fastMRI: An Open Dataset and Benchmarks for Accelerated MRI
Paper • 1811.08839 • Published • 1
The dataset viewer is not available because its heuristics could not detect any supported data files. You can try uploading some data files, or configuring the data files location manually.
YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
@article{zbontar2018fastmri, title={fastMRI: An Open Dataset and Benchmarks for Accelerated MRI, author={Zbontar, Jure and others}, journal={arXiv preprint arXiv:1811.08839}, year={2018} } This dataset contains preprocessed slices from the fastMRI Knee Single-Coil Validation set.
The original data was converted from .h5 format to normalized .npy slices for faster loading and training.
.npy (NumPy arrays)(2, 320, 320) representing the Real and Imaginary components of the k-space.train/: 180 volumes (approx. 90%)val/: 19 volumes (approx. 10%)(320, 320) in the image domain.float32 complex-valued components (Real, Imag).import numpy as np
import os
# Load a sample slice
data = np.load('s0000_000.npy')
print(data.shape) # Output: (2, 320, 320)
@article{zbontar2018fastmri, title={fastMRI: An Open Dataset and Benchmarks for Accelerated MRI, author={Zbontar, Jure and others}, journal={arXiv preprint arXiv:1811.08839}, year={2018} }