臺灣醫院資料庫的儲存量是相當龐大的，資料也非常多元並且複雜，而這些資料並非所有資料都被需要，把大資料使用個別獨立分析的小型資料集，然後再將各個資料群集合併進行分析，找出資料的關聯性以及從而尋找出隱藏在大型資料庫中的有價值的資訊。
    在本次研究之中，整合主動脈瘤與周邊動脈阻塞性疾病兩種疾病，依照疾病的不同建立相對應的資料表，醫護人員能夠利用數據分析管理系統更簡單的紀錄病患的病歷資料，減少醫護人員的工作量使醫護人員能夠運用更多的時間在照護病人上面，並且醫護人員能夠利用本研究所建立的數據分析管理系統，進行相對應疾病的個別分析與雙疾病交叉分析，根據疾病與疾病所記錄相同的資料項目進行交叉分析並產生分析圖表，醫護人員則能夠觀察圖表找出資料與資料之間的關聯性，進而獲取隱藏在資料背後的資訊。

Taiwan hospital database is very large. Data which stored in the hospital database is also very diverse and complex, and these data are not all data are needed. The individual data set will be analyzed in the big data, and the data set will be integrated and then analyze again to find the relationship and the hidden knowledge in the database.        
      In the research, we integrated the aortic aneurysm with peripheral arterial occlusive disease in the database, and established the corresponding data table with the different diseases.  Medical staff can use the data analysis and management system to record medical data of patients more easily, and also reduce the workload of medical staff. Medical staff can spend more time in caring for patients. Medical staff can make the use of the data analysis and management system established by the Institute to do the individual analysis of the corresponding diseases and the cross analysis of the two diseases. According to the different diseases recorded in the same data item cross analysis and chart analysis, the medical staff can observe the charts to identify correlation between data and data, and then obtain the information hidden in the data behind. Medical staff observe and analyze charts to obtain information hidden behind the data.

Chapter 1 Introduction	1
1.1 Background	1
1.2 Motivation	2
1.3 Purpose	3
Chapter 2 Literature Review	4
2.1 Cloud Clinical Diagnostics Aid System of Division of Cardiovascular Surgery	4
2.2 Data Mining	6
2.2.1 Introduction to data mining	6
2.2.2 Knowledge Discovery in Database	6
2.2.3 Data Mining Techniques	9
2.2.4 Data Mining Approach	10
2.2.5 Application of data mining in the medical	20
2.3 Aortic aneurysm	20
2.3.1 Introduction of the Aortic aneurysm	20
2.3.2 Classification of aortic aneurysms	22
2.4 Peripheral Arterial Occlusive Disease	24
2.4.1 Fontaine Classification	25
2.4.2 Clinical diagnosis and treatment	26
Chapter 3 Method	28
3.1 Research Process	28
3.2 System Architecture	30
3.2.1 Development Environment	30
3.2.2 Design of the System Architecture	32
3.2.3 Design of system model and table of database	33
3.3 System Analysis	36
Chapter 4 Results and case analysis	40
4.1 System establishment	40
4.1.1 Data input and modification module	40
4.1.2 Database search module	44
4.1.3 Analysis module	50
4.1.4 Export module	56
4.2 Case analysis	56
4.2.1 Overall data analysis	57
4.2.2 Preoperative and postoperative track analysis	61
4.2.3 PAOD/AA data analysis	68
Chapter 5 Conclusions	71
5.1 Conclusions	71
5.2 Future work	72
Reference	74

List of Figures
Figure 1. KDD Process	8
Figure 2 Apriori flow chart	15
Figure 3 the flow charts of genetic algorithm	18
Figure 4 K-means flow chart	19
Figure 5 Guidelines for the diagnosis of ESC aortic disease, 2014	22
Figure 6 Process of the system development	30
Figure 7 three tiered network architecture	33
Figure 8 DB table of the AAA	35
Figure 9 DB table of the DAA	35
Figure 10 DB Table of the TAA	36
Figure 11 DB Table of the PAOD	36
Figure 12 Process of user’s operating	39
Figure 13 Select disease type and method of the medical record input (AAA as an example)	42
Figure 14 medical record input	42
Figure 15 Enter the date of operation and establish a medical record	42
Figure 16 Enter medical records (take AAA as an example)	43
Figure 17 Medical record number search engine	45
Figure 18 the medical record result of the search sample ID	46
Figure 19 select the disease (make DAA as a sample)	46
Figure 20 the list of result search of the 2 times or more	47
Figure 21 select condition items (make PAOD as a sample)	48
Figure 22 Enter the range value of the item(PAOD)	48
Figure 23 the result of the search	48
Figure 24 select date range and the type of disease (make DAA as the sample)	49
Figure 25 the result of the search	50
Figure 26 select disease and data type (make AAA as the sample)	51
Figure 27 select analysis condition (make AAA as the sample)	52
Figure 28 enter the data condition of the search	52
Figure 29 select data attribute	52
Figure 30 the relationship between WBC and age	52
Figure 31 select disease and the analysis object	54
Figure 32 select patient data (individual patient analysis)	54
Figure 33 select data attribute	54
Figure 34 result chart	55
Figure 35  PAOD and AA data analysis	56
Figure 36 export interface	56
Figure 37 Age and WBC line chart	59
Figure 38 Age and CHOL line chart	60
Figure 39 WBC average values before surgery and after surgery (straight bar chart)	64
Figure 40 Platelet values before and after surgery (straight bar chart)	65
Figure 41 WBC data before and after surgery (line chart)	66
Figure 42 Platelet value before and after surgery (line chart)	67
Figure 43 PAOD and AAA's Age/WBC line chart	69
Figure 44 PAOD/AAA’s age and triglyceride line chart	70

