HL7 Soup

DICOM Structured Reports (SR) contain a wealth of clinical findings, but their nested structure makes them difficult to parse. This tutorial showcases how to easily extract this complex data into a clean CSV file using only Integration Soup's visual transformer tools.

How to Create a CSV from a DICOM SR

This workflow uses the DICOM Receiver's ability to automatically create a human-readable JSON structure for SR files, allowing you to simply drag-and-drop the findings you need.

1. Configure the DICOM Receiver
   Add a DICOM Receiver activity. To see the report structure, drag and drop a sample DICOM SR file (e.g., from an ultrasound) into the "Message Template". Notice that in addition to the standard tags, a "Report" item appears in the binding tree, containing the cleanly organized clinical findings.
2. Add the File Writer Activity
   Add a File Writer activity. In the "File to write" field, create a dynamic filename by dragging a tag from the binding tree, for example: C:\Reports\${AccessionNumber}.csv. Set the Message Type to CSV.
3. Build the CSV Visually
   In the File Writer's "Message Template", simply drag and drop the fields you want from the binding tree, separating each with a comma. As you drop them, the fields will be mapped to variables and inserted into the template. You can mix data from the root and from the Report section. For example:

   ${PatientID},${Aorta/Height},${LiverGallBladderPancreas/LiverDistance}

4. Add CSV Headers
   In the Header field, type in the column titles for your report, corresponding to the fields you added in the template. For instance: PatientID,AortaHeight,LiverDistance. When the workflow runs, it will write this header line at the top of your new CSV file.