Marcus A. Brubaker, Jiqiang Guo, Peter Li, and Allen
Riddell In press for Journal of Statistical Software. Stan: A Probabilistic
Programming Language, 2015.
[15]Stan : http://mc-stan.org/interfaces/rstan
[16]Andrew Gelman, Daniel Lee, and Jiqiang Guo Stan: A probabilistic
programming language for Bayesian inference and optimization. In press,
Journal of Educational and Behavior Science, 2015.
[17]RStan: the R interface to Stan The Stan Development Team, 2016.
[18]Niragire, François, et al. Bayesian Mapping of HIV Infection among Women of Reproductive Age in Rwanda. PloS one 10.3, 2015.
[19]王濟川,郭志綱. Logistic迴歸模型-方法及應用, 2010.
[20]Darwiche, Adnan. Bayesian networks. Communications of the ACM 53.12 2010: 80-90.
[21] Kai-Fang Hu, Yu-Hsiang Chou, Yen-Hsia Wen, Kun-Pin Hsieh, Jui-Hsiu Tsai,
Pinchen Yang, Yi-Hsin Yang, Chun-Hung Richard Lin (2016, Oct). Antipsychotic medications and dental caries in newly diagnosed schizophrenia: A nationwide cohort study. Psychiatry Research (SCI), 245 (2016) 45-50.
[22] Yen-Hsia Wen, Shihn-Sheng Wu, Chun-Hung Richard Lin, Jui-Hsiu Tsai,
Pinchen Yang, Yang-Pei Chang, Kuan-Hua Tseng (2016, May). A Bayesian Approach to Identifying New Risk Factors for Dementia. Medicine (Impact Factor=5.723), Volume 95, Number 21, May 2016
[23] Hoffman MD, Gelman A (2014). “The No-U-Turn Sampler: Adaptively Setting
Path Lengths in Hamiltonian Monte Carlo.” Journal of Machine Learning Research, 15(Apr), 1593–1623.
[24] Lunn D, Thomas A Best N, Spiegelhalter D (2000). “WinBUGS—A Bayesian
Modelling Framework: Concepts, Structure, and Extensibility.” Statistics and Computing, 10, 325– 337. URL www.mrc-bsu.cam.ac.uk/bugs/
[25] Neal R (2011). “MCMC Using Hamiltonian Dynamics.” In S Brooks, A Gelman,
GL Jones, XL Meng (eds.), Handbook of Markov Chain Monte Carlo, pp. 116–162. Chapman and Hall/CRC.
[26] Plummer, M. (2003). JAGS: A program for analysis of Bayesian graphical
models using Gibbs sampling. In Proceedings of the 3rd international workshop on distributed statistical computing (Vol. 124, p. 125).
[27] Ott, Alewijn, et al. "Atrial fibrillation and dementia in a population-based study
the Rotterdam Study." Stroke 28.2 (1997): 316-321.
[28] Feinberg, William M., et al. "Prevalence, age distribution, and gender of patients
with atrial fibrillation: analysis and implications." Archives of internal medicine155.5 (1995): 469-473.
[29] Fang, Margaret C., et al. "Gender differences in the risk of ischemic stroke and
peripheral embolism in atrial fibrillation the anticoagulation and risk factors in atrial fibrillation (ATRIA) study." Circulation 112.12 (2005): 1687-1691.
[30] Ott, Alewijn, et al. "Diabetes mellitus and the risk of dementia The Rotterdam
Study." Neurology 53.9 (1999): 1937-1937.
[31] Azad, Nahid A., Muneerah Al Bugami, and Inge Loy-English. "Gender
differences in dementia risk factors." Gender Medicine 4.2 (2007): 120-129.