請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/52592
標題: | 使用乳房超音波影像量化分析乳癌預測指標 Quantitative Analysis of Predictive Markers for Breast Cancer Using Ultrasound Images |
作者: | Yu-Ling Hou 侯玉翎 |
指導教授: | 張瑞峰 |
關鍵字: | 乳癌,預測指標,微鈣化,回音組織樣本, breast cancer,predictive markers,microcalcification,background echotexture pattern, |
出版年 : | 2015 |
學位: | 博士 |
摘要: | 根據統計,乳癌是全球女性因癌症死亡的第二大主因,而早期發現並治療是目前降低致死率最有效的方式。為了能在早期階段篩檢出乳癌,許多關於乳癌預測指標的相關研究被提出。而在所有非侵入式的篩檢方式當中,超音波是一種簡單又有效的技術。近年來,超音波影像不論是在臨床或是研究應用中,都扮演了深具應用價值的角色。儘管目前傳統的二維(2-D)超音波技術已被廣泛運用,但單靠2D 影像卻難以傳達乳房的整體特徵,因此,三維(3-D)超音波的影像裝置被提出。三維(3-D)超音波可以全方位完整呈現乳房的結構,有能力提供一個更廣闊的方向來描述病變特徵,不論是應用在乳癌早期預測指標或是偵測或診斷的研究上,都是很好的影像系統,可提供更多細節資訊及幫助醫生減少誤判的機率。本篇研究主要的目的是利用超音波影像量化分析各種乳癌預測指標,包含微鈣化(microcalcification)及回音組織樣本 (background echotexture pattern),使超音波影像在常規的檢查中,能篩檢出早期乳房病變,以便早期治療,降低致死率。 Breast cancer is the global second-leading cause of cancer death among women. If any possible cancer symptom could be detected in early stages, we can prevent it from getting into an advanced state and reduced lethality. Many studies of predictive markers were proposed in order to inspect breast cancer in early stage. Ultrasound (US) is a simple and effective modality in all non-invasive inspect and plays an important role whether in clinical or research. Although the conventional two-dimensional (2-D) US techniques of the breast at present have been widely used, 2-D images are not enough to transmit the entire characteristics of breast. Therefore, the three-dimensional (3-D) breast US is proposed to improve drawbacks of 2-D breast US. The three-dimensional (3-D) US is also called automated breast ultrasound (ABUS). ABUS can fully provide the architecture of breast in all aspects and therefore is capable of offering a more comprehensive way to characterize pathological features of breast cancer. ABUS is a useful image system, whether it uses in detection, diagnosis or predictive markers in early stage. It provides more detail image information of breast cancer and assists physicians to reduce misdiagnosis. In this study, we present quantitative analysis of predictive markers which include microcalcification and background echotexture for breast cancer using US images. We expected the quantitative analysis of predictive markers for breast cancer could detect the non-palpate breast lesion in early stage in routine examination. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/52592 |
全文授權: | 有償授權 |
顯示於系所單位: | 生醫電子與資訊學研究所 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-104-1.pdf 目前未授權公開取用 | 7.53 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。