深度学习重建改善乳腺多激发DWI质量并缩短扫描时间
回顾性研究纳入146例新诊断乳腺癌女性,比较标准读出分段EPI与深度学习重建多激发DWI。结果显示DLR序列在b=800及合成b=1500时SNR、CNR及病灶对比度均显著提高,扫描时间缩短29%。独立验证队列39例同样观察到SNR、CNR和整体图像质量改善。
要点
- DLR多激发DWI较rs-EPI扫描时间缩短29%,病灶对比度和整体图像质量更高。
- DLR序列肿瘤ADC略低、纤维腺体组织ADC略高,阅片判读时需注意序列间差异。
- b=800时伪影抑制更优,b=1500合成图像同样显示质量改善。
核心结论
- 深度学习重建可提高乳腺多激发DWI的定量与定性图像质量,同时缩短扫描时间。
- 研究结果支持将DLR多激发DWI整合用于临床乳腺MRI检查。
研究设计与样本
- 回顾性研究,纳入146例新诊断乳腺癌女性,平均年龄55.8岁,均行术前乳腺MRI。
- 每例同时采集rs-EPI和原型DLR多激发DWI序列。
- 独立验证队列纳入39例连续患者,采用相同方法分析。
- 定量指标包括SNR、CNR、病灶对比度和ADC值,两名放射科医师独立进行定性评分。
- 定性评分采用5分Likert量表,评价整体图像质量、伪影抑制、病灶显著性和纤维腺体组织抑制。
关键影像表现/征象
- DLR多激发DWI在b=800及合成b=1500时SNR、CNR和病灶对比度均显著高于rs-EPI。
- DLR序列b=800时伪影抑制评分更高,整体图像质量和病灶显著性评分均优于rs-EPI。
- DLR序列肿瘤ADC值略低于rs-EPI,纤维腺体组织ADC值略高于rs-EPI。
- 验证队列中DLR的SNR、CNR和整体图像质量改善仍可观察到,但ADC值在两种技术间无显著差异。
诊断标准与关键数值
- 观察者间一致性为高度至几乎完全一致,加权κ值为0.753至0.981。
- DLR多激发DWI较rs-EPI采集时间缩短29%。
- b=800及合成b=1500时SNR、CNR、病灶对比度比较均p<0.05。
- 肿瘤ADC值及纤维腺体组织ADC值在两种序列间差异均p<0.05。
临床应用与阅片要点
- 对乳腺DWI图像质量不满意或有明显伪影的病例,可考虑采用DLR多激发DWI重建。
- 报告ADC值时需注意DLR序列与rs-EPI序列间肿瘤及腺体ADC存在系统性差异。
- 合成b=1500图像在DLR序列下病灶对比度更高,可用于提高DWI对病灶的显示能力。
局限性
- 回顾性单中心研究设计。
- DLR为原型重建技术,需要进一步验证其在不同设备和中心的通用性。
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
PURPOSE: To evaluate the effectiveness of deep learning (DL)-based reconstruction for improving image quality and reducing scan time in multi-shot diffusion-weighted imaging (DWI) of the breast, compared with standard readout-segmented echo-planar imaging (rs-EPI).
MATERIALS AND METHODS: This retrospective study included 146 women (mean age: 55.8 years) with newly diagnosed breast cancer who underwent preoperative breast MRI with both rs-EPI and prototype DL-reconstructed (DLR) multi-shot DWI. Quantitative metrics-signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), lesion contrast, and apparent diffusion coefficient (ADC)-were compared between sequences. Two radiologists independently assessed overall image quality, artifact suppression, lesion conspicuity, and fibroglandular tissue suppression using a 5-point Likert scale. An independent validation cohort of 39 consecutive patients was analyzed using the same methodology.
RESULTS: Compared with rs-EPI, DLR multi-shot DWI demonstrated significantly higher SNR, CNR, and lesion contrast at both b = 800 and synthetic b = 1500 s/mm2 (all p < 0.05). Tumor ADC values were slightly lower and fibroglandular tissue ADC values slightly higher with DLR (both p < 0.05). Qualitative assessments demonstrated higher scores for overall image quality and lesion conspicuity with DLR, with improved artifact suppression at b = 800 s/mm2. Interobserver agreement was substantial to almost perfect (weighted κ = 0.753-0.981). DL reconstruction reduced acquisition time by 29% compared with rs-EPI. In the validation cohort, improvements in SNR, CNR, and overall image quality were also observed, whereas ADC values did not differ significantly between techniques.
CONCLUSION: DL-based reconstruction improves quantitative and qualitative image quality in multi-shot breast DWI while reducing scan time. These findings support its potential integration into clinical breast MRI.
引用
- Kim, Kim, Hwangbo, et al. Deep learning-based reconstruction for multi-shot breast DWI: Quantitative and qualitative improvements. Eur J Radiol. 2026. PMID: 42594608. doi:10.1016/j.ejrad.2026.113159
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。