README.md

metadata

language: en
tags:
  - MRI image priors
  - Generative models
  - Diffusion models
  - TensorFlow
  - PixelCNN

Generative pretrained models on MRI images.

The prior distribution of MRI images learned with generative models has proven to be effective in MRI image reconstruction. Here, we include four PixelCNN models and two diffusion models, one is SMLD and the another one is DDPM. These models are trained with spreco. For more details on how these models were trained, please find them in our paper and the related codes.

How to use

The Berkeley Advanced Reconstruction Toolbox, (BART), provides many functionalities for MRI image reconstruction. It introduced the application of the TensorFlow graph as regularization in this paper. You can try it on colab.

Prior	Model	Phase	Size	Contrast	Subscript
${\mathtt{\text{P}}}_{\mathrm{S}\mathrm{C}}\backslash texttt\{P\}\_\backslash mathrm\{SC\}$	PixelCNN	preserved	1000	T1, T2, T2-FLAIR, ${\mathtt{\text{T}}}_2^{\ast }\backslash texttt\{T\}^*\_\backslash mathrm\{2\}$	SC - Small, complex
${\mathtt{\text{P}}}_{\mathrm{S}\mathrm{M}}\backslash texttt\{P\}\_\backslash mathrm\{SM\}$	PixelCNN	unknown	1000	T1, T2, T2-FLAIR, ${\mathtt{\text{T}}}_2^{\ast }\backslash texttt\{T\}^*\_\backslash mathrm\{2\}$	SM - Small, magnitude
${\mathtt{\text{P}}}_{\mathrm{L}\mathrm{M}}\backslash texttt\{P\}\_\backslash mathrm\{LM\}$	PixelCNN	unknown	~20000	MPRAGE	LM - Large, magnitude
${\mathtt{\text{P}}}_{\mathrm{L}\mathrm{C}}\backslash texttt\{P\}\_\backslash mathrm\{LC\}$	PixelCNN	generated	~20000	MPRAGE	LC - Large, complex
${\mathtt{\text{D}}}_{\mathrm{S}\mathrm{C}}\backslash texttt\{D\}\_\backslash mathrm\{SC\}$	Diffusion	generated	~80000	MPRAGE	SC - SMLD, complex
${\mathtt{\text{D}}}_{\mathrm{P}\mathrm{C}}\backslash texttt\{D\}\_\backslash mathrm\{PC\}$	Diffusion	generated	~80000	MPRAGE	PC - DDPM, complex

Citation

1. Luo, G, Blumenthal, M, Heide, M, Uecker, M. Bayesian MRI reconstruction with joint uncertainty estimation using diffusion models. Magn Reson Med. 2023; 1-17
2. Blumenthal, M, Luo, G, Schilling, M, Holme, HCM, Uecker, M. Deep, deep learning with BART. Magn Reson Med. 2023; 89: 678- 693.
3. Luo, G, Zhao, N, Jiang, W, Hui, ES, Cao, P. MRI reconstruction using deep Bayesian estimation. Magn Reson Med. 2020; 84: 2246-2261.