Jie Gao, Xinxin Li, Zhonghai Chi, Zhixin Sun, Zhongyu Chen, Yakui Wang, Yi Yi, Yingyi Qi, Zhiang Wang, Qian Zhao, Zhuozhao Zheng
In the phantom, SNR was 196.14 versus 183.34, PIU 92.47% versus 71.30%, and ghost-to-signal ratio 0.0093 versus 0.0145 (metasurface vs. commercial coil). Measured flip angle was 44.7° ± 1.31° (0.67% deviation from nominal 45°); B1 field uniformity was 91.14%. In volunteers, subjective image quality scores were slightly higher for the 12-channel coil on FS PDW FSE (p < 0.05), whereas scores were comparable on T1W SE and 3D FS T1W GRE (p = 0.06, 0.38). Convenience was markedly better for the metasurface (5.0 vs. 4.0, p < 0.05). In patients, diagnostic sensitivity (72.7%-83.3% vs. 72.7%-81.5%), specificity (86.7%-96.4% vs. 85.6%-96.4%), and accuracy (82.1%-90.4% vs. 81.4%-88.5%) were comparable between coils, inter-coil kappa was 0.884-1.000. "DATA" CONCLUSION: The wireless metasurface coil was more convenient and provided image quality and diagnostic performance comparable to the 12-channel commercial knee coil at 1.5 T.
BACKGROUND: Knee MRI image quality depends largely on the radiofrequency coil; metasurfaces offer a new approach to improving coil performance.
PURPOSE: To evaluate the imaging performance and diagnostic capability of a metasurface coil for knee MRI.
STUDY TYPE: Prospective.
PHANTOM/POPULATION: A cylindrical phantom, 60 healthy volunteers (36 females; median age, 33), and 52 patients with knee injuries (28 females; median age, 47).
FIELD STRENGTH/SEQUENCES: T1 Weighted Spin Echo (T1W SE), Fat-Saturated Proton Density-Weighted Fast Spin-Echo (FS PDW FSE), T1-Weighted Gradient Recalled Echo (T1W GRE) at 1.5 T.
ASSESSMENT: All subjects underwent paired scanning with a wireless metasurface coil (plus built-in spine array) and a commercial 12-channel knee coil. Phantom metrics included signal-to-noise ratio (SNR), percent image uniformity (PIU), ghost-to-signal ratio, and B1 mapping. Volunteers were scored for image quality (5-point), SNR, contrast-to-noise ratio (CNR), and handling convenience. Patients were assessed for ACL, meniscal, and cartilage injuries against arthroscopy.
STATISTICAL TESTS: T-test and Wilcoxon signed-rank test for quantitative and semi-quantitative data; McNemar test for diagnostic efficacy. Kendall's coefficient of concordance and kappa, with p < 0.05 considered significant.
RESULTS: In the phantom, SNR was 196.14 versus 183.34, PIU 92.47% versus 71.30%, and ghost-to-signal ratio 0.0093 versus 0.0145 (metasurface vs. commercial coil). Measured flip angle was 44.7° ± 1.31° (0.67% deviation from nominal 45°); B1 field uniformity was 91.14%. In volunteers, subjective image quality scores were slightly higher for the 12-channel coil on FS PDW FSE (p < 0.05), whereas scores were comparable on T1W SE and 3D FS T1W GRE (p = 0.06, 0.38). Convenience was markedly better for the metasurface (5.0 vs. 4.0, p < 0.05). In patients, diagnostic sensitivity (72.7%-83.3% vs. 72.7%-81.5%), specificity (86.7%-96.4% vs. 85.6%-96.4%), and accuracy (82.1%-90.4% vs. 81.4%-88.5%) were comparable between coils, inter-coil kappa was 0.884-1.000. "DATA" CONCLUSION: The wireless metasurface coil was more convenient and provided image quality and diagnostic performance comparable to the 12-channel commercial knee coil at 1.5 T.
EVIDENCE LEVEL: 2.
TECHNICAL EFFICACY: Stage 3.