腦中風患者之心房顫動預測模型開發與驗證：使用自然語言處理與機器學習技術

NCUIR > School of Management at National Central University > Graduate Institute of Information Management > Electronic Thesis & Dissertation > Item 987654321/95529

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/95529

Title:	腦中風患者之心房顫動預測模型開發與驗證：使用自然語言處理與機器學習技術
Authors:	楊于璇;Yang, Yu-Hsuan
Contributors:	資訊管理學系
Keywords:	心房顫動;腦中風;電子病歷;文字探勘;臨床支援決策;Atrial Fibrillation;Stroke;Electronic Medical Records;Text Mining;Clinical decision support system
Date:	2024-07-19
Issue Date:	2024-10-09 16:59:23 (UTC+8)
Publisher:	國立中央大學
Abstract:	突發性腦血管疾病，俗稱中風，是全球第二大致命原因和第三大導致失能的原因，而在台灣則是第五大致命原因。其中，最常復發者為缺血性中風。心房顫動是中風的潛在風險因素之一，由於其具有陣發性或無症狀的特點，難以在短時間內進行檢測，這可能導致患者未能及時採取應對措施，進一步增加再次中風的風險。考慮到醫療資源有限，因此需要建立一個良好的風險預測模型，以幫助醫師更詳細地對高風險患者進行檢查。本研究的目的是利用電子病歷中的結構化和非結構化資料，透過不同的分類技術建立預測模型，並使用實際的電子病歷數據進行驗證。除了在聯新國際醫院進行內部驗證外，考慮到模型的通用性，另外將使用嘉義基督教醫院的資料進行外部驗證。;Acute cerebrovascular disease, commonly known as stroke, ranks as the second leading cause of death globally and the third leading cause of disability. In Taiwan, it is the fifth leading cause of death. The most common recurrence among these cases is ischemic stroke. Atrial fibrillation is identified as one of the key risk factors for stroke. However, due to its intermittent or asymptomatic presentation, detecting it promptly is challenging, potentially delaying necessary interventions and heightening the risk of recurrent stroke. Given these challenges and the constraints of medical resources, it is imperative to develop a robust risk prediction model. Such a model would aid physicians in conducting thorough assessments of high-risk patients. This study aims to leverage both structured and unstructured data from electronic health records to develop predictive models using various classification techniques. Validation will be performed using real-world electronic health record data. Internal validation will initially take place within the same hospital, with external validation planned using data from other hospitals to ensure the model′s generalizability across different settings.
Appears in Collections:	[Graduate Institute of Information Management] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	17	View/Open

社群 sharing

Loading...