請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/76603
標題: | 腹部電腦斷層掃描影像之骨骼肌自動判別方法及臨床應用研究 Automatic Detection of Skeletal Muscle at Abdominal CT Images |
作者: | Yun-Hsuan Hsu 許芸瑄 |
指導教授: | 陳正剛(Argon Chen) |
關鍵字: | 骨骼肌,自動判別方法,Canny邊緣偵測, Skeletal muscle,Automatic detection,Canny Edge, |
出版年 : | 2016 |
學位: | 碩士 |
摘要: | 骨骼肌指標(Skeletal Muscle Index, SMI)、骨骼肌密度指標是放射科醫師常作為判斷病人身體狀況的重要指標,而人體的骨骼肌含量可經由多種方法測量,其中以影像分析的用途最為廣泛且結果也最為精準,因此本研究採用腹部電腦斷層掃描橫切面之影像。目前電腦斷層影像的骨骼肌範圍仍仰賴醫師手動圈選,程序繁瑣又費時,因此本研究希望能利用骨骼肌的特徵自動判別骨骼肌,並利用自動化判別結果進行臨床研究分析。
本研究所採用的骨骼肌自動化判別流程主要是由外向內,先定義最外層的皮膚層,次為由皮膚向內搜尋骨骼肌外圈邊緣,再由骨骼肌外圈邊緣向內定義骨骼肌內圈邊緣,最後在骨骼肌外圈、內圈之間取得的骨骼肌範圍。本研究提出之骨骼肌自動判別方法主要利用Canny邊緣偵測法取得的腹部電腦斷層影像邊緣。經由本研究所提出之邊緣選取及處理方法得到最後區域。 自動判別骨骼肌方法所得之區域與手繪進行結果驗證,迴歸分析結果與醫師手繪區域內骨骼肌面積 可達0.96。自動判別骨骼肌方法所得之指標與台大醫院所提供的92筆資料進行評估病人健康狀況之臨床分析,其中病人的年齡、骨骼肌密度、皮下脂肪密度三項指標之AUC分別為 0.856、0.897、0.766。 Skeletal Muscle Index (SMI) and Skeletal Muscle Density Index are widely used as important indexes to evaluate health condition of patients by radiologists. Among all common evaluation methods, image analysis, which provides more accurate measurement results, is widely adopted. As such, our research focuses on image analysis of abdominal CT images. So far, the region of skeletal muscle on the image has to be manually circumscribed by doctors with tedious and time-consuming procedures. Therefore, our research aimes to segment the skeletal muscle region automatically by identifying critical morphology and Hounsfield Unit (HU) features of the abdominal CT image and apply the detection results to clinical analysis. The procedure for automatic segmentation of the skeletal muscle is to identify the margins of tissue layers from the outer part gradually towards the inner part of the human body. First of all, the outer margin of the skin layer is defined by the high contrast between the human body portion and the non-human body portion on the CT images. Secondly, we search inward to find the outer boundary of the skeletal muscle and then inner boundary of the skeletal muscle. The true region of the skeletal muscle between the outer and inner boundaries is then segmented using a proposed edge detection method adapted from the well-known Canny Edge Detection method. The results of automatic segmentation of the skeletal muscle were compared to the results of the skeletal muscle manually circumscribed by the doctor. Regression analysis is performed and the resulted R2 reaches 0.96. The indexes calculated from the automatic segmentation are then used in clinical analysis of 92 cases collected at NTUH. Results of clinical research indicate that the Area Under ROC curve (AUC) of the skeletal muscle density index and the subcutaneous adipose tissue density index and the age are 0.856, 0.897 and 0.766, respectively. Keywords: Skeletal muscle, Automatic detection, Canny Edge |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/76603 |
DOI: | 10.6342/NTU201700591 |
全文授權: | 未授權 |
顯示於系所單位: | 工業工程學研究所 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-105-R03546012-1.pdf 目前未授權公開取用 | 3.37 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。