三维酰胺质子转移成像在原发性肝癌临床扫描可行性及图像质量评估中的初步研究

Preliminary study of 3D-APTw imaging in the evaluation of clinical scanning feasibility and image quality of primary liver cancer

目的 评估参数优化的三维酰胺质子转移加权(3 dimensions-amide proton transfer weighted,3D-APTw)成像在原发性肝癌扫描中的图像质量并分析其临床扫描的可行性。材料与方法 前瞻性招募2020年10月至2022年2月临床怀疑原发性肝癌患者,共109例。采用飞利浦3.0 T MRI设备进行扫描,分别采集T1WI、T2WI、弥散加权成像、酰胺质子转移加权(amide proton transfer weighted,APTw)及多期增强图像。机器自动生成经过B0校正的肝脏APTw图像。计算3D-APTw成像的扫描成功率及计算成功率。采用Kendall一致性系数对观察者内及观察者间主观评分进行一致性分析,对主观评分3分以上的病例测量3D-APTw图像的信噪比(signal-to-noise ratio,SNR)、肿瘤对比噪声比(contrast-to-noise ratio,CNR)、肿瘤及肝实质的APTw值并计算各参数值的变异系数,采用组内相关系数进行观察者内及观察者间可重复性测量的一致性分析。结果 109例患者中有11例患者3D-APTw图像中病变区域存在信号缺失,扫描成功率为89.91%(98/109),剩余98例患者中78例3D-APTw图像的主观评分为3分以上,计算成功率为71.56%(78/109)。观察者内及观察者间图像质量主观评分一致性分别为0.771和0.692,P<0.01。肿瘤组织APTw值观察者内及观察者间ICC分别为0.822和0.811,肝实质APTw值、SNR及CNR观察者ICC分别为0.675、0.634、0.666,肝实质APTw值、SNR及CNR观察者间ICC分别为0.614、0.290、0.560。肿瘤组织APTw值高于肝实质[(2.55±0.08)%vs.(1.45±0.07)%],两者间差异有统计学意义(P<0.001)。肿瘤组织和肝实质APTw值的变异系数分别为30.40%和44.40%。SNR和CNR为25.92±18.50和3.35±3.20。SNR变异系数为71.40%,CNR变异系数为90.00%。结论 参数优化后的3D-APTw成像可以用于原发性肝癌的临床扫描,但容易受呼吸运动、胆囊或血管的影响而出现信号缺失或伪影,在参数设置及成像技术上仍有需要改进和提高的空间,以进一步优化其图像质量。

Objective:To investigate the feasibility of parameter optimized 3 dimensions amide proton transfer weighted(3D-APTw)imaging for image quality evaluation and clinical scanning in primary liver cancer.Materials and Methods:A total of 109 patients with suspected primary liver cancer were prospectively collected from October 2020 to February 2022.Philips 3.0 T MRI equipment was used for scanning,and T1WI,T2WI,diffusion-weighted imaging,amide proton transfer weighted(ATPw)and multi-phase enhanced images were collected.The machine automatically generates an 3D-APTw image of the liver corrected by B0.The scanning success rate and calculation success rate of 3D-APTw imaging were calculated.Kendall agreement consistency test was used to analyze the subjective score of inter-observer and intra-observer APTw image quality.SNR and CNR,APTw of tumor and liver were measured for cases with subjective scores above 3 and calculated and the coefficient of variation was calculated for each parameter value.Intra-group correlation coefficient was used to analyze the consistency of intra-observer and inter-observer repeatability measurements.Results:Among the 109 patients,11 patients had signal loss in the lesion area in 3D-APTw images,and the scanning success rate was 89.91%(98/109).Among the remaining 98 patients,78 patients had APTw image subjective score of more than 3 points,and the calculation success rate was 71.56%(78/109).The intraobserver and interobserver subjective scores of 3D-APTw image quality with a correlation coefficient of 0.771 and 0.692,P<0.01.ICC of intraobserver and interobserver agreement with APTw values in tumor tissue was 0.822 and 0.811,while ICC of intraobserver agreement with APTw,SNR and CNR in hepatic tissue was 0.675,0.634,0.666.ICC of interobserver agreement with APTw,SNR and CNR in hepatic tissue was 0.614,0.290,0.560.APTw value of tumor tissue was higher than that of liver tissue[(2.55±0.08)%vs.(1.45±0.07)%],and the difference was statistically significant(P<0.001).The coefficients of variation of APTw values in tumor tissue and liver were 30.40%and 44.40%,respectively.SNR and CNR were 25.92±18.50 and 3.35±3.20.The coefficient of variation was 71.40% for SNR and 90.00% for CNR. Conclusions: Parameter-optimized 3D-APTw imaging can be used for clinicalscanning of primary liver cancer, but it is prone to signal loss or artifact due to the influence of respiratory movement, gallbladder orblood vessels, so there is still room for improvement in parameter setting and imaging technology to further optimize its image quality.