應用類神經網路模型預測髖關節骨折一年後存活率

Taipei Medical University Institutional Repository > 醫學科技學院 > 醫學資訊研究所 > 博碩士論文 > Item 987654321/53906

請使用永久網址來引用或連結此文件: http://libir.tmu.edu.tw/handle/987654321/53906

題名:	應用類神經網路模型預測髖關節骨折一年後存活率 Using artificial neural network models for predicting survival rate of hip fracture in elderly patients
作者:	簡嘉良 Chien, Chia-Liang
關鍵詞:	類神經網路;髖關節骨折;存活預測;artificial neural network;hip fracture;survival prediction
日期:	2013-06-14
上傳時間:	2018-11-15 10:41:30 (UTC+8)
摘要:	近年來由於工業化、生活型態的改變及醫學的進步，使得平均餘命增加，高齡族群已占總人口的7.1%。因應高齡人口的增加，骨質疏鬆所帶來的骨折風險也日益增加。而髖關節的骨折容易發生在骨質疏鬆的年老族群，且創傷後一年內的死亡率根據文獻的統計可高達20%至30%。自1980年代之後資料探勘技術已廣泛使用於醫學領域，其中應用類神經網路預測心肌梗塞、乳癌及大腸癌病人存活率更是有良好的表現。但是在預測髖關節骨折存活率方面仍少見相關研究及文獻。本研究的目的是建立一類神經網路模型來預測髖關節骨折手術後一年的存活率。利用年齡、性別、骨折型態、手術方式、住院天數、住院金額、手術醫師及其他內科病症，如糖尿病、高血壓、心臟病、腦血管意外、身心症、腎衰竭及阻塞性肺病等14項變數建立類神經網路預測模型，藉以預測髖關節骨折手術後一年存活率。本研究使用準確度、靈敏度、特異度以及接收者操作特徵曲線下面積等指標評估該預測模型效能，並與邏輯斯迴歸模型的預測效能做比較，藉以評估及證明類神經網路為預測髖關節骨折一年後存活率的良好決策支援工具。 In the aging population society, osteoporosis is a popular issue due to the associated complications, like hip fracture which have 20% to 30% mortality rate one year after surgery. Data mining techniques have been used in many medical applications since 1980s. Artificial neural networks are considered good alternatives to conventional statistical methods for the prediction of myocardial infarction, breast and colorectal cancer patients’ survival. Otherwise, there are few applications for predicting survival rate in elderly patients with hip fracture. The purpose of this study is to establish a predictive model to assess the one-year survival rate in the treatment of older patients with hip fracture. All of the chosen cases are older than 65 years old with regular follow up for more 1 year and excluding carcinoma or high-energy trauma cases. The artificial neural network (ANN) models will be developed with 14 variables, like age, sex, fracture pattern, surgery, mental status, hospital days, cost, surgeon and medical conditions. Cases are randomly assigned to training and testing datasets. The performances of prediction models are evaluated according to parameters such as accuracy, sensitivity, specificity, and the area under receiver operating characteristic curve. Otherwise, the performances of the artificial neural network (ANN) models also compare with the performances of the logistic regression models. This study wants to prove this prediction system is a useful toll in the predicting survival rate in elderly patients with hip fracture.
描述:	碩士指導教授-徐建業委員-邱泓文委員-張博論
資料類型:	thesis
顯示於類別:	[醫學資訊研究所] 博碩士論文

文件中的檔案:

檔案	描述	大小	格式	瀏覽次數
index.html		0Kb	HTML	119	檢視/開啟

檢視Licence

在TMUIR中所有的資料項目都受到原著作權保護.

TAIR相關文章

資料載入中.....