Does FIRMM store DICOMs?
Any DICOMs used by FIRMM will be automatically deleted from the FIRMM computer after two days. Therefore, it is very important that users never count on FIRMM for long term data storage.
How do I get data from my scanner to FIRMM?
To use FIRMM effectively, DICOMs need to be transferred as fast as possible to the incoming DICOM directory on the FIRMM host computer.
- SIEMENS: Set up and run
ideacmdtoolor the FIRMM start/stop shortcuts on your scanner. Instructions for this are available in our ideacmdtool README or shortcuts README, respectively.
- GE: Please email us for help.
- PHILIPS: Please email us for help.
When the FIRMM installation script is run, it makes two Windows batch files on the FIRMM machine. They are called
FIRMM_session_stop.bat. Getting these to the scanner from the FIRMM host PC will allow SIEMENS users to use DICOM streaming start/stop shortcuts. Read our shortcuts documentation for more information.
Does FIRMM work with GE or PHILIPS scanners?
FIRMM is designed to work with any scanner as long as DICOM data can be sent to a SAMBA shared network directory on the FIRMM host computer. All of our documentation and DICOM streaming shortcuts, etc., are currently built for ease of use with SIEMENS scanners and SIEMENS' real-time DICOM transfer. We are still working on the PHILIPS and GE modes for FIRMM.
How do I test FIRMM?
After connecting to the FIRMM host computer via
ssh -X firmm_host (where
firmm_host is your FIRMM Linux system's name), run
FIRMM -t. This will start FIRMM on the
firmm_host computer and copy a few test DICOM series to the incoming DICOM directory specified in the settings. Remember to click Start FIRMM in the browser window. FIRMM will close automatically a little after the test is finished.
How do I change the FD thresholds?
The FD thresholds can be adjusted before beginning a session by using the settings tab in the FIRMM GUI. See our Usage documentation for more information for more information.
Can I revert to a previously installed version of FIRMM if needed?
Users can revert to a previous version of FIRMM if the minor version is the same (e.g. reverting from 2.1.1 to 2.1.0). This would normally occur only if a bugfix that was introduced within the minor version caused problems running FIRMM on the user's system.
To revert to a previous version, find the previous version of the
run.sh file that was created by the FIRMM automatic updater. It will be stored in the same directory as your current
run.sh and its name will contain the previous version number (e.g.
run.sh.2.1.0). Re-save that previous file under the name
run.sh, replacing or removing your current file. Then you can run FIRMM commands as normal and the desired version will be used.
Does FIRMM work with structural data?
Not currently, but we plan to add this capability in an upcoming release.
Where has FIRMM been tested?
As of the creation date of this document, FIRMM has been tested on the following systems/scanners:
- Intel Xeon E5-2640v3 (16GB RAM, HDD) with Siemens Prisma scanner
- Dell Optiplex with i5 processor (4GB RAM, HDD) with Siemens Skyra scanner
- Core i7-4790K (16GB RAM, SSD) with Siemens Prisma scanner
What is the FIRMM FD filter?
New changes in MRI acquisition procedures bring new opportunities and challenges to BOLD imaging. One of the most drastic changes in acquisition procedures in recent years is the introduction of multiband imaging. However, an unintended consequence of the improved temporal and spatial resolution that accompanies multiband imaging is artifacts in motion estimates from post-acquisition frame alignment procedures, caused primarily by chest motion during respiration. Chest motion, secondary to respiration, changes the magnetic field (B0) and 'tricks' any frame-to-frame alignment procedure used in real-time motion monitoring into correcting a 'head movement' even though no actual head movement existed. In the newest version of FIRMM, an optional band-stop (or notch) filter to remove such respiration-related artifacts from motion estimates is available, thus giving a more accurate real-time representation of motion. For more detail, see our upcoming publication.