TMII XNAT FAQ and User Manual

This is a place for you to find and ask TMII XNAT related questions.

Q: What is the URL of TMII XNAT? 

A:

https://tmii.mssm.edu/xnat

   When you type the url to your web browser, it looks like as follows. 

   

Q: What are username and password? 

 A: If you have a mssm.edu (mssmcampus) ID, then just type the school ID and password. 

     If you have a mountsinai.org (msnyuhealth) ID, first select mountsinai.org in the login via drop down box, then type your hospital ID and password. 

     If you don't the above two accounts, select Database in the login via drop down box, then you can email chen.yang@mssm.edu to request an account.

     For example, login as mssmcampus account. 

     

Q: Dose XNAT store my password? 

A: No. All your user information is stored in LDAP servers. 

Q: What if my mssmcampus or msnyuhealth account password changes? 

A: When you login the TMII XNAT, then just type your username and new password. 

Q: What it looks like when I login? 

A: A snapshot on August 1, 2013

Q: How does XNAT organize my data? 

A: Projects,Subjects,and Experiments 

   Project --- Your study project (e.g. IRB approved project)

   Subject --- a project can enroll many subjects

   Experiment ---each subject can have many experiments (e.g. MR session)

Q: How to create a project?

A: Click New->Project


Fill the forms as follows and Click Submit. 

Here is an example page for the created project. 

Choose and save auto-archive if you want XNAT automatically archive your images as following:


 

Q: What are the project accessibility? 

A: It's a good idea to keep your data private. You can then add members to your project to access the data. 

   Public---All users will be able to access study data for analysis and download.

   Protected --- All users will be able to see your study title and description, but only collaborators you approve will be able to analyze and download data.

   Private -- Only you and study members will be able to access study data.

Q: How to add a user to access my project data? 

A: You must be a project owner.

    1. Go to your project and click Access tab

     
 2. Add the user's email or user's ID as one of the following roles to your project

   Collaborator (Can access and download data)

   Member (Can add and change your data)

   Owner (Can add, change, and delete)

Click OK.  When the invited user login, the user will have access to your project data as the role your selected. 

Q: How to create a subject in my project?

A: In your project, Click New -> Subject

Fill the information as your need and Submit. 

Q: How does my MR session data go to the correct project and subject from scanner? 

A:  

Provide project, subject,and experiment label for XNAT to automatically archive your images.

 

Click Yes for the following window.

Please add the project, subject and session info in the "Additional info" textbox as following: 

Project: your_project_id;Subject:your_subject_label_string, Session: your_session_label 

The valid name should be a string without space. 

The last step is to choose your MR session from the image browser on the scanner, from the menu, click Transfer -> Send to ..., choose XNAT from the nodes list.

A (Technical): *Choose your MR session label (good label should be unique like label subject_label_modality_visit_number, e.g. *test_2_MR_1

    Setup the scanner and send MR data directly to XNAT as follows. 

    Put Project: {Project_ID},Subject: {Subject_ID}, Session: {Session_ID} AA:true  in DICOM tag: Patient Comments attribute (0010,4000) 

    XNAT DICOM server info:

    Remote host: localhost

    Port: PortNumber

    Remote AE title: YouAETitle

    Local AE title: XNAT-SCU 

   After finishing the above setting, then the MR session data will auto-archive to your XNAT project. 

Q: What does a MR session looks like in XNAT? 

A:

Q: How to download one MR session? 

A:

 

Q: How to add image processing pipelines to my project? 

A: DICOM to Nifti pipeline has been set up for you to add into your project by following the following steps. 


Click Add and change in the pop up window

Close the pop up window.

Q: How to convert my DICOM images to a nii file? 

A: For example, convert PROJECT: TMII_TEST > SUBJECT:Test_01 > Test_01_MR_1>scan 8 's DICOM files to nii.


When the pipeline finishes to run, you will receive a email saying NIFTI files have been generated for Test_01_MR_1 . Details of the session are available here.

 Your file will be available as the following:

 

Q: Will new protocols help to save my time? 

A:  Yes. The new protocol follows the following pattern: 

RESEARCH-[YR]^GCO-[YR][1234](D)(something else without dashes)
YR = always two digits. e.g. 10, 11, 12, 13
[1234] = always four digits
(D) = optional "D" for development/pilot

It will map you project gco number to DICOM study description field (0008,1030).

For example, RESEARCH-12^GCO-12-0991-V.Gabbay is from one dcm file. The project id we want to extract is 12-0991. 
Most of our XNAT project ids look like 10-1362, 13-8882, and a few of them look like 05-1110-1001. 
StudyComments (0032,4000)  is the second field that people types their project number like 05-1110-1001 in our old scanners.  

