Announcements
July, 2011
CNI Software update
On Tue, Jul 5, 2011 at 11:42 AM, Gunnar Schaefer <gsfr@stanford.edu>
Dear all:
I am very pleased to announce that all new data collected at CNI will now be available for secure download at https://cni.stanford.edu/nims/. New data will not be available on the CNI Linux workstation that you have been using, so you will need to update your data transfer procedures. This new data access method is a significant improvement over the previous system. Details are provided below. As always, feel free to ask me or any of the CNI staff for help.
To improve data security and subject privacy concerns, the data download website is secured via Stanford's WebAuth system. You will need to log in with your SUNet ID and password, and you will only be authorized to access your lab's data (more details below).
All data collected at CNI from now on will be permanently archived and will be available through this web interface. To keep the permanent archive clean and tidy, old data will not be automatically copied there. However, if you want any previously collected data to be put into the permanent archive we are happy to do so. Just send me a note indicating specific exams, or come by to see me.
To facilitate the organization of this data archive, we have implemented a system based on the scanner's "Patient ID" field for you to indicate that a scan belongs to your lab and specific project. You can also indicate that you do not want a certain scan to be archived. In general, squash and phantom scans would fall into this category and should be flagged as such, so that they don't end up in the permanent archive.
What follows are details on the new "Patient ID" data attribution and sorting (and discarding) scheme. It is important that you follow these guidelines precisely, so that your data can be automatically attributed to your lab and made available to you via the web interface. Otherwise, manual intervention by CNI staff will be necessary before your data are accessible for download.
The format requirements for the "Patient ID" field are as follows (and are also posted by the GE console):
(Name list suppressed)
Our system also recognizes the special Patient ID value "discard". Enter "discard" for scans that you do not want, such as phantom or squash scans. The data for any scan with a Patient ID of "discard" will not be archived in our system and will not be available for download. Note that "discarded" scan data are NOT immediately deleted. They will linger in the GE database for about a month. So, if you decide that you do want to keep a "discarded" scan after all, just let us know and we can retrieve it.
Lastly, for technical reasons, data from the last series of each exam are typically not available in this data access system for a few minutes. All data should be available for download by the time you get back to your office, though.
Cheers,
Gunnar
June, 2011
CNI update
From: bobd at stanford.edu (Robert Dougherty) Date: Tue, 07 Jun 2011 11:38:21 -0700 Subject: [cni-mrusers] CNI update
Hi All,
Just a little update on how things are going down here at the CNI MR facility.
1. We finally got our signage approved and installed on the wall next to our main entrance! This should make it easier for scan subjects to find us. (And yes, we think the colon looks silly too.)
2. The mock scanner is ready for use. We had a double mirror built for it, so the display system is now a nice replica of the real scanner. Thanks to Kalanit and Alina for helping us get it up and running.
3. The 8-channel coil now also has a double-mirror like the one in the 32-channel coil, so stimuli viewed through it are properly oriented.
4. The prototype of the bone-conduction auditory system is working and several researchers are using it to present auditory stimuli in the 32-channel coil. We are working on an improved version with a more comfortable and easier to use form-factor, but that will probably take several months to develop and install. If you want to use the current system, please contact me and I'll show you how to set it up.
5. We now have the scan triggering sorted out in what we think is a cleaner, more organized system. Note that the "Serial Trigger" device has been updated and is now labelled "Eprime trigger". This device can also be used to receive the timing pulses in from the scanner, if you need those. You might need to install a new driver to use the new device. The old serial trigger device is still available for use (labelled "Defunct serial trigger"), but we hope to remove it soon, so please update your drivers to use the new trigger device. See the wiki http://cniweb.stanford.edu/wiki/MR_Hardware#Scan_Triggers) for details, and don't hesitate to ask us for help in getting your triggering sorted out.
6. We finished building a device needed for EEG balistocardiogram artifact removal. This is based on Gary Glover's idea of using the PPG (pulse oximetry) rather than ECG leads. Let us know if you'd like to do EEG and we can give you a tour of the system.
