Algorithms, robotics: healing potential

Robotic tentacles are on Ron Alterovitz’s mind. The UNC computer scientist and his research group have been investigating new algorithms to enable a robotic device to go deep inside the human body to precisely treat abnormal growths, tumors and other ailments.

The snake-like set of concentric tubes is the focus of Alterovitz’s quest to send the device to targets throughout the body. Deployed through a bronchoscope, it would allow physicians to reach farther than ever into the lungs to diagnose abnormal growths called nodules. Deployed via the nose, it could surgically access tumors in the brain or nearby structures in the head.

The interdisciplinary, NIH-funded surgical device has the potential to move throughout the body so precisely that it can avoid anatomical obstacles and reach its target