基于X射线相衬显微CT的肝组织脱水优化研究

Optimization of liver tissue dehydration by X-ray phase contrast computed tomography

目的采用X射线相衬显微CT(synchrotron radiation-based X-ray phase-contrast computed tomography,SR-XPCT),对处理后正常C57BL/6小鼠肝脏组织进行成像研究,以阐明不同组织样品处理方法间成像的差异,进而为SR-XPCT实验样品前处理提供参考。方法72个正常C57BL/6小鼠肝脏样品随机分为4%多聚甲醛组(P组,亚组P1~P4)和10%中性福尔马林溶液组(F组,亚组F1~F4),亚组采用4种脱水浓度处理,干燥后进行X射线成像,观察各组软组织结构并进行比较。结果F组的血管边缘灰度及图像信噪比均高于对应P组,差异有统计学意义(P<0.05)。在P组内,F2血管边缘灰度差均低于P3、P4,差异有统计学意义(P<0.05);P2血管直径均高于P3、P4,差异有统计学意义(P<0.05);P2断层感兴趣区域(ROI,region of interest)中对比度噪声比(Signal to Noise Ratio,CNR)均高于其余P1、P3和P4,差异有统计学意义(P<0.05)。在F组内,F2血管边缘灰度差高于F1、F2和F3,差异有统计学意义(P<0.05);F2血管直径高于F4,差异有统计学意义(P<0.05);F2断层ROI中CNR均高于F1、F4,差异有统计学意义(P<0.05)。结论室温下采用10%中性福尔马林固定液固定,40%低起始脱水梯度缓慢过渡到100%的处理,实验样品具有脱水形变小,血管及软组织重构精度高、图像信噪比较高的特点,对诊断和分析生物医学软组织和器官的微小病理差异具有相对优势。

Objective Imaging of normal C57 BL/6 mice liver tissue was performed with synchrotron radiation-based X-ray phase-contrast computed tomography(SR-XPCT) to elucidate different tissue sample processing methods. The difference in imaging is to provide a reference for the pretreatment of SR-XPCT experimental samples. Methods The liver samples of 72 normal C57 BL/6 mice were randomly divided into paraformaldehyde group(P group, subgroup P1-P4) and formalin group(F group, subgroup F1-F4). Subgroups were dehydrated by 4 concentrations. X-ray imaging was performed after that, and the soft tissue structures of each group were observed and compared. Results The vascular edge gray scale and image signal-to-noise ratio of group F were higher than those of corresponding group P(P<0.05). In group P, the gray scale difference of F2 blood vessel edge was lower than that of P3 and P4, and the differences were statistically significant(P<0.05);the diameter of P2 blood vessel was higher than that of P3 and P4, and the differences were statistically significant(P<0.05);the CNRs in the P2 of the P2 fault were higher than those of the P1, P3 and P4, and the differences were statistically significant(P<0.05). In group F, the gray level difference of F2 blood vessel edge was higher than that of F1, F2 and F3, and the differences were statistically significant(P<0.05);the blood vessel diameter of F2 was higher than that of F4, and the difference was statistically significant(P<0.05);the CNRs in ROI of F2 fault were higher than that of F1 and F4, and the differences were statistically significant(P<0.05). Conclusion The results of SR-XPCT imaging experiments show that the best condition is that using 10% neutral formalin fixed solution at a starting temperature of 4℃ with a 40% low initial dehydration gradient and slowly transiting to 100% treatment(ie. F2 group). The characteristics of small dehydration deformation, high precision of vascular and soft tissue reconstruction, and high image signal-to-noise have a comparative advantage in the diagnosis and analysis of microscopic pathological differences in biomedical soft tissues and organs.