研究生: |
張哲愷 Che-Kai Chang |
---|---|
論文名稱: |
穿戴式脈波量測系統結合機器學習與脈波變異度分析應用於新冠肺炎疫苗之心血管副作用監測 Monitoring cardiovascular side effects of COVID-19 vaccine by using wearable pulse measurement system with machine learning and pulse variability analysis |
指導教授: |
許昕
Hsin Hsiu |
口試委員: |
劉如濟
Ju-Chi Liu 鮑興國 Hsing-Kuo Pao |
學位類別: |
碩士 Master |
系所名稱: |
應用科技學院 - 醫學工程研究所 Graduate Institute of Biomedical Engineering |
論文出版年: | 2022 |
畢業學年度: | 110 |
語文別: | 中文 |
論文頁數: | 89 |
中文關鍵詞: | 新冠肺炎疫苗 、疫苗副作用 、穿戴式裝置 、循環系統 、機器學習 、脈波變異度分析 |
外文關鍵詞: | COVID-19 vaccine, vaccine side effect, wearable system, circulatory system, machine learning, pulse variability analysis |
相關次數: | 點閱:174 下載:0 |
分享至: |
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
2019年以來新冠肺炎已對全球經貿、人民健康造成極大衝擊,而新冠肺炎疫苗是現有比較有效堵住疫情惡化的預防方法,不過疫苗也存在著一些副作用隱憂。根據研究,目前疫苗嚴重副作用大多與心血管系統有關,因此心血管副作用監測的需求是與日俱增且必要的,不過現今常見的心血管副作用臨床檢驗方式仍有諸多不便。
本研究將藉由實驗室自行開發之穿戴式脈波量測系統結合機器學習、脈波變異度分析,探討穿戴式裝置可否觀察出疫苗副作用對心血管系統造成的生理影響,並且是否可對此影響具有區別能力。
研究與雙和醫院心臟內科聯合收案,納入本研究總人數為194位。以臨床檢驗結果將BNT受試者分為健康組(N=28)、心血管副作用組(N=39)、血管副作用組(N=11)、心臟副作用組(N=34)。實驗會對受試者進行一分鐘的脈波量測取得BPW、PPG數據,再以機器學習、脈波變異度分析強化對脈波變動的解析能力,進而區別出疫苗心血管副作用的有無。
實驗結果顯示,此系統確實可區別出疫苗有無心血管副作用的差異,機器學習(LDA)與脈波變異度分析各自可對心血管副作用的判別達到準確率0.69(AUC=0.67)與準確率0.67(AUC=0.66)的表現。若排除模糊地帶則脈波變異度分析可對心血管組做到準確率0.72、AUC=0.75(適用範圍65.67%)的判別;對血管組做到準確率0.92、AUC=0.94(適用範圍64.1%)的判別。
研究結論可以看出,穿戴式脈波量測系統未來可望實際運用於臨床監測,也方便於社區居家環境中使用,至此本篇研究已為穿戴式裝置應用於疫苗心血管副作用的監測建立了一個新方向與基礎。
Since 2019, COVID-19 has significantly impacted the global economy, trade, and people's health. The COVID-19 vaccine is a more effective method to block the deterioration of the epidemic. However, there are some side effects of the vaccine. According to research, most of the severe side effects of vaccines are related to the cardiovascular system. Therefore, the demand for monitoring cardiovascular side effects is increasing and necessary. However, there are still many inconveniences in the typical clinical testing methods for cardiovascular side effects.
This study will use our laboratory's wearable pulse measurement system to combine machine learning and pulse variability analysis. Exploring whether the wearable device can observe the physiological effects of vaccine side effects on the cardiovascular system and whether it can this effect be discriminatory.
This study is cooperated with The Cardiology Department of Shuang-Ho Hospital, and the total number of people included was 194. BNT subjects were divided into the healthy group (N=28), cardiovascular side effect group (N=39), vascular side effect group (N=11) and cardiac side effect group (N=34) according to the clinical test results. In the experiment, subjects will be measured by pulse wave for one minute to obtain BPW and PPG data. Then machine learning and pulse variability analysis will be used to strengthen the ability to analyze pulse wave changes and then distinguish whether the subject have cardiovascular side effects.
