Department of Diagnostic and Interventional Radiology, Medical Faculty, RWTH Aachen University, Aachen, Germany.
Eur Radiol Exp. 2024 May 1;8(1):53. doi: 10.1186/s41747-024-00451-3.
To compare denoising diffusion probabilistic models (DDPM) and generative adversarial networks (GAN) for recovering contrast-enhanced breast magnetic resonance imaging (MRI) subtraction images from virtual low-dose subtraction images.
Retrospective, ethically approved study. DDPM- and GAN-reconstructed single-slice subtraction images of 50 breasts with enhancing lesions were compared to original ones at three dose levels (25%, 10%, 5%) using quantitative measures and radiologic evaluations. Two radiologists stated their preference based on the reconstruction quality and scored the lesion conspicuity as compared to the original, blinded to the model. Fifty lesion-free maximum intensity projections were evaluated for the presence of false-positives. Results were compared between models and dose levels, using generalized linear mixed models.
At 5% dose, both radiologists preferred the GAN-generated images, whereas at 25% dose, both radiologists preferred the DDPM-generated images. Median lesion conspicuity scores did not differ between GAN and DDPM at 25% dose (5 versus 5, p = 1.000) and 10% dose (4 versus 4, p = 1.000). At 5% dose, both readers assigned higher conspicuity to the GAN than to the DDPM (3 versus 2, p = 0.007). In the lesion-free examinations, DDPM and GAN showed no differences in the false-positive rate at 5% (15% versus 22%), 10% (10% versus 6%), and 25% (6% versus 4%) (p = 1.000).
Both GAN and DDPM yielded promising results in low-dose image reconstruction. However, neither of them showed superior results over the other model for all dose levels and evaluation metrics. Further development is needed to counteract false-positives.
For MRI-based breast cancer screening, reducing the contrast agent dose is desirable. Diffusion probabilistic models and generative adversarial networks were capable of retrospectively enhancing the signal of low-dose images. Hence, they may supplement imaging with reduced doses in the future.
• Deep learning may help recover signal in low-dose contrast-enhanced breast MRI. • Two models (DDPM and GAN) were trained at different dose levels. • Radiologists preferred DDPM at 25%, and GAN images at 5% dose. • Lesion conspicuity between DDPM and GAN was similar, except at 5% dose. • GAN and DDPM yield promising results in low-dose image reconstruction.
比较去噪扩散概率模型（DDPM）和生成对抗网络（GAN）在从虚拟低剂量减影图像中恢复增强型乳腺磁共振成像（MRI）减影图像方面的性能。
回顾性、伦理批准的研究。使用定量测量和放射学评估，将 50 例具有增强病变的乳房的 DDPM 和 GAN 重建的单张减影图像与原始图像在三个剂量水平（25%、10%、5%）进行比较。两位放射科医生根据重建质量对其进行偏好，并将病变的显著性与原始图像进行比较，对模型进行盲评。对 50 例无病变的最大强度投影进行假阳性评估。使用广义线性混合模型比较模型和剂量水平之间的结果。
在 5%的剂量下，两位放射科医生都更喜欢 GAN 生成的图像，而在 25%的剂量下，两位放射科医生都更喜欢 DDPM 生成的图像。在 25%和 10%的剂量下，GAN 和 DDPM 的病变显著性评分无差异（5 与 5，p=1.000；4 与 4，p=1.000）。在 5%的剂量下，两位读者均认为 GAN 的病变显著性高于 DDPM（3 与 2，p=0.007）。在无病变的检查中，DDPM 和 GAN 在 5%（15%与 22%）、10%（10%与 6%）和 25%（6%与 4%）的假阳性率方面无差异（p=1.000）。
GAN 和 DDPM 均在低剂量图像重建中取得了有希望的结果。然而，在所有剂量水平和评估指标下，它们都没有表现出优于其他模型的结果。需要进一步开发来减少假阳性。
对于基于 MRI 的乳腺癌筛查，降低造影剂剂量是理想的。扩散概率模型和生成对抗网络能够对低剂量图像的信号进行回顾性增强。因此，它们可能会在未来补充低剂量成像。
•深度学习可能有助于恢复低剂量对比增强型乳腺 MRI 的信号。•两种模型（DDPM 和 GAN）在不同剂量水平进行训练。•放射科医生在 25%的剂量下更喜欢 DDPM，在 5%的剂量下更喜欢 GAN 图像。•DDPM 和 GAN 之间的病变显著性相似，除了在 5%的剂量下。•GAN 和 DDPM 在低剂量图像重建中均取得了有希望的结果。
