October 2010 Newsletter
Dear Friends of CITRIS:
It is always an honor to be associated with CITRIS, an organization as dedicated to excellence in research as it is to putting that research to work for the good of humankind. But I was especially grateful last month at the third annual Scientific Colloquium for Healthcare, Engineering and Medicine (SCHEME III). The meeting, organized by my colleagues Dr. Javeed Siddiqui, CITRIS Medical Director, and Steven DeMello, Director of Health Care, was simply magnificent. Those of you who were unable to make it to the UC Davis Medical Center can read about some of the highlights in the first newsletter story below, or watch much of the conference itself, which is streamable on CITRIS's YouTube channel.
Jonah Frohlich, Deputy Secretary for Health Information Technology at the California Health and Human Services Agency (CHHS), kicked the conference off with a potent keynote, establishing at the outset why the healthcare related work we are doing on all four CITRIS campuses is so important. By 2050, he explained, as the number of senior citizens in the state triples compared to 2000, and the number of people with two or more chronic health conditions skyrockets, the State’s already chronic medical problems will be potentially catastrophic. Due to the recession, there are two million more uninsured Californians today than there were just two years ago, says Frohlich, and though the implementation of major new healthcare reforms will change that, exactly how, how quickly, or how much are still unclear. One thing we do know, however, is that developing and implementing new healthcare technologies is both a national and state priority. The Health Information Technology for Economic and Clinical Health (HITECH) Act, part of President Obama’s 2009 economic stimulus package, is expected to bring about three billion dollars to California’s physicians and hospitals trying to implement electronic records systems. In that effort, cyber security of the sort pursued by the CITRIS-affiliated Team for Research in Ubiquitous Secure Technology (TRUST) will be essential.
In addition to helping to move information about patients more efficiently, the State of California, with help from CITRIS, has just launched an effort to effectively move doctors’ expertise to patients in remote regions of the state. The California Telehealth Network (CTN) is working with 22 million dollars to link top urban medical centers, such as the UC Davis Medical Center, to rural clinics and doctors’ offices. In addition to acting as a secure communications superhighway for California healthcare, CTN will also be a testbed for exactly the kinds of technologies being incubated in CITRIS’s labs.
An example of such a project is a collaboration between UC Berkeley's Ruzena Bajcsy and UC Davis's Oliver Kreylos. Their project, covered in the newsletter’s second story, creates a virtual, three-dimensional space in which researchers or clinicians in different locations can meet, along with an “objectified” 3D data set, such as an MRI. The work literally finds new dimensions in which great scientific--and clinical—work can occur.
CITRIS has focused its sites on another of California’s most pressing problems. So far, we have invested $1.2 million in 27 active health care related projects and, I’m grateful to say, we are going to make a big difference.
Thank you for your continued interest and support of CITRIS.
Paul K. Wright
Director, CITRIS and the Banatao Institute@CITRIS Berkeley
by Gordy Slack
In the next few decades, demographic changes in California will put a major strain on the California healthcare system’s ability to provide adequate care for its citizens, said Jonah Frohlich, Deputy Secretary for Health Information Technology at the California Health and Human Services Agency. In his keynote address at CITRIS’s third Scientific Colloquium for Healthcare, Engineering and Medicine (SCHEME) conference on September 8, Frohlich said that unless the system is revised and modernized it will be overwhelmed by an older and sicker population of patients with an ever more diverse range of needs.
“The years between 2000 and 2050 will see a tripling of the state’s senior citizens,” said Frohlich. And as we age, we get sicker. For example, about 22 percent of those senior citizens have diabetes and 20 percent have two or more chronic diseases. Such patients account for 60 percent of healthcare dollars spent. As we age, that number will grow, compounding the state’s economic and moral healthcare burdens.
California already suffers from great disparities both in the quality of healthcare and its accessibility. The poorest Californians, and those living in parts of the state far from medical centers, get much less care, and have lower overall health status.
