Summary of A Novel Image Space Formalism Of Fourier Domain Interpolation Neural Networks For Noise Propagation Analysis, by Peter Dawood et al.
A novel image space formalism of Fourier domain interpolation neural networks for noise propagation analysis
by Peter Dawood, Felix Breuer, Istvan Homolya, Jannik Stebani, Maximilian Gram, Peter M. Jakob, Moritz Zaiss, Martin Blaimer
First submitted to arxiv on: 27 Feb 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Medical Physics (physics.med-ph)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper proposes a novel framework for developing an image space formalism of multi-layer convolutional neural networks (CNNs) for Fourier domain interpolation in MRI reconstructions. The authors aim to analytically estimate noise propagation during CNN inference, enabling the description of noise characteristics. They achieve this by expressing nonlinear activations in the Fourier domain using complex-valued Rectifier Linear Units and transforming it into a convolution in the image space. This allows them to derive an algebraic expression for the derivative of the reconstructed image with respect to the aliased coil images, which serves as input tensors to the network in the image space. The framework is validated through Monte-Carlo simulations and numerical approaches based on auto-differentiation, and tested on retrospectively undersampled invivo brain images. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary In simple terms, this paper creates a new way to use neural networks for MRI reconstruction that allows us to understand how noise affects the results. They do this by treating the neural network as if it were working in a different space, which lets them calculate the uncertainty of the results. This is important because it can help make MRI scans more accurate and reliable. |
Keywords
* Artificial intelligence * Cnn * Inference * Neural network
