基于深度学习的扩散磁共振成像降噪方法研究进展

马素超, 王远军

Research Progress on Diffusion Magnetic Resonance Imaging Noise Reduction Methods Based on Deep Learning

MA Suchao, WANG Yuanjun

图5 1D-CNN降噪网络结构示意图. 网络以同一空间位置的N个高噪声体素构成的一维序列为输入，依次经过两层一维卷积（Conv）与池化操作（Pooling）进行特征提取. 第一层卷积通道数为16，核尺寸为1；第二层卷积通道数为32，核尺寸为8. 特征展平（Flatten）后通过全连接层（Dense）完成信号重构，并以均方根误差（Root Mean Squared Error, RMSE）损失作为损失函数进行监督训练

Fig. 5 Schematic diagram of the 1D-CNN denoising network architecture. The network takes a one-dimensional sequence composed of N highly noisy voxels at the same spatial location as input. It sequentially undergoes two layers of one-dimensional convolution (Conv) and Pooling operations for feature extraction. The first convolutional layer has 16 channels and a kernel size of 1; the second convolutional layer has 32 channels and a kernel size of 8. After feature Flatten, signal reconstruction is performed via a dense connected layer (Dense), with root mean squared error (RMSE) loss serving as the loss function for supervised training