臺灣博碩士論文加值系統

心血管疾病（Cardiovascular diseases，CVD）根據世界衛生組織統計[1] ，是目前世界上致死率最高的疾病，在2016年有1790萬的人死於CVD，總共占了全球死亡總數的31%；在本研究中，使用了人工智慧（Artificial Intelligence，AI）來對CVD進行建模，將CVD中的關鍵診療數據-心電圖（Electrocardiography，ECG）輸入到AI模型中進行分析。在使用類神經網路（Artificial Neural Network、ANN）來進行建模後，相關實驗結果進一步藉由交叉比對驗證法對患有CVD的患者數據進行分析以及偵測；最終實驗數據顯示：本研究CVD檢測的精度達到了96%。

According to the statistics of the World Health Organization [1] , Cardiovascular Disease (CVD) is currently the disease with the highest fatality rate in the world. In 2016, 17.9 million people died of CVD, accounting for 31% of the total number of global deaths. In this study, Artificial Intelligence (AI) was used to model CVD, and the key diagnosis and treatment data in CVD - Electrocardiogram (ECG) was input into the AI model for further analysis, and the cross-validation method was applied. Simulation results positively shown that, by applying artificial neural network (ANN) to model the corresponding datasets, the accuracy of the proposed model for CVD detection can reach 96%.

目錄
摘要 ........................................................................................................................ I
Abstract ............................................................................................................... II
目錄 ..................................................................................................................... III
表目錄 .................................................................................................................. V
圖目錄 .............................................................................................................. VIII
第一章 緒論 ......................................................................................................... 1
1.1 研究背景 ..................................................................................................... 1
1.2 研究動機以及目的 ..................................................................................... 1
1.3 研究總結 ..................................................................................................... 2
第二章 心血管疾病 ............................................................................................. 4
2.1 心血管疾病（CVD） ................................................................................ 4
2.2 心電圖（Electrocardiogram、ECG） ................................................... 5
2.3 麻省理工心率失常資料庫MIT-BIH ....................................................... 9
第三章 文獻探討 ............................................................................................... 10
第四章 研究架構及實現 ................................................................................... 27
4.1 資料集與預處理 ....................................................................................... 27
4.2 模擬架構及環境 ....................................................................................... 28
4.3 BFGS 演算法 ............................................................................................ 30
4.4 模擬結果與討論 ....................................................................................... 32
第五章 結論與未來展望 ................................................................................... 45
參考文獻 ............................................................................................................. 46
附錄 ..................................................................................................................... 52

表目錄

表1. 使用特徵提取-MLP方法文獻....................17
表2. 使用特徵提取-BPNN方法文獻...................18
表3. 使用特徵提取-ANN方法文獻....................20
表4. 使用特徵提取-LVQ、FGNN方法文獻..............21
表5. 未使用特徵提取-NN方法文獻...................23
表6. 未使用特徵提取-非NN方法文獻.................25
表7. BFGS演算法的參數設定.......................31
表8. SCG演算法的參數設定........................31
表9. CGP演算法的參數設定........................31
表10. CGB演算法的參數設定.......................32
表11. UCI 100-fold Cross-Validation平均結果....35
表12. 4層隱藏層節點為1-105x63平均模擬結果........36
表13. 4層隱藏層節點為2-105x63平均模擬結果........36
表14. 4層隱藏層節點為5-105x63平均模擬結果........37
表15. 4層隱藏層節點為10-105x63平均模擬結果.......37
表16. 4層隱藏層節點為15-105x63平均模擬結果.......38
表17. 4層隱藏層節點為1-63x63平均模擬結果.........39
表18. 4層隱藏層節點為2-63x63平均模擬結果.........39
表19. 4層隱藏層節點為5-63x63平均模擬結果.........40
表20. 4層隱藏層節點為10-63x63平均模擬結果........40
表21. 4層隱藏層節點為15-63x63平均模擬結果........41
表22. 4層隱藏層節點為1-525x63平均模擬結果........42
表23. 4層隱藏層節點為2-525x63平均模擬結果........42
表24. 4層隱藏層節點為5-525x63平均模擬結果........43
表25. 資料集平均結果比較........................43
表26. 4層隱藏層節點為1-105x63模擬結果............52
表27. 4層隱藏層節點為2-105x63模擬結果............55
表28. 4層隱藏層節點為5-105x63模擬結果............58
表29. 4層隱藏層節點為10-105x63模擬結果...........61
表30. 4層隱藏層節點為15-105x63模擬結果...........64
表31. 4層隱藏層節點為1-63x63模擬結果.............66
表32. 4層隱藏層節點為2-63x63模擬結果.............69
表33. 4層隱藏層節點為5-63x63模擬結果.............72
表34. 4層隱藏層節點為10-63x63模擬結果............75
表35. 4層隱藏層節點為15-63x63模擬結果............78
表36. UCI資料集模擬結果.........................80
表37. 4層隱藏層節點為1-525x63模擬結果............82
表38. 4層隱藏層節點為2-525x63模擬結果............85
表39. 4層隱藏層節點為5-525x63模擬結果............88