Furthermore, as the population ages, says Steven DeMello, CITRIS Director of Healthcare, and as the incidence of chronic disease grows, the state’s shrinking number of professional healthcare providers will be swamped with patients. And the insurance and state agencies that are responsible for funding their care will be overwhelmed with expenses as well.
And finally, as the state’s population becomes ever more diverse, and as the capacity to do personalized medicine grows, the kind of medical care citizens expect will diversify, too, requiring healthcare providers to tailor their approach to the individual needs of patients. “One model is no longer going to satisfy all,” says DeMello.
Participants in the SCHEME conference, which met at the UC Davis Medical Center in Sacramento, explored several ways that CITRIS--through its collaborations of engineers, doctors, and health administrators—is finding ways to address these challenging trends by delivering more, better, and more equitable healthcare to all Californians. And to deliver it for less money.
Dr. Javeed Siddiqui, CITRIS Medical Director, and Steven DeMello, Director of Healthcare, outlined a CITRIS vision of California healthcare where all patients have access to “care anywhere.”
“Users will have access to healthcare at home, in clinics, hospitals, and on their cell phones, 24/7,” says DeMello.
A big step toward “care anywhere” came with the launch on August 17 of the California Telehealth Network (CTN), which will eventually link over 800 hospitals, clinics, and doctor’s offices around the state with a secure, dedicated, high-speed computer network, shrinking and in many cases effectively deleting the distance between healthcare providers all over the state.
Empowering Tools for Home Care
As the number of healthcare givers shrinks and the number of healthcare recipients grows, more people will be taking care of themselves as well as their sick and elderly friends. Healthcare is going to be a growing part of everyday life, and we need tools that can help non-professionals to manage it. Much of this transition will be aided by monitors and sensors under development at CITRIS that can take and transmit vital readings, such as glucose levels, heart rate, blood pressure, and temperature.
“Heathcare is a trajectory, not a condition,” says UC Berkeley engineering professor John Canny. “We want to explore ways of tracking that trajectory to give people a softer landing.”
Sensing Depression, Promoting Exercise
Sensor technology can be used in surprising ways to measure health status. Depression, for example, often a co-morbid condition, may be detectable in its early stages, says Canny, who is developing a cell-phone-based analyzer that can pick up the guttural changes in the voices of people suffering from depression. The tell-tale changes are physiological responses to the biochemistry of depression, Canny says, and are detectable with a surprisingly high degree of accuracy. Catching depression before it becomes debilitating or starts to impact other health-related conditions could save a lot of suffering, and could preempt a lot of healthcare, says Canny.
Canny, whose lab is also conducting wireless monitoring studies on edema, stress, and Crohn’s disease, is also exploring ways to promote good health by providing feedback to users on their activities and lifestyles before they get sick.
Ruzena Bajcsy’s lab at UC Berkeley is also developing wireless body sensors to measure different variables. She and her colleagues are working on the serious privacy challenges that must be addressed before the new technologies can be widely adopted. There are social and even aesthetic challenges, as well. Some people, Bajcsy says, do not want the monitors to be visible, for example. “There may be some stigma,” she says, “so the design of small camouflaged sensors may be useful.”
The most effective and cost-efficient way to address diabetes is to keep people from getting it in the first place. Fighting obesity is a key part of that battle. Edmund Seto, a CITRIS researcher and lecturer at the UC Berkeley School of Public Health, also gave a presentation about his use of cell phones to monitor physical activity, including exercise among teens who increasingly struggle with obesity. He is hoping to create programs that could use cell-phone-based GPS to let users know not only how much exercise they are getting, but also where they are getting it and how best to reinforce it. Soheil Ghiasi, an engineer at UC Davis, is looking at healthcare from a preventative perspective as well. He is using advanced imaging technology and pressure monitoring boards to analyze whether athletes are at risk for one of the most common injuries in sports. The incidence of anterior cruciate ligament injury, or ACL, is especially high among young female athletes; it is between three and six times more common than among their male counterparts. By recording and analyzing images of athletes running, Ghiasi can both detect dangerous running styles and train young runners, using immersive images, to practice drills and build muscle memory, thereby preventing a broad stream of accidents from entering the river of Emergency Department visits.
Douglas Boyd, a heart surgeon, robotic surgery pioneer, and professor of surgery at UC Davis Medical School, presented his work at SCHEME III, and talked about the importance of forging relationships between doctors and engineers. Budding collaborations between Boyd and UC Santa Cruz robotics specialists Jacob Rosen and UC Berkeley engineer Ken Goldberg will be examining new ways to make surgery easier to do remotely and less invasive and more precise overall. Goldberg, for example, is looking for a way to guide a bevel-tipped and flexible needle through tissue to aid surgery and make targeted drug delivery simpler and less invasive.
With the right technology, de-professionalizing some treatments and bringing them closer to home can be liberating and empowering, for patients as well as less expensive for healthcare providers. UC Santa Cruz robotics bioengineer Rosen, for example, director of the Bionics Lab, gave a presentation on his work with therapeutic robotics, which could eventually be used in a patient’s home to help stroke victims and others to do the kind of physical therapy that now requires a dedicated professional physical therapist. Long sessions of repeating a particular, precisely constrained set of movements can help patients relearn to use their limbs. Today these sessions must be overseen by physical therapists, but they could be carefully guided by pre-programmed exoskeletal robotic arms that could gather and feed progress data back to a human practitioner. This way, a physical therapist could oversee therapy for dozens of patients a day instead of two or three.
by Gordy Slack
At the Institute for Data Analysis and Visualization (IDAV) at UC Davis, Oliver Kreylos has developed modeling and visualization tools that help put scientists inside their own work. Whether they are building proteins, conducting archaeological research, or exploring deep-earth geology, Kreylos and his team take huge data sets and develop them into three-dimensional models that can be projected like holograms into virtual space, where they can be spun around, flipped over, magnified, and even entered.
Three-dimensional images grab people, says Kreylos. They seem to engage a part of their brain devoted to the analysis of objects rather than images, a part that has greater bandwidth and greater imaginative and analytic power.
One geologist Kreylos worked with had always insisted on going into the field before doing an analysis. He had to see the “thing in itself” to grasp its full dimensionality. “No number of two-dimensional pictures would make the work come together for him,” says Kreylos. But entering the CAVE, the central immersive visualization facility of UC Davis' W. M. Keck Center for Active Visualization in the Earth Sciences with a 3D representation (the Davis lab’s immersion space) “allowed his brain to believe that he was dealing with the real thing, that he was in the field.” More than that, he says, it allowed him to see the data in new ways.
The CAVE has been running for several years. But through a collaboration with engineering professor and former CITRIS executive director Ruzena Bajcsy and her Tele-Immersion Lab at UC Berkeley, it now can bring users “across the street or across the continent,” Kreylos says, and into the virtual environment together. This fusion of teleconferencing and immersive imaging is called tele-immersion.
Tele-immersion could be particularly useful to doctors who want to look at medical data and to collaborate with colleagues in other locations. New partnerships have formed between the engineering teams at UC Davis and Berkeley and doctors at the UC Davis Medical Center. Kreylos’s lab is working with two radiologists to improve the examination of MRIs and other medical images, which are currently visualized as a series of two-dimensional slices representing the brain at different depths.
“Doctors have to look at these and build a mental model of the brain being examined,” says Kreylos.
But if they could look at the patient’s brain as a 3-D object, with the ability to penetrate and examine it at any depth, they would be much more likely to reach consensus in their collaborations with other doctors. Rather than relying on separate analyses, the doctors could look at the brain together, pointing out salient features. And what better place for two neurologists to conduct a consultation than inside a digitized representation of their patient’s brain?