7. Reno has completed and installed a new version of the automated data transfer system. We call it "MR Steward". MR Steward in now more stable and reliable than the previous versions, and adds a couple of nice new features. As in the previous version, it automatically copies your DICOM files and P-files to the Linux box to the right of the console and generates NIFTI files for you. It now also finds and copies physiological data that you might have recorded and puts it in the same directory as your other data. Finally, it automatically starts the recon for any P-files produced by Atsushi's spiral PSD. This should all happen while you scan, so that your data should be available soon after you finish scanning.
8. Finally, Atsushi has make several nice improvements to the cni_epi pulse sequence. Details on this will follow in a separate email. We encourage everyone doing BOLD EPI to ensure that they are using this PSD. (Contact me or Atsushi for help.)
cheers, bob
BOLD EPI PSD update
From: bobd at stanford.edu (Robert Dougherty) Date: Tue, 07 Jun 2011 12:35:14 -0700 Subject: [cni-mrusers] BOLD EPI PSD update
Hi All,
Atsushi has just completed testing an updated BOLD EPI PSD. It is now installed on our scanner as "cni_epi". The basic imaging functionality is the same as with previous versions of our EPI psd, however Atsushi has made several important improvements around the edges.
1. The 10,000 image limit has been removed. You can now set up scans to run for as long as you like!
2. ASSET acceleration limits have been removed. Acceleration was previously locked to only two options: 1 (no acceleration), or 2. You can now set the ASSET acceleration factor to just about anything that you like, including fractional values, like 1.5.
3. There are now several new UserCVs under the "Advanced" tab:
- Scope trigger: This controls the timing pulses that come out of the scanner through the "scope trigger" connection. These pulses can be captured by our serial port trigger device and used to record the exact timing of each slice (set this CV to 0) or each volume (set this CV to 1). See the wiki (http://cniweb.stanford.edu/wiki/MR_Hardware#Capturing_scan_timing_triggers) for details on capturing these pulses.
- Reconstruct image at native resolution: This CV allows you to force the image reconstruction to occur at the acquired matrix size (set this CV to 1) rather than at the next higher power of two. E.g., if you set a matrix size of 74, the image is reconned at 74x74 rather than interpolated up to 128x128. This makes for faster recons and smaller files, and avoids confusion about "effective resolution" vs. the image voxel size. This is now the default behavior! If you liked your images to be interpolated up to the next higher power of two, you can set this CV to 0 to get that old behavior back.
- Save physio data: If you have connected the PPG, respiration belt, and/or ECG leads to your scan subject and want to save the data that are measured, set this to 1. Reno's MR Steward data transfer system should automatically find and transfer the physio files for you. However, this new feature of MR Steward is experimental. If you think that you saved physio data but can't find it, let us know and we'll help you get it out.
The start scan trigger CV is still there and functions as before, allowing no trigger (0), external triggering (1), or cardiac gating (2).
Please make sure that your BOLD EPI protocol uses this new PSD. If you are unsure of how to check this, just let us know and we can have a look with you.
We don't expect this update to interfere with any ongoing studies. The core imaging logic has not been touched, so images acquired with this new PSD should be the same as before. The only change that you might notice is that the resulting image size is no longer interpolated to the next power of two. If you think this might cause problems for your analysis pipeline, be sure to change this new CV to 0 to get the old behavior back.
