Grant Proposal Language

Quotations from previous proposals using CINL resources are provided below. Please email kmv@uab.edu to discuss exact budget amounts.

Also, if you have a good write up about this, please add it to the wiki

general resources
In addition to expertise available within the lab, we will also make use of UAB’s Visual Brain Core, which provides development and help with neuroimaging tools (docs.uabgrid.uab.edu/wiki/VisualBrainCore). Access to computing infrastructure (192 3.0GHz Intel-based compute cores with 386GB of RAM interconnected via a DDR Infiniband network) and supplemental data analysis help (a masters level computing systems administrator and PhD-level algorithms specialist) will ensure the accuracy and validity of the analysis methods planned above, but will also ensure that methods that may arise during the grant period can be applied to the analyses of our data.

MRI facility resources
Civitan International Research Center’s Functional Neuroimaging Laboratory: fMRI DATA ACQUISITION

A state-of-the-art 3 Tesla Siemens Prisma magnet is at our Civitan International Neuroimaging Laboratory at UAB Highlands (see https://www.uab.edu/medicine/circ/neuroimaging for more details). The Prisma is the most current upgrade to Siemens’ research line of scanners with excellent field stability that is optimized for MPRAGE, EPI and DTI scan acquisition as is required for many protocols including the “Human Connectome Project” protocols which we use regularly. We have a 20 channel and a 64 channel coil system, with many WIP agreements for performing cutting edge research quality scans. The facility includes ‘landing space’ for six trainees to work, two fully equipped testing rooms for working with participants, an equipment room including eye tracking equipment, a Bold++ display screen for showing high quality stimuli to participants, high quality audio equipment. CINL Director Dr. Mark Bolding oversees the facility, with help from Co-Director Dr. Kristina Visscher. They are assisted by Dr. John Totenhagen who spends full time at the magnet working with users and keeping the facility running smoothly.

Quality Assurance for MRI
Periodic QA assessments are performed on the UAB Prisma 3T MRI system. A standard spherical agar phantom is scanned weekly using the fMRI quality assurance methodology described by Friedman & Glover (1). Four quantitative measures of SNR, signal fluctuation, and signal drift are calculated and graphed on a website available to all facility users. Values which differ from measured historical means by more than 2 standard deviations are flagged and investigated by lab personnel and referred to MRI system manufacturer service engineers as needed. An American College of Radiology (ACR) large MRI geometry phantom is scanned every two weeks. Seven quantitative parameters are measured: geometric accuracy, high-contrast spatial resolution, slice thickness accuracy, slice position accuracy, image intensity uniformity, percent-signal ghosting, and low-contrast object detectability. These values are given on a website that is available to all facility users. Values which differ from measured historical means by more than 2 standard deviations are flagged and investigated by lab personnel and referred to MRI system manufacturer service engineers as needed. When QA procedures are flagged by lab personnel, users will be notified so that they can make note of anomalies that may affect their data.

Friedman, L. and Glover, G. H. (2006), Report on a multicenter fMRI quality assurance protocol. J. Magn. Reson. Imaging, 23: 827–839. doi:10.1002/jmri.20583

personnel resources
An example budget justification from a previous grant:

NAME HERE, Postdoctoral Researcher/ Analyst (1.2 months) A high level researcher will be needed to assist with programming data analysis and stimulus presentation scripts, as well as with overseeing our data backup system. Dr. NAMEis an experienced programmer for fMRI data analysis, and holds a position shared among several labs within the UAB neuroimaging community, as part of the Visual Brain Core (see docs.uabgrid.uab.edu/wiki/VisualBrainCore). She will devote 1.2 calendar months to this project.

computing resources
The following is one way to describe the available computing resources in a proposal. As a VBC user, if you've written a proposal that describes this better, please add that language to this page.

--

High performance Computing Cluster (Cheaha):

Neuroimaging datasets are large and modern analysis techniques require state-of-the-art processing capabilities. UAB's High Performance Computing cluster has the capability to perform the complex processing we propose here. **The following section contains older information -- you may want to refer to the cheaha site for the most up-to-date specs on the system. UAB IT Research Computing maintains high performance compute and storage resources for investigators. The Cheaha compute cluster includes 192 3.0GHz Intel-based compute cores with 386GB of RAM interconnected via a DDR Infiniband network. A high-performance, 180TB Lustre parallel file system built on a Direct Data Network (DDN) hardware platform is also connected to these cores via the Infiniband fabric. An additional 20TB of traditional SAN storage is available via a 1GigE network fabric. This general access compute fabric is available to all UAB investigators. Additionally, NIH funded investigators are granted priority access to the NIH SIG award acquired compute and storage pool that includes an additional 576 2.66GHz Intel-based compute cores, 2.3TB RAM and 180TB in the high-performance Lustre parallel file system all interconnected via a QDR Infiniband network fabric.**  Our research group has been working with UAB IT research computing to develop an efficient workflow for neuroimaging using the Cheaha system. Support for this work is provided in part through the Visual Brain Core at UAB (https://docs.uabgrid.uab.edu/wiki/VisualBrainCore). More information about Cheaha is available at: docs.uabgrid.uab.edu/wiki/Cheaha

Location: Rust Building, accessible remotely

back to VBC page
back to Visual Brain Core page