A Deep Learning Based Anti-aliasing Self Super-resolution Algorithm for MRI.
A Deep Learning Based Anti-aliasing Self Super-resolution Algorithm for MRI.
复制标题
DOI：
10.1007/978-3-030-00928-1_12
复制
发表时间：
2018-09
期刊：
影响因子：
--
通讯作者：
中科院分区：
文献类型：
--
作者：
在许多临床应用中需要高分辨率磁共振（MR）图像，然而获得具有足够信噪比的此类数据需要很长时间，使其成本高昂且容易受到运动伪影的影响。部分实现这一目标的一种常见方法是获得具有良好平面内分辨率和较差平面分辨率（即大切片厚度）的MR图像。对于这种二维成像协议，还引入了通平面方向的混叠，并且这些高频伪影无法通过传统的插值去除。超分辨率（SR）算法可以减少混叠伪影，提高空间分辨率。最先进的SR方法大多是基于学习的，需要外部训练数据，包括配对的低分辨率（LR）和高分辨率（HR） MR图像。然而，由于扫描仪的限制，这样的训练数据往往是不可用的。提出了一种不需要外部训练数据的抗混叠和自超分辨率算法。它利用了这些MR图像的面内切片包含高频信息的事实。我们的算法包括三个步骤：1)构建自AA （SAA）深度网络，2)SSR深度网络，这两个深度网络都可以在原始图像的不同方向上应用，3)使用傅立叶突发积累对步骤1和步骤2的多个方向输出进行重组。我们将SAA+SSR算法应用于多种MR数据，除了N4非均匀性校正外，无需修改或预处理，结果表明与竞争对手的SSR方法相比，SAA+SSR算法有显著改进。
High resolution magnetic resonance (MR) images are desired in many clinical applications, yet acquiring such data with an adequate signal-to-noise ratio requires a long time, making them costly and susceptible to motion artifacts. A common way to partly achieve this goal is to acquire MR images with good in-plane resolution and poor through-plane resolution (i.e., large slice thickness). For such 2D imaging protocols, aliasing is also introduced in the through-plane direction, and these high-frequency artifacts cannot be removed by conventional interpolation. Super-resolution (SR) algorithms which can reduce aliasing artifacts and improve spatial resolution have previously been reported. State-of-the-art SR methods are mostly learning-based and require external training data consisting of paired low resolution (LR) and high resolution (HR) MR images. However, due to scanner limitations, such training data are often unavailable. This paper presents an anti-aliasing (AA) and self super-resolution (SSR) algorithm that needs no external training data. It takes advantage of the fact that the in-plane slices of those MR images contain high frequency information. Our algorithm consists of three steps: 1) We build a self AA (SAA) deep network followed by 2) an SSR deep network, both of which can be applied along different orientations within the original images, and 3) recombine the multiple orientations output from Steps 1 and 2 using Fourier burst accumulation. We perform our SAA+SSR algorithm on a diverse collection of MR data without modification or preprocessing other than N4 inhomogeneity correction, and demonstrate significant improvement compared to competing SSR methods.
登录
查看更多内容
DOI：
10.1007/978-3-319-46726-9_64
发表时间：
2016-10
期刊：
Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention
影响因子：
--
作者：
Jog A;Carass A;Prince JL
通讯作者：
Prince JL
影响因子：
10.6
作者：
Tustison NJ;Avants BB;Cook PA;Zheng Y;Egan A;Yushkevich PA;Gee JC
通讯作者：
Gee JC
影响因子：
5.7
作者：
Luesebrink, Falk;Wollrab, Astrid;Speck, Oliver
通讯作者：
Speck, Oliver
DOI：
10.1109/tbme.2012.2218246
发表时间：
2012-12
期刊：
IEEE transactions on bio-medical engineering
影响因子：
--
作者：
Woo J;Murano EZ;Stone M;Prince JL
通讯作者：
Prince JL
影响因子：
1.4
作者：
Greenspan, Hayit
通讯作者：
Greenspan, Hayit