透過您的圖書館登入
IP:18.119.126.80
  • 學位論文

應用能量加權於X光相位對比影像之觀察者效能評估

Application of Energy-Weighting on Observer Performance Assessment in X-ray Phase-Contrast Imaging

指導教授 : 周呈霙
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


相位對比成像法藉由取得相位資訊,以達到提高影像對比度的效果。許多相位對比成像的理論是以單色光作為光源,限制了在實際應用上的可行性,因為傳統的X光管的輸出為含有各種不同能量的多色光。藉由光子計數探測器的可分辨能量特性,讓我們可使用多色光源,並藉由妥善運用多色光含有的資訊,可提升影像的品質。本研究之目標為應用能量加權於相位對比影像之觀察者效能的評估。 比較三種方法計算出的能量加權影像、平均影像和能量積分影像之觀察者效能。能量加權影像和平均影像是由光子計數探測器來測量光強度後,做相位擷取並加權,不同之處在於能量加權影像使用計算出之權重,而平均影像在不同能量之權重皆相等。而能量積分探測器之輸出只有一個光強度數值,相位擷取後直接計算觀察者效能。 計算加權後之相位和吸收影像以及用能量積分探測器計算得到相位和吸收影像之觀察者效能,我們可以評估訊號偵測的效能。以能量加權的方式計算相位和吸收影像,可得到比平均影像較佳的觀察者效能,亦優於使用能量積分探測器所得到的影像。

關鍵字

X光 相位對比 能量加權 訊號偵測

並列摘要


The phase contrast imaging methods utilize phase information to improve image contrast. Many phase-contrast imaging formulas are based on monochromatic light as the light source. Because traditional X-ray tube output is polychromatic, it limits the usage in practical applications. Photon counting detectors can distinguish between different photon energies and their outputs are energy-bin intensity data. The energy-bin of each output channel can be determined by setting the energy threshold of the photon counting detector. The image quality can be enhanced by proper use of the information contained in polychromatic light. The objective of this study is to use energy-weighting technique on phase contrast images, and assess the performance of the observer. The observer performance of the energy-weighted image, the averaged image and image with energy integrating detectors were compared. We used the weight derived from phase retrieval formulas for a particular energy-bin phase image to get the energy-weighted phase images. The weights of all energy-bin images were the same in the averaged image. By calculating the observer performance of weighted and averaged images with photon counting detectors and the images with the energy integrating detectors, we could evaluate the observer performance of the signal detection. The energy-weighted image was better than the averaged image in observer performance, and was better than the images obtained with the energy integrating detectors.

參考文獻


Anastasio, M.A., C.Y. Chou, A.M. Zysk, and J.G. Brankov. 2010. Contributions to ideal observer SNRs in propagation-based x-ray phase-contrast imaging. In "Proc. Medical Imaging 2010: Physics of Medical Imaging", 76220S. San Diego, California, USA: SPIE.
Anastasio, M.A., C.Y. Chou, A.M. Zysk, and J.G. Brankov. 2010. Ideal observer analysis of signal detectability in phase-contrast imaging employing linear shift-invariant optical systems. Journal of the Optical Society of America A: Optics and Image Science, and Vision. 27(12): 2648-2659.
Ballabriga, R., M. Campbell, E.H.M. Heijne, X. Llopart, and L. Tlustos. 2006. The Medipix3 Prototype, a Pixel Readout Chip Working in Single Photon Counting Mode with Improved Spectrometric Performance. In "Proc. IEEE Nuclear Science Symposium Conference Record", 3557-3561. San Diego, CA, USA: IEEE.
Barrett, H.H., C.K. Abbey, and E. Clarkson. 1998. Objective assessment of image quality. III. ROC metrics, ideal observers, and likelihood-generating functions. Journal of the Optical Society of America A: Optics and Image Science, and Vision. 15(6): 1520-1535.
Barrett, H.H. 1990. Objective assessment of image quality: effects of quantum noise and object variability. Journal of the Optical Society of America A: Optics and Image Science, and Vision. 7(7): 1266-1278.

延伸閱讀