REC_LPDNet_CC359_12_channel_poisson2d_5x_10x_NNEstimationCSM

Model Overview

Learned Primal Dual Network (LPDNet) for 5x & 10x accelerated MRI Reconstruction on the CC359 dataset.

ATOMMIC: Training

To train, fine-tune, or test the model you will need to install ATOMMIC. We recommend you install it after you've installed latest Pytorch version.

pip install atommic['all']

How to Use this Model

The model is available for use in ATOMMIC, and can be used as a pre-trained checkpoint for inference or for fine-tuning on another dataset.

Corresponding configuration YAML files can be found here.

Automatically instantiate the model

pretrained: true
checkpoint: https://huggingface.co/wdika/REC_LPDNet_CC359_12_channel_poisson2d_5x_10x_NNEstimationCSM/blob/main/REC_LPDNet_CC359_12_channel_poisson2d_5x_10x_NNEstimationCSM.atommic
mode: test

Usage

You need to download the CC359 dataset to effectively use this model. Check the CC359 page for more information.

Model Architecture

model:
  model_name: LPDNet
  num_primal: 5
  num_dual: 5
  num_iter: 5
  primal_model_architecture: UNET
  primal_in_channels: 2
  primal_out_channels: 2
  primal_unet_num_filters: 16
  primal_unet_num_pool_layers: 2
  primal_unet_dropout_probability: 0.0
  primal_unet_padding_size: 11
  primal_unet_normalize: true
  dual_model_architecture: UNET
  dual_in_channels: 2
  dual_out_channels: 2
  dual_unet_num_filters: 16
  dual_unet_num_pool_layers: 2
  dual_unet_dropout_probability: 0.0
  dual_unet_padding_size: 11
  dual_unet_normalize: true
  dimensionality: 2
  reconstruction_loss:
    l1: 0.1
    ssim: 0.9
  estimate_coil_sensitivity_maps_with_nn: true

Training

  optim:
    name: adamw
    lr: 1e-4
    betas:
      - 0.9
      - 0.999
    weight_decay: 0.0
    sched:
        name: CosineAnnealing
        min_lr: 0.0
        last_epoch: -1
        warmup_ratio: 0.1

trainer:
  strategy: ddp_find_unused_parameters_false
  accelerator: gpu
  devices: 1
  num_nodes: 1
  max_epochs: 20
  precision: 16-mixed
  enable_checkpointing: false
  logger: false
  log_every_n_steps: 50
  check_val_every_n_epoch: -1
  max_steps: -1

Performance

To compute the targets using the raw k-space and the chosen coil combination method, accompanied with the chosen coil sensitivity maps estimation method, you can use targets configuration files.

Evaluation can be performed using the evaluation script for the reconstruction task, with --evaluation_type per_slice.

Results

Evaluation against RSS targets

5x: MSE = 0.001668 +/- 0.001584 NMSE = 0.02567 +/- 0.0265 PSNR = 28.26 +/- 4.222 SSIM = 0.8493 +/- 0.07524

10x: MSE = 0.002367 +/- 0.002247 NMSE = 0.03687 +/- 0.03859 PSNR = 26.73 +/- 4.229 SSIM = 0.8096 +/- 0.09866

Limitations

This model was trained on the CC359 using a UNet coil sensitivity maps estimation and might differ from the results reported on the challenge leaderboard.

References

[1] ATOMMIC

[2] Beauferris, Y., Teuwen, J., Karkalousos, D., Moriakov, N., Caan, M., Yiasemis, G., Rodrigues, L., Lopes, A., Pedrini, H., Rittner, L., Dannecker, M., Studenyak, V., Gröger, F., Vyas, D., Faghih-Roohi, S., Kumar Jethi, A., Chandra Raju, J., Sivaprakasam, M., Lasby, M., … Souza, R. (2022). Multi-Coil MRI Reconstruction Challenge—Assessing Brain MRI Reconstruction Models and Their Generalizability to Varying Coil Configurations. Frontiers in Neuroscience, 16. https://doi.org/10.3389/fnins.2022.919186