Guide Importing Data into DASHER
Home | Installation Guide | User Guide | Admin Guide | User Tests
DASHER requires certain DICOM header information to be present, as described below. If data has been previously anonymised it may have been modified such that the required header information is missing or set incorrectly set, which can cause problems when importing the data into DASHER. Data exported directly from a treatment planning system should be imported without any problems, but if problems are encountered when importing data please report these via DASHER gitlab repo Issues page.
Importing data takes time! It can take 10s of minutes to import and process data. However, multiple datasets from different subjects can be imported simultaneously - it is not necessary to let one dataset finish importing before starting the next.
The main steps for importing data are
- Using an import mechanism
- Pre-sorting according to a protocol
- Conformity check in the Prearchive
- Archiving into hospital project of non-anonymised XNAT
DASHER expects non-anonymised data with certain DICOM header fields present in order to organize and label the data. When data is imported into the non-anonymised XNAT, the following DICOM header entries are translated into XNAT identifiers as follows:
-
(\[0010,0010\]), Patient's Name-> XNAT Subject ID -
(\[0010,0020\]), Patient's ID-> XNAT Session ID
RT data (RTStruct, RTDose, RTPlan) has to be imported at the same time as the referenced imaging data (usually the planning CT-scan) and the files must be linked correctly by UIDs. Some anonymisers break the links between files.
If sending multiple RT files, ensure they have separate Series Instance UIDs and Series Numbers.
If RT data (RTStruct, RTDose, RTPlan) is imported without the associated scan the RT data will be deleted during the import process. You will have to import again along with the corresponding scan.
Two mechanisms are available to import data into DASHER which are described in the following sections.
During installation of DASHER an import folder was defined, which is continually monitored by DASHER for subfolders containing DICOM data. Any folder with DICOM data copied into the import folder will be automatically imported into the non-anonymised XNAT and then deleted from the import folder. Any non-DICOM files will be deleted. The location of the import folder is set during the installation process and can be a network drive.
Note: The DICOM data has to be organised in folders. Please do not copy .dcm files directly into the import folder. Data is imported by folder - therefore a folder/subfolder cannot contain files belonging to separate sessions (i.e. all files should have the same Patient's Name, Patient's ID and Study UID)
A session can contain multiple folders, one for each scan, and should belong in the same top folder. However a top folder can contain mixed sessions. For example, the following is permitted:
[import_folder]/folder1/
../session1/scan1/
../session1/scan2/
../session2/scan1/
../session2/scan2/
/folder2/
../session3/
../session4/
DASHER also implements a DICOM-reciever, so you can use any tool capable of performing a DICOM-push operation - such as a PACS - to import data into DASHER. The DICOM receiver redirects data to the import folder, not into XNAT directly. To perform a DICOM-push use the following settings:
- Remote Host:
[xnat_host], i.e. the URL defined during the installation. Do not include http:// or https://. - Port:
8104 - AE title:
XNAT
Regardless of the import mechanism used, any incoming data is sorted according to the protocol which defines what data is allowed into the main hospital project and what data needs to be diverted into the QUARANTNE project (see The QUARANTINE Project and Setting the Quarantine Protocol for further information).
After sorting incoming data into target projects of the non-anonymised XNAT, all data will arrive in the Prearchive. This allows checks to be performed before data is archived into the hospital project or the QUARANTINE project.
To navigate to the Prearchive, log into the non-anonymised XNAT instance and in the top bar of the non-anonymised XNAT select Browse -> Go to Prearchive. You will see a list of sessions currently residing in the Prearchive. Note, it can take 1-2 minutes for a session to appear in the Prearchive. For each session you will see the target project, the subject ID, the session ID as well as the current status listed. Sessions will remain in the "Receiving" status for a significant amount of time. However, more data can be imported in the meantime.
You may need to use the Refresh button located over the Prearchive session list to see an update of sessions and their status.
If no conflicts arise, sessions will be archived automatically and will disappear from the session list and will appear in the hospital or QUARANTINE project as described here in more detail.
If a session with the same Subject and Session IDs already exists within the target project, the session will stay in the Prearchive with a ‘Conflict’ status. To resolve this, the session can be (a) edited and given a new session ID or (b) deleted if it was imported accidentally.
(a) To give the imported data a new session ID (no special characters permitted) select the tick-box next to the session in the Prearchive. Click on ‘Review and Archive’. Replace the conflicting Session ID with a new Session ID. Click on Submit.
(b) If you simply want to delete the import, select 'Delete' instead of 'Review and Archive' and the import will be cancelled.
Next Pseudonymising Data