Regular teleconferencing, where users at remote sites see 2-D video images of each other and perhaps of a patient, is good for many projects, Kreylos says, but there is a “formality” in the communication. In the traditional teleconference, users tend to focus on one another’s words and faces, and to neglect the data. In the immersive conference, users are fully aware of their colleagues’ presence (which is represented by a 3-D image of them “in” the space, but the data is the focus. Collaborators tend start exploring the data object immediately, says Kreylos.
The immersive 3-D is also better than 3-D visualizations that are made for traditional computer screens, which distort distances and angles to get their 3-D effect, says Kreylos. It is no help that a representation of data “looks right” if its not strictly accurate. In the immersive system, the virtual scale model has no distance or angle distortions.
The Teleimmersion project involves three components, says Gregorij Kurillo, Assistant Research Engineer at UC Berkeley. First, there is the collaboration infrastructure, which provides the 3-D workspace itself. The most extreme and immersive workspace is the UC Davis CAVE. A less extreme version, which still allows for a high degree of immersion, is a large 3-D TV, such as the one at the Teleimmersion lab in Berkeley.
Part of the project between the Berkeley and Davis labs is to develop simple and efficient ways to allow one user to be in the CAVE, or a CAVE-like environment in a hospital, and for another user to have a less elaborate (more affordable) setup in a doctors office.
The second component is the real-time video capturing system, which was developed and continues to be perfected in Bajcsy’s Tele-I mmersion Lab. This component captures three-dimensional images of the users and transmits them over a broadband connection to the collaborators, allowing them all to perceive each other and the shared space. The points-of-view and manipulations of all of the users are visible to all other users; when one person grabs the image of the brain and flips it over, everyone experiences the flip from their own vantage point. If someone zooms in to get a closer look, everyone else can see that, too.
This and other applications will be employing as a testbed the newly launched California Telehealth Network (CTN), a private, fast, and secure network overseen by the CITRIS-affiliated Center for Health and Technology, centered at the UC Davis Medical Center. The CTN will connect more than 800 California health care facilities to a statewide medical-grade network of health care and emergency services.
In addition to the examination of large data objects, the engineers are developing other medical applications. For instance, Bajcsy’s group is working with a UC Davis neurologist, Dr. Jay Han, who specializes in treating neurodegenerative diseases. Together they have begun to develop an application that would precisely measure the range of a patient’s arm movement from a remote site, and then record and analyze that at the Medical Center. The system would make it much easier to monitor the progression of diseases that inhibit motion without constant patient visits to the hospital.
“It could also be used for telerehabilitation,” says Kurillo, “where patients could meet with the therapist in a virtual environment and train over the network.”
Because the teleimmersive system registers the movement of the body’s surface, and not specific geometric points, the researchers need to extract the kinematics for the underlying skeleton, which means applying new algorithms and processing a lot of data in real time, says Kurillo.
Located near Bajcsy’s new Tele-Immersion Lab in the basement of Sutardja Dai Hall will be a telemedicine-wired mock medical examination room that will allow doctors and engineers to experiment with the CTN and various teleimmersive applications still under development.
Maintaining high-resolution images in real time without overloading processors or jamming transmission bandwidth is big challenge. To address this, Bajcsy and her colleagues are developing segmentation algorithms that help recognize and divide human forms into individual limbs for efficient reconstruction. They are also making mathematical models of human movement in order to identify actions, refining stereo algorithms to more quickly discern depth information, and customizing hardware architectures to optimize and coordinate all of these algorithms on their many parallel processors.
And finally, the system also requires a 3D visualizer, a programming tool kit for converting, developing, and processing data sets, like MRIs or seismic tomographic data, into 3D visualizations. These are also the domain of Kreylos’s team at Davis.
Between the three campuses, a whole new way of visualizing data is emerging and a whole new way of penetrating more deeply into data-driven work.