電機資訊學院: 電子工程學研究所指導教授: 呂學士黃彥銘Huang, Yen-MingYen-MingHuang2017-03-062018-07-102017-03-062018-07-102015http://ntur.lib.ntu.edu.tw//handle/246246/276094血壓量測是檢視心臟健康的一個重要指標。在診斷以及治療高血壓的問題時，必須要能先準確的量測血壓。一般使用聽診法的傳統水銀血壓計，被視為是最準確的非侵入式量測血壓的方法，但是使用方法較為複雜，只適合專業醫護人員使用。所以現在市場上以及家庭照護上，全自動的自我血壓監測儀器成為主流。這些儀器大部分是使用共振法的原理去設計製造，它的優點是不需要經過專業的訓練即可操作以及儀器本身不易受到環境的雜訊影響。但是大部分這些儀器在量測年輕健康的族群時，有比較高的精確度，而在中老年齡層時，準確度就不是這麼的理想。現在市面上能買得到的血壓量測儀器，大多數都是非連續性的量測，因此只能提供單一的時間點的血壓量測值，看不出血壓的連續性變化，因此使用連續性的血壓量測比傳統非連續式血壓量測能得到更多的生理資訊。 因此，在本論文中，我們提出了新式的血壓感測器以及用於過濾雜訊的血壓訊號處理演算法。新型血壓感測器不會受到受試者的年齡層影響且容易使用。訊號處理是利用HHT的演算法來達到去除感測器的雜訊。HHT本身的特性就是用來處理非穩態的訊號，因此用在處理血壓訊號上面非常適合。利用HHT這個方法，可以有效率的把原本的訊號拆成很多個IMFs，藉此方法，可以從感測器輸出的訊號把雜訊分離開來。Blood pressure (BP) is one of the most important signs of human cardiovascular health. The precision measurement of the blood pressure is necessary in diagnosis and treatment of hypertension and the risks related blood pressure. While the traditional auscultatory method using mercury sphygmomanometer is still viewed as the most accurate non-invasive blood pressure measurement method, it is complicated and only suitable for medical personnel. Currently, self-blood pressure monitoring devices are popular in the market and widely used in homecare. Most of those devices are based on the oscillometric method, as it requires less professional training and is less sensitive to external noise. However, most of these work well on young healthy subjects, but show less precision in some cases such as older people. Most of the devices in the market can only provide single time BP value, it’s unable to see the continuous change in BP. However, by continuous way, we can get more physiological information than traditional non-continuous measurement. As a result, a novel blood pressure sensor and signal processing algorithm for removing noise have been developed in this study. It can accurately determine blood pressure non-invasively for all age group. The effective signal processing is based on Ensemble Empirical Mode Decomposition (EEMD) method to remove the noise from the sensor. Due to the non-stationary characteristics of BP, EEMD is practical to achieve accurate decomposition. The signal can be decomposed into several Intrinsic Mode Functions (IMFs) by EEMD. The results suggest that that the proposed EEMD can indeed effective separate the pure BP from the sensor output.論文使用權限: 不同意授權非侵入式血壓量測,訊號處理,希爾伯特黃轉換Non-invasive blood pressure measurement,Signal Processing, HHT[SDGs]SDG3thesis