logo

Facilities

Neuroscape is located at the University of California, San Francisco Mission Bay Campus.

Neuroscape facilities are headquartered in the Sandler Neurosciences Center and consist of an Analytical Suite, a Development Studio, a MRI Lab and Neuroscape Research Labs.

Mission Bay

UCSF Mission Bay opened its first research building in 2003. A growing hub for the SF Bay Area biotechnology industry, it now includes seven world-class research facilitates and the UCSF Medical Center at Mission Bay (Children’s Hospital, Women’s Hospital, Cancer Hospital).
Mission Bay is home to one of the largest neuroscience complexes in the world, comprised of the Sandler Neurosciences Center – headquarters to Neuroscape – and Arthur and Toni Rembe Rock Hall. Two additional buildings will continue to expand the neuroscience community on this campus over the next decade – the UCSF Weill Institute for Neurosciences (WIN) and a new Psychiatry building. These two new sites will house Neuroscape facilities.

Sandler Neurosciences Center

A 237,000 square foot research facility, located on the Mission Bay campus was completed in 2012. It houses about 100 principal investigators and more than 500 additional researchers and staff. It is a home for the UCSF Department of Neurology, Memory and Aging Center (MAC), the Institute for Neurodegenerative Diseases (IND), and the Center for Integrative Neuroscience (CIN).

The Sandler Building is the headquarters for Neuroscape, which includes Analytical Suites, a Development Studio, a MRI Lab and Neuroscape Research Labs.

Analytical Suites

Our Analytical Suites are located on the 5th floor of the Sandler Neurosciences Center and serve as the primary sites for computational activities. It includes three rooms with twenty workstations for our Core team, as well as dedicated rack space in both on- and off-site secure server rooms. All other resources are available to facilitate research and writing (e.g., printer, copier, email, fax, on-line library catalogs).

Neuroscape ‘supercomputer’ for CUDA-enabled data processing

  • Customized Microway 2 Intel Xeon E5-2643 Sandy Bridge 3.3 Ghz Quad Core CPUs, 2 Nvidia Quadro 5000 GPUs, 1 Nvidia K40 GPUs, 64 GB DDR3 1600 Mhz, 512GB Samsung 850 Pro SATA

Workstation computers

  • 6 Quad-core (3.9GHz) Intel Core i7 iMacs with NVIDIA GeForce GT 750M 1GB GDDR5, 16GB 1600MHz DDR3 SDRAM, 1TB Fusion Drive
  • 4 MacBook Pros with 2.8GHz Quad-core Intel Core i7, 16GB 1600MHz DDR3L SDRAM, 1TB PCIe-based Flash Storage, NVIDIA GeForce GT 750M 2GB GDDR5 memory
  • 3 dual quad 2.8GHz Core Duo Intel Mac with 10GB RAM
  • 1 quad 2.8GHz Core Duo Intel Mac with 6GB RAM
  • 2 quad 2.66GHz Core Duo Intel Mac with 4GB RAM
  • 3 quad 2.5GHz G5 PowerMac with 4GB RAM
  • 2 quad 3.60GHz Intel Pentium® Dell with 2 GB RAM
  • 1 dual 2.24GHz HP Intel Core Duo with 3GB RAM (laptop)
  • 1 dual quad 2.0 GHz core Intel Xeon® Dell with 16 GB RAM

Computational Core

  • Computational Cluster: Dual Quad-Core 2GHz XEON XSERVE with 64GB RAM; 40TB XSERVE RAID; w/ an additional 16 virtual SGE nodes for parallel computational processing
  • Data Server: Dual Xeon E5-2620 Six-Core 2GHz processors with 64GB RAM; 24-disk chassis with 24 Seagate Enterprise Constellation 4TB disks (total 70+ TB storage) and 2 2.5” SSD

Neuroscape ‘supercomputer’ for CUDA-enabled data processing

  • Customized Microway 2 Intel Xeon E5-2643 Sandy Bridge 3.3 Ghz Quad Core CPUs, 2 Nvidia Quadro 5000 GPUs, 1 Nvidia K40 GPUs, 64 GB DDR3 1600 Mhz, 512GB Samsung 850 Pro SATA

Workstation computers

  • 6 Quad-core (3.9GHz) Intel Core i7 iMacs with NVIDIA GeForce GT 750M 1GB GDDR5, 16GB 1600MHz DDR3 SDRAM, 1TB Fusion Drive
  • 4 MacBook Pros with 2.8GHz Quad-core Intel Core i7, 16GB 1600MHz DDR3L SDRAM, 1TB PCIe-based Flash Storage, NVIDIA GeForce GT 750M 2GB GDDR5 memory
  • 3 dual quad 2.8GHz Core Duo Intel Mac with 10GB RAM
  • 1 quad 2.8GHz Core Duo Intel Mac with 6GB RAM
  • 2 quad 2.66GHz Core Duo Intel Mac with 4GB RAM
  • 3 quad 2.5GHz G5 PowerMac with 4GB RAM
  • 2 quad 3.60GHz Intel Pentium® Dell with 2 GB RAM
  • 1 dual 2.24GHz HP Intel Core Duo with 3GB RAM (laptop)
  • 1 dual quad 2.0 GHz core Intel Xeon® Dell with 16 GB RAM

