Projects:RegistrationDocumentation:UseCaseInventory

From NAMIC Wiki
Jump to: navigation, search
Home < Projects:RegistrationDocumentation:UseCaseInventory

Back to ARRA main page

Back to Registration main page

The Slicer Registration Case Library

Welcome to the 3DSlicer Registration Case Library. This page is under continuous development as we add and refine case examples of image registration within 3DSlicer. The goal is for you to find a case here similar to yours that will address the issues of your particular image registration problem. You will find 3 sets of content with each case: 1) a step-by-step tutorial that will show you how to register images of this type, 2) the example dataset used in the tutorial so you can try for yourself , 3) a custom Registration Presets file that contains the optimized registration parameters that you can load directly into slicer.

So the procedure would be as follows:

  1. browse/search the library below for a case similar to yours
  2. if you cannot find one consider our Call for Datasets.
  3. if you find a good match, download the data, preset file and tutorials
  4. run the tutorial on your machine with your installation of slicer
  5. load the preset file and try that on your own data
  6. if you do not get a satisfactory match with the presets, have a look at the Discussion section on the download page of each case. You will find recommendations there on how to venture forth
  7. if still unsuccessful you may have a case of interest to the library. Also post a message to the slicer user group,


Source

Data is collected from a variety of sources. Because we want to focus on the registration problem and not be distracted by image format or other data management issues, the datasets listed here are copied and reformatted. As the data becomes available a download link is added.

  • Call for Datasets
  • Options for the download location is a direct link to a download page from the main 3Dslicer user page or XNAT Central. The former seems preferable, since the general XNAT GUI appears too complex for a simple 1-case download. We expect most users to download only 1 or 2 files closest to their particular use case, not entire study sets. Such download should be acessible with 1-2 clicks, as with other slicer resources.
  • Disk usage: Likely these example sets will grow no larger than 10GB. Estimating 50-80MB per case would allow 60-100 cases within a 5GB space.
  • Since we post reformatted and preprocessed data, where possible the link to the original source is provided.
  • Links here also from the NA-MIC resource page.To be updated when moving.

Categories & Formatting

  • Ideally we seek 2 sets for each use case to define a range of settings, i.e. one case that is average and works reasonably well, and one case that is more challenging. Both cases together then define a range. The two cases would be posted together, provided the difficult case can be solved with different settings and does not require a different approach. The latter case would then be better listed in a Troubleshooting category.
  • The main categories will follow along the hierarchies outlined here.
  • We seek to have all datasets in a single-volume file format, most likely .nii or .nrrd. This will greatly facilitate file management and case documentation and will also guarantee full anonymization. The NRRD format is defined here and the NIFTI format is defined here.
  • Anonymization: Anonymized data is imperative. In a first pass posting single volume files only that do not contain subject-specific meta data is the safest. Relying on the provider/user to properly anonymize is likely insufficient. It's easy to overlook single DICOM fields. Defacing algorithms are avail. thru MBIRN but will affect the result and require the mask to be avail. to the registration.
  • Once formatting, anonymization and description are complete, listing here will be replaced with a download link. Temporarily listed are links to source databases, where avail.
  • Speed: include speed measurement with each case. Include note in tutorial that speed concerns are built into the default settings, i.e. if results do not look right on first run, there's many options to fine tune, hence the library concept

Case Inventory Brain

Case Inventory Non-Brain

Case Inventory Non-human

Self-validation Sets

These are datasets with artificial misalignment, i.e. the perfect alignment is known.

  • testing PET-CT-MRI on the Vanderbuild database: RIRE
  • Dual Echo PD/T2 intra subject different contrast self validation
  • Isotropic Brain SPGR T1 with single axis rotation

First 10 cases (not in sequential order)

  • Brain intra-subject, same contrast T1, change detection follow-up: tumor growth
  • Brain MRI, intra-subject, same contrast PD&T2, change detection follow-up: new MS lesions
  • Brain MRI, intra-subject, different contrast, co-register all series of the same exam: T1 SPGR, FLAIR, T2
  • Brain MRI, intra-subject, DTI to reference, apply Xform to 25-direction DTI tensor
  • Brain MRI, intra-subject, example with clipped FOV, example of masking required
  • Brain MRI, inter-subject, co-register T1 SPGR to atlas dataset (ICBM). Resample a label map.
  • Liver intra-subject, pre-procedural MRI to intra-procedural CT
  • Prostate MRI, intra-subject
  • Knee inter-subject registration to initialize segmentation
  • PET to CT Whole Body single timepoint
  • Breast Cancer: feasible with affine focus?