深度学习增强的0.55T单次屏气腹部MRI——技术可行性与图像质量评估
本研究纳入33名健康志愿者,在0.55T下比较常规与深度学习增强的T2 HASTE和T1 VIBE-Dixon腹部MRI图像质量;深度学习增强单次屏气序列图像质量优于常规序列(均P<0.001),图像噪声和空间分辨率更佳(P<0.001),单次与多次屏气深度学习序列间无质量差异(P>0.572),偶然病灶检出数量与大小一致。
要点
- 深度学习增强使0.55T腹部T2 HASTE和T1 VIBE-Dixon可在单次屏气下完成,图像质量优于常规序列。
- 单次屏气与多次屏气深度学习HASTE序列图像质量无差异(均4[4-4],P>0.572)。
- 深度学习与常规序列检出的偶然病灶数量和大小一致(16个,平均直径8±5 mm,P=1.000)。
核心结论
- 基于深度学习的图像重建使0.55T腹部T2 HASTE和T1 VIBE-Dixon成像可在单次屏气下完成,图像质量优于常规MRI。
研究设计与样本
- 2024年7月至2024年9月纳入33名健康志愿者(平均年龄30±4岁),在0.55T下进行常规与深度学习增强的T2 HASTE、脂肪抑制T2 HASTE(HASTE FS)和T1 VIBE-Dixon采集,由三名腹部放射科医生以Likert 1-5分量表评估图像质量参数、伪影和偶然发现,并进行读者间一致性和比较分析。
关键影像表现/征象
- 深度学习增强单次屏气T2 HASTE图像质量评分4[IQR:4-4]优于常规T2 HASTE的3[3-3](P<0.001)。
- 深度学习增强单次屏气脂肪抑制T2 HASTE图像质量评分4[4-4]优于常规HASTE FS的3[3-3](P<0.001)。
- 深度学习增强T1 VIBE-Dixon图像质量评分4[4-5]优于常规T1 VIBE-Dixon的4[3-4](P<0.001)。
- 深度学习MRI扫描的图像噪声和空间分辨率均优于常规扫描(P<0.001)。
诊断标准与关键数值
- 图像质量采用Likert 1-5分量表评估,读者间一致性良好或更好(κ≥0.61)。
- 深度学习增强单次屏气与多次屏气HASTE或HASTE FS序列图像质量均评为4[4-4],无差异(P>0.572)。
- 偶然病灶检出数量与大小在技术间一致(16个病灶;平均直径8±5 mm;P=1.000)。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
RATIONALE AND OBJECTIVES: Inherently lower signal-to-noise ratios hamper the broad clinical use of low-field abdominal MRI. This study aimed to investigate the technical feasibility and image quality of deep learning (DL)-enhanced T2 HASTE and T1 VIBE-Dixon abdominal MRI at 0.55 T.
MATERIALS AND METHODS: From July 2024 to September 2024, healthy volunteers underwent conventional and DL-enhanced 0.55 T abdominal MRI, including conventional T2 HASTE, fat-suppressed T2 HASTE (HASTE FS), and T1 VIBE-Dixon acquisitions, and DL-enhanced single- (HASTE DLSBH) and multi-breath-hold HASTE (HASTE DLMBH), fat-suppressed single- (HASTE FS DLSBH) and multi-breath-hold HASTE (HASTE FS DLMBH), and T1 VIBE-Dixon (VIBE-DixonDL) acquisitions. Three abdominal radiologists evaluated the scans for quality parameters and artifacts (Likert scale 1-5), and incidental findings. Interreader agreement and comparative analyses were conducted.
RESULTS: 33 healthy volunteers (mean age: 30±4years) were evaluated. Image quality was better for single breath-hold DL-enhanced MRI (all P<0.001) with good or better interreader agreement (κ≥0.61), including T2 HASTE (HASTE DLSBH: 4 [IQR: 4-4] vs. HASTE: 3 [3-3]), T2 HASTE FS (4 [4-4] vs. 3 [3-3]), and T1 VIBE-Dixon (4 [4-5] vs. 4 [3-4]). Similarly, image noise and spatial resolution were better for DL-MRI scans (P<0.001). No quality differences were found between single- and multi-breath-hold HASTE DL or HASTE FS DL (both: 4 [4-4]; P>0.572). The number and size of incidental lesions were identical between techniques (16 lesions; mean diameter 8±5 mm; P=1.000).
CONCLUSION: DL-based image reconstruction enables single breath-hold T2 HASTE and T1 VIBE-Dixon abdominal imaging at 0.55 T with better image quality than conventional MRI.
引用
- Seifert AC, Breit HC, Obmann MM, et al. Deep Learning-Enhanced Single Breath-Hold Abdominal MRI at 0.55 T-Technical Feasibility and Image Quality Assessment. Acad Radiol. 2025. PMID: 40846585. doi:10.1016/j.acra.2025.08.013
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。