i3-sponsored Calit2 Strategic Research Opportunities

UCSD Calit2 is a non-academic university research unit where researchers from different disciplines come together to study and develop new technology to address needs and issues in our economy and society.

The Calit2 Strategic Research Opportunities (CSRO) program provides one-year awards of cash and in-kind support via a peer-reviewed process established by the University of California, San Diego division of the California Institute for Telecommunications and Information Technology (Calit2).

Calit2 hopes to encourage high-impact, near-term research by faculty and research scientists who are already affiliated with the institute at UC San Diego, or who would like to be so affiliated. Calit2 also uses the CSRO process to systematically deploy Calit2 resources (services, equipment and personnel) to ensure maximum usage and return on its investment in these resources. The program also encourages new research initiatives that reflect the priorities outlined in the Calit2 Strategic Plan adopted in 2010, and the production of hardware, software artifacts, publications, external research support and other tangible results.

All funding for the CSRO program comes from private sources such as the International Innovation Initiative (i3). Our sponsorship of selected projects provides an opportunity for Calit2 leadership to showcase specific research projects that could not have been undertaken without donor support to the institute.

The International Innovation Initiative (i3) has provided sponsored support to the following CSRO research projects for 2012-2013:

Quantum-Structure Enabled Thermal Energy Harvester for Self-Powered Electronics (Green Energy Tech).

Conduct a proof-of-concept study of a thermal energy harvesting device with high performance and small footprint, which can be readily integrated into self-powered electronics.

PI: Renkun Chen and Zhaowei Liu


The objective of this project is to conduct a proof-of-concept study of a thermal energy harvesting device with high performance and small footprint, which can be readily integrated into self-powered electronics. The proposed work is built upon the PIs’ recent fundamental discovery on thermoelectric transport in multilayer quantum structures. This project, if successful, will enable breakthroughs for powering a variety of devices including embedded sensor network, nanoelectronics, and personal health devices, among others.

People: Renkun Chen (rec001@ucsd.edu), Zhaowei Liu (z4liu@ucsd.edu)

Cognitive Healthcare: An Enabling Android

Developing a detection system to monitor a subject’s physiological state in order to promote better health

PI: Tara Javidi and Giorgio Quer


The recent emergence of medical devices that can collect time series for long durations at very high sampling rates makes it possible to study temporal patterns in biophysical signals. There is a huge amount of data that can be sensed with tiny and inexpensive wireless sensor devices, which can become part of the everyday life of many people in the future. The data open new research opportunities that can have a dramatic impact in the world of healthcare. We seek novel techniques to extract information from this data, interpret this information, control certain vital parameters and provide to the user a precious real-time feedback on his/her state of wellness. The enabling technology is a cyberinfrastructure that include wireless sensors (to measure physiological parameters), mobile devices (such as a smartphone or tablet), a specific application (running on the mobile devices), a remote server (to store the data and run the most computationally intensive algorithms), and a web interface. The system is not designed to find new ways to cure diseases. Instead it is designed to assist healthy subjects and patients with chronic diseases making everyday decisions that may affect their health. The system provides the subject with rich physiological information helping the subject understand his/her physiological state. The system can train the subject to be fully aware of his/her physiological state, suggesting personalized ways to improve efficiency at work and all-day wellness.

People: Tara Javidi (tjavidi@ucsd.edu), Giorgio Quer (gquer@ucsd.edu)

Automatic Radiotherapy Treatment Plan (Health & Wireless & Cloud Computing)

Utilizing computation to develop tailored treatment plans for cancer patients

PI: Yuanyuan Zhou and Steven Jiang


Cancer is the second cause of death in US. About 2/3 of cancer patients are treated with radiotherapy since it is a proven effective treatment for various cancer types. The treatment is complex and very patient specific. The radiation beam direction, fluence profile, energy, size, exposure time, and many other parameters need to be tailored to each patient’s case through a process called treatment planning, where an optimal treatment plan is designed for each individual patient based on his/her CT image and physiology. Treatment planning is conducted by planners using dedicated software systems called treatment planning systems (TPS). A finished treatment plan needs to be carefully inspected and approved by the attending physician before it can be used to configure the radiotherapy settings. To improve the efficiency and effectiveness of radiotherapy, we propose a GPU-based next generation of treatment planning cloud infrastructure. In our proposed solution, the computation and storage is hosted in the cloud and the user can use the service via commodity computing devices, such as desktops, tablets or even smart phones. The final treatment plan will be downloaded to the treatment machine and used for patient treatment. Such process can be done within 10-30 minutes, without wasting patients’ precious time fighting against cancer.

People: Yuanyuan Zhou (y4zhou@ucsd.edu), Steve Jiang (sbjiang@ucsd.edu)

Flexible EEG “Patch” for Point-of-Care Health Monitoring (Medical Devices & Health)

Developing a compact and portable EEG device

PI: Deli Wang


Electroencephalography (EEG) monitors the brain function and activities, which has been playing extremely important role in both clinical diagnosis and cognitive neuroscience research applications. However, the current EEG technology is bulky and can be only performed in hospital or specially equipped facilities, which has limited the use of EEG in modern and future healthcare and does not allow point-of-care health monitoring application. We propose the development of a prototype device of an EEG “patch”, which, like a band-it, can be put to behind our ear (non-hairy area), and constantly monitor our brain activity. The tested rests can be read out via either wired or wireless connection to a portable reader or smart phone. The proposed device has significant clinical and research value, as well as great commercialization potential and market value.

People: Deli Wang (dwang@ece.ucsd.edu)

Camelot: A Multi-Camera Stereo Video Environment Recorder

Developing a module for applications in advanced robotic education Curricula

PI: Truong Nguyen, Tom Defanti, and Jurgen Schulze | International Partner: KETI


Camelot is an ultra-wide angle, multiple-lens stereo camera for shooting super-high resolution video for display on Calit2’s virtual reality (VR) systems. It will be a significant improvement over the CAVEcam stereo camera system currently in use, as it will add motion/video capability and an ability to capture more dynamic scenes by using synchronized arrays of cameras. Camelot is a challenging research and technology development project because the necessary means to synchronize these cameras is not published, nor for sale, nor obvious, and the temporal and spatial stitching of so many 24-60fps images is a daunting task for which new code will be needed running on Calit2 visualization clusters using ~10,000 streaming multiprocessors in the graphics processor units (GPUs). Camera choice and discovery of optimum mounting strategies are critical and difficult research tasks for Camelot’s success. As a continuation of the cyber-archaeology and tourism work at Calit2, many cultural applications are anticipated.

People: Tom DeFanti (tdefanti@ucsd.edu), Jurgen Schulze (jschulze@ucsd.edu)
PhD Students: Zucheul Lee (zucheul.lee@gmail.com)