以平扫图像噪声设置毫安秒降低肝脏CT增强扫描辐射剂量的研究

Reducing radiation dose in liver enhanced CT scan by setting mAs according to plain scan noise

目的探讨在固定管电流CT上根据图像噪声设置毫安秒，行肝脏增强扫描，降低辐射剂量的可行性。方法前瞻性连续收集在我科行肝脏增强CT扫描的患者100例（A组），采用固定250mAs进行平扫和多期增强扫描，测量平扫和静脉期图像噪声，对图像质量进行评分，并推导出增强扫描毫安秒和平扫图像噪声的方程。再连续收集在我科行肝脏增强CT扫描的患者100例（B组），按照上述方程，由平扫图像噪声确定增强扫描毫安秒，测量静脉期图像噪声，并对图像质量进行评分。按照体质量指数（BMI），把A、B两组患者分为BMI〈18．5kg／m2、18．5kg／m2≤BMI〈25．0kg／m2和BMII〉25．0kg／m2 3组，分别对静脉期图像质量评分进行秩和检验，对CT剂量指数（CTDI）和有效剂量（ED）进行2个独立样本t或f’检验。结果通过A组患者推导出的增强扫描毫安秒（mAsX）与平扫图像噪声（SDp）的方程为：mAsX=mAsl×[（0．989×SDp＋1．06）／SDx]。，其中mAsl=250mAs，靶噪声（SDx）=13。BMI〈18．5kg／m2患者，A组（12例）、B组（10例）ED分别为（6．86±0．38）和（2．66±0．46）mSv，差异有统计学意义（t=18．52，P〈0．01）。18．5kg／m2≤BMI〈25．0kg／m2患者，A组（66例）、B组（73例）ED分别为（7．08±0．91）和（4．50±1．41）mSv，差异有统计学意义（t’=10．57，P〈0．01）。BMI≥25．0kg／m2患者，A组（22例）、B组（17例）ED分别为（7．54±0．62）和（8．19±3．16）mSv，差异无统计学意义（t’=0．89，P=0．39）。静脉期图像质量评分A组5例患者低于3分，图像无法诊断，B组所有患者图像均满足诊断要求。结论在固定管电流CT上行肝脏增强扫描，根据平扫图像噪声大小设置增强扫描毫安秒，可以在保证图像质量的同时降低辐射剂量。

Objective To investigate the feasibility of setting mAs in liver enhanced CT scan according to plain scan noise with fixed mA CT scanner, in order to reduce the radiation dose. Methods One hundred continuous patients underwent liver enhanced CT scan （group A） prospectively. Two hundred and fifty mAs was used in plain and enhanced CT scans. Noises of plain and venous phase CT images were measured, and the image quality was evaluated. The equation between mAs of enhanced scan and noise of plain scan image was derived. Another 100 continuous patients underwent liver enhanced CT scan （ group B）. Enhanced scan mAs was calculated from noise on plain scan by using the equation above. Noises on venous phase images were measured and the image quality was measured. Based on body mass index （BMI） , patients in groups A and B were divided into three subgroups respectively： BMI 〈 18.5 kg/m2, 18.5 kg/m2 ≤ BMI 〈 25.0 kg/m2and BMI ≥ 25.0 kg/m2. Image quality score was compared with nonparametrie rank sum test, CT dose index （CTDI） and effective dose （ED） were measured and compared between each subgroup with 2 independent samples t or t＇test. Results The equation between enhanced scan mAs （mAsX） and plain scan noise （SDp） was as follows ： mAsX = mAsl x [ （0. 989 x SDp ＋ 1.06 ） / SDx]2, mAsl = 250 mAs, SDx = 13. In patients with BMI 〈 18.5 kg/m2, ED of group A [ （6. 86± 0. 38） mSv,n = 12] was significantly higher than group B [ （2. 66 ±0. 46） mSv,n = 10） ] （t = 18.52,P 〈 0. 01 ）. In patients with 18.5 kg/m2≤BMI 〈25.0 kg/m2 ,ED of group A [ （7. 08 ±0. 91 ） mSv,n =66] was significantly higher than group B [ （4. 50± 1.41 ） mSv, n = 73 ] （ t＇ = 10. 57, P 〈 O. 01 ）. In patients with BMI≥25.0 kg/m2 ,there was no significant difference between EDs of group A （7.54 ± 0. 62 mSv, n = 22） and group B E （8. 19 ± 3.16） mSv, n = 17 ]（ t＇ = 0. 89, P = 0. 39）. Image quality of 5 patients in group A and none in group B did not meet the diagnostic requirement. Conclusion Setting mAs of enhanced scan according to plain scan noise could rednee the radiation dose with maintainence of image quality.