圖目錄

圖 1. 常見心電圖................................5
圖 2. 具有QRS偏離的QRS波群[2]...................6
圖 3. 具有QR偏離的QRS波群[2]....................6
圖 4. 具有RS偏離的QRS波群 [2]...................7
圖 5. 僅有R波的QRS波群 [2]......................7
圖 6. 具有RSR’偏離的QRS波群 [2].................7
圖 7. 表現為QS波的QRS波群 [2]...................7
圖 8. ST波段位置[2].............................8
圖 9. 心電圖-心率測量[2]........................9
圖 10. ANN架構................................11
圖 11. 專家系統架構............................14
圖 12. 性能比對：105x63........................32
圖 13. 性能比對：525x63........................33
圖 14. 性能比較：UCI...........................34
圖 15. 誤差曲線................................34

[1] World Health Organization, Cardiovascular Diseases, https://www.who.int/zh/news-room/fact-sheets/detail/cardiovascular-diseases-(cvds) （最後查詢日期：2022.06.01）
[2] 看圖識字ECG, http://www.medicine-on-line.com/html/ecg/e0001ct.htm. （最後查詢日期：2022.06.15）
[3] “MIT/BIH database distributor”, Beth Israel Hospital, Biomedical Engineering, Division, available in the address: http://ecg.mit.edu/, USA, 1979. （最後查詢日期：2022.06.01）
[4] UCI Repository of Machine Learning Data bases, Cleveland Heart Disease Dataset. available in the address: https://uci.edu/.（最後查詢日期：2022.06.01）
[5] Hassoun, M. H. (1995). Fundamentals of Artificial Neural Networks. MIT press.
[6] Silva, I., Moody, B., & Moody, G. (2021). Waveform Database Software Package (WFDB) for MATLAB and Octave (version 0.10.0). PhysioNet. https://doi.org/10.13026/6zcz-e163.
[7] Sammut, C., & Webb, G. I. (Eds.). (2011). Encyclopedia of machine learning. Springer Science & Business Media.
[8] Ghongade, R. (2007, October). A brief performance evaluation of ECG feature extraction techniques for artificial neural network based classification. In Proceedings of the TENCON 2007-2007 IEEE Region 10 Conference (pp. 1-4).
[9] Debnath, T., Hasan, M. M., & Biswas, T. (2016, December). Analysis of ECG signal and classification of heart abnormalities using Artificial Neural Network. In Proceedings of the 2016 9th International Conference on Electrical and Computer Engineering (ICECE) (pp. 353-356).
[10] Jalil, M. M., Saaid, M. F., Ahmad, A., & Ali, M. M. (2014, December). Arrhythmia modelling via ECG characteristic frequencies and artificial neural network. In Proceedings of the 2014 IEEE Conference on Systems, Process and Control (ICSPC 2014) (pp. 121-126).
[11] Haseena, H., Joseph, P. K., & Mathew, A. T. (2009). Artificial neural network based ECG arrhythmia classification. Journal of Mechanics in Medicine and Biology, 9(04), 507-525.
[12] Raj, A. A. S., Dheetsith, N., Nair, S. S., & Ghosh, D. (2014, November). Auto analysis of ECG signals using artificial neural network. In Proceedings of the 2014 International Conference on Science Engineering and Management Research (ICSEMR) (pp. 1-4).
[13] Gothwal, H., Kedawat, S., & Kumar, R. (2011). Cardiac arrhythmias detection in an ECG beat signal using fast Fourier transform and artificial neural network. Journal of Biomedical Science and Engineering, 4(04), 289-296.
[14] Ceylan, R., & Özbay, Y. (2007). Comparison of FCM, PCA and WT techniques for classification ECG arrhythmias using artificial neural network. Expert Systems with Applications, 33(2), 286-295.
[15] Roza, V. C. C., de Almeida, A. M., & Postolache, O. A. (2017, May). Design of an artificial neural network and feature extraction to identify arrhythmias from ECG. In Proceedings of the 2017 IEEE International Symposium on Medical Measurements and Applications (MeMeA) (pp. 391-396).
[16] Dewangan, N. K., & Shukla, S. P. (2016, May). ECG arrhythmia classification using discrete wavelet transform and artificial neural network. In Proceedings of the 2016 IEEE International Conference on recent Trends in Electronics, Information & Communication Technology (RTEICT) (pp. 1892-1896).
[17] Das, M. K., & Ari, S. (2013, December). ECG arrhythmia recognition using artificial neural network with S-transform based effective features. In Proceedings of the 2013 Annual IEEE India Conference (INDICON) (pp. 1-6).
