正子斷層掃描在核醫治療上，扮演了很重要的角色，目前發展的TOF PET加入飛行時間的資訊，因此可更準確的估計互毀光子真正發生的位置，具有降低雜訊，提升訊雜比(SNR)等優勢。由於TOF PET常用在體型較大的物體，但隨掃描物體的增大會使得散射量增加。
目前臨床多使用飛行時間單次散射光子模擬法去除影像中的散射事件，但此方式是使用non-TOF來估算散射擬合的比例常數，因此在最尾端的時間分裝(time bin)會產生散射量高估的情形，因此影響散射修正的準確度。射束擋塊修正法是利用高原子序物質的射束擋塊放置在受測物體周圍，來阻擋部分真實事件，藉由有/無射束擋塊的掃描資訊可直接計算出散射事件的分布。本篇研究目的是提出TOF BS方法修正TOF PET影像中的散射並針對每個TOF bin 散射分率不相同的特性，分別對物體內外散射點做多項式擬合來評估散射分布。
本實驗使用蒙地卡羅進行假體模擬，並與飛行時間單次散射光子模擬法做比較，驗證TOF BS修正效果。目前使用TOF MLEM進行重建，並利用對比、變異係數和均方誤差方式做影像品質評估，結果顯示在對比和均方誤差評估的結果TOF BS方法較TOF SSS修正法能準確的修正整體的散射事件。

Positron emission tomography (PET) plays a dominant role in nuclear medicine. Due to the extremely high time resolution, the TOF PET technology is valuable clinically. Generally, the SNR could be promoted by Gaussian distribution weighting that means the two photons annihilating position of PET with Time-of-flight information could be located by calculating LOR.
Principally, the TOF gain is proportional to the size of the object; however, the scattering photons are also related with the object’s size so that the scatter events will be increasing. Currently the time of flight single scatter simulation (TOF SSS) is usually adopted to remove the scattering events. However, TOF SSS algorithm uses the single scattering quantity of non-TOF to evaluate scaling factor, so it will overestimate the scatter for the both side of tails in TOF bins.
We have developed a scatter correction method for 3D PET using beam stopper device. Several stoppers were placed surrounding the object to attenuate primary events. The object was scanned twice (with and w/o BS), and the information of acquisition could accurately calculate the amount of the total scatter of the object. In this work, we extended the BS method to correct scatter events in TOF PET. Owing to each TOF bins with different scatter distribution and scatter ratio in TOF PET imaging, the TOF BS method could correct it for the different scatter properties of TOF bin.
The Monte Carlo simulation is performed to study the proposed method using two digital phantoms. The corrected effect of the images reconstructed by TOF MLEM algorithm was evaluated using contrast, CV, and MSE parameters. The preliminary result showed that the proposed method is better than TOF SSS on contrast and MSE of images. The proposed TOF BS method could be a feasible way to correct scattered events and it will be applied in clinic experiment in our future work.

[1]Bai B, Li Q, Holdsworth C H, Asma E, Tai Y C,Chatziioannou A and Leahy R M 2002 Model-based normalization for iterative 3D PET image reconstruction. Phys. Med. Biol. 47 2773-84
[2]Badawi R D and Marsden P K 1999 Developments in component-based normalization for 3D PET Phys. Med. Biol. 44 571–94.
[3]Bailey D L and Meikle S R 1994 A convolution-subtraction scatter correction method for 3D PET Phys. Med. Biol. 39 411-24
[4]Beyer T, Townsend D Wand Czernin J 2011The future of hybrid imaging—part 2: PET/CT. Insights Imaging 2 225–34
[5]Chen C L, Wang Y, Lee, J J S, Tsui B M W 2008 Integration of SimSET photon history generator in GATE for efficient Monte Carlo simulations of pinhole SPECT Med. Phys. 35 3278-84
[6]Chuang K S , Wu J, Jan M L , Chen S and Hsu C H 2005 Novel scatter correction for three-dimensional positron emission tomography by use of a beam stopper device Nuclear Instruments and Methods in Physics Research A 551 540–52
[7]Conti M, Bendriem B, Casey M, et al 2005 First experimental results of t ime-of-flight reconstruction on an LSO PET scanner. Phys. Med. Biol. 50 4507–26
[8]Couturier O, Luxen A, Chatal J F et al 2004 Fluorinated tracers for imaging cancer with positron emission tomography Eur. J. Nucl. Med. Mol. Imaging 31 1182–1206
[9]Fletcher J W, Djulbegovic B, Soare H P, et al. 2008 Recommendations on the Use of 18F-FDG PET in Oncology J. Nucl. Med. 49 (3) 480-508
[10]Gonias P., Bertsekas N., and Karakatsanis N., et al. 2007 Validation of a GATE model for the simulation of the Siemens biographTM 6 PET scanner. Nucl. Instr. Meth. 571 263-66