深度学习合成虚拟T2加权抑脂MR图像:多中心研究
本研究开发生成对抗网络(GAN),从标准T1和T2加权图像生成虚拟T2抑脂(T2FS)序列,用于脊柱肿瘤评估以减少MRI扫描时间。回顾性纳入两家机构1389例连续脊柱肿瘤患者,分为训练集1026例、内部验证集257例和外部测试集106例。外部测试集MSE为0.0060±0.0038,SSIM为0.667±0.120,PSNR为23.368±4.298 dB;病灶分割DSC内部0.820±0.176、外部0.807±0.188。合成与真实图像SNR和CNR相当,合成图像抑脂更优,真实图像肿瘤内部细节更佳。
要点
- GAN可从标准T1/T2加权图像生成虚拟T2FS图像,外部测试集MSE 0.0060±0.0038、SSIM 0.667±0.120。
- 合成与真实图像病灶分割一致性高,DSC内部0.820±0.176、外部0.807±0.188。
- 合成图像抑脂表现优于真实图像,但真实图像肿瘤内部细节显示略优。
核心结论
- 所提出的基于GAN的方法生成的虚拟T2FS图像具有诊断价值,可在不损害脊柱肿瘤评估诊断价值的前提下减少MRI扫描时间。
研究设计与样本
- 回顾性研究,纳入两家机构1389例连续脊柱肿瘤患者,分为训练集1026例(49.2±16.4岁,男540例)、内部验证集257例(48.2±17.2岁,男140例)和外部测试集106例(52.8±17.0岁,男59例);模型以T1和T2加权图像为输入生成T2FS图像。
关键影像表现/征象
- 外部测试集合成图像MSE为0.0060±0.0038,SSIM为0.667±0.120,PSNR为23.368±4.298 dB。
- 合成与真实图像在内部和外部数据集中SNR和CNR相当。
- 定性评估显示合成与真实图像整体质量和伪影相当。
- 合成图像抑脂表现优于真实图像,真实图像肿瘤内部细节显示优于合成图像。
诊断标准与关键数值
- 病灶分割DSC内部验证集为0.820±0.176,外部测试集为0.807±0.188。
- 两名经验丰富的放射科医生独立采用5分制评分,评估整体质量、肿瘤细节保留、抑脂表现和伪影。
- 定量图像保真度评估包括MSE、SSIM和PSNR;客观质量采用SNR和CNR测量。
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
OBJECTIVES: To develop a Generative Adversarial Network (GAN) for generating virtual T2 fat-suppressed (T2FS) sequences from standard T1- and T2-weighted images, with the clinical objective of reducing MRI scan time without compromising diagnostic value for spinal tumor assessment.
MATERIALS AND METHODS: This retrospective study included 1,389 consecutive patients with spinal tumors from two institutions, divided into training (n = 1,026; 49.2 ± 16.4 years; 540 males), internal validation (n = 257; 48.2 ± 17.2 years; 140 males), and external test (n = 106; 52.8 ± 17.0 years; 59 males) sets. The model used T1- and T2-weighted images as input to generate T2FS images. Quantitative image fidelity evaluations included mean squared error (MSE), structural similarity index measure (SSIM), and peak signal-to-noise ratio (PSNR). The Dice similarity coefficient (DSC) assessed lesion segmentation. Signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) measured objective quality. Two experienced radiologists independently rated the images on a 5-point scale, evaluating overall quality, tumor detail preservation, fat suppression performance, and artifacts.
RESULTS: The external test set exhibited an MSE of 0.0060 ± 0.0038, SSIM of 0.667 ± 0.120, and PSNR of 23.368 ± 4.298 dB. The real and synthetic images agreed strongly in lesion segmentation, with mean DSC of 0.820 ± 0.176 (internal) and 0.807 ± 0.188 (external). SNR and CNR were comparable between real and synthetic images in both datasets. Qualitative assessments indicated equivalent overall image quality and artifacts. Synthetic images showed superior fat suppression, while real images offered better tumor internal detail.
CONCLUSIONS: The proposed GAN-based method generated diagnostically valuable virtual T2FS images.
KEY POINTS: 1. The proposed deep learning model successfully generated virtual T2FS images from standard T1/T2 MRI, demonstrating favorable quantitative agreement (MSE 0.0060, SSIM 0.667). 2. Synthetic and real images demonstrated strong consistency in lesion segmentation (DSC 0.809-0.824) and comparable SNR/CNR values. 3. While synthetic images provided superior fat suppression, real images maintained slightly better tumor internal detail visualization.
引用
- Wang C, Xu J, Zhang M, et al. Deep learning synthesis of virtual T2-weighted fat-suppressed MR images: a multi-center study. Eur J Radiol. 2026. PMID: 41855768. doi:10.1016/j.ejrad.2026.112797
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。