数字乳腺X线摄影中管电压与滤过材料的优化选择研究

A study for the optimization of tube voltage and filter in digital mammography

目的探讨数字乳腺摄影中不同乳房压迫厚度下管电压与滤过材料与图像质量和辐射剂量的关系。方法全野数字乳腺摄影机中，采用钼（Mo）-Mo和Mo-铑（Rh）靶滤过组合，对2～7 cm厚度的CDMAM模体选择合适的管电流量，管电压范围为21～39 kVp，进行手动曝光，计算图像影像质量因子（IQF）、对比度噪声比（CNR），记录平均腺体剂量（AGD）。通过计算品质因子（FOM），找出各种厚度下最优化滤过材料和管电压，同时对不同厚度下的Mo和Rh滤过的IQF、CNR、FOM和AGD结果进行配对样本秩和检验。结果在Mo和Rh滤过时，2～7 cm压迫厚度下，固定相应的管电流，随着管电压增加，AGD、CNR和IQF呈递增趋势，AGD明显增加而CNR和IQF逐渐增加并到平衡。在各厚度下，Mo和Rh滤过间的AGD差异均有统计学意义（P均〈0.05）。压迫厚度为2、3 cm时，Mo滤过的CNR高于Rh滤过，差异有统计学意义（P〈0.05），Mo和Rh滤过间的IQF差异无统计学意义（P〉0.05）；压迫厚度为4 cm时，2种滤过间的CNR差异无统计学意义（P〈0.05），Mo滤过的IQF高于Rh滤过，差异有统计学意义（P〈0.05）；压迫厚度为5～7 cm时，Mo滤过的CNR低于Rh滤过，但仅在6 cm下差异有统计学意义（P〈0.05），5、7 cm下2种滤过间的CNR差异无统计学意义（P〉0.05），2种滤过间的IQF差异均无统计学意义（P〉0.05）。在Mo、Rh滤过下各厚度的FOM差异均有统计学意义（P〈0.05），FOMRh〉FOMMo。不同压迫厚度时，管电压和滤过材料的适宜匹配结果为Rh滤过，2 cm和27 kVp、3 cm和29 kVp、4 cm和29 kVp、5 cm和30 kVp、6 cm和31 kVp、7 cm和32 kVp。结论Rh滤过是图像质量和辐射剂量优化的最佳滤过，在各种厚度时都有对应的最佳能量响应。

ObjectiveTo investigate the image quality and average glandular dose from different breast compression thickness, and to provide recommendations for filtration material and tube voltage in digital mammography. MethodsThe CDMAM phantom of 2 to 7 cm thickness were exposed by manual exposure modes（proper mAs, 21 to 39 kVp range） using Mo/Mo and Mo/Rh anode/filer combinations at full-field digital mammography. The image quality figure （IQF）, contrast to noise ratio （CNR）, figure of merit （FOM） and the average glandular dose （AGD） were obtained from images. The optimum filtration and kVp for each breast thickness were found from the calculated FOMs. And the paired sample rank sum test was used to analyze the difference of AGD, CNR, IQF and FOM for Mo and Rh filters. ResultsFor both Mo and Rh as used filter, under the compression of 2 to 7 cm thickness and the fixed mAs, with the increasing of tube voltage, AGD, CNR and IQF will gradually increase as following： AGD had the significant increasing while CNR and IQF gradually increasing to balance. Under different compression thickness, the AGD for Mo and Rh were statistically different （P〈0.05）. When the compression thickness were 2 and 3 cm, CNR for Mo was higher than Rh, with the statistically significant difference （P〈0.05） while the IQF for Mo and Rh with no statistical significance （P〉0.05）.When the compression thickness was 4 cm, CNR for Mo and Rh were with no statistical significance （P〈0.05） while IQF for Mo was higher than Rh, with statistically significant difference （P〈0.05）. When the compression thickness were 5 to 7 cm, CNR for Mo was lower than Rh, but only under the thickness of 6 cm with statistically significant difference （P〈0.05）, while there was no statistical significance （P〉0.05） under the thickness of 5 cm and 7 cm; the IQF for Mo and Rh with no statistical significance （P〉0.05）. The FOM in each thickness for Mo and Rh filtration had statistical significance （P〈0.05）, and FOMRh〉FOMMo. For different compression thickness, the suitable matching of tube voltage and filter material were Rh filtration, 2 cm and 27 kVp, 3 cm and 29 kVp, 4 cm and 29 kVp, 5 cm and 30 kVp, 6 cm and 31 kVp, 7 cm and 32 kVp. ConclusionConsidering both image quality and radiation dose, Rh filtration is the optimum selection material and there is a corresponding optimal energy response in all kinds of thickness.