Data

Data sources:

Multi-channel k-space data from

• 5+ medical centers 
• 10+ MRI scanners from GE, Philips, Siemens, United Imaging. 
• Field strengths including 1.5T, 3.0T and 5.0T (for special tasks) 

Image acquisition:

• Modalities and sequences: 
• TrueFISP: Used for cine, phase-contrast (PC), and tagging 
• FLASH: Used for mapping and dark-blood imaging 
• TSE: for T2-weighted imaging 
• Anatomical views: 
• Long-axis (2-chamber, 3-chamber, 4-chamber).
• Short-axis
• Outflow tract
• Aortic views (cross-sectional and sagittal).
• Whole heart 3D
• Diseases: 
• Hypertrophic cardiomyopathy
• Dilated cardiomyopathy
• Myocardial infarction
• Coronary artery disease
• Arrhythmias
• ...

Dataset seperation:

Total of 600 volunteers with multi-parametric CMR imaging. The dataset is seperated into:

• 200 cases for training (5 centers, both healthy volunteers and patients).
• 100 cases for validation (5 derivation centers and 1 new centers).
• 300 cases for testing (5 derivation centers and 5+ new centers).

Pre-processing:

The raw k-space data exported from the scanner will be processed and transformed to '.mat' format using the script provided by each vendor. A readme file will be provided to describe the content and usage of the data.

Details of raw data description for each modality and view

Note: Undersampled and the fully sampled data share the same dimension. nx: matrix size in x-axis (k-space); ny: matrix size in y-axis (k-space); nc: coil array number; nz: slice number (for SAX) or slice group (for LAX, i.e., 3CH, 2CH, and 4CH views); nt: temporal/parametric frame.

Details of different mask types and acceleration factors

Note: % represents the task number (1 or 2). # represents varied acceleration factors (8x, 16x, 24x), ACS not included for calculations. nx: matrix size in x-axis (k-space); ny: matrix size in y-axis (k-space); nt: temporal/parametric frame. If the data has no temporal/parametric dimension, the corresponding mask dimension becomes (nx, ny).

Taking multi-coil long-axis (LAX) cine in Task 1 as an example, the format of data is as follows:

1) Data in folder 'FullSample': cine_lax.mat
# cine with long-axis view (including 3CH, 2CH, and 4CH views within the nz dimension).
# variable name:
# 'kspace' for fully sampled kspace
# data type: complex kspace data with the dimensions (nx,ny,nc,nz,nt)
-nx: matrix size in x-axis (kspace)
-ny: matrix size in y-axis (kspace)
-nc: coil array number (compressed to 10)
-nz: slice number (for SAX) or slice group (for LAX, i.e., 3CH, 2CH, and 4CH views)
-nt: temporal/parametric frame

2) Data in folder 'UnderSample_Task1': cine_lax_kus_xxx.mat
# cine with long-axis view (including 3CH, 2CH, and 4CH views within the nz dimension).
# 'xxx' is the corresponding undersampling scenarios. For example, 'ktUniform8' means uniform undersampling with temporal interleaving at acceleration factor 8x (ACS not included for calculations)
# variable name:
# 'kus' for undersampled kspace
# data type: complex kspace data with the dimensions (nx,ny,nc,nz,nt), the central 20 lines (ny) are always fully sampled to be used as autocalibration signals (ACS)
-nx: matrix size in x-axis (kspace)
-ny: matrix size in y-axis (kspace)
-nc: coil array number (compressed to 10)
-nz: slice number (for SAX) or slice group (for LAX, i.e., 3CH, 2CH, and 4CH views)
-nt: temporal/parametric frame

3) Data in folder 'Mask_Task1': cine_lax_mask_xxx.mat
# undersampling mask with long-axis view, the mask is fixed among different nc and nz. But the mask is interleaved along nt.
# 'xxx' is the corresponding undersampling scenarios. For example, 'ktGaussian8' means Gaussian undersampling with temporal interleaving at acceleration factor 8x (ACS not included for calculations)
# variable name:
# "mask" for undersampling mask
# data type: 3D binary data with the dimensions (nx,ny,nt), the central 20 lines (ny) or central 20*20 regions (nx*ny) are always fully sampled to be used as autocalibration signals (ACS)
-nx: matrix size in x-axis (kspace)
-ny: matrix size in y-axis (kspace)
-nt: temporal/parametric frame