基于同轴相衬成像方法的人体膝关节软骨可视化

Visualization of knee cartilage based on in-line phase contrast imaging

目的同轴相位衬度成像对软组织具有高衬度和高分辨率的特性。利用该成像技术,通过纹理特征参数提取进行定量分析,可为骨性关节炎的早期诊断提供可行性的依据。方法研究样本来自人工关节置换手术患者的膝关节软骨。样本分为正常组与早期骨性关节炎(osteoarthritis,OA)组,每组18例膝关节软骨组织。利用同轴相衬成像(in-line phase contrast imaging,IL-PCI)技术,观察两组软骨组织的纹理差异,并运用行程长度矩阵与分形维数方法提取纹理特征参数长行程加重(long run emphasis,LRE)、短行程加重(short run emphasis,SRE)、程分数(run percentage,RP)、灰度级的非均匀性(gray level non-uniformity,GLN)、行程长度的非均匀性(run length non-uniformity,RLN)以及分形维数。通过独立样本t检验方法比较上述参数,量化分析正常与病变软骨组织纹理差异。结果利用IL-PCI技术可以清晰观察到正常与早期OA软骨中微结构的变化。通过纹理参数提取比较两组的软骨纹理特征,结果表明正常组的LRE明显高于早期OA组(P<0.01),而早期OA组的GLN、RLN远高于正常组(P<0.01),正常组的分形维数小于早期OA组(P<0.01)。结论同轴相衬技术可为骨性关节炎的早期诊断提供定量分析手段。

Objective The in-line phase contrast imaging(IL-PCI)technology has the properties of high contrast and resolution for soft tissue.In this work,the imaging of knee cartilage with this technique is used to provide a feasible basis for the early diagnosis of osteoarthritis through quantitative analysis.Methods The samples were in vitro cartilage of human knee joints,which were obtained from patients undergoing artificial joint replacement surgery.The samples were divided into the normal group and the early osteoarthritis(OA)group,with 18 cases of isolated knee cartilage in each group.IL-PCI technology was used to observe the differences between the two groups of cartilage texture,and the stroke length matrix and fractal dimension methods were used to extract the cartilage texture features including long run emphasis(LRE),short run emphasis(SRE)and run percentage(RP),gray level non-uniformity(GLN),run length non-uniformity(RLN),and fractal dimension.The above parameters were compared with the independent-samples t-test method,and the differences were analyzed quantitatively for the measure of normal and pathological cartilage tissue texture.Results The changes of microstructure in early OA cartilage could be clearly observed by IL-PCI.The results showed that LRE in the normal group was significantly higher than that in the early OA group(P<0.01).The GLN and RLN of the early OA group were significantly higher than those of the normal group(P<0.01),the fractal dimension of the normal group was lower than that of the early OA group(P<0.01).Conclusions IL-PCI technique can provide quantitative analysis for early diagnosis of osteoarthritis.