[18] Sharma, A., & Sharma, T. (2011). ECG Beat Recognition using Principal Components Analysis and Artificial Neural Network. International Journal of Electronics Engineering, 3(1), 55-58.
[19] Thabit, R., Zeyad, T., & Abou-Loukh, S. J. (2010). Ecg Classification Using Slantlet Transform and Artificial Neural Artificial Network. Journal of Engineering, 16(1), 4510-4526.
[20] Rai, H. M., Trivedi, A., & Shukla, S. (2013). ECG signal processing for abnormalities detection using multi-resolution wavelet transform and Artificial Neural Network classifier. Measurement, 46(9), 3238-3246.
[21] Pantelopoulos, A., & Bourbakis, N. (2010, November). Efficient single-lead ECG beat classification using matching pursuit based features and an artificial neural network. In Proceedings of the 10th IEEE International Conference on Information Technology and Applications in Biomedicine (pp. 1-4).
[22] Taha, Z. K. (2018). Feature Extraction and Classification for ECG signals Processing based on Stationary Multiwavelet Transform and Artificial Neural Network. AL-MANSOUR Journal, (29), 85-101.
[23] Ghongade, R., & Ghatol, A. A. (2007, December). Performance analysis of feature extraction schemes for artificial neural network based ECG classification. In Proceedings of the IEEE International Conference on Computational Intelligence and Multimedia Applications (ICCIMA 2007) (Vol. 2, pp. 486-490).
[24] He, L., Hou, W., Zhen, X., & Peng, C. (2006, October). Recognition of ECG patterns using artificial neural network. In Proceedings of the Sixth IEEE International Conference on Intelligent Systems Design and Applications (Vol. 2, pp. 477-481).
[25] Neagoe, V. E., Iatan, I. F., & Grunwald, S. (2003). A neuro-fuzzy approach to classification of ECG signals for ischemic heart disease diagnosis. In AMIA Annual Symposium Proceedings (Vol. 2003, p. 494). American Medical Informatics Association.
[26] Ozbay, Y., & Karlik, B. (2001, October). A recognition of ECG arrhytihemias using artificial neural networks. In Proceedings of the 2001 23rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society (Vol. 2, pp. 1680-1683).
[27] Ghongade, R., & Ghatol, A. (2008, November). A robust and reliable ECG pattern classification using QRS morphological features and ANN. In Proceedings of the TENCON 2008-2008 IEEE Region 10 Conference (pp. 1-6).
[28] Silipo, R., & Marchesi, C. (1998). Artificial neural networks for automatic ECG analysis. IEEE Transactions on Signal Processing, 46(5), 1417-1425.
[29] Nayak, S. G., & Nayak, D. (2012). Classification of ECG signals using ANN with resilient back propagation algorithm. International Journal of Computer Applications, 54(6), 20-23.
[30] Vishwa, A., Lal, M. K., Dixit, S., & Vardwaj, P. (2011). Classification of arrhythmic ECG data using machine learning techniques. International Journal of Interactive Multimedia and Artificial Intelligence, 1(4), 67-70.
[31] Amma, N. B. (2012, February). Cardiovascular disease prediction system using genetic algorithm and neural network. In Proceedings of the 2012 IEEE International Conference on Computing, Communication and Applications (pp. 1-5).
[32] Priyatharshini, R., & Chitrakala, S. (2019). A self-learning fuzzy rule-based system for risk-level assessment of coronary heart disease. IETE Journal of Research, 65(3), 288-297.
[33] Sihem, N. I. T. A., Bitam, S., & Mellouk, A. (2018, June). An enhanced random forest for cardiac diseases identification based on ECG signal. In Proceedings of the 2018 14th International Wireless Communications & Mobile Computing Conference (IWCMC) (pp. 1339-1344).
[34] Gao, D., Madden, M., Chambers, D., & Lyons, G. (2005, July). Bayesian ANN classifier for ECG arrhythmia diagnostic system: A comparison study. In Proceedings of the 2005 IEEE International Joint Conference on Neural Networks, 2005. (Vol. 4, pp. 2383-2388).
[35] Darmawahyuni, A., Nurmaini, S., & Firdaus, F. (2019). Coronary heart disease interpretation based on deep neural network. Computer Engineering and Applications Journal, 8(1), 1-12.
[36] Wiharto, W., Kusnanto, H., & Herianto, H. (2017). Hybrid system of tiered multivariate analysis and artificial neural network for coronary heart disease diagnosis. International Journal of Electrical and Computer Engineering, 7(2), 1023.
