Define your cases

When the basic configuration of your reader study is ready, it is time to collect your data. Grand Challenge supports the following image formats: .mha, .mhd, .raw, .zraw, .dcm, .tiff, .png, .jpeg and .jpg. Besides images, it is also possible to use other objects in your reader study, such as regions of interest to guide your readers to certain areas in an image, or other values to provide default answers to the questions, see default answers. You need at least one image per case.

In Grand Challenge terminology, a case is called a display set. A single display set contains all the images and potential other objects for a case and each image / object for a case is defined by a unique interface.

You can upload your data by using the menu on the left hand side of the screen: Side menu → Cases as shown below. You have two options: batch uploading images or creating each display set seperately, one by one.

Batch uploading images

Uploading images in batch creates one display set for each image in the batch with the interface chosen by you on the upload form. In the example below, 2 display sets will be created, with each display set containing a CT image.

The overview table will then look as follows:

You can monitor the status of your uploads on this page.

Cases with more than 1 image or other data types

If you wanted to add a secondary image or another object to each display set, you can do so by clicking on the edit icon for each case and then choose "Add value for new interface". Type the interface you would like to add, for example "Airway segmentation", upload the secondary image and click "save".

Note that you could also batch upload the secondary images with the respective interface. Doing so however, will create a new display set for each secondary image, leaving you with double the amount of display sets as before. To get them organized properly, you will still need to edit each display set seperately and ultimately delete any leftover, duplicate display sets. So batch uploading the secondary images leaves you with more work ultimately.

Instead of bulk uploading and then editing, you could of course also create the cases one by one. To do so, just choose "Add Display Set (single)" on the cases overview page.

Finally, if you have a lot of cases to define, you might be best advised to use our API client and create the display sets programmatically. Read more about that here.

Good to know

  • Images can be used in more than one case.

  • The order field is used to determine the order in which your cases are presented to your readers. This field is automatically populated, but can be changed.

  • Use the View Display Set button to open the reader study at a particular case.

  • 💡Note that if the View Display Set button is used in a reader study that has Allow case navigation disabled, the viewer will only allow answering a case if that case is the first unanswered case. Otherwise it will show the case in a read-only view. Normal functionality will resume once the Continue button is clicked.

  • 💡Once a case has been answered by any participant you cannot edit the case anymore. You can make the case editable again by removing all answers (Users progress → Remove Answers).

Default answers

If you have defined an interface in any of the questions of the reader study, this interface will be shown in each display set. This allows you to set the default answer to that question. If the question has answer type Mask, you first need to upload the masks, before you can select the specific mask. For example, the display set below contains one image, the other values are default answers for a number of questions:

Define your layout

When your cases are set up, you can define the layout for your reader study. For an extensive description of the various layout options see this section.

You only need to define a layout if you have more than a single image per display set. If you only have a single image (optionally with an overlay), you can leave the view content and hanging protocol empty, and the viewer will use sensible defaults to show your image.

If you have several images, you will need to at least define the view content. This will ensure each image will be loaded in a separate view. You can leave the hanging protocol empty if you would like side views for 3D images, or a minimap overview for multi-resolution images. The viewer will then show all your images at the same size. If you want full control over how images are shown (for example one large view and three smaller views below the main view), you will need to define a hanging protocol. See existing protocols.

Case text

If you would like to display some information specific to an image you can configure Case text. Use the image name as key, for example:

  "image1.mhd": "Explanatory text for image 1",
  "image2.mhd": "Explanatory text for image 2"

The case text for the viewed images will be displayed in the viewer: