以多項性生理訊號早期預測急性中風病患之預後與以光體積描述訊號長期偵測心房顫動

中風是造成死亡與失能的主要原因，若能早期預測中風病患的預後，對於他們的治療會是有幫助的。然而，目前的診斷設備，例如電腦斷層掃描與核磁共振成像，其缺點為昂貴、不可攜、可能導致副作用；目前的診斷量表，例如美國國家衛生院中風量表，其缺點為無法連續不斷地監測。因此，本論文提出以多項性生理訊號，包括：心電圖、動脈血壓、光體積描述訊號，來早期預測中風病患的預後。透過多項性生理訊號分析架構，本論文預測非心房顫動中風預後的準確率達到77.8%，該準確率較單一種生理訊號預測準確率佳，且跟美國國家衛生院中風量表預測準確率相近，這意味著本論文提出的多項性生理訊號分析架構有潛力用於預測中風病患的預後。此外，心房顫動則是提高中風風險約五倍的危險因素，長期偵測心房顫動是一個重要的議題，尤其對疑似心房顫動的病患而言。目前最有效的診斷為心電圖，然而，各種心電圖監測設備都有一些缺點，例如，監測時間較短、需要由受測者觸發設備、費用較高、或侵入式。光體積描述訊號的量測較心電圖的量測簡便，然而，本論文中提到的光體積描述訊號現有著作，其缺點為樣本數較少、訊號量測環境較標準化、量測時間較短、且只考量單一的光體積描述訊號參數。因此，為了長期偵測心房顫動，本論文提出基於光體積描述訊號的心房顫動偵測架構，其接收者操作特徵曲線的曲線下面積達到92.4%，該面積較光體積描述訊號現有著作佳，且跟心電圖的心房顫動偵測架構相近，這意味著本論文提出的光體積描述訊號架構有潛力用於長期偵測心房顫動。

關鍵字

中風；心房顫動；心電圖；動脈血壓；光體積描述訊號；多項性生理訊號分析架構

並列摘要

Stroke is a leading cause of death and disability. Early prediction of stroke patients’ functional outcomes is helpful for their treatments. However, current diagnosis machines, such as computed tomography (CT) and magnetic resonance imaging (MRI), are expensive, not portable, and may cause side effects. Additionally, current diagnosis scales, such as National Institutes of Health Stroke Scale (NIHSS), should be evaluated by professional medical staff, and thus cannot be conducted continuously. In this thesis, we propose a multi-modal analysis methodology to early predict a stroke patient’s functional outcome based on physiological signals, including electrocardiogram (EKG), arterial blood pressure (ABP), and photoplethysmogram (PPG). By using the multi-modal framework to analyze the stroke patients’ physiological signals in intensive care unit (ICU), we find that the accuracy of non-atrial fibrillation (non-AF) stroke outcome predictions achieves 77.8%, which performs better than a single-modal framework built by any single phase. In addition, the joint EKG-ABP-PPG analysis achieves performance comparable to NIHSS, implying that the proposed multi-modal analysis framework has potential for predicting functional outcomes of stroke. Furthermore, atrial fibrillation (AF) is a risk factor for stroke, increasing risk for approximately five-fold. Long-term AF detection becomes an important issue, especially to suspected AF patients. Currently, the most useful test for diagnosing AF is EKG. However, each EKG monitoring device has its limitations or drawbacks, such as short monitoring period, requiring patients to trigger the recorder, high cost, or invasive examination. Additionally, compared to EKG examination, PPG examination is more convenient. However, current studies of the PPG-based AF detection mentioned in this thesis have some limitations or drawbacks, such as small sample size, standardized data collection environments, short measurement time, and considering only one PPG parameter. Therefore, we propose a long-term PPG-based AF detection framework. The area under the receiver operating characteristic (ROC) curves of the proposed PPG-based AF detection framework achieve 94.2% (training) and 92.4% (cross-validation), which are close to those of the EKG-based AF detection framework and higher than those of the PPG related works. In short, besides the convenience of PPG measurements, the performances and the similarity compared to EKG imply the potential of the proposed PPG-based AF detection framework for long-term AF detection.

並列關鍵字

stroke ； atrial fibrillation (AF) ； electrocardiogram (EKG) ； arterial blood pressure (ABP) ； photoplethysmogram (PPG) ； multi-modal analysis framework

參考文獻

[1] The Internet Stroke Center. http://www.strokecenter.org/patients/about-stroke/what-is-a-stroke/

Google Scholar

[2] World Health Organization (WHO), Stroke, Cerebrovascular accident. http://www.who.int/topics/cerebrovascular_accident/en/

Google Scholar

[3] Neurology, Stroke. http://www.neurodoc.in/stroke

Google Scholar

[4] National Institutes of Health (NIH). http://www.nhlbi.nih.gov/health/health-topics/topics/af

Google Scholar

[5] MAYO CLINIC. http://www.mayoclinic.org/diseases-conditions/atrial-fibrillation/multimedia/img-20096412

Google Scholar

國際替代計量

以多項性生理訊號早期預測急性中風病患之預後與以光體積描述訊號長期偵測心房顫動

全文下載

主題瀏覽