請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/87873
標題: | 適用於長期生理監控之光體積描述訊號處理機制 Quality-aware Signal Processing Mechanism of PPG signal for Long-term Vital Sign Monitoring |
作者: | 馬咏治 Win-Ken Beh |
指導教授: | 吳安宇 An-Yeu Wu |
關鍵字: | 光體積描述訊號訊號,訊號品質,訊號前處理,生理監測,演算法選擇, photoplethysmography,Signal Quality,Signal Preprocessing,Vital Sign Monitoring,Algorithm Selection, |
出版年 : | 2023 |
學位: | 博士 |
摘要: | 隨著穿戴式裝置的普及,透過光體積描述訊號(Photoplethysmography, PPG)進行長期生理(Vital Sign)監控有著極大潛力能夠改善民眾健康,減緩逐年攀升的肥胖率問題及醫療負擔。
然而,光體積描述訊號在實現長期生理監控面對了兩大挑戰:1.)訊號品質的不穩定性造成不准確的生理監控, 2.)穿戴式裝置在長期生理監控下續航力不足的問題。 針對光體積描述訊號訊號品質的不穩定性造成不准確資料的量測,本文基於生理監控的應用,提出了一個利用心電訊號訊號、動脈血壓訊號與呼吸訊號,來輔助訊號品質評估系統的設計,稱為機器反饋輔助下的光體積描述訊號品質評估系統(Machine-Aided PPG Signal Quality Assessment System)。透過不同生理訊號與光體積描述訊號的生理關係,本方法可獲取更為準確的訊號品質標籤,進而達到能夠準確去除高預測誤差訊號的效果。 針對穿戴式裝置在長期生理監控下續航力不足的問題,本文從優化訊號前處理流程進行著手,首先提出了適用於光體積描述訊號的新型訊號處理機制,旨在改善傳統訊號處理作法中單一前處理下對運算資源的浪費。與傳統方式不同,本機制會基於每筆訊號的訊號品質進行前處理演算法的推薦,高品質訊號會被推薦使用低複雜度的前處理演算法,透過有效的推薦與訊號處理使我們能夠有效避免運算資源的浪費,降低整體功耗,延長電池續航力。 With the widespread use of wearable devices, photoplethysmography (PPG) has great potential for long-term vital sign monitoring to improve public health and alleviate the increasing obesity rate problem and medical burden. However, PPG faces two major challenges in achieving long-term vital sign monitoring: 1) the instability of signal quality resulting in inaccurate vital sign monitoring, and 2) the insufficient battery capacity of wearable devices for long-term monitoring. To address the issue of unstable signal quality in PPG measurements, this article proposes a machine-aided PPG signal quality assessment system that utilizes electrocardiogram signals, arterial blood pressure signals, and respiratory signals to assist in signal quality evaluation. By leveraging the physiological relationships between different signals and PPG, this method can provide more accurate signal quality labels and effectively remove high prediction error signals. To address the problem of insufficient battery capacity of wearable devices during long-term vital sign monitoring, this article starts with optimizing the signal pre-processing process and proposes a novel signal processing mechanism applicable to PPG signals. The mechanism aims to reduce the waste of computing resources under traditional signal processing methods using a single pre-processing algorithm. Unlike traditional methods, this mechanism recommends pre-processing algorithms based on the quality of each signal. For example, low-complexity pre-processing algorithms recommended for high-quality signals. By effectively recommending and processing signals, this approach can effectively avoid waste of computing resources, reduce overall power consumption, and extend battery life. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/87873 |
DOI: | 10.6342/NTU202301048 |
全文授權: | 同意授權(全球公開) |
顯示於系所單位: | 電子工程學研究所 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-111-2.pdf 此日期後於網路公開 2028-06-15 | 6.61 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。