The dVRK Allows Worldwide Research in Robotic Surgery

February 8, 2017

Peter Kazanzides, a computer science research professor, and Russell Taylor, John C. Malone Professor in computer science and director of the LCSR, have created an open-source software and electronics kit called the da Vinci Research Kit (dVRK) that is making minimally-invasive robotic surgery more accessible – and affordable – to research universities and institutions worldwide. The Taylor-Kazanzides team recently received a three-year National Science Foundation grant aimed at supporting their consortium’s developing work in this area.

The da Vinci robot is a revolutionary medical robotics system that uses tele-operation and stereo cameras for the research and application of minimally-invasive surgery. Many hospitals use the most recent (fourth generation) da Vinci robots, and the system is an increasingly common sight on TV medical dramas.

The dVRK is an “open-source mechatronics” system – consisting of electronics, firmware, and software that are used to control systems based on the first-generation da Vinci system. Research institutions can access the open-source software and electronics through GitHub. To date, researchers at 25 institutions –from Canada and Italy, to Hungary, Israel and Hong Kong – are using the dVRK.

What’s more, the dVRK is making retired and outdated da Vinci systems useful again.

“Until now, research into more sophisticated control strategies using the da Vinci system was hampered by a lack of a robust common platform, leaving [researchers] to choose between building an expensive, one-of-a-kind system, or using an overly simplistic platform that reduced research impact,” explained Kazanzides.

Many institutions had obtained their da Vinci robots from manufacturer Intuitive Surgical, Inc., which retired its first-generation systems as they became outdated. The company donated the system’s mechanical components to research institutions interested in pursuing tele-robotic medical research. (Others got their retired systems from hospitals or online.)

However, an issue arose: universities had only received the mechanical parts, and not the electronics or software that made the systems work. This is where Hopkins came in.

When Hopkins acquired its da Vinci back in 2009, a team of researchers led by Kazanzides and Allison Okamura (now at Stanford University) decided to build electronics and write custom software for the system tailored to their research objectives. Colleagues at other institutions took notice.

“I started getting these emails from people I knew at other universities doing this type of research with medical robotics, saying ‘Hey, we heard you have a working set-up, could we copy it?’” said Kazanzides.

In 2012, the Kazanzides team joined with Worcester Polytechnic Institute, Stanford, and University of British Columbia to better manage their da Vinci related work. The result, according to Kazanzides, is that many research universities around the world now have access to a robotic surgical system that would be extremely expensive to buy and to customize.

Researchers at the University of Washington developed a similar system, complete with mechanics, electronics, and software, called Raven. Raven is similar to the parts of the dVRK that operate on the patient, but does not include master controllers. Seattle-based company Applied Dexterity has sold about 18 of these systems to other research universities and institutions. Between those using the dVRK and those using Raven, there are about 40 institutions around the world using these systems for minimally invasive medical robotics applications, and that community continues to grow.

With the support of the $969,800 National Science Foundation grant, which is administered through the NSF’s National Robotics Initiative, the Kazanzides team plans to expand the consortium each year. (According to Kazanzides, at least eight more institutions have expressed interest in obtaining a dVRK.) The team also plans to improve the technical details of the da Vinci and is devising a way to consolidate and merge the technology in various surgical tele-robotic units, including the Raven and the dVRK.

“Right now, it’s like Raven and dVRK speak different languages. Let’s say one’s French; the other’s English, but we have to come up with a common language. So we have to come up with a common way of talking to them together,” Kazanzides said.

Hopkins researchers have also been working with NASA on ways to use the da Vinci master console for research in satellite servicing. The dVRK’s good human/computer interface means that it is easy to use the da Vinci master console from the ground to control remote robots to refuel satellites in space.

Finally, over the next three years, the team members also plan to work more with simulation, especially for educational purposes.

“Even though there’s a lot of these da Vinci systems out there now, they’re still limited, and you can’t take a class of 40 students and have them all have their own da Vinci. So the thought is you can do a lot of work in simulation and then you can do a lot of de-bugging, testing in a simulated environment, and then maybe have scheduled a few blocks of time on the real hardware to try it out,” Kazanzides said.

He notes that the da Vinci’s master controller is expensive, bulky, and difficult to simulate. The Hopkins team is exploring lower cost alternatives including input devices such space mouses or joysticks.

“If we have these low cost input devices with simulated robots, then we can broaden the outreach to hundreds or thousands of people who are able to work with this type of technology and learn about it,” Kazanzides said.


Johns Hopkins University

Johns Hopkins University, Whiting School of Engineering

3400 North Charles Street, Baltimore, MD 21218-2608

Laboratory for Computational Sensing + Robotics