Domain transformation learning for MR image reconstruction from dual domain input

• Published in: Computers in biology and medicine Vol. 170; p. 108098
• Main Authors: Oh, Changheun, Chung, Jun-Young, Han, Yeji
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.03.2024; Elsevier Limited
• Subjects: Algorithms; Deep learning; End to end; Fourier Analysis; Fourier transforms; Image acquisition; Image processing; Image Processing, Computer-Assisted - methods; Image reconstruction; Irregular sampling; Magnetic resonance; Magnetic resonance imaging; Magnetic Resonance Imaging - methods; Medical imaging; Methods; Neural networks; Parallel imaging; Phantoms, Imaging; Recurrent neural network; Sampling; Sensors; Deep learning; Recurrent neural network; End to end; Parallel imaging; Magnetic resonance; Image reconstruction
• ISSN: 0010-4825; 1879-0534
• DOI: 10.1016/j.compbiomed.2024.108098

Medical images are acquired through diverse imaging systems, with each system employing specific image reconstruction techniques to transform sensor data into images. In MRI, sensor data (i.e., k-space data) is encoded in the frequency domain, and fully sampled k-space data is transformed into an image using the inverse Fourier Transform. However, in efforts to reduce acquisition time, k-space is often subsampled, necessitating a sophisticated image reconstruction method beyond a simple transform. The proposed approach addresses this challenge by training a model to learn domain transform, generating the final image directly from undersampled k-space input. Significantly, to improve the stability of reconstruction from randomly subsampled k-space data, folded images are incorporated as supplementary inputs in the dual-input ETER-net. Moreover, modifications are made to the formation of inputs for the bi-RNN stages to accommodate non-fixed k-space trajectories. Experimental validation, encompassing both regular and irregular sampling trajectories, validates the method's effectiveness. The results demonstrated superior performance, measured by PSNR, SSIM, and VIF, across acceleration factors of 4 and 8. In summary, the dual-input ETER-net emerges as an effective both regular and irregular sampling trajectories, and accommodating diverse acceleration factors. •For MR image reconstruction, the domain-transform learning approach was rarely studied.•The most appealing aspect of these approaches is that it is capable of correcting errors in both the image domain and the k-space domain•The dual-input ETER-net can reduce the complexity of the sensor-to-image domain transformation network•The dual-input ETER-net can successfully handle data acquired with random sampling trajectories, further expanding its versatility and applicability.