Photo: David Clugston

THE DOCTOR IS IN: Developed at the University of Washington’s BioRobotics Lab, a remote-controlled two-armed surgical robot “operates” on a plastic and rubber anatomical model of a human torso.

On a hot morning this past June, our research group at the University of Washington, in Seattle, crammed into two cargo vans and drove 2000 kilometers south to the rangeland north of Simi Valley, in southwestern California. In the back of one of the vans was our latest creation: a prototype surgical robot we’d been developing for the past four years. Our mission was to field-test the robot—by operating not on a person but rather on latex objects mimicking human organs, with a surgeon commanding the robot from a control console 100 meters away.

At the test site, we met the rest of our team—surgeons, aerospace engineers, networking experts—and set up a base camp on a flat expanse circled by undulating hills. Under a blistering sun, some of us assembled the robot, a portable surgery table, and a video camera under a canopy tent, while another group installed the surgical control console and a video monitor in a second tent. With all systems ready, we waited for the communications channel to be set up to link the two locations.

Remote surgery technologies like the ones we were testing have long interested military groups all over the world. On the battlefield, medical response time often determines who lives and who dies. A recent study of combat casualties found that in nearly two-thirds of fatal battlefield injuries, death comes within 30 minutes. There’s precious little time to perform even simple life-saving procedures, such as controlling bleeding.

Surgical robots offer a tantalizing possibility [see "Submerging Surgery "]. They would allow military doctors, stationed safely distant from the front line, to perform operations without once putting their hands on patients. Medical vehicles equipped with such remote-controlled robots could get surgical care to soldiers in a lot less time than it would take to evacuate them to the nearest base or hospital.

For that vision to become reality, however, surgical robots need plenty of improvement. One challenge is designing systems that can work under conditions very different from those of pristine operating rooms. Indeed, what was new about our trial in California was the fact that the whole setup was compact, rugged, and easy to transport. Local conditions for the experiment, organized by Gerald Moses and Timothy Broderick of the U.S. Army’s Telemedicine and Advanced Technology Research Center, in Frederick, Md., included sandy gusts of wind, high humidity, and an ambient temperature crawling toward 40 °C. Gasoline-fueled generators were all we had to power our electronics.

Our system is one of a new generation of surgical robots that may one day bring advanced medical care not only to soldiers but also to people in remote locations lacking specialized physicians. Disaster relief, too, could be revolutionized. Last year’s Hurricane Katrina in the United States, the October 2005 earthquake in northern Pakistan, and the December 2004 Indian Ocean tsunami overwhelmed the medical resources available in those places. The hope is that in the not-so-distant future, fleets of ambulances or helicopters equipped with surgical robots would rush to such areas, allowing faraway doctors to save lives and limbs.

Indeed, some of the military projects seem to be quite easily convertible to civilian use. Our group is participating in another U.S. military–funded initiative, the US $12 million Trauma Pod program, launched last year by the Defense Advanced Research Projects Agency (DARPA). Managed by SRI International, in Menlo Park, Calif., the program aims to develop an unmanned, mobile operating room that is equipped with a host of automated surgical systems and could be quickly dispatched anywhere in a war zone.

But giving surgical robots the necessary mobility requires meeting another challenge: the reliable transmission of the surgeon’s commands to a system often roaming in far-flung places. Surgeons have remotely commanded surgical robots before, even with real patients under the robots’ scalpels. But those setups took place in well-equipped hospitals and relied on dedicated, wired communications channels. How, then, to break free from wires?

Geosynchronous satellites have good data bandwidth, but they don’t cover all regions of the world and their delayed transmissions make surgery difficult. With our experiment in California [see illustration, “Far-Flung Fingers”], we wanted to test an alternative that could be heaved into the air—literally—in a matter of seconds: a wireless link enabled by an unmanned aircraft. Launched from the ground, a small drone was sent to fly in lazy circles above us. Video from the camera near the robot was compressed by special hardware into MPEG format and beamed to the plane, which relayed the feed to the monitor below. At the same time, motion commands from the surgeon’s console were bounced through the plane to the robot, which responded only a fraction of a second later, performing such tasks as tying suture knots.

Photo: David Clugston

Sewing IT up: The University of Washington’s surgical robot ties suture knots, one of the most complex tasks in minimally invasive surgery.

Surgery has come a long way from the prehistoric days when operations were done with sharpened flints and bones or when the barber-surgeons of the Middle Ages in Europe did ­haircuts as well as surgery. Modern surgery has benefited from many technological advances, notably the relatively recent invention of minimally invasive instrumentation. These tools include laparoscopic devices, which are the size of chopsticks, with tiny surgical tools at their tips. They enter the body through small incisions—0.25 to 1 centimeter, compared with several centi­meters in conventional open surgery—and surgeons maneuver them guided by images from tiny camera probes.

Nevertheless, surgery today remains fundamentally the same as it was centuries ago in its basic aspects: it relies heavily on the experience of the surgeons and the dexterity of their hands.

The idea of robot surgeons may conjure visions of C-3PO–like androids clad in scrubs excising appendixes, but existing systems and those being developed by our group and others are actually more like robotized laparoscopic instruments. They function as a surgeon’s miniaturized and electromechanically enhanced hands, maneuvering tools with greater dexterity, accuracy, and stability than humans could ever achieve. The tools, positioned by high-precision motors, can reach spaces, such as those around the prostate gland or in the tiny bodies of infants, that surgeons often can’t reach using their hands.

Indeed, robotically assisted surgery is already a bona fide, if small, category of robotics. In its “World Robotics 2005” report, the International Federation of Robotics estimated the number of robots used in surgery and therapy—including electromechanical arms to position instruments, robotic bone drills, and computer-controlled radiotherapy equipment—at 2800 units worldwide. These units include the da Vinci surgical system, by Intuitive Surgical, of Sunnyvale, Calif., currently the only commercially available surgical robot approved by the U.S. Food and Drug Administration. It has two or three arms equipped with surgical tools and an extra arm with a stereoscopic video camera probe. A surgeon controls the machine from a console located in the same room as the patient. Currently, some 400 da Vinci units are installed worldwide, a new unit costing about $1.5 million.

For at least one procedure, the da Vinci system is beginning to show advantages over conventional surgical techniques: the laparoscopic radical prostatectomy, in which the prostate gland is removed through tiny incisions in the lower abdomen. Surgeons have reported that patients undergoing the robotic procedure had less blood loss, fewer complications, and shorter hospital stays. Some experts say that cardiac and gynecological procedures could be next, but further studies have yet to show the benefits of robotic surgery—considering its higher costs—over conventional techniques.

And despite its achievements, the da Vinci system wasn’t engineered to stand up to the extremes of work in the field. It weighs nearly half a ton, and controlling it remotely requires substantial electronic modifications. Moreover, the system is designed specifically for minimally invasive surgery, precluding significant use for trauma procedures.