zenodoembargoed

SynthRAD2023 Grand Challenge test dataset: synthetizing computed tomography for radiotherapy

<div> DATASET STRUCTURE The dataset can be downloaded from <a href="https://doi.org/10.5281/zenodo.7260705">https://doi.org/10.5281/zenodo.10514185</a> and a detailed description is offered at "synthRAD2023_dataset_description.pdf". The test datasets for Task1 is in Task1.tar.zst, while for Task2 in Task2.tar.zst. After unzipping, each Task is organized according to the following folder structure: Task1 └── mr │   ├── brain │   │   ├── 1BA003 │   │   │   ├── mask.nii.gz │   │   │   └── mr.nii.gz │   │   ├── 1BA029 │   │   │   ├── mask.nii.gz │   │   │   └── mr.nii.gz │   │   ├── ... │   └── pelvis │       ├── 1PA003 │       │   ├── mask.nii.gz │       │   └── mr.nii.gz │       ├── 1PA006 │       │   ├── mask.nii.gz │       │   └── mr.nii.gz │       ├── ... ├── ct │   ├── 1BA003.nii.gz │   ├── 1BA029.nii.gz │   ├── 1BA063.nii.gz │   ├── ... ├── doseplanning │   ├── 1BA003 │   │   ├── 1BA003_RTplan_photons.mat │   │   └── 1BA003_RTplan_protons.mat │   ├── 1BA029 │   │   ├── 1BA029_RTplan_photons.mat │   │   ├── 1BA029_RTplan_protons.mat │   ├── ... Task2 ├── cbct │   ├── brain │   │   ├── 2BA019 │   │   │   ├── cbct.nii.gz │   │   │   └── mask.nii.gz │   │   ├── 2BA021 │   │   │   ├── cbct.nii.gz │   │   │   └── mask.nii.gz │   │   ├── ... │   └── pelvis │       ├── 2PA022 │       │   ├── cbct.nii.gz │       │   └── mask.nii.gz │       ├── 2PA023 │       │   ├── cbct.nii.gz │       │   └── mask.nii.gz │       ├── ... ├── ct │   ├── 2BA019.nii.gz │   ├── 2BA021.nii.gz │   ├── 2BA022.nii.gz │   ├── .... ├── doseplanning │   ├── 2BA019 │   │   ├── 2BA019_RTplan_photons.mat │   │   └── 2BA019_RTplan_protons.mat │   ├── 2BA021 │   │   ├── 2BA021_RTplan_photons.mat │   │   └── 2BA021_RTplan_protons.mat │   ├── ... Each patient folder has a unique name that contains information about the task, anatomy, center and a patient ID. The naming follows the convention below: <table> <tbody> <tr> <td>[Task]</td> <td>[Anatomy]</td> <td>[Center]</td> <td>[PatientID]</td> </tr> <tr> <td>1</td> <td>B</td> <td>A</td> <td>001</td> </tr> </tbody> </table> The zip contains the following structure:  <ul> <li> ct/<patient code>.nii.gz: The gold-standard CT image. </li> <li> doseplanning/<patient code>/<patient code>_RTplan_[photons/protons].mat: The matRad photon and proton doseplans for this particular patient.  </li> <li> [mr/cbct]/[brain/pelvis]/<patient code>/[mr/cbct].nii.gz: The corresponding CBCT or MR image. </li> <li> [mr/cbct]/[brain/pelvis]/<patient code>/mask.nii.gz: image containing a binary mask of the dilated patient outline. </li> </ul> DATASET DESCRIPTION This challenge dataset contains imaging data of patients who underwent radiotherapy in the brain or pelvis region. Overall, the population is predominantly adult and no gender restrictions were considered during data collection. For Task 1, the inclusion criteria were the acquisition of a CT and MRI during treatment planning while for task 2, acquisitions of a CT and CBCT, used for patient positioning, were required. Datasets for task 1 and 2 do not necessarily contain the same patients, given the different image acquisitions for the different tasks. Data was collected at 3 Dutch university medical centers: <ul> <li> Radboud University Medical Center </li> <li> University Medical Center Utrecht </li> <li> University Medical Center Groningen </li> </ul> For anonymization purposes, from here on, institution names are substituted with A, B and C, without specifying which institute each letter refers to. The following number of patients is available in the training set. Training <table> <tbody> <tr> <td> </td> <td> Brain </td> <td> Pelvis </td> </tr> <tr> <td> </td> <td> Center A </td> <td> Center B </td> <td> Center C </td> <td> Total </td> <td> Center A </td> <td> Center B </td> <td> Center C </td> <td> Total </td> </tr> <tr> <td> Task 1 </td> <td> 60 </td> <td> 60 </td> <td> 60 </td> <td> 180 </td> <td> 120 </td> <td> 0 </td> <td> 60 </td> <td> 180 </td> </tr> <tr> <td> Task 2 </td> <td> 60 </td> <td> 60 </td> <td> 60 </td> <td> 180 </td> <td> 60 </td> <td> 60 </td> <td> 60 </td> <td> 180 </td> </tr> </tbody> </table> Each subset generally contains equal amounts of patients from each center, except for task 1 brain, where center B had no MR scans available. To compensate for this, center A provided twice the number of patients than in other subsets. Validation <table> <tbody> <tr> <td> </td> <td> Brain </td> <td> Pelvis </td> </tr> <tr> <td> </td> <td> Center A </td> <td> Center B </td> <td> Center C </td> <td> Total </td> <td> Center A </td> <td> Center