The Deep Learning Model of Osteoporosis Screening by Combining Use of EHR and Chest X-ray Images

English | 正體中文 | 简体中文 | 全文笔数/总笔数 : 45422/58598 (78%)
造访人次 : 2559762 在线人数 : 169

RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.

Taipei Medical University Institutional Repository > 醫學科技學院 > 醫學資訊研究所 > 博碩士論文 > Item 987654321/61507

jsp.display-item.identifier=請使用永久網址來引用或連結此文件: http://libir.tmu.edu.tw/handle/987654321/61507

题名:	The Deep Learning Model of Osteoporosis Screening by Combining Use of EHR and Chest X-ray Images
作者:	DEWI, SARI RAHMAWATI KUSUMA
贡献者:	醫學資訊研究所碩士班林明錦
关键词:	deep learning;osteoporosis;screening;chest x-ray;EHR
日期:	2021-08-19
上传时间:	2022-03-28 19:08:52 (UTC+8)
摘要:	Title of Thesis : The Deep Learning Model of Osteoporosis Screening by Combining Use of EHR and Chest X-ray Images Author : Sari Rahmawati Kusuma Dewi Thesis advised by : Associate. Professor. Ming-Chin Lin, MD, PhD Background: Osteoporosis is the one of common disease in the elderly worldwide. The incidence of osteoporosis was increased hand in hand with increasing the elderly population at some country even though the prevention action and clinical treatment had been done. It is caused by asymptomatic osteoporosis only can be diagnosed until the symptoms happen. Therefore, we need the approach for detecting osteoporosis as early as possible for minimalizing the burden of osteoporosis. Research Aims: The aim of this study is to develop the osteoporosis screening model for supporting osteoporosis diagnosis and preventive action. Material and Methods: We were collected patient data from Shuang Ho Hospital, consist of image data and tabular data. Osteoporosis diagnosis was established using T-score refer to WHO criteria. The deep learning (DL) model have been developed through data pre-processing, generating deep learning models, and evaluating the performance models. In this study, the fusion deep learning model had been compared with image-only and electronic health record (EHR)-only model. The DL model predictive performance had been evaluated using accuracy, sensitivity (recall), precision, F-1 score, and area under ROC curve (AUC). Results: The 1148 patient data were collected from Shuang-Ho Hospital Information System, including chest x-ray images and patient meta data. The result of this study showed that the xi fusion model highest average predictive performance, including 0.80 of accuracy, 0.80 of recall, 0.81 of precision, 0.80 of F-1 score, and 0.80 of AUC. Conclusions: The performance of fusion model is higher than image-only and EHR-only model. Our proposed model can be alternative for osteoporosis screening. The further research is ongoing process that is aimed to improve the deep learning performance.
描述:	碩士指導教授：林明錦委員：邱泓文委員：蘇家玉
数据类型:	thesis
显示于类别:	[醫學資訊研究所] 博碩士論文

文件中的档案:

档案	描述	大小	格式	浏览次数
index.html		0Kb	HTML	995	检视/开启

检视Licence

在TMUIR中所有的数据项都受到原著作权保护.

TAIR相关文章

著作權聲明 Copyright Notice

本平台之數位內容為臺北醫學大學所收錄之機構典藏，包含體系內各式學術著作及學術產出。秉持開放取用的精神，提供使用者進行資料檢索、下載與取用，惟仍請適度、合理地於合法範圍內使用本平台之內容，以尊重著作權人之權益。商業上之利用，請先取得著作權人之授權。
The digital content on this platform is part of the Taipei Medical University Institutional Repository, featuring various academic works and outputs from the institution. It offers free access to academic research and public education for non-commercial use. Please use the content appropriately and within legal boundaries to respect copyright owners' rights. For commercial use, please obtain prior authorization from the copyright owner.
瀏覽或使用本平台，視同使用者已完全接受並瞭解聲明中所有規範、中華民國相關法規、一切國際網路規定及使用慣例，並不得為任何不法目的使用TMUIR。
By utilising the platform, users are deemed to have fully accepted and understood all the regulations set out in the statement, relevant laws of the Republic of China, all international internet regulations, and usage conventions. Furthermore, users must not use TMUIR for any illegal purposes.
本平台盡力防止侵害著作權人之權益。若發現本平台之數位內容有侵害著作權人權益情事者，煩請權利人通知本平台維護人員([email protected])，將立即採取移除該數位著作等補救措施。
TMUIR is made to protect the interests of copyright owners. If you believe that any material on the website infringes copyright, please contact our staff([email protected]). We will remove the work from the repository.

数据加载中.....