Abstract:

To remove Gibbs artifacts in Magnetic Resonance Imaging （MRI）， a Self-attention connection UNet based on Self-Distillation training （SD-SacUNet） algorithm was proposed. In order to reduce the semantic gap between the encoding and decoding features at both ends of the skip connection in the UNet framework and help to capture the location information of artifacts， the output features of each down-sampling layer at the UNet encoding end was input to the corresponding self-attention connection module for the calculation of the self-attention mechanism， then they were fused with the decoding features to participate in the reconstruction of the features. Self-distillation training was performed on the network decoding end， by establishing the loss function between the deep and shallow features， the feature information of the deep reconstruction network was used to guide the training of the shallow network， and at the same time， the entire network was optimized to improve the level of image reconstruction quality. The performance of SD-SacUNet algorithm was evaluated on the public MRI dataset CC359， with the Peak Signal-to-Noise Ratio （PSNR） of 30.261 dB and the Structure Similarity Index Measure （SSIM） of 0.917 9. Compared with GRACNN （Gibbs-Ringing Artifact reduction using Convolutional Neural Network）， the proposed algorithm had the PSNR increased by 0.77 dB and SSIM increased by 0.018 3； compared with SwinIR （Image Restoration using Swin Transformer）， the proposed algorithm had the PSNR increased by 0.14 dB and SSIM increased by 0.003 3. Experimental results show that SD-SacUNet algorithm improves the image reconstruction performance of MRI with Gibbs artifacts removal and has potential application values.

Key words: Magnetic Resonance Imaging (MRI) reconstruction, deep learning, self-distillation, Transformer, UNet, attention mechanism

摘要：

为去除磁共振成像（MRI）中的吉布斯伪影，提出一种基于自蒸馏训练的自注意力连接UNet （SD-SacUNet）算法。为了缩小UNet框架中跳连接两端编码和解码特征之间的语义差距，帮助捕捉伪影的位置信息，将UNet编码端每个下采样层的输出特征分别输入各自的自注意力连接模块进行自注意力机制的运算，而后与解码特征进行融合，参与特征的重建；在网络解码端进行自蒸馏训练，通过建立深层与浅层特征之间的损失函数，使深层重建网络的特征信息可以用于指导浅层网络的训练，同时优化整个网络，提升图像重建水平。在公开的MRI数据集CC359上评估SD-SacUNet算法的性能，获得的峰值信噪比（PSNR）为30.26 dB，结构相似性（SSIM）为0.917 9；与GRACNN （Gibbs-Ringing Artifact reduction using Convolutional Neural Network）、SwinIR （Image Restoration using Swin Transformer）相比，SD-SacUNet的PSNR分别提高了0.77 dB、0.14 dB，SSIM分别提高了0.018 3、0.003 3。实验结果表明，SD-SacUNet算法提升了MRI去除吉布斯伪影的图像重建性能，具备潜在的应用价值。

关键词: 磁共振成像重建, 深度学习, 自蒸馏, Transformer, UNet, 注意力机制