| 黄超,胡军武,李丽,方纪成.人工智能辅助压缩感知技术用于腰椎MRI[J].中国医学影像技术,2023,39(7):1080~1083 |
| 人工智能辅助压缩感知技术用于腰椎MRI |
| Artificial intelligence-assisted compressed sensing technology for lumbar MR scanning |
| 投稿时间：2022-12-06 修订日期：2023-05-14 |
| DOI：10.13929/j.issn.1003-3289.2023.07.029 |
| 中文关键词: 腰椎 人工智能 压缩感知 磁共振成像 |
| 英文关键词:lumbar vertebrae artificial intelligence compressed sensing magnetic resonance imaging |
| 基金项目: |
|
| 摘要点击次数: 865 |
| 全文下载次数: 530 |
| 中文摘要: |
| 目的 观察人工智能辅助压缩感知(ACS)技术用于腰椎MRI的价值。方法 前瞻性纳入80例拟接受腰椎MR检查患者,随机将其均分为ACS组及对照组,每组各40例。对ACS组行基于ACS的腰椎MR扫描,对照组行常规序列扫描;比较组间图像信噪比(SNR)、对比度噪声比(CNR)及质量评分,观察ACS技术用于腰椎MR检查的价值。结果 ACS组及对照组扫描时间分别为487 s及232 s。ACS组矢状位快速自旋回波(FSE)脂肪抑制(FS)T2WI、轴位FSE T2WI和FSE FS T2WI的SNR均高于对照组(P均<0.05),组间其余序列图像SNR差异、全部序列图像CNR差异,以及图像质量评分差异均无统计学意义(P均>0.05)。结论 ACS技术用于腰椎MR检查可在保证图像质量的同时缩短扫描时间,值得临床推广。 |
| 英文摘要: |
| Objective To observe the value of artificial intelligence-assisted compressed sensing (ACS) technology for lumbar MR scanning. Methods A total of 80 patients who would undergo lumbar MR examination were prospectively enrolled and randomly divided into ACS group or control group (each n=40). Lumbar MR scanning based on ACS was performed in ACS group, while conventional lumbar MR scanning was performed in control group. The signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR) and quality scores of images were compared between groups, and the value of ACS technology for lumbar MR examination was analyzed. Results The scanning time was 487 s in ACS group and 232 s in control group, respectively. SNR of sagittal section fast spin echo (FSE) fat suppression (FS) T2WI, transverse section FSE T2WI and FSE FS T2WI in ACS group were all higher than those in control group (all P<0.05), while there was no significant difference of SNR of other sequence images, CNR between all sequence images, nor of imaging quality scores between groups (all P>0.05). Conclusion Using ACS technology could ensure image quality and shorten scanning time of lumbar MR examination. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