By using the new protocol, TMII XNAT will automatically extract your project GCO number and save your time to type it. 

Q: How to download one scan's images via REST API? 

A: https://tmii.mssm.edu/xnat/data/projects/TMII_TEST/projects/project_id/subjects/subject_label/experiments/experiment_label/scans/series_number/resources/DICOM/files?format=zip

  For example, here is the REST call to download all DICOM files just for one scan with series number 4. 

  https://tmii.mssm.edu/xnat/data/projects/TMII_TEST/subjects/Sample_Patient/experiments/20061214/scans/4/resources/DICOM/files?format=zip

Q: What version of XNAT is TMII_XNAT running for production?

A: XNAT 1.6.3. TMII_XNAT was upgraded from XNAT1.6.2.1 to XNAT 1.6.3 on 09/27/2013.

Q: What is the default image session label? 

A: Image session labels in a certain project must be unique. It's the folder name for XNAT to archive images. 

In order to reduce the conflict rates, we change the default image label from study date to accession number. 

If there is no accession number for a specific image session, patient ID will be used. 

Q: How to set up project specific anonymization? 

A:  Source: https://wiki.xnat.org/display/XNAT16/Anonymization.

      To set up Project Specific anonymization:

  1. From Home screen click a project
  2. Navigate to the Manage tab on the project screen
  3. Enable Project Specific anonymization and paste in anonymization script
  4. click Save button to initialize script
  5. You can also disable this script as needed by choosing the Disable radio button and clicking the SAVE button
  6. Here is an example anonymization script. 
  1. Anonymization script
    
-(0008,0050) // Accession Number
-(0008,1048) // Physicians of record
-(0008,1050) // Performing Physicians Name
-(0032,1032) // Requesting Physician
-(0008,1070) // Operators Name
-(0010,0021) // Issuer of Patient ID
-(0010,0030) // Patient's Birth Date
-(0010,0032) // Patient's Birth Time
-(0010,0050) // Patient's Insurance Plan Code SQ
-(0010,0101) // Patient's Primary Language Code SQ
-(0010,1000) // Other Patient IDs
-(0010,1001) // Other Patient Names
-(0010,1002) // Other Patient IDs SQ
-(0010,1005) // Patient's Birth Name
-(0010,1010) // Patient's Age
-(0010,1040) // Patient's Address
-(0010,1060) // Patient's Mother's Birth Name
-(0010,1020) //Patient's Size
-(0010,1030) // Patient's Weight
-(0010,1080) // Military Rank
-(0010,1081) // Branch of Service
-(0010,1090) // Medical Record Locator
-(0010,2000) // Medical Alerts
-(0010,2110) // Allergies
-(0010,2150) // Country of Residence
-(0010,2152) // Region of Residence
-(0010,2154) // Patient's Telephone Numbers
-(0010,2160) // Ethnic Group
-(0010,2180) // Occupation
-(0010,21A0) // Smoking Status
-(0010,21B0) // Additional Patient History
-(0010,21C0) // Pregnancy Status
-(0010,21D0) // Last Menstrual Date
-(0010,21F0) // Patient's Religious Preference
-(0010,2203) // Patient's Sex Neutered
-(0010,2297) // Responsible Person
-(0010,2298) // Responsible Person Role
-(0010,2299) // Responsible Organization
(0012,0063) := "CR common deidentification v001"
-(0038,0500) // Patient State
-(0038,0100) // Pertinent Documents SQ
-(0040,1400) // Requested Procedure Comments
-(7053,1000) // Philips Private Tag
-(7053,1001) // Philips Private Tag
-(7053,1002) // Philips Private Tag
-(7053,1003) // Philips Private Tag
(0008,1030) := project
(0010,0010) := subject
(0010,0020) := session
// END Common CR anonymization script

Q: How to write project specific anonymization script?

A: Please study the materials in this page: http://nrg.wustl.edu/software/dicomedit/dicomedit-reference/

Q: How to access the scanner auto phantom quality control results? 

A:  Auto Phantom QC is performed on research Skyra regularly. Please contact XNAT team to access the Agar_1372_Skyra_Research QCs.



Q: How to upload non-image data like Eprime, behavioral, and other files to XNAT? 

A:

    1. Go to your project->subject->session, e.g. 

    

2. Click Manage Files, Add Folder -> choose Level and item(scan is Level is Scans) -> input Folder name
 

Upload files to the corresponding folder in XNAT. 

