English  |  正體中文  |  简体中文  |  Items with full text/Total items : 889/889 (100%)
Visitors : 15414579      Online Users : 52
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
Scope Tips:
  • please add "double quotation mark" for query phrases to get precise results
  • please goto advance search for comprehansive author search
  • Adv. Search
    HomeLoginUploadHelpAboutAdminister Goto mobile version
    Please use this identifier to cite or link to this item: http://ccur.lib.ccu.edu.tw/handle/A095B0000Q/483

    Title: 結合脊椎影像分析技術及機器學習建構脊椎壓迫性骨折接受椎體增強手術術後發生鄰近節椎體骨折之預測模式;Combine Image-Analysis Techniques of Spine X-ray Images with Machine Learning to Predict Adjacent Fracture after Vertebral augmentation procedures
    Authors: 林愈鈞;LIN, YU-CHUN
    Contributors: 資訊管理系醫療資訊管理研究所
    Keywords: 鄰近節椎體骨折;機器學習;影像分析技術;椎體增強手術;Adjacent Fracture;Machine Learning;Image-Analysis Techniques;Vertebral Augmentation Procedures
    Date: 2017
    Issue Date: 2019-07-17
    Publisher: 資訊管理系醫療資訊管理研究所
    Abstract: 椎體增強手術常用於治療治療因骨質疏鬆症引起之脊椎壓迫性骨折,而鄰近節椎體骨折為手術後常見的併發症之一。然其成因在醫學上仍是一個高度爭論與熱門研究的議題。近年隨著資訊技術的快速發展,在醫學領域,諸如疾病的診斷、治療、或公共衛生的議題,已有許多學者嘗試著利用機器學習與資料探勘的技術來進行研究。而結合醫學影像特徵萃取分析技術與機器學習,在近來也使熱門的研究方向之一。就收集及可查閱的文獻顯示,尚未有報告以機器學習之技術來研究此一議題。60歲以上因骨質疏鬆性脊椎壓迫性骨折且接受單一節椎體增強手術之男性及女性病患共516人(女性占384人,男性132人,其平均年齡74.77歲)。在接受手術的516名患者中,於術後一年內有發生鄰近節椎體骨折的病人有69人(13.4%)。本研究以回溯分析病患之臨床風險因子及利用主動形狀模型於脊椎X光影像之分割與特徵值角度萃取。應用類神經網路、支援向量機、決策樹、及邏輯迴歸的分類技術,透過人工智慧的演算來建構病人接受椎體增強手術後一年之內發生鄰近節椎體骨折的預測模型。經評比各項分類器之優劣並分析各變項,以類神經網路為最佳分類器,其Sensitivity達99.13%,Specificity達90.09%,Accuracy為94.60%,而AUC則是97.31%。 分析27項自變項,年齡、身高、體重、骨水泥滲漏至椎間盤、骨水泥注入量、骨折椎體術後前柱高度恢復率、骨折椎體術後駝背角度恢復率、骨折椎體術後前後緣高度比、骨折椎體術後駝背角度為統計上顯著風險因子。以多元邏輯回歸之結果,年齡、骨水泥滲漏至椎間盤為顯著風險因子。
    Vertebral augmentation procedure is usually used to treat the osteoporotic compression fracture, and adjacent fracture is a common complication after this procedure. The causes of it is still a hot and controversial Issue. With the rapid development of information technology, the disease diagnosis, treatment, and issues of public health have been studied by many scholars using machine learning and data mining technology. To our knowledge, no study has been published to research the issue by using machine learning technology. Male and female individuals with ages over 60 years old, who have accepted single-level vertebral augmentation procedure because of osteoporotic compression fracture are enrolled. A total of 516 patients (384 women, 132 men, with a mean age of 74.77 years) are enrolled in this study. Of the 516 patients, 69 (13.4%) sustained adjacent fractures after the index procedure. This study retrospectively collects clinical risk and applies active shape model to segment the spinal X-Ray images and extracts the angle features. By using artificial neural network (ANN), support vector machine (SVM), decision tree (DT), and logistic regression (LR), these scores along with important risk factors are analyzed by machine learning algorithm to construct a prediction model for adjacent fracture during 1 year postoperatively. The results show that the best prediction model is artificial neural network (ANN). Regarding the prediction model, the sensitivity, specificity, accuracy, and AUC are 99.13%, 90.09%, 94.60% and 97.31%, respectively.Total 27 variates are analyzed. Statistically significant differences in age, body weight, body height, intradiscal cement leakage, volume of bone cement injected, rate of anterior vertebral height restoration, wedge angle change, postoperative anterior-to-posterior ratio of vertebral height, wedge angle after surgery are found between the adjacent fracture and nonadjacent fracture groups. By using the multivariate logistic regression analysis, age and intradiscal cement leakage are significant risk factors.
    Appears in Collections:[資訊管理系醫療資訊管理研究所] 學位論文

    Files in This Item:

    File Description SizeFormat

    All items in CCUR are protected by copyright, with all rights reserved.

    版權聲明 © 國立中正大學圖書館網頁內容著作權屬國立中正大學圖書館


    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - Feedback