Graf Robert, Platzek Paul-Sören, Riedel Evamaria Olga, Kim Su Hwan, Lenhart Nicolas, Ramschütz Constanze, Paprottka Karolin Johanna, Kertels Olivia Ruriko, Möller Hendrik Kristian, Atad Matan, Bülow Robin, Werner Nicole, Völzke Henry, Schmidt Carsten Oliver, Wiestler Benedikt, Paetzold Johannes C, Rueckert Daniel, Kirschke Jan Stefan
Department of Diagnostic and Interventional Neuroradiology, School of Medicine, Technical University of Munich, Munich, Germany.
Institut für KI und Informatik in der Medizin, Klinikum rechts der Isar, Technical University of Munich, Munich, Germany.
Eur Radiol. 2025 Apr;35(4):1761-1771. doi: 10.1007/s00330-024-11047-1. Epub 2024 Sep 4.
To generate sagittal T1-weighted fast spin echo (T1w FSE) and short tau inversion recovery (STIR) images from sagittal T2-weighted (T2w) FSE and axial T1w gradient echo Dixon technique (T1w-Dixon) sequences.
This retrospective study used three existing datasets: "Study of Health in Pomerania" (SHIP, 3142 subjects, 1.5 Tesla), "German National Cohort" (NAKO, 2000 subjects, 3 Tesla), and an internal dataset (157 patients 1.5/3 Tesla). We generated synthetic sagittal T1w FSE and STIR images from sagittal T2w FSE and low-resolution axial T1w-Dixon sequences based on two successively applied 3D Pix2Pix deep learning models. "Peak signal-to-noise ratio" (PSNR) and "structural similarity index metric" (SSIM) were used to evaluate the generated image quality on an ablations test. A Turing test, where seven radiologists rated 240 images as either natively acquired or generated, was evaluated using misclassification rate and Fleiss kappa interrater agreement.
Including axial T1w-Dixon or T1w FSE images resulted in higher image quality in generated T1w FSE (PSNR = 26.942, SSIM = 0.965) and STIR (PSNR = 28.86, SSIM = 0.948) images compared to using only single T2w images as input (PSNR = 23.076/24.677 SSIM = 0.952/0.928). Radiologists had difficulty identifying generated images (misclassification rate: 0.39 ± 0.09 for T1w FSE, 0.42 ± 0.18 for STIR) and showed low interrater agreement on suspicious images (Fleiss kappa: 0.09 for T1w/STIR).
Axial T1w-Dixon and sagittal T2w FSE images contain sufficient information to generate sagittal T1w FSE and STIR images.
T1w fast spin echo and short tau inversion recovery can be retroactively added to existing datasets, saving MRI time and enabling retrospective analysis, such as evaluating bone marrow pathologies.
Sagittal T2-weighted images alone were insufficient for differentiating fat and water and to generate T1-weighted images. Axial T1w Dixon technique, together with a T2-weighted sequence, produced realistic sagittal T1-weighted images. Our approach can be used to retrospectively generate STIR and T1-weighted fast spin echo sequences.
从矢状位T2加权（T2w）快速自旋回波序列和轴位T1加权梯度回波狄克逊技术（T1w-Dixon）序列生成矢状位T1加权快速自旋回波（T1w FSE）和短反转时间反转恢复（STIR）图像。
这项回顾性研究使用了三个现有数据集：“波美拉尼亚健康研究”（SHIP，3142名受试者，1.5特斯拉）、“德国国家队列”（NAKO，2000名受试者，3特斯拉）和一个内部数据集（157名患者，1.5/3特斯拉）。我们基于两个相继应用的3D Pix2Pix深度学习模型，从矢状位T2w FSE和低分辨率轴位T1w-Dixon序列生成合成矢状位T1w FSE和STIR图像。在消融测试中，使用“峰值信噪比”（PSNR）和“结构相似性指数度量”（SSIM）来评估生成的图像质量。进行了一项图灵测试，由七位放射科医生将240张图像评定为是原始采集的还是生成的，使用错误分类率和弗莱iss卡帕评分者间一致性进行评估。
与仅使用单个T2w图像作为输入相比（T1w FSE的PSNR = 23.076/24.677，SSIM = 0.952/0.928；STIR的PSNR = 23.076/24.677，SSIM = 0.952/0.928），纳入轴位T1w-Dixon或T1w FSE图像可使生成的T1w FSE（PSNR = 26.942，SSIM = 0.965）和STIR（PSNR = 28.86，SSIM = 0.948）图像具有更高的图像质量。放射科医生难以识别生成的图像（T1w FSE的错误分类率：0.39±0.09，STIR的错误分类率：0.42±0.18），并且在可疑图像上显示出较低的评分者间一致性（T1w/STIR的弗莱iss卡帕值：0.09）。
轴位T1w-Dixon和矢状位T2w FSE图像包含足够的信息来生成矢状位T1w FSE和STIR图像。
T1w快速自旋回波和短反转时间反转恢复可以追溯添加到现有数据集中，节省MRI时间并实现回顾性分析，例如评估骨髓病变。
单独的矢状位T2加权图像不足以区分脂肪和水以及生成T1加权图像。轴位T1w狄克逊技术与T2加权序列一起可生成逼真的矢状位T1加权图像。我们的方法可用于回顾性生成STIR和T1加权快速自旋回波序列。
