Frequency-regularized Neural Representation Method for Sparse-view Tomographic Reconstruction
This work addresses overfitting in medical imaging reconstruction for reducing radiation dose, but it appears incremental as it builds on prior neural methods with a regularization tweak.
The paper tackles the problem of sparse-view tomographic reconstruction, where existing models overfit to high-frequency information, by introducing a frequency-regularized neural method that balances high and low frequencies, achieving state-of-the-art accuracy on CBCT and SPECT datasets.
Sparse-view tomographic reconstruction is a pivotal direction for reducing radiation dose and augmenting clinical applicability. While many research works have proposed the reconstruction of tomographic images from sparse 2D projections, existing models tend to excessively focus on high-frequency information while overlooking low-frequency components within the sparse input images. This bias towards high-frequency information often leads to overfitting, particularly intense at edges and boundaries in the reconstructed slices. In this paper, we introduce the Frequency Regularized Neural Attenuation/Activity Field (Freq-NAF) for self-supervised sparse-view tomographic reconstruction. Freq-NAF mitigates overfitting by incorporating frequency regularization, directly controlling the visible frequency bands in the neural network input. This approach effectively balances high-frequency and low-frequency information. We conducted numerical experiments on CBCT and SPECT datasets, and our method demonstrates state-of-the-art accuracy.