ABSTRACT
Objective:To evaluate the feasibility and the application values of quantitative susceptibility mapping (QSM) for the assessment of meniscal injury and in distinguishing meniscus degeneration and tears.Methods:The clinical and imaging data of 70 patients suspected of meniscus injury and scheduled for arthroscopy in Shandong Medical Imaging Research Institute, Cheeloo College of Medicine, Shandong University from November 2019 to June 2020 were analyzed retrospectively. Thirty age-and sex-matched healthy subjects were also examined as controls. All subjects received knee joint QSM and routine MR imaging. According to the results of arthroscopy, the patients was divided into meniscus degeneration and meniscus tear groups, respectively. The conventional MR was evaluated by two radiologists. The meniscus injury area was delineated on the original QSM magnitude images (the central area of the posterior corner of the lateral meniscus was selected in the healthy controls) and mapped to the corresponding QSM maps, and the magnetic susceptibility values were measured. Kruskal-Wallis H test was used to analyze the magnetic sensitivity values of meniscal degeneration, meniscal tear and healthy control groups; and Bonferroni was used to correct the pairwise comparison. ROC curve was established to evaluate the threshold and efficacy of magnetic susceptibility value in the diagnosis of meniscal tear. The results were compared with those of conventional MRI. Results:The magnetic susceptibility values of meniscus of healthy controls, meniscal degeneration and meniscal tear groups were (0.035±0.016)ppm, -0.031(-0.040,-0.005)ppm, and(-0.122±0.115)ppm, respectively, with significant difference found among the three groups (χ2=44.419, P<0.05). The magnetic susceptibility values of meniscus of healthy controls was significantly higher than those of meniscus degeneration patients and meniscus tear patients (χ2=-23.843, -48.253, P<0.05). The magnetic susceptibility values of meniscus of meniscus tear group was significant lower than those of meniscus degeneration group (χ2=-24.410, P<0.05). Taking magnetic susceptibility values of -0.062 5 ppm as threshold, the area under the ROC curve for the diagnosis of meniscal tears was 0.949, with the sensitivity as 87% and the specificity as 100%. The sensitivity and specificity of conventional MRI in the diagnosis of meniscal tears were 86.8% and 87.5%, respectively. Conclusion:QSM can quantitatively evaluate meniscus injury and can be used as an effective supplement method to conventional MRI, which is helpful to improve the diagnosis of meniscus tear.
ABSTRACT
Objective To explore the advantages of susceptibility weighted imaging (SWI) in depiction of normal fetal vertebra and vertebral anomalies.Methods This prospective study was approved by our institutional review board, and written informed consent was obtained from every participant, Fifty-eight pregnant women (gestation age 22 to 39 weeks, average 29 ± 3 weeks) who were suspected of carrying babies with vertebral anomalies by ultrasound screening underwent 1.5 T fetal spine MRI[including half-fourier acquisition single-shot turbo spin-echo(HASTE),true fast imaging with steady-state(True FISP) and SWI sequences]. MR images were reviewed for their quality by two radiologists independently. The image scores in HASTE, True FISP and SWI were compared by using Kruskal-Wallis test and Mann-Whitney U test. Three segments (cervical, thoracic and lumbosacral segments, respectively) of 15 fetuses were, at random, collected to compare among HASTE,True FISP and SWI and then evaluated by ANOVA analysis.The diagnostic accuracy of the three sequences among 32 cases with follow-up results was calculated respectively and compared by using Chi-square test. Results There was statistical differences among three sequences(χ2=50.685,P<0.05).The scores of SWI was higher than that of True FISP, and the scores of True FISP was higher than that of HASTE in the evaluation of the fetal vertebra(P all<0.05).The differences among cervical,thoracic and lumbosacral segments on True FISP and HASTE showed significant difference statistically (P all<0.05), also the image quality of cervical segment could not meet the requirement of clinical diagnosis. The image quality of SWI was favorable clinically and no statistical difference among three segments was found(P>0.05).A total of 32 fetal vertebral anomalies were identified by follow-up after birth including hemivertebra (n=14), fusion of vertebrae (n=1), butterfly vertebra (n=1), multiple vertebral malformations(n=9),spinal bifida(n=5),caudal regression syndrome(n=2).The diagnostic accuracy of SWI, True FISP and HASTE was 93.75% (30/32), 56.25% (18/32) and 37.50% (12/32) respectively.The diagnostic accuracy of SWI was the best compared to that of True-FISP and HASTE(χ2=10.083,20.017;P<0.01). Conclusion SWI proved to be the optimal technique in depiction of fetal vertebra and vertebral anomalies than True FISP and HASTE,especially in depiction of cervical vertebra.
ABSTRACT
Objective To assess the optimal visiualization capacity of brachial plexus with three-dimensional nerve-sheath signal increased with inked rest-tissue rapid acquisition of relaxation imaging (3D SHINKEI), exploring the feasibility of preliminary diagnostic value on brachial plexus diseases. Methods MRI scans were performed on 24 healthy volunteers with no history of brachial plexus injury, and 46 patients whose outcomes of lesions had been verified as post-ganglionic brachial plexus injuries by surgery or clinical follows-up . The scan series consist 3D SHINKEI, STIR in the coronal plane as well as DW-MRN in the axial plane using a 3.0 T MR system. The source and post-processed images of 3D SHINKEI and DW-MRN were scored according to the optimal visibility on brachial plexus, in the meanwhile, contrast-to-noise ratio of the original images in the 3D SHINKEI and STIR sequences were calculated separately. Two radiologists blindly compared the detection rate of positive brachial plexus injuries between 3D SHINKEI and STIR in 46 patients. And then analyze the outcomes by means of Kappa test, Mann-Whitney test , independent sample t test, and Chi-square test. Results Post-ganglionic brachial plexus showed high intensity in the 3D SHINKEI sequence. In the 24 healthy volunteers, the scores by the two radiologists were 3.6 ± 0.6, 3.5 ± 0.6, 3.0 ± 0.2, 2.9 ± 0.1, respectively. There was statistical difference between the two sequences (Z=2.667,P=0.008,P<0.05). And the Kappa was 0.8 and 0.6 with favorable consistency. The CNR of 3D SHINKEI and STIR were 0.61 ± 0.07, 0.42 ± 0.03 (t=12.78, P=0.001, P<0.05). The positive detection rates of post-ganglionic brachial plexus injuries on 3D SKINKEI and STIR were, 78.3%, 52.2%(χ2=9.421, P<0.05). Conclusions 3D SHINKEI sequence demonstrates robust visibility consistently and can clearly display the structures and signals of post-ganglionic abnormality, compared with DW-MRN and STIR. This technique can be helpful to provide more complementary information to further confirm the diagnosis of brachial plexus injuries.