超分辨率深度学习重建改善脑FLAIR图像质量并保持WMH容积分析
结果显示SR-DLR相比高斯滤波重建(GA)和去噪DLR(dDLR)显著降低噪声、提高SNR、CNR和锐利度,且WMH容积与高分辨率参考标准一致。
要点
- SR-DLR可显著提升2D脑FLAIR的图像质量,适合日常阅片应用。
- SR-DLR重建不改变自动分割的WMH容积,可用于随访比较。
- SR-DLR对噪声、锐利度和整体质量的定性评分显著优于GA和dDLR。
核心结论
- SR-DLR在3T二维脑FLAIR上显著降低噪声、提高SNR、CNR和锐利度,同时保持与高分辨率参考标准一致的WMH容积。
研究设计与样本
- 图像分别采用GA、dDLR和SR-DLR(两倍平面内升采样)重建,并在标准化感兴趣区测量噪声、SNR、CNR和锐利度。
- WMH阳性亚组中使用基于transformer的U-Net分割模型在GA、dDLR、SR-DLR和高分辨率采集(HR-dDLR)上获取WMH容积。
- 定量指标采用单因素方差分析,WMH容积采用配对t检验,观察者间一致性采用加权kappa。
关键影像表现/征象
- SR-DLR在定量指标上噪声最低,SNR、CNR和锐利度最高,均显著优于GA和dDLR(均p<0.001)。
- 定性评分中噪声、锐利度和整体图像质量在SR-DLR上显著更高(p<0.001),锐利度和整体质量观察者间一致性完全一致。
- 各重建方式间平均WMH容积差异无统计学意义,SR-DLR容积与HR-dDLR高度接近。
诊断标准与关键数值
- 定量指标比较采用单因素方差分析,各组间噪声、SNR、CNR和锐利度均p<0.001。
- 定性评分中噪声、锐利度和整体图像质量SR-DLR显著高于GA和dDLR,p<0.001。
- 锐利度和整体质量加权kappa提示完全一致,但原文未给出具体数值。
- WMH容积各重建间差异无统计学意义,SR-DLR与HR-dDLR容积密切匹配。
临床应用与阅片要点
- SR-DLR可用于日常脑2D FLAIR重建,在提升图像质量的同时不影响WMH定量随访。
- 阅片时SR-DLR图像噪声更低、锐利度更高,有助于更清晰显示脑白质病变边界。
- WMH容积随访可基于SR-DLR重建进行,与高分辨率标准具有一致性。
局限性
- 单中心研究,缺乏外部验证和多厂商设备对比。
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 whether super-resolution deep learning reconstruction (SR-DLR) improves two-dimensional (2D) brain fluid-attenuated inversion recovery (FLAIR) image quality while preserving automated white matter hyperintensity (WMH) volumetry, compared with Gaussian-filtered reconstruction (GA) and denoising DLR (dDLR).
METHODS: Thirty-six healthy volunteers underwent 3T axial 2D FLAIR. Images were reconstructed using GA, dDLR, and SR-DLR (twofold in-plane upscaling). Quantitative metrics (noise, SNR, CNR, sharpness) were measured in standardized regions of interest. Two radiologists independently scored qualitative image quality. In a WMH-positive subgroup, WMH volumes were obtained using a transformer-based U-Net segmentation model on GA, dDLR, SR-DLR, and a high-resolution acquisition (HR-dDLR). Quantitative image-quality metrics were compared using one-way ANOVA, WMH volumes using paired t-tests, and interobserver agreement using weighted kappa.
RESULTS: SR-DLR demonstrated the lowest image noise and the highest SNR, CNR, and sharpness compared with GA and dDLR (all p < 0.001). Qualitative scores for noise, sharpness, and overall image quality were significantly higher for SR-DLR (p < 0.001), with perfect inter-observer agreement for sharpness and overall quality. Mean WMH volumes did not differ significantly across reconstructions, and SR-DLR volumes closely matched those of HR-dDLR.
CONCLUSION: SR-DLR substantially improves 2D brain FLAIR image quality at 3T by reducing noise and increasing SNR, CNR, and sharpness while preserving WMH volumetry consistent with high-resolution reference standards.
引用
- Ryu, Jang, Otgonbaatar, et al. Super-resolution deep learning reconstruction for brain fluid-attenuated inversion recovery: image quality and white matter hyperintensity volume analysis. Neuroradiology. 2026. PMID: 42406027. doi:10.1007/s00234-026-04104-7
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。