Check your files

Q: How to view image files in XNAT?

A: Here are the instructions from the page: https://wiki.xnat.org/display/XNAT/Viewing+Image+Files

XNAT comes bundled with a Java applet that allows users to view images from an MR Session in the browser window. Here is a step-by-step process for getting this to work:

  1. From within a project, go to the subject and open the session
  2. Go to the Actions box, select View - View Images
  3. BE SURE to click "NO" on the Warning - Security box that inevitably comes up when a Java applet is launched. Accidentally clicking "Yes" here (as you might if you think "Yes, I want to run this applet") may prevent the applet from running correctly in the future, at least until you have cleared your browser cache. 
  4. A selection window will pop up when the applet successfully launches. From within the selection window:## Under Type: select the type of file you wish to view
    1. Under Image: pick the exact image you wish to view
    2. Click GO
  5. The Image viewer will open as shown below. You can use the slider to navigate from frame to frame of your image session. 
  6. To make the image larger, type "2" on your keyboard for 2 times bigger, "3" for three times, "4" for four times

At the very bottom of this window, move the slider to the left and right to see the image movement

Q: What is the BOLD_Preprocessing pipeline and how to install it? 

A: The BOLD Preprocessing pipeline performs gradient non-linearity distortion correction, motion correction, and field map distortion correction for your BOLD scans if it uses BIC protocol. 

   Install it. Ask your project owner to do: Project>Pipelines>Add More Pipelines>Add BOLD_preprocessing.xml> in the pop up window click submit>close as follows.

  

Q: How to run the BOLD Preprocessing pipeline in your project? 

A: Run it. Note that you can only launch one job per image session.  


Then wait for your results to be processed and uploaded to XNAT as follows. 

Q: What is DTI preprocessing pipeline and how to install it? 

A:  The DTI Preprocessing pipeline performs b0 intensity normalization, EPI distortion correction(topup), eddy current + subject motion correction and Gradient Nonlinearity correction for your DWI scans if it uses BIC protocol. 

   Install it. Ask your project owner to do: Project>Pipelines>Add More Pipelines>Add dti_preprocessing.xml> in the pop up window click submit>close as follows.

Then, please click Submit in the popup window as follows.

Close the following window. 

Q: How to run DTI preprocessing pipeline in your XNAT image session? 

A: Run it.

 

Your results will be available on SOMA and Minerva.

Q: How to run the structural pipeline (PreFreeSurfer, FreeSurfer, and PostFreeSurfer) with or without readout distortion correction? 

A: Click build and select Structual BIC pipeline

     
   

        If you have fieldmaps close to T1 and T2 images in the same image session, then put the scan ID e.g. for PA_or_RL_ScanId put 53 for AP_or_LR_ScanID  put 52.

        If you don't have fieldmaps or you don't want to do the readout distortion correction just put NONE in the fieldmaps box as in the picture below.


        

Q: How to run Functional Preprocessing pipeline (FMRIVolume and FMRISurface)? 

A: 

Q: What is XNAT REST API? 

A:  

Please check the following web page for details:


https://wiki.xnat.org/display/XNAT16/XNAT+REST+API+Directory

https://wiki.xnat.org/display/XNAT14/XNAT+REST+API

Q: What is pyxnat? 

A: The homepage of pyxnat is http://packages.python.org/pyxnat/ .

     The code is in github: https://github.com/pyxnat/pyxnat

Q: I sent two scanning sessions with the same session label to XNAT but only one was archived - how do I find the second one? 

A: When you send the image sessions with the same session label to XNAT several times, only the first one will be archived automatically.

The other later sessions will be placed in the prearchive as conflict in XNAT. 

You need to resolve the conflict by merging sessions or archiving them as new sessions or deleting them manually.

 

For the best practice, please choose the unique label for your image sessions in each project e.g. accession number or 2014_12_29_v1. 

Q: What is a Functional QC pipeline? How to run it?

A: A Functional QC pipeline performs rapid quality control for fMRI scans in XNAT.  The pipeline uses the Minerva computer cluster to run QC jobs and the results will be uploaded to XNAT. 

The snapshots below are showing one Resting state scan QC results. 


Here are the steps to run the pipeline for your fMRI scans. 
1. Install the FunctionalQC pipeline to your project. 
2. Click Build > Select FunctionalQC > Submit

3.Put the BOLD scan id in the comma separated values in the BOLD_list_id text box and submit. 

4. Check your results on XNAT(Note the estimated run time per scan is one hour)

Enter labels to add to this page:
Please wait 
Looking for a label? Just start typing.