Facilities & Resources

Patient monitoring within all of the magnets includes non-invasive measures of heart rate, blood pressure, pulse-oximetry and end tidal CO2 levels.

We have 3 Siemens 3.0T Prisma Fit scanners. This system is the latest state-of-the-art scanner available and has 80mT/m gradients, a 64-channel, 32-channel, and 20-channel head-coils, and simultaneous multislice capabilities for generating protocols consistent with the Human Connectome Protocol. This system is dedicated to research studies, primarily focused on the development of functional MRI and applications in development, neuroscience, psychology, psychiatry, and neurosurgery.

This scanner is dedicated to both body and brain imaging research applications. With the wider, 70cm, bore this MRI is aimed at facilitating studies of obesity as it can accommodate much larger subjects than the Prisma scanners. Like the Prismas, this scanner is equipped with a 64-channel head/neck array-coil, and multiband and parallel imaging capabilities.

Dedicated to in vivo spectroscopy, the 4T system uses a new Bruker Avance console with two SGI Octane computers as hosts. The magnet is equipped with two actively shielded gradient sets, a 72cm id bore body set (Magnex Scientific) and a 38cm id head gradient insert (Bruker Instruments). The head gradient insert has a maximum strength of 45mT/m (X, Y, and Z directions) with a rise time of less than 150 msec for all three axes. It is equipped with all second order shims and a shielded Z2 and Z0 coil to allow rapid shim updating. There are four receivers, allowing phased array applications.

For a more detailed technical description of the Human Systems, please see Magnet Specifications.

MRRC has two MR compatible EEG systems for research use:

This system is capable of 24 bit maximum sampling at 20000 Hz. The MRRC has built an in-house EEG cap with carbon-wire electrodes that ensure MRI transparency and safety, with attached artifact detectors that are used to remove gradient artifacts and the ballistocardiogram from the EEG signal. In addition, the center has built a front-end anti-polarization circuit that reduces low-frequency drift of the carbon-wire electrodes. The Synamps2 also contains digital TTL inputs for triggering.

This system is capable of 22 bit maximum sampling at 1000 Hz. The MRRC has built a front-end analog filters to ensure that any gradient artifacts above the 500Hz sampling rate are not aliased into lower frequencies. The analog filtering is performed prior to any digitization. The Nu-amps also contains digital TTL inputs for triggering.

MRRC has 3 behavioral testing rooms and a mock scanner room available for use to all users. Locations are:

1 - Behavioral testing room (TAC-NL05)
2 - Behavioral testing room (TAC-NL03A)
3 - Behavioral testing room (TAC-NL33A)
Mock Scanner Room (TAC-NL33)

All 3 behavioral testing rooms are available for use by researchers who wish to bring in their own laptops or other hardware for behavioral tests. No in-house computers are available in any of the testing rooms. Researchers who wish to use the in-house button boxes should contact our Research MR Development Technician, Cheryl Lacadie for further information.

Room Reservations

The behavioral testing room and the mock scanner room can be reserved using the online booking system. Yahoo calendars have been created for booking time slots on each of these rooms. For reserving time in the behavioral testing rooms at the MRRC, visit https://calendar.yahoo.com/.

Please Note: Usernames and passwords for these calendars are not posted here for security reasons. Please contact Cheryl Lacadie to request this information for the room that you are interested in reserving.

Once you log in to the calendar, you will be on the schedule page. You can either add only specific dates, or, if you have regularly scheduled times each week/month, you can make the event "recurring" and it will come up each week/month.

General policies concerning the use of this calendar system:

• Include your name, phone number and/or email address so people can find you in case of a conflict. If you do not include your contact information, your reservation is VOID.
• Never delete other people's scheduled times without permission.
• Keep the calendar current by deleting times you have decided not to use, so that someone else can reserve the room.

• Biochemistry Lab for biochemical and isotopic analyses of plasma and tissue samples
• Neurophysiology Lab. The long-term goal of the Computational Neurophysiology Laboratory (CNL) is to understand seizure generation. The tools we employ towards this are the visual and quantitative analysis of the scalp and intracranial EEG.
• The Molecular and Cellular MR Imaging Lab (MCIL), which encompasses 1600 square feet of space adjacent to the animal MRI systems. This space is dedicated to projects within the field of Molecular and Cellular Imaging. The MCIL has dedicated wet lab space for tissue/cell culture and immunohistochemistry, a dedicated animal surgical suite, and office space for fellows and students.
• Electronics And Machine Shop (3 full time staff). Dedicated to the design and fabrication of MR hardware, RF coils, and associate electronics for the development of novel imaging methods and applications.
• Patient waiting room, and preparation rooms adjacent to the scanner bays.
• A two bed clinical suite Patient Recovery Room for out-patient style studies and subject monitoring located immediately adjacent to the human whole body magnets described above. The clinical suite is maintained in collaboration with the Hospital Research Unit of the Yale School of Medicine, and contains adult and pediatric crash-carts.
• Surgical Suites for Animal Studies

Data Storage: All MRI imaging data is stored in the data center at the MRRC, which houses several fault-tolerant servers with a capacity to store over 14 terabytes of data. The high-capacity file server ensures secure storage for all MRRC project data and is backed up to the cloud via a Gigabit Ethernet Backplane. Key features of the data center include high availability, reliability, scalability, and speed. The computing environment is regularly upgraded and securely maintained to be HIPAA compliant, with access restricted to MRRC members through password protection.