B </td> <td> Center C </td> <td> Total </td> </tr> <tr> <td> Task 1 </td> <td> 10 </td> <td> 10 </td> <td> 10 </td> <td> 30 </td> <td> 20 </td> <td> 0 </td> <td> 10 </td> <td> 30 </td> </tr> <tr> <td> Task 2 </td> <td> 10 </td> <td> 10 </td> <td> 10 </td> <td> 30 </td> <td> 10 </td> <td> 10 </td> <td> 10 </td> <td> 30 </td> </tr> </tbody> </table> Testing <table> <tbody> <tr> <td> </td> <td> Brain </td> <td> Pelvis </td> </tr> <tr> <td> </td> <td> Center A </td> <td> Center B </td> <td> Center C </td> <td> Total </td> <td> Center A </td> <td> Center B </td> <td> Center C </td> <td> Total </td> </tr> <tr> <td> Task 1 </td> <td> 20 </td> <td> 20 </td> <td> 20 </td> <td> 60 </td> <td> 40 </td> <td> 0 </td> <td> 20 </td> <td> 60 </td> </tr> <tr> <td> Task 2 </td> <td> 20 </td> <td> 20 </td> <td> 20 </td> <td> 60 </td> <td> 20 </td> <td> 20 </td> <td> 20 </td> <td> 60 </td> </tr> </tbody> </table> In total, for all tasks and anatomies combined, 1080 image pairs (720 training, 120 validation, 240 testing) are available in this dataset. This repository only contains the test data. All images were acquired with the clinically used scanners and imaging protocols of the respective centers and reflect typical images found in clinical routine. As a result, imaging protocols and scanner can vary between patients. A detailed description of the imaging protocol for each image, can be found in spreadsheets that are part of the dataset release (see dataset structure). Data was acquired with the following scanners: <ul> <li> Center A: <ul> <li> MRI: Philips Ingenia 1.5T/3.0T </li> <li> CT: Philips Brilliance Big Bore or Siemens Biograph20 PET-CT </li> <li> CBCT: Elekta XVI </li> </ul> </li> <li> Center B: <ul> <li> MRI: Siemens MAGNETOM Aera 1.5T or MAGNETOM Avanto_fit 1.5T </li> <li> CT: Siemens SOMATOM Definition AS </li> <li> CBCT: IBA Proteus+ or Elekta XVI </li> </ul> </li> <li> Center C: <ul> <li> MRI: Siemens Avanto fit 1.5T or Siemens MAGNETOM Vida fit 3.0T </li> <li> CT: Philips Brilliance Big Bore </li> <li> CBCT: Elekta XVI </li> </ul> </li> </ul> For task 1, MRIs were acquired with a T1-weighted gradient echo or an inversion prepared - turbo field echo (TFE) sequence and collected along with the corresponding planning CTs for all subjects. The exact acquisition parameters vary between patients and centers. For centers B and C, selected MRIs were acquired with Gadolinium contrast, while the selected MRIs of center A were acquired without contrast. For task 2, the CBCTs used for image-guided radiotherapy ensuring accurate patient position were selected for all subjects along with the corresponding planning CT. The following pre-processing steps were performed on the data: <ul> <li> Conversion from dicom to compressed nifti (nii.gz) </li> <li> Rigid registration between CT and MR/CBCT </li> <li> Anonymization (face removal, only for brain patients) </li> <li> Patient outline segmentation (provided as a binary mask) </li> <li> Crop MR/CBCT, CT and mask to remove background and reduce file sizes </li> </ul> The code used to preprocess the images can be found at: <a href="https://github.com/SynthRAD2023/">https://github.com/SynthRAD2023/</a>. Detailed information about the dataset are provided in SynthRAD2023_dataset_description.pdf published here along with the data and will also be submitted to Medical Physics. ETHICAL APPROVAL Each institution received ethical approval from their internal review board/Medical Ethical committee: <ul> <li> UMC Utrecht approved not-WMO on 4/03/2022 with number 22/474 entitled: “Synthetizing computed tomography for radiotherapy Grand Challenge (SynthRAD)”. </li> <li> UMC Groningen approved not-WMO on 20/07/2022 with number 202200310 entitled: “Synthesizing computed tomography for radiotherapy - Grand Challenge”. </li> <li> Radboud UMC declared the study not-WMO on 17/10/2022 with number 2022-15950 entitled “Synthetizing computed tomography for radiotherapy Grand Challenge”. </li> </ul> CHALLENGE DESIGN The overall challenge design can be found at <a href="https://doi.org/10.5281/zenodo.7746020">https://doi.org/10.5281/zenodo.7746020</a>.  <h2>Notes</h2> <div>FUNDING BODIES: The challenge has been funded thanks to the support of the Seed Fund provided by the " EWUU Alliance TU/e, WUR, UU, UMCU" https://ewuu.nl/en/collaboration/seed-fund/.</div> </div>

ShareScore

20/100

Overall dataset sharing score

Score breakdown

These five areas show where the dataset supports — or may limit — practical reuse.

Stewardship: 8
Harmonization: 4
Access: 8
Reuse readiness: 0
Engagement: 0

Topics

medical images radiotherapy magnetic resonance imaging mri computed tomography cbct ct image synthesis