Computational Core

  • Computational Cluster: Dual Quad-Core 2GHz XEON XSERVE with 64GB RAM; 40TB XSERVE RAID; w/ an additional 16 virtual SGE nodes for parallel computational processing
  • Data Server: Dual Xeon E5-2620 Six-Core 2GHz processors with 64GB RAM; 24-disk chassis with 24 Seagate Enterprise Constellation 4TB disks (total 70+ TB storage) and 2 2.5” SSD

Development Studio

The Development Studio, located on the 1st floor of the Sandler Neurosciences Center, is the primary site of our tech development projects tasked with both creating new technologies and advancing them to our Neuroscience, ClinicalEducation and Arts divisions.

Projects primarily include video game production, VR content creation, CUDA-enabled programming, BCI algorithm creation, MRI processing and website/video content generation and maintenance.

Equipment

PC1

  • CPU: Quad-core i7-4820K, 3.7Ghz
  • RAM: 32GB
  • Graphics: Nvidia Gefore Titan Black / GTX 1080 (soon)
  • Data: 1x 256GB SSD
  • Data: 2x 2TB Hli

PC2

  • CPU: Six-core i7-4930K 3.4Ghz
  • RAM: 64GB
  • Graphics: Nvidia Quadro K6000
  • Computational: Nvidia Tesla K40
  • Data: 1x 256GB SSD
  • Data: 2x 2TB Hli

PC3

  • CPU: Six-core E5-2630 2.3Ghz
  • RAM 32GB
  • Graphics: Nvidia Geforce 780Ti / GTX 1080 (soon)
  • Data: 1x 256GB SSD
  • Data: 1x 2TB Hli

Monitors

  • Overlord 2560×1440 @120Hz

Wireless Mobile Recording and Gameplay Devices

  • 20 Basis Peak watches for mobile physiology (heart rate, EDA, skin temp, and sleep monitoring)
  • 2 Empatica E4 wristbands for research-grade physiology data (HRV, EDA, accelerometer, temperature)
  • 52 iPad mini 3s, Wi-Fi – 128GB
  • 50 iPad 2s, Wi-Fi – 32GB
  • 12 Surface 3 tablets and 1 Surface Pro 3 tablet.

MRI Lab

The Neuroscape MRI Lab features a new 3T Siemens Prisma Fit MRI scanner with 64- and 20-channel head+neck coils, along with capabilities for simultaneous multi- slice (SMS) diffusion and BOLD imaging, 3D pCASL perfusion, SWI, and high-resolution anatomical imaging. The upgrade included the addition of a new XR 80/200 gradient coil capable of 80 mT/m @ 200 T/m/s simultaneously on all three axes, resulting in faster imaging in general and significantly improved SNR for diffusion imaging. The new TIM RF system with a 64 channel head+neck coil is capable of cutting-edge parallel acquisitions, such as multiband imaging, that enable higher resolution scans in reduced time with minimal artifacts. These improvements enable us to do connectome-level diffusion imaging and provide a huge improvement in the resolution of our echo-planar based scans, such as fMRI.

The MRI Lab has hardware and software for presentation of visual and auditory stimuli, as well as recording within the scanner of manual responses, eye-movements, galvanic skin response, pulse oximetry, respiration, EEG and EKG.

Lara Stables
Manager, Neuroscape Scanner

Team:
Dr. Sinyeob Ahn, Siemens physicist
Beth Pierce, MRI Technologist

Neuroscape Research Labs

The Neuroscape Research Labs (NRLs) are unique, state-of-the-art research laboratories that are designed to study novel neuro-diagnostic and neuro-therapeutic approaches.

Currently, three NRLs are located on the first floor of the Sandler Neurosciences Center at Mission Bay. One is specialized for full-body motion capture (MoCap NRL) and two for multimodal biosensing (BioSense NRLs). Our plan is to serve the greater UCSF scientific community with the creation of multiple NRLs located throughout the many UCSF campuses.

In the NRLs, research participants engage in customized, interactive environments aimed at more closely mimicking the natural world, but within the controlled setting of a modern laboratory. They perform experiments that merge closed-loop video games with recent technological advances in multi-sensory presentation, transcranial brain stimulation, and cognitive brain computer interfaces.