cheers, bob
May, 2011
Sound delivery for the 32-channel coil
From: bobd at stanford.edu (Robert Dougherty) Date: Wed May 11 21:25:16 PDT 2011 Subject: [cni-mrusers] Sound delivery for the 32-channel coil For those interested in presenting auditory stimuli at the CNI, please read on. As many of you know, the Resonance Tech headphones that are currently installed at the CNI do not fit comfortably in the 32-channel head coil for many subjects. We are currently testing a prototype system that will fit in this coil. It uses MR-safe piezoelectric transducers (kindly provided by Resonance Tech.), that are identical to the ones used in our current headphones. But instead of conducting sound via air, these transducers are pressed against the subject's cheeks and conduct sound via direct bone conduction. The transducers are very thin and thus easily fit in the 32-channel coil. I just finished a pilot test, and the sound quality was good; speech-range stimuli were clearly audible during all our typical fMRI scan sequences (EPI and spiral), as well as the anatomical sequences. The main issue to be worked out is consistent positioning of the transducers to deliver clear sound while maintaining subject comfort. I think the best way to learn how to do this is to come down and try it out on yourself! I can show you what I did to get it working for me. So, if you want to present auditory stimuli in the 32-channel coil, let me know when you can come by to see the new system. cheers, bob -- Robert F. Dougherty, PhD Research Director Stanford Center for Cognitive and Neurobiological Imaging 70 Jordan Hall * Stanford CA 94305 * 650-725-0051 http://www.stanford.edu/~bobd
April, 2011
Flat panel display out for repairs
From: bobd at stanford.edu (Robert Dougherty) Date: Thu Apr 7 17:24:11 PDT 2011 Subject: [cni-mrusers] Flat panel display out for repairs... Hi All, The flat panel display that sits at the back of the bore will be going out for repairs tomorrow (4/8) at 11am. It will be back in place by 10am Wednesday (4/13) morning. The goggle system will be available for visual stimuli during this time. When the display returns next Wednesday, the sticky power button issue will be fixed and we will be able to adjust the display to make it brighter (it is currently stuck at its lowest brightness setting). We will also be able to move the display closer to the bore for a larger field of view. We'll mark the current position so if you can keep the viewing distance the same as it is now if you like. Also, I'd like to you all know that we are planning to remove the current mirror-reversal that subjects experience while viewing the flat panel display through the single mirror by switching to a dual-mirror and mounting the display upside-down. This will happen in about two weeks. (I will send another note next week with a more specific date for the switch.) I apologize to those who have already adapted their stimulus display software to the current configuration by left-right flipping your stimuli-- you will have to undo this in your software after we implement the new configuration. This new arrangement will fix the mirror-reversal in a way that is compatible with all stimulus display packages. thanks, bob -- Robert F. Dougherty, PhD Research Director Stanford Center for Cognitive and Neurobiological Imaging 70 Jordan Hall * Stanford CA 94305 * 650-725-0051 http://www.stanford.edu/~bobd
LCD display, bite bar
From: bobd at stanford.edu (Robert Dougherty) Date: Mon Apr 25 22:06:25 PDT 2011 Subject: [cni-mrusers] LCD display, bite bar Hi All, There have been a couple of recent improvements at the CNI: 1. The 32 channel coil now has a bite bar holder. It's same design as the one on the 8-channel (see the wiki: http://cniweb.stanford.edu/wiki/MR_Peripherals#Subject_Positioning). We have several mouth-pieces that can be used in either holder (see me for details). We don't yet have a supply of dental wax at the CNI, so bring your own if you need to use the bite bar. 2. The LCD display is now mounted upside-down. No, Gunnar and I didn't just screw up when re-mounting it. This, coupled with the new double-mirror on the 32-channel coil, means that images presented on the screen are no longer mirror reversed. Stimuli will appear to the subject exactly as they appear to you looking at the stimulus display monitor in the control room. Double-mirrors for the 8-channel coil and the 16-channel visual array are currently being built an should be in place by next week. See me if you need a mirror solution for these coils before then. Note that the mirrors are first-surface, which means that the reflected images are very nice, but also means that the mirror coating is exposed and more fragile than a standard (rear-surface) mirror. Please use care when cleaning them. There is a box of lens cleaners in the scan room near the positioning pads. 3. The LCD display has been been fixed so that it no longer needs to be power-cycled to wake up from sleep mode. Just turn it on once when you arrive and it should wake up whenever a video signal is sent. 4. There is a new response device available for use. It's a 4-button cylinder. See the wiki (http://cniweb.stanford.edu/wiki/MR_Peripherals#Response_Box) for details. Ask me if you need help using this or any of the other response devices. 5. We now have two scan trigger methods installed. See the wiki (http://cniweb.stanford.edu/wiki/MR_Peripherals#Scan_Triggers) for details, or just ask us for help getting your triggers working. Please don't hesitate to contact me or just drop by to get help using anything. Also let us know right away if something doesn't work as expected. cheers, bob -- Robert F. Dougherty, PhD Research Director Stanford Center for Cognitive& Neurobiological Imaging 474 Jordan Hall * Stanford CA 94305 * 650-725-0051 http://www.stanford.edu/~bobd