完整後設資料紀錄
DC 欄位語言
dc.contributor.author依高曼en_US
dc.contributor.authorEvgeny, Goldmanen_US
dc.contributor.author邱俊誠en_US
dc.contributor.authorChiou, Jin Chernen_US
dc.date.accessioned2014-12-12T02:43:46Z-
dc.date.available2014-12-12T02:43:46Z-
dc.date.issued2013en_US
dc.identifier.urihttp://140.113.39.130/cdrfb3/record/nctu/#GT070060089en_US
dc.identifier.urihttp://hdl.handle.net/11536/75630-
dc.description.abstract摘要 在世界上的頭號死亡原因是心臟疾病 - 有 67%心臟疾病死亡是由突如其來的心臟猝死所導致。這些案件大部分是由於心臟的異常電訊號觸發,如果沒有使用電擊治療,則會中斷血液循環而導致即時死亡。自動的心律失常分析有助於加快除顫治療以糾 正在心電訊號的干擾。植入式除顫器分析每一次心跳,且適當的分類是每年拯救成 千上萬人的生命的關鍵。 本研究提出的方法主要集中在心電圖疾病形式化。研究的結果可應用於醫療導向決 策系統分類工具。希爾伯特變換、豪斯多夫空間與傅立葉級數等修改過的基本方法,組合在一起進行過濾、模糊化與心電圖特徵波形(P、Q、R、S 和 T 波)複雜的定義, 作為重新建立心臟疾病分類的一個里程碑。詳細的解釋呈現在算法和方法的篇章。 本研究目標在於針對不同類型的疾病建立不同功能範圍,如心肌梗塞、心肌症/心臟衰竭、束支傳導阻滯、心律不整、心肌肥厚和瓣膜性心臟病,成功地達到目標,並在結果和結論章節介紹。所使用的資料庫是由聯邦物理技術研究院(PTB)所提供,提供的數據具有不同心臟病的心電圖。 通過我的數值實驗,我得出了結論,目的導向的演算法提供快速且精確的心律失常 分類方式,並能夠自動分析,且無人為疏失。zh_TW
dc.description.abstractABSTRACT 67% It’s the number cases of heart disease death cases is brought on by sudden cardiac death. Automatic heart disease analysis helps prevent an early death. Proper and early classification is the key to saving millions of lives every year. Proposed methods in this research are focused on ECG disease formalizing. Results of thesis can be used as classification tool in medical oriented decision-making system. Hilbert transformation, Hausdorff space, Fourier series and other fundamental approaches modified and combined together for filtration, fuzzification, PQRST complex identification and function reconstruction as a milestone in classification of heart disease. Detailed explanation presented in algorithms and methods chapter. The goal of creating the range of functions for different type of disease such as Myocardial infarction, Cardiomyopathy/Heart failure, Bundle branch block, Dysrhythmia, Myocardial hypertrophy and Valvular heart disease, successfully reached and presented in results and conclusion chapter. Database provided by Physikalisch-Technische Bundesanstalt (PTB), providing data with different heart disease ECG. Numerical experiments showed that purposed algorithm provides fast and precise arrhythmia classification on the expert decision level allowing automatic analysis.en_US
dc.language.isoen_USen_US
dc.subject分类zh_TW
dc.subject心电图zh_TW
dc.subject傅立叶级数zh_TW
dc.subjectHausdorff度量zh_TW
dc.subjectclassificationen_US
dc.subjectECGen_US
dc.subjectFourier seriesen_US
dc.subjectHausdorff metricen_US
dc.title以 Hausdorff 距離及 Fourier 級數從事心臟疾病分類zh_TW
dc.titleHausdorff metric and Fourier series as methods for Heart disease classificationen_US
dc.typeThesisen_US
dc.contributor.department電控工程研究所zh_TW
顯示於類別:畢業論文