The MRRC computing lab offers a diverse computing environment with workstations running Windows, Mac, and Linux. Most data analysis is done on 24 dedicated quad-processor workstations with RedHat Linux, used for programming pulse sequences, image reconstruction, data analysis, and algorithm development. Shared workstations support tasks like image preprocessing, registration, motion correction, and functional MRI analysis. A variety of commercial and in-house MRI analysis software is available. The lab is managed by an experienced IT manager with over 12 years of experience supporting MRRC scientists. Data analysis and storage for this proposal will take place here.

BioImage Suite, developed at Yale over the past 12-15 years, has been widely used in peer-reviewed publications across neuroimaging and other fields such as diabetes, molecular imaging, and tissue engineering. It is a modular collection of programs designed for both interactive (GUI) and batch-mode processing, supporting platforms including Windows, Linux, and Mac OS. BioImage Suite includes graphical applications and command-line utilities, with nightly regression testing ensuring stability.

The unified framework allows for easy development and deployment of image analysis algorithms. New algorithms can be added, and custom workflow pipelines can be created, where each step processes inputs to produce output images and transformations. Written in C++ and using GPU-accelerated CUDA functionality, BioImage Suite also enables the creation of custom applications, such as cardiac segmentation editors or fMRI tools. It offers extensive neuroimaging tools, including segmentation, registration, and fMRI processing (task-based and connectivity). A unique data tree tool helps manage multimodal image analysis, tracking images and transformations across different coordinate spaces.

CWave is a software developed by Professor Graeme Mason for analyzing ¹³C isotopic labeling in brain studies. It is used to measure metabolic rates, observe substrate consumption, and track the appearance of metabolites in the brain. Provided free of charge with a license agreement, CWave includes a 76-page manual and allows flexible modeling of isotopic flow in metabolic pathways. Key components include:

• Pools: Chemicals with isotopic labeling (e.g., astroglial glutamate, neuronal glutamate, GABA).
• Rates: Speed of chemical conversion between pools (e.g., tricarboxylic acid cycle, neurotransmitter cycling).
• Drivers: Substrates that supply isotopes to pools (e.g., glucose and acetate).
• Combination Pools: Represent combined pools when individual pools cannot be discriminated (e.g., total tissue glutamate).
• Targets: Measured data, typically ¹³C enrichment timecourses.

CWave allows users to adjust rates and pool sizes to match simulated isotopic labeling or concentration time courses to target data, and can accommodate time-dependent changes for kinetic analysis.

DCMRsoft, developed at Yale by Daniel Coman, is used for processing MRI and MRS data from both single-coil and multi-coil systems. It can fit or simulate Z-spectra using Bloch equations, and analyze BIRDS or MRI datasets. For CEST applications, it fits Z-spectra for 2-pool or 3-pool systems to extract parameters like chemical shifts, water relaxation times (T₁, T₂), and exchange rates. It also simulates Z-spectra based on Bloch equations. Datasets from BIRDS can be processed to create 3D temperature and pH maps, while MRI data can generate 3D T₁/T₂ maps or functional activation maps using t-test analysis.

NMRWizard NMRWizard was designed at Yale by Robin de Graaf for MRI and MRS data processing. For MRS data analysis, NMRWizard contains a module for 1D/2D/3D CSI data processing, a module for individual spectral processing and a module for LCModeling. For MRI data processing, NMRWizard contains a module for image analysis and one for B0/B1 mapping, designed for both single-coil and multi-coil image processing.

FHstat, a 32-bit Matlab-based fMRI processing software developed by Fahmeed Hyder and Remco Renken, is designed for ease of use, particularly for Matlab users. It processes native files from Paravision (Bruker) or VnmrJ (Agilent), converting them into its own file format. FHstat supports voxel interrogation or ROI-based analysis and calculates t-maps from fMRI sessions, which can be overlaid on anatomical images. The analysis process can be highly automated with simple scripts, making it efficient for users.

Spike2, purchased from Cambridge Electronic Devices (CED), is used to control animal experiments by recording signals with sampling frequencies up to 166 kHz. It features a basic control language (sequences) that executes with high time precision (0.1 ms), ideal for synchronizing somatosensory stimulations with pulse sequence evaluations in high-field magnets. The software also supports signal analysis, including online or offline spike sorting, and offers a macro language for task automation. Recorded data can be easily exported to Matlab for further analysis.

The MRRC features 11,000 square feet of dedicated office space for the faculty, research staff, and personnel, including postdoctoral trainees, students, and MRRC business office. They all have secretarial, administrative, and IT support in MRRC.