摘要
目的:探讨临床环境中通过优化扫描参数结合基于深度学习的复合超分辨率重建算法在提升膝关节MRI扫描效率和图像质量的可行性。方法:前瞻性搜集110例行膝关节MRI平扫的患者,先后进行常规(常规组)与复合超分辨率重建算法扫描(复合组),采用双盲法比较两组主客观图像质量。结果:相较常规组,复合组PD和T1序列的骨髓、软骨、半月板、韧带、肌肉、脂肪、关节液的SNR分别提升89.3%、52.5%、65.3%、73.8%、60.3%、103.9%、58.9%和78.0%、172.9%、78.0%、72.5%、75.4%、63.4%、97.0%。相较常规组,复合组PD和T1序列的软骨-关节液、软骨-骨髓、半月板-关节液、韧带-关节液、骨髓-关节液、脂肪-关节液、肌肉-关节液的CNR分别提升119.5%、83.3%、116.2%、109.2%、109.2%、99.3%、116.8%和61.7%、23.1%、78.7%、32.5%、161.7%、44.9%、39.2%。复合组的峰值信噪比(PSNR)相较常规组显著提高(P<0.05),结构相似度(SSIM)均>0.999。主观图像质量评价中复合组病灶边缘区分度、运动伪影和综合诊断度的主观评分显著高于常规组(P<0.05),两组病灶辨别度的主观评分差异无统计学意义(P>0.05)。结论:合理优化扫描参数并结合基于深度学习的复合超分辨率重建算法可在提升扫描效率的同时显著提高膝关节MRI的图像质量和综合诊断效果。
Objective:The purpose of this study was to explore the feasibility of optimizing scanning parameters combined with a deep learning-based composite super-resolution reconstruction algorithm in improving the efficiency and image quality of knee joint MRI scanning in clinical settings.Methods:A prospective collection of 110 cases of knee joint MRI plain scans was conducted,followed by routine and composite super-resolution reconstruction algorithm scans.The subjective and objective image quality of the two groups was compared using a double-blind method.Results:The SNR of bone marrow,cartilage,meniscus,ligament,muscle,fat,and joint fluid in the super group PdW and T 1W was significantly increased compared with the conventional group.The CNR of cartilage synovial fluid,cartilage bone marrow,meniscus synovial fluid,ligament synovial fluid,bone marrow synovial fluid,fat synovial fluid,and muscle synovial fluid in the super grouped PdW and T 1W groups was significantly increased compared with the conventional group.The PSNR of the super group was significantly improved compared with the conventional group,and there was no statistically significant difference in SSIM.The subjective scores of edge discrimination,motion artifacts,and comprehensive diagnosis in the supergroup lesion were significantly higher than those in the conventional group in subjective image quality,and there was no difference in subjective scores of lesion discrimination.Conclusion:Reasonable optimization of scanning parameters combined with a composite super-resolution reconstruction algorithm based on deep learning can significantly improve the image quality and comprehensive diagnostic performance of knee joint MRI while improving scanning efficiency.
作者
王超
谢晓亮
朱熹
黄文诺
尚松安
叶靖
王志军
WANG Chao;XIE Xiao-liang;ZHU Xi(Department of Radiology,Northern Jiangsu People′s Hospital,Jiangsu 225000,China)
出处
《放射学实践》
北大核心
2025年第1期67-72,共6页
Radiologic Practice
基金
国家自然科学基金项目(82202120)
扬州市科技计划(社会发展)项目(YZ2023142)。
关键词
卷积神经网络
深度学习
膝关节
磁共振成像
超分辨率重建
Convolutional neural network
Deep learning
Knee joint
Magnetic resonance imaging
Super-resolution reconstruction
作者简介
王超(1986-),男,江苏扬州人,硕士,主管技师,主要从事医学影像新技术与人工智能研究工作;通讯作者:王志军,E-mail:wangzhijun20042024@163.com。
