Zongpai Zhang1, Huiyuan Yang1, Yanchen Guo1, Lijun Yin1, David C. Alsop2, and Weiying Dai1
1State University of New York at Binghamton, Binghamton, NY, United States, 2Beth Israel Deaconess Medical Center & Harvard Medical School, Boston, MA, United States
1State University of New York at Binghamton, Binghamton, NY, United States, 2Beth Israel Deaconess Medical Center & Harvard Medical School, Boston, MA, United States
We designed an affine registration network (ARN) to explore its feasibility on image registration of perfusion fMRI. The results demonstrated that our ARN markedly outperforms the iteration-based SPM algorithm both in simulated and real data.
Figure 1. The ARN (CNN + FCN) architecture for image registration. The input layer took the moving image (128x128x40 matrix) as an input, 20 channels were used for the 5 CNN layers. Six parameters that contain x, y, z shifts, and x, y, z rotations were generated after the 2 FCN layers. Bilinear interpolation was used to obtain the moved image by applying 6 parameters to the moving image. The ARN was trained to minimize the loss function, which included the MSE loss, pixel-wise L1 loss, and SSIM loss between the fixed and moved images.
Figure 2. Comparison of loss between ARN and SPM registration. (a) MSE loss, (b) Pixel-wise L1 loss, (c) SSIM loss using simulated perfusion data, and (d)MSE loss, (e) Pixel-wise L1 loss, (f) SSIM loss using real perfusion data were compared between ARN and SPM registration.