Neuroscape researchers monitor these complex interactions in a Control Room via multimodal data collection devices that leverage the latest breakthroughs in consumer motion capture and physiological wearable sensors (e.g., wireless EEG, electrodermal activity, heart rate, respiration, face expression capture and eye tracking).

We favor consumer-facing, mobile devices and cloud computing. The wireless and mobile approach is critical for seamless interactions and for offering the greatest likelihood of transfer of experiments to the home, clinic and classroom.

The NRLs are the first of their kind; offering investigators the most powerful lab yet to study complex, dynamic interactions between individuals and the environment.

Why did we create NRLS?

 

Current approaches to studying how the human brain functions and why it goes awry in disease states usually take place in unnatural laboratory settings using contrived experimental paradigms. This hinders us from learning how the brain actually functions in our daily lives.

A high-tech approach to neuroscience, which integrates richly simulated environments, full-body interactivity, and the latest in physiological sensor/motion detection technology, all within a world-class neurological/psychiatric research center will also serve as a unique platform to create and validate novel therapeutics.

MoCap NRL

The motion capture NRL (12′ x 24′) is primarily used in experiments that involve the integration of immersive sensory environments with full-body interactions.

  • 85” and 65” Plasma Screens
  • 5.1 Meyer Surround Sound systems
  • Oculus Rift Crescent Bay
  • Oculus Rift DK2
  • Sixsense Stem
  • HTC Vive
  • AKAI MPK Mini Midi keyboard
  • XBOX 360 controllers
  • Microsoft Kinect
  • Leap Motion
  • Myo Band
  • Subpac M2 and S2
  • Wearable physiological sensors (e.g., mobile EEG and transcranial electrical stimulation, electrodermal activity, cardiac function, respiration and eye tracking)

BioSense NRLs

Two acoustically-shielded BioSense NRLs (10’ X 12’) are primarily used for multimodal data collection (stress metrics, cognitive assessments, brain activity, eye-movements….). These data often serve as outcome measures to assess the impact of our training studies.

  • Biosemi Active Two EEG/ERP Acquisition Systems (128-channel and 64-channel) (http://www.biosemi.com/products.htm)
  • Rogue Research Frameless Stereotaxy system with BrainSight version 2 software and Polaris Vicra position sensor (https://www.rogue-research.com)
  • Magstim Rapid magnetic pulse generator with one 70mm rapid-pulse TMS coil and one 70mm air-cooled repetitive-TMS coil (http://www.magstim.com)
  • Two Neuroelectrics Starstim tCS devices (8 channel and 32 channel) (http://www.neuroelectrics.com)
  • One Neuroelectrics Enobio EEG (20 channel) (http://www.neuroelectrics.com)
  • Chattanooga Ionto tDCS device (http://www.caputron.com)
  • ASL and Tobii (www.tobii.com) eye-tracking systems
  • Shimmer GSR (www.shimmersensing.com) and Empatica (www.empatica.com) E4 and Embrace wrist-worn physiological sensors for collecting electrodermal, heart rate, skin temperature, accelerometer data both in-lab (Shimmer and E4) and remotely in participants’ homes (Embrace
  • iMotions Biometric Research Platform (imotions.com) to collect synchronized biometric data (EEG, GSR, ECG, and eye-tracking) time-locked to stimulus presentation.

Control Room

An adjoining, common control room (10’ X 24’) with audio/visual connectivity is used to direct the research experiments that are performed in all three NRLS.

  • 2 Meyersound MPS-488HP
  • FSR SPC-20
  • Crestron Pro2
  • Extron HDMI DA2, HAE 100, SSP Surround Sound Processor, XTP Crosspoint 1600
  • Soundweb London BLU-101
  • MeyerSound Galileo 616
  • 2 Meyersound MPS-488HP
  • FSR SPC-20
  • Crestron Pro2
  • Extron HDMI DA2, HAE 100, SSP Surround Sound Processor, XTP Crosspoint 1600
  • Soundweb London BLU-101
  • MeyerSound Galileo 616
  • 2 custom built quad core Intel i7-4820K (3.7GHz) with 32GB RAM and GTX 1070 / GTX 1080 for HD graphical display and full-body motion capture
  • 4 custom built quad core Intel i7-4820K (3.7GHz) with 32GB RAM (for EEG recording, stimulus presentation, eyetracking, and physiological monitoring)
  • 2 custom built quad core Intel Xeon E5-2537 (3.5GHz) with 32GB RAM and Tesla K20 for CUDA computation
  • 2 custom built six core Intel i7-5930K (3.5GHz) with 64GB RAM for biosensor computation
Get Involved

Neuroscape relies on support and participation from you to make this research possible.

Become a participant, donor, or volunteer.

Join Us