We analyzed the data from 36 patients (18 non-RA patients and 18 RA patients) who complained of pain and underwent MRI between April 2008 and August 2021. From the MRI scans of these patients, 279 radiomics features were extracted using STIR image data of the ROIs on the lateral pterygoid muscle of patients with RA and analyzed using MaZda ver. 3.3. Seven gray-level co-occurrence matrix features (Sum entropy, Sum variance) were picked up using the Fisher coefficient, for comparison between the RA and non-RA groups. Data analysis was performed using the Mann-Whitney U test A P value of < 0.05 was considered as statistically significant.
All seven lateral pterygoid muscle radiomic features indicated significant differences between the non-RA and RA groups (P < 0.05).
MRI texture analysis shows potential for application in radiomics diagnosis of RA in TMJ.
方法:我们分析了36例患者(18例非RA患者和18例RA患者)的数据,这些患者在2008年4月至2021年8月期间主诉疼痛并接受了MRI检查。从这些病人的核磁共振扫描结果来看,使用RA患者翼外肌ROI的STIR图像数据提取279个影像组学特征,并使用MaZdaver进行分析。3.3.七个灰度共生矩阵特征(和熵,Sum方差)使用Fisher系数,RA和非RA组之间的比较。使用Mann-WhitneyU检验进行数据分析。〈0.05的P值被认为是统计学上显著的。
结果:所有7个翼外肌X线组学特征均表明非RA组和RA组之间存在显着差异(P<0.05)。
结论:MRI纹理分析显示在TMJRA的影像组学诊断中的应用潜力。