基于稀疏变换学习的改进灵敏度编码重建算法

灵敏度编码(SENSE)是一种利用多个接收线圈的灵敏度信息来减少扫描时间的技术。 基于 SENSE 算法的磁共振成像重建算法的重建图像存在模糊伪影和细节缺失等问题,不利于临床医学诊断。 为减少磁共振重建图像伪影并提高重建图像质量,将数据驱动的自适应稀疏变换学习(TL)引入 SENSE 算法中,得到一种 TL-SENSE 算法。该算法利用交替方向乘子法进行求解,通过变换更新、硬阈值去噪和图像更新实现并行磁共振成像重建。 仿真实验结果表明,所提算法对图像去噪和修复效果较好,能完整保留纹理细节和边缘轮廓信息,目标图像与原始图像的一致性较高。 对所选 48 组数据, TL-SENSE 算法重建图像的平均信噪比相比 SENSE 算法、 L1-SENSE 算法、TV-SENSE算法和 LpTV-SENSE 算法,分别提高了 4.62 dB、1.91 dB、1.30 dB 和 0.89 dB。

An Improved Sensitivity Encoding Reconstruction Algorithm Based onSparse Transform Learning

The sensitivity encoding ( SENSE) technique utilizes sensitivity informationfrom multiplereceiving coils to reduce scan time. The existing SENSE-based parallel MRI reconstruction methods haveproblems of artifacts and missing details, which is not conducive to clinical diagnosis. By introducingdata-driven adaptive sparse transform learning (TL) into the SENSE algorithm, TL-SENSE algorithm isproposed, that reduce the artifacts and improve the quality of parallel MRI reconstruction. The proposedalgorithm employs the alternating direction method of multipliers (ADMM) to solve the target optimizationproblem. And The proposed algorithm comprises three steps: transform updating, hard thresholddenoising and image updating. The simulation results show that the proposed algorithm performs well inimage denoising and restoration and preserves the texture details and edge information. It also achieveshigher consistency between the reconstructed image and the original image. For the selected 48 sets ofdata, the average signal noise ratio of TL-SENSE increased by 4.62 dB, 1.91 dB, 1.30 dB and 0.89 dBcompared with that of SENSE, L1-SENSE, TV-SENSE, and LpTV-SENSE, respectively.