wdika's picture
Upload config
5483448
|
raw
history blame
3.59 kB
metadata
language:
  - en
license: apache-2.0
library_name: atommic
datasets:
  - StanfordKnees2019
thumbnail: null
tags:
  - image-reconstruction
  - RVN
  - ATOMMIC
  - pytorch
model-index:
  - name: REC_RVN_StanfordKnees2019_gaussian2d_12x_AutoEstimationCSM
    results: []

Model Overview

Recurrent Variational Network (RVN) for 12x accelerated MRI Reconstruction on the StanfordKnees2019 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_RVN_StanfordKnees2019_gaussian2d_12x_AutoEstimationCSM/blob/main/REC_RVN_StanfordKnees2019_gaussian2d_12x_AutoEstimationCSM.atommic
mode: test

Usage

You need to download the Stanford Knees 2019 dataset to effectively use this model. Check the StanfordKnees2019 page for more information.

Model Architecture

model:
  model_name: RVN
  in_channels: 2
  recurrent_hidden_channels: 64
  recurrent_num_layers: 4
  num_steps: 8
  no_parameter_sharing: true
  learned_initializer: true
  initializer_initialization: "SENSEe"
  initializer_channels:
    - 32
    - 32
    - 64
    - 64
  initializer_dilations:
    - 1
    - 1
    - 2
    - 4
  initializer_multiscale: 1
  accumulate_predictions: false
  dimensionality: 2
  reconstruction_loss:
    wasserstein: 1.0

Training

  optim:
    name: adamw
    lr: 1e-4
    betas:
      - 0.9
      - 0.999
    weight_decay: 0.0
    sched:
        name: InverseSquareRootAnnealing
        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 SENSE targets

12x: MSE = 0.001201 +/- 0.005875 NMSE = 0.04067 +/- 0.1203 PSNR = 31.96 +/- 6.899 SSIM = 0.7781 +/- 0.3002

Limitations

This model was trained on the StanfordKnees2019 batch0 using a UNet coil sensitivity maps estimation and Geometric Decomposition Coil-Compressions to 1-coil, and might differ from the results reported on the challenge leaderboard.

References

[1] ATOMMIC

[2] Epperson K, Rt R, Sawyer AM, et al. Creation of Fully Sampled MR Data Repository for Compressed SENSEing of the Knee. SMRT Conference 2013;2013:1