利用射束阻擋裝置處理小動物錐形電腦斷層影像的散射修正

Scatter correction in animal cone beam computed tomography (CBCT) using a beam stopper device

程言鼎(Yan-Ding Cheng)

指導教授 ： 莊克士

國立清華大學/原子科學院/生醫工程與環境科學系/碩士(2006年)

https://doi.org/10.6843/NTHU.2006.00617

摘要

目的：在傳統電腦斷層造影中，散射是影響影像品質的一大要素，近年來CBCT大量的被應用，且CBCT的散射量較傳統電腦斷層造影來得大，散射會造成影像的對比度降低、影像像素值低估、雜訊增加、生成杯狀假影，本研究先量測出核研所小動物錐形電腦斷層儀的散射量，再利用射束阻擋裝置處理小動物錐形電腦斷層影像的散射修正，並期望日後可應用在臨床的CBCT。
材料與方法：使用純壓克力假體及鉛擋塊求該系統的SPR，再用對比假體、純壓克力假體及數種BSD預測INER’s Micro CBCT的散射光子分佈，之後再進行CBCT影像修正。BSD的方法，主要是利用 有/無 BSD的兩次投影資料或是單次投影資料預測散射光子分佈，再將散射光子訊號與主射束光子訊號分開，將主射束光子訊號進行重建，即可得到沒有散射光子的CBCT影像。文中除了BSD散射修正法外，尚與Curve Fitting的方法做比較，包括經過散射修正後的散射光子分佈正確性、影像對比度、雜訊、杯狀假影改善程度。
結果與結論：INER’s Micro CBCT的SPR高達37.47%。Point BSD的方法所預測的散射光子分佈較接近真實散射光子分佈，CBCT的影像經過Point BSD方法修正後，影像的對比度最大可增加15%，影像像素值可正確被修正10%以上，杯狀假影可減少約2.5倍，雜訊增加方面與Curve Fitting的方法相比並不會對影像造成多餘負荷。CBCT的影像經BSD散射修正，可有效且正確的被修正，並優於Curve Fitting的方法，若可正確預估T值，可將此方法應用在臨床的CBCT上。

關鍵字

錐形射束電腦斷層掃描 ； 小動物電腦斷層掃描 ； 散射修正法 ； 散射與主射束比值

並列摘要

Introduction: Scatter is an important factor to effect CT image quality, especially in CBCT. X-ray scatter reduces image contrast, increases image noise, introduces pixel value inaccuracy and some image artifact. In this study, we use a Beam Stopper Devices (BSD) to correct the scatter of INER’s Micro CBCT images and expect that the method can be apply to the clinical use of CBCT in the future.
Materials and methods: First, We investigate the scatter to primary ratio (SPR) of INER’s Micro CBCT with a cylinder PMMA phantom and different diameter lead foils. The two cylinder phantoms, include a contrast phantom and a PMMA phantom, were used. And we use beam stopper device (BSD) to estimate the scatter of the projection data. By using two projection data, with/without BSD, we can calculate the scatter distribution. And the scatter in the projection data could be removed exactly . Furthermore, we also use the curve fitting method to correct the CBCT images. Some image index, such like contrast index, cup index, normalized standard deviation, and mean pixel value, were evaluated to compare the correction effect between the BSD and curve fitting method.
Results and conclusions: The SPR of INER’s Micro CBCT is up to 37.47%. Point BSD method can correctly estimate real scatter distribution of projection data. After Point BSD corrected, image contrast could increased 15%, pixel value were restored to original value nearly, and cup index were reduced 2.5 times. The added noise from Point BSD method is less than Curve Fitting method. Overall, BSD method is better than Curve Fitting method according to our experiments. When T-values are estimated correctly, the BSD method used in CBCT will be very valuable in the future.

並列關鍵字

CBCT ； Micro CBCT ； scatter correction ； SPR

參考文獻

A. L. C. Kwan, J. M. Boone, N. Shah 2005 Evaluation of x-ray scatter properties in a dedicated cone-beam breast CT scanner Med. Phys. 32(9) 2967-2975

C. Badea, L.W. Hedlund, and G.A. Johnson 2004 Micro-CT with respiratory and cardiac gating Med. Phys. 31 (12) 3324-3329

D. A. Jaffray and J. H. Siewerdsen 2000 Cone-beam computed tomography with a flat-panel imager : Initial performance characterization Med. Phys. 27(6) 1311-1323

E. Y Sidky, Y. Zou and X. Pan 2004 Impact of polychromatic x-ray sources on helical, cone-beam computed tomography and dual-energy methods Phys. Med. Biol. 49 2293-2303

G. H. Glover 1982 Compton scatter effects in CT reconstructions Med. Phys. 9(6) 860-867