小乳房的乳房攝影影像品質改進之研究

The Improvement of Image Quality for Patient with Small Breast Size in Mammography

葉砡秀(Yu-Hsiu Yeh)

指導教授 ： 蘇振隆

中原大學/工學院/醫學工程研究所/碩士(2007年)

https://doi.org/10.6840/cycu200700349

摘要

乳房攝影是以自動暴露控制系統調控影像品質，乳房完全覆蓋底片匣下方的游離腔就能維持正常影像品質。然而當乳房太小無法完全覆蓋游離腔時，以AEC攝影會產生曝光不足的現象。本研究的目的是探討底片式乳房攝影(FSM)與電腦放射乳房攝影系統(CRM)對小乳房女性攝影時之影像變化，並建立一套攝影技術以解決小乳房女性在臨床乳房攝影上所面臨之問題。
本研究採用各種假體(BR12、PMMA與CDMAM)進行模擬攝影，評估FSM與CRM的表現。並以影像解析度、對比度、對比雜訊比(CNR)與影像品質係數(IQFinv)等指標量化分析小乳房攝影的影像品質變化。最後藉由影像直方圖正規化與均等化評估降低CRM攝影輻射量的可行性。此外推導一數學模式，針對不同乳腺比例之乳房等效假體進行模擬攝影，以建立小乳房女性在FSM與CRM之攝影技術。並分別回溯分析7位臨床小乳房女性接受乳房攝影之資料及應用至新的病人，評估本攝影技術之臨床可行性。
初步結果顯示，FSM與CRM受到小乳房女性的乳房長度縮短確會導致曝光不足。對假體而言，平壓後長度低於4.5公分時會產生影像品質下降的現象。就對比度而言，FSM與CRM的分佈範圍分別為2%-23%與24%-25%。CRM影像經過影像處理後，可以提升高管負載攝影的品質，然無助於輻射量的降低。回溯分析中，放射師依據經驗手動設定條件時，FSM與CRM之IQFinv分別為54.6b8.8與67.9b18.1；而採用本研究建立攝影技術，IQFinv則提昇為60.3b3.2與69.9b3.8。當應用在臨床受檢者乳房攝影時，影像品質也有顯著的改善。
本研究建立一套小乳房攝影技術，可依據受檢者乳房厚度提供合理攝影條件，有效提升小乳房女性之乳房影像品質。此技術無須對攝影設備進行修改即可立即應用在臨床乳房攝影上，臨床上確可對小乳房女性之乳房攝影有實質上的助益。

關鍵字

電腦放射乳房攝影 ； 傳統底片式乳房攝影 ； 數學模式 ； 影像品質 ； 自動暴露控制系統

並列摘要

The image quality which usually could optimize with automatic exposure control (AEC) system is very important issue in mammography. However, imaging patients with small breast size may cause system underexposure and induce image quality problem. The purpose of this study is to develop an imaging technique to solve the problem for patients with small breast size during mammography.
In this study, several phantoms (such as BR12, PMMA and CDMAM) are employed to investigate the imaging performances for both film-screen mammography (FSM) and computed radiography mammography (CRM) systems. Indexes such as resolution, contrast, contrast-to-noise ratio (CNR) and image quality factor (IQFinv) are used to evaluate system performances respect to varied breast sizes. Histogram equalization and histogram normalization techniques are applied to investigate the possibility of reducing dose during imaging. A mathematical model is employed to develop a better imaging method for patient with small breast size. The parameters in the mathematical model are determined from simulated imaging with several of breast-tissue equivalent phantoms with different thicknesses. Later, this method was evaluated by applying retrospectively analyzed study with 7 patients and prospectively to image a clinic patient.
Results show that underexposure of images are observed, while imaged with patient with insufficient breast length for both imaging systems. For the phantom imaging, the image quality is decreased with the breast phantom length under 4.5 cm. The ranges of contrast are 2%-23% and 24%-25% respective for FSM and CRM system. For image of CRM, the image qualities of proceeded image are improved for the images with high tube loading. Moreover, the image processing algorithms employed in this work may not suitable for reducing radiation exposure in clinic routine procedure. For retrospective study, the IQFinvs of FSM and CRM are 54.6b8.8 and 67.9b18.1 when imaged with the parameters determined by radiographers. However, the IQFinvs of FSM and CRM are improved to 54.6b8.8 and 67.9b18.1 by applied the new developed method.
A novel imaging technique which is helpful in improving image quality of patient with small breast size in mammography is developed. A suitable range of imaging setting is recommended depending on the compressed breast thickness of individual patient. Moreover, this imaging technique can be easily apply to clinic imaging procedure without modified the mammographic equipments.

並列關鍵字

mathematical model ； Computed Radiography Mammography ； Image quality ； Auto exposure control ； Film-Screen Mammography

參考文獻

[17] K. C. Young, "NHSBSP Equipment Report 0501," 2005.

[1] "The report of Department of Health Executive Yuang Republic of China. ," 2006 http://www.doh.gov.tw.

[25] K. C. Young, "NHSBSP Equipment Report 0504," 2005.

[2] L. Tabar, M. F. Yen, B. Vitak, H. H. Chen, R. A. Smith, and S. W. Duffy, "Mammography service screening and mortality in breast cancer patients: 20-year follow-up before and after introduction of screening," Lancet, vol. 361, pp. 1405-10, Apr 26 2003.

[3] "European Guidelines for Quality Assurance in Mammography Screening 3rd edn Report EUR No 14821, Luxembourg," 1999.