加速MRI数据集
收藏资源简介:
加速MRI数据集是由浙江大学的研究人员创建的,包含来自18个公共来源的原始k空间数据,共计110万张图像。该数据集旨在研究数据筛选策略对MRI重建的影响,并构建了一个包含48个测试集的评估集,以捕捉解剖结构、对比度、线圈数量等因素的变化。数据集创建过程中,研究人员提出了不同的数据筛选策略,以提高当前最先进的神经网络在加速MRI重建方面的性能。该数据集适用于研究MRI重建、图像恢复等领域,旨在解决加速MRI重建过程中图像质量下降的问题。
The accelerated MRI dataset was created by researchers from Zhejiang University. It contains raw k-space data from 18 public sources, with a total of 1.1 million images. This dataset aims to investigate the impact of data filtering strategies on MRI reconstruction, and an evaluation set consisting of 48 test subsets was constructed to capture variations in factors such as anatomical structures, image contrasts, and coil counts. During the dataset creation process, researchers proposed various data filtering strategies to improve the performance of state-of-the-art neural networks for accelerated MRI reconstruction. This dataset is applicable to research in fields including MRI reconstruction and image restoration, and targets the issue of degraded image quality during accelerated MRI reconstruction.