List of Table
Table 1. KDD Process	7
Table 2 Data mining Techniques	9
Table 3 Fontaine classification	25
Table 4 Data Item of the medical record	43

[1] 余博淵, ”關聯式法則在主動脈瘤臨床病歷上的應用 以主動脈內人工血管支架置放手術為例”,碩士,資訊工程學系碩士班,淡江大學,2011,1-77.
[2] 陳毅展,”心臟血管外科雲端臨床輔助診斷系統”,碩士,資訊工程學系碩士班,淡江大學,新北市,2015
[3] 黃勝崇,”資料探勘應用於醫療院所輔助病患看診指引之研究”,碩士,資訊管理學系碩士班,南華大學,嘉義縣,2001
[4] 吳淑儀,”應用資料探勘技術於多重死因資料之疾病關聯分析”,碩士,公共衛生研究所,國立成功大學,臺南市,2008
[5] 黃雅芳,”運用資料探勘方法評估冠狀動脈疾病之研究”,碩士,醫務管理學研究所碩士在職專班,高雄醫學大學,高雄市,2009
[6] 陳章友,”類神經網路在醫學檢驗的應用──以肝病為例”,碩士,工業工程與管理系,國立交通大學,新竹市,2000
[7] 羅玉惠,”以類神經網路做系統的控制及錯誤診斷”,碩士,控制工程研究所,國立交通大學,新竹市,1992
[8] 陳俊龍,”以資料探勘技術建立病患就醫導引－以胃腸科病患為例”,碩士,資訊管理系,南台科技大學,臺南市,2004
[9] 蔡明哲,”運用資料探勘法探討台灣老年人口全民健保醫療資源利用之研究”,碩士,資訊管理學系,國立中央大學,桃園市,2013
[10] 李文瑞,”運用基因演算法建構糖尿病前期診斷模型之研究”,碩士,資訊管理學系,輔仁大學,新北市,2004
[11] 楊正宏,何長軒,温政浩,張學偉,and莊麗月,”基因演算法應用於骨質疏鬆症之不同染色體基因多型性群之關聯性分析”,國立金門技術學院,金門,2006
[12] William J. Frawley, Gregory Piatetsky-Shapiro, Christopher J. Matheus ,“Knowledge Discovery in Databases: An Overview”
[13] Agrawal R. and Srikant R., ”Fast Algorithms for Mining Association Rules”, IBM Almaden Research Center 650 Harry Road, San Jose, CA 95120
[14] Chen CK, Chang HT, Chen YC, Chen TJ, Chen IM, Shih CC*. Surgeon elective abdominal aortic aneurysm repair volume and outcomes of ruptured abdominal aortic aneurysm repair: A 12-year nationwide study. World journal of surgery. 2013;37:2360-2371
[15]  Lee CW, Sung SH, Chen CK, Chen IM, Cheng HM, Yu WC, Shih CC, Chen CH. Measures of carotid-femoral pulse wave velocity and augmentation index are not reliable in patients with abdominal aortic aneurysm. J Hypertens. 2013 Jun 7.

[16] Sung SH, Wu TC, Chen JS, Chen YH, Huang PH, Lin SJ, Shih CC*, Chen JW*. Reduced number and impaired function of circulating endothelial progenitor cells in patients with abdominal aortic aneurysm. Int J Cardiol. 2013;168:1070-1077 
[17] Chen IM, Chang HH, Hsu CP, Lai ST, Shih CC*.  10-year Experience with Surgical Repair of Mycotic Aortic Aneurysms  J Chin Med Assoc 2005;68(6):265–271.
[18] Kan CB, Wang JS, Yu TJ, Huang CH, Shih CC, Yung MC, Lai ST.  Is combined abdominal aortic aneurysm repair and coronary artery bypass grafting feasible ? Chung Hua I Hsueh Tsa Chih- Chinese Medical Journal  2002;65(1):29-33.
[19] Wei-Lung Tseng, Hsu-Hung Tseng, Chiu Yang Lee. Sigmoid colon perforation following abdominal aortic aneurysm surgery. Formosan Journal of surgery.2009.Vol
[20] 資料探勘-維基百科
https://zh.wikipedia.org/wiki/%E6%95%B0%E6%8D%AE%E6%8C%96%E6%8E%98#cite_note-acm-1
[21] ASP.net-維基百科  
https://zh.wikipedia.org/wiki/ASP.NET