Compressed Sensing Technology for Meniscus Injury Assessment
Objective:To explore the feasibility of compressed sensing(CS)for meniscus imaging and the value of diagnosis and evaluation for meniscus injury.Methods:MRI examination of knee joint was performed in 31 patients with initially knee injury(injury group)and 15 healthy volunteers without knee injury(control group),whose CS-3D-MRI(compression ratio of 9)were collected,while proton den-sity weighted spectral attenuated inversion recovery(PDW-SPAIR)images were collected from the con-trol group for comparison,the scanning time for these two sequences were 6 min and 2 min 49s,respec-tively,the results of arthroscopic examination in the injury group were recorded.Patients in the injury group received MRI examination before the arthroscopy,and the CS-3D-MRI were generated using the reconstruction of CS-DSstrong,and the consistency between the diagnosis results of reconstructed CS im-ages and those of arthroscopic was analyzed.The image quality of control group was evaluated by appar-ent signal-to-noise ratio(SNR),contrast-to-noise ratio(CNR),and image quality score.Results:The SNR,CNR and image quality scores of reconstructed CS images and PDW-SPAIR images were 2.71± 0.71 and 2.43±0.68,-19.96±4.47 and-17.39±4.53,2.90±0.28 and 2.93±0.18,respectively.All the recon-structed CS images showed meniscus injuries in the injury group,which had substantial consistency with the arthroscopic diagnosis results(Kappa=0.67,P＜0.05).Conclusion:CS-3D-MRI can be used for knee meniscus imaging,the image signal-to-noise ratio is similar to that of the conventional PDW-SPAIR image,and the imaging time can be significantly shortened,which can be used to evaluate knee meniscus injury.
NETL
NSTL
万方数据
