Unlike conventional open surgery, the laparoscopic approach uses a few very small incisions to access the internal organs. It has several advantages over open surgery: less trauma for the patient, less risk of infection, and a recovery period that's significantly shorter. But it also means the surgeons cannot see what they are doing and so must rely on their instruments to guide them around the body cavity. Because these instruments need to fit through the tiny incisions, this necessarily limits their technical specifications.
"Surgeons currently use optical endoscopes or 2D ultrasound when conducting minimally invasive surgery," explained Stephen Smith, a professor of biomedical engineering at Duke. Interpreting the images produced is not easy, as they lack tactile or depth information, and provide a very limited field of view.
Interpretation could soon become easier, however, thanks to a 3D ultrasound laparoscope developed by Smith's team in 2005 (Ultrasonic Imaging 27 129). "With our scanner, doctors could see the target lesion or a portion of an organ in a real-time 3D scan," said Smith. "They would have the option of viewing the tissue in three perpendicular cross-sectional slices simultaneously or in the same way a camera would see it – except that a camera can't see through blood and tissue."
The device not only lets surgeons see beneath the surfaces of organs in real time; because it produces 3D data it can also reveal many features that would be missed by 2D ultrasound and optical laparoscopes. Now the Duke team has taken the next step by linking it to a surgical robot.
"An additional advantage that real-time 3D laparoscopic ultrasound provides over conventional 2D LUS is the ability to establish a true 3D coordinate system for measurement and guidance," explained the team in the November issue of IEEE Trans. Ultrasound, Ferroelectrics & Frequency Control.
Surgical robots have been infiltrating more and more operating theatres over the last few years. At the moment, though, they are still very much a tool for the surgeon rather than a replacement. To become fully autonomous, they need is an accurate guidance system and some form of artificial intelligence. The latter is some way off, but the Duke team has showed that its 3D LUS system can fulfil the first requirement.
With their latest work, Smith and colleagues reported promising results from tests of a simple robotic biopsy system. The 3D ultrasound laparoscope successfully guided a robotically controlled needle to targets in a tissue-mimicking phantom and a real (but deceased) dog. "Once the robot takes over, it sends the needle to within about 1.5 mm of the centre of the target," said Smith. "That's pretty good accuracy."
A surgeon would still be needed to oversee procedures with a system like this, but even a semi-autonomous robot is well worth having. "It will enable faster, more accurate surgical procedures and open up new surgical niches, benefiting both patients and surgeons," Smith told medicalphysicsweb. He envisages the system initially being used for biopsies and ablations, but eventually a fully autonomous surgical robot could be invaluable for medical care in extreme environments such as space and the battlefield.
It might not be too long before semi-autonomous systems make it into hospital operating theatres. "All the technology is available," said Smith. "We just need to make the connections between the ultrasound scanner and the robots now in use by surgeons. There are no technological barriers to doing that right away."
The next step for the team is to begin trials of the technology on live animals. Smith reckons clinical use is only three years away.