获取途径

作者

Cheng, Hang Hou, Xuewen Huang, Gang Jia, Shouqiang Yang, Guang Nie, Shengdong   

作者单位

8) (ISNI:0000 0000 9188 055X)  6) (ISNI:0000 0004 1764 1277)  ¹  University of Shanghai for Science and Technology, School of Health Science and Engineering, Shanghai, China (GRID:grid.26713 3f) (ISNI:0000 0004 0369 6365)  ⁵  East China Normal University, Shanghai Key Laboratory of Magnetic Resonance, Department of Physics, Shanghai, China (GRID:grid.2206 ³  Shanghai University of Medicine and Health Sciences, Shanghai Key Laboratory of Molecular Imaging, Shanghai, China (GRID:grid.50703 ⁴  Jinan People’s Hospital Affiliated to Shandong First Medical University, Department of Radiology, Jinan Shandong, China (GRID:grid.41058 f)  ²  Shanghai United Imaging Healthcare Co., Ltd., Radiotherapy Business Unit, Shanghai, China (GRID:grid.49784

刊名

Journal of Digital Imaging

年份

2024

卷号

Vol.37 No.4

页码

1969-1979

ISSN

0897-1889

摘要

Magnetic resonance imaging occupies a pivotal position within contemporary diagnostic imaging modalities, offering non-invasive and radiation-free scanning. Despite its significance, MRI’s principal limitation is the protracted data acquisition time, which hampers broader practical application. Promising deep learning methods for undersampled magnetic resonance image reconstruction outperform the traditional approaches in terms of speed and image quality. However, the intricate inter-coil cor...更多

Magnetic resonance imaging occupies a pivotal position within contemporary diagnostic imaging modalities, offering non-invasive and radiation-free scanning. Despite its significance, MRI’s principal limitation is the protracted data acquisition time, which hampers broader practical application. Promising deep learning methods for undersampled magnetic resonance image reconstruction outperform the traditional approaches in terms of speed and image quality. However, the intricate inter-coil correlations have been insufficiently addressed, leading to an underexploitation of the rich information inherent in multi-coil acquisitions. In this article, we proposed a method called “Multi-coil Feature Fusion Variation Network” , which introduces an encoder to extract the feature from multi-coil MR image directly and explicitly, followed by a feature fusion operation. Coil reshaping enables the 2D network to achieve satisfactory reconstruction results, while avoiding the introduction of a significant number of parameters and preserving inter-coil information. Compared with VN, MFFVN yields an improvement in the average PSNR and SSIM of the test set, registering enhancements of 0.2622 dB and 0.0021 dB respectively. This uplift can be attributed to the integration of feature extraction and fusion stages into the network’s architecture, thereby effectively leveraging and combining the multi-coil information for enhanced image reconstruction quality. The proposed method outperforms the state-of-the-art methods on fastMRI dataset of multi-coil brains under a fourfold acceleration factor without incurring substantial computation overhead.收起

文献类型

期刊