The experimental results show that the system can distinguish whether people have cardiovascular side effects. Machine learning (LDA) and pulse variability analysis can determine cardiovascular side effects with an accuracy rate of 0.69 (AUC=0.67) and an accuracy rate of 0.67. (AUC=0.66) performance. If the fuzzy area is excluded, the pulse variability analysis can achieve an accuracy rate of 0.72 and AUC=0.75 for the cardiovascular group (65.67% of the applicable range); for the blood vessel group, the accuracy rate is 0.92 and AUC=0.94 (the useful range is 64.1%). ) judgment.
From the result we can see that the wearable pulse measurement system is expected to be used in clinical monitoring in the future, and it is also convenient for use in the community home environment. So far, this study has established a system for wearable devices to monitor the cardiovascular side effects of vaccines. A new direction and foundation.
[1] Timeline: WHO’s COVID-19 responses
https://www.who.int/emergencies/diseases/novel-coronavirus-2019/interactive-timeline
[2] 疾病介紹-衛生福利部疾病管制署
https://www.cdc.gov.tw/Category/Page/vleOMKqwuEbIMgqaTeXG8A
[3] 1110623 COVID-19疫苗接種後不良事件通報
https://www.cdc.gov.tw/File/Get/dntI5iYqJt5M7gUwGHOL0A
[4] Yoganathan A, Sajjad MS, Harky A. Cardiovascular disease and the impact of COVID-19. J Card Surg. 2020 Aug;35(8):2113. doi: 10.1111/jocs.14682. Epub 2020 Jun 16. PMID: 32542799; PMCID: PMC7323179.
[5] Ravichandran B, Grimm D, Krüger M, Kopp S, Infanger M, Wehland M. SARS-CoV-2 and hypertension. Physiol Rep. 2021 Jun;9(11):e14800. doi: 10.14814/phy2.14800. PMID: 34121359; PMCID: PMC8198473.
[6] Wu Z, McGoogan JM. Characteristics of and Important Lessons From the Coronavirus Disease 2019 (COVID-19) Outbreak in China: Summary of a Report of 72 314 Cases From the Chinese Center for Disease Control and Prevention. JAMA 2020 Apr 7; 323(13): 1239–42.
[7] 新型冠狀病毒(SARS-CoV-2)感染臨床處置指引
https://www.cdc.gov.tw/Category/Page/xCSwc5oznwcqunujPc-qmQ#
[8] Fazlollahi A, Zahmatyar M, Noori M, Nejadghaderi SA, Sullman MJM, Shekarriz-Foumani R, Kolahi AA, Singh K, Safiri S. Cardiac complications following mRNA COVID-19 vaccines: A systematic review of case reports and case series. Rev Med Virol. 2021 Dec 17:e2318. doi: 10.1002/rmv.2318. Epub ahead of print. PMID: 34921468
[9] Side effect worry grows for AstraZeneca vaccine
Science. 2 Apr 2021. Vol 372, issue 6537. pp.14-15. DOI:10.1126/science.372.6537.14
[10] 陳均曄(2019)。非侵入式循環訊號量測系統應用於不同癌症之偵測與評估。國立臺灣科技大學醫學工程研究所碩士論文,台北市。
[11] 陳冠彰(2016)。循環訊號量測應用於乳癌患者病程監控之研究。國立臺灣科技大學醫學工程研究所碩士論文,台北市。
[12] 楊舒涵(2015)。都卜勒微循環血流與動脈血壓波形指標 應用於代謝症候群之研究。國立臺灣科技大學醫學工程研究所碩士論文,台北市。
[13] 許家良(2014)。血壓波形的諧波分析應用於多囊性卵巢及中風之偵測及療效評估研究。國立臺灣科技大學應用科技研究所博士論文,台北市。
[14] 劉人銘(2018)。循環量測應用於服用高血壓藥物患者之研究。國立臺灣科技大學醫學工程研究所碩士論文,台北市。