摘要
目的利用体外模型实验,确定PET/CT图像中生物靶区边界的阈值及不同PET图像采集条件下对阈值的可能影响,为放射治疗计划中应用生物靶区奠定方法学基础。方法利用Flanged Jaszczak Ect Phantom及等比稀释法建立不同直径与靶本比浓度的热区模型,应用Profile曲线软件,将在CT图像上显示的热区边界准确复制到PET图像中,通过最大阈值法计算勾画生物靶区体积的阈值;并研究在不同总计数率、靶本比浓度、直径大小和采集模式下对热区边界阈值变化的影响。结果靶本比对PET图像中靶区边界阈值大小的确定存在明显差异,当靶本比分别为2∶1和≥4∶1时,其阈值分别为0.54±0.10和0.41±0.09(P<0.05)。热区直径、总计数率及采集模式对靶区边界阈值的计算没有明显差异。结论使用PET/CT图像确定生物靶区体积时,应当根据图像中靶本比大小选择合适的阈值进行勾画,可近似准确地计算靶区体积;总计数率、直径大小与采集模式对靶区边界的确定没有明显影响。
Objective After a series of phantom studies to determine an accurate and universal applicable method for defining a biological target volume contour with integrated PET/CT. Methods A method was tested by a phantom study to determine a threshold based on target/background ratio and diameters of “hot” spot for PET/CT based GTV definition. The degree to which target/background ratio and diameters of “hot” spot influenced that GTV definition was evalu-ated. Results T/N ratio can affect the limit of the threshold, when T/N ratio equal to 2: 1 and ≥ 4: 1, respec-tively, limit value of the threshold was 0.54 ± 0.10 and 0.41 ± 0.09 ( P 〈 0.05 ). Total counts, diameter of sphere and imaging mode,offered no significant difference of the threshold value. Conclusion T/N ratio can affect the limit of the threshold. Diameters of sphere, total counts and imaging mode were not independent factors to affect the threshold.
出处
《上海第二医科大学学报》
CSCD
北大核心
2005年第12期1225-1228,共4页
Acta Universitatis Medicinalis Secondae Shanghai
基金
上海市教委重点发展基金(04BA03)
上海市科委重点基金(04JC14044)
上海市"百人计划"第三期(97BR012)资助项目.