[37] Azuaje, F., Dubitzky, W., Wu, X., Lopes, P., Black, N., Adamson, K., & White, J. A. (1997, September). A neural network approach to coronary heart disease risk assessment based on short-term measurement of RR intervals. In Computers in Cardiology 1997 (pp. 53-56).
[38] Kaur, J., & Raina, J. P. S. (2012). An Intelligent Diagnosis System for Electrocardiogram (ECG) Images Using Artificial Neural Network (ANN). International Journal of Electrical, Electronics and Computer Engineering, 1(1), 47-51.
[39] Chen, J., Xi, G., Xing, Y., Wang, J., & Zheng, C. (2007, November). An unsupervised multi-valued stochastic neural network algorithm to cluster in Coronary Heart Disease data. In Proceedings of the 2007 International Conference on Convergence Information Technology (ICCIT 2007) (pp. 640-644). IEEE.
[40] Sanghavi, M. A., & Bojewar, S. M. (2014). Classification of Cardiovascular Disease from ECG using Artificial Neural Network and Hidden Markov Model. IOSR Journal of Computer Engineering, (3) (pp. 92-98).
[41] Prasad, G. K., & Sahambi, J. S. (2003, October). Classification of ECG arrhythmias using multi-resolution analysis and neural networks. In Proceedings of the IEEE TENCON 2003. Conference on convergent Technologies for Asia-Pacific Region (Vol. 1, pp. 227-231).
[42] Madani, M. (2016). Comparing learning algorithms in neural network for diagnosing cardiovascular disease. arXiv preprint arXiv:1611.01678.
[43] Kim, J., Lee, J., & Lee, Y. (2015). Data-mining-based coronary heart disease risk prediction model using fuzzy logic and decision tree. Healthcare Informatics Research, 21(3), 167-174.
[44] Cinetha, K., & Maheswari, P. U. (2014). Decision support system for precluding coronary heart disease (CHD) using fuzzy logic. International Journal of Computer Science Trends and Technology, 2(2), 2347-857.
[45] Allahverdi, N., Torun, S., & Saritas, I. (2007, June). Design of a fuzzy expert system for determination of coronary heart disease risk. In Proceedings of the 2007 International Conference on Computer Systems and Technologies (pp. 1-8).
[46] Yang, M. Y., Hu, W. C., & Shyu, L. Y. (1997, October). ECG events detection and classification using wavelet and neural networks. In Proceedings of the 19th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 'Magnificent Milestones and Emerging Opportunities in Medical Engineering'(Cat. No. 97CH36136) (Vol. 1, pp. 280-281).
[47] Poungponsri, S., & Yu, X. H. (2009, August). Electrocardiogram (ECG) signal modeling and noise reduction using wavelet neural networks. In Proceedings of the 2009 IEEE International Conference on Automation and Logistics (pp. 394-398).
[48] Kim, J. K., & Kang, S. (2017). Neural network-based coronary heart disease risk prediction using feature correlation analysis. Journal of Healthcare Engineering.
[49] Zand, M. D., Ansari, A. H., Lucas, C., & Zoroofi, R. A. Z. (2010, November). Risk assessment of coronary arteries heart disease based on neuro-fuzzy classifiers. In Proceedings of the 2010 17th Iranian Conference of Biomedical Engineering (ICBME) (pp. 1-4).
[50] Wiharto, W., Suryani, E., & Cahyawati, V. (2019). The methods of duo output neural network ensemble for prediction of coronary heart disease. Indonesian Journal of Electrical Engineering and Informatics (IJEEI), 7(1), 51-58.
[51] Abidin, B., Dom, R. M., Rahman, A. R. A., Bakar, R. A., Demiralp, M., Baykara, N., & Mastorakis, N. (2009, May). Use of fuzzy neural network to predict coronary heart disease in a Malaysian sample. In Proceedings of the WSEAS International Conference Mathematics and Computers in Science and Engineering (Vol. 8, pp. 76-80). World Scientific and Engineering Academy and Society.
[52] Fletcher, R. (2013). Practical methods of optimization. John Wiley & Sons.
[53] Satheeskumaran, S., Venkatesan, C., & Saravanan, S. (2021). Real-time ECG signal pre-processing and neuro fuzzy-based CHD risk prediction. International Journal of Computational Science and Engineering, 24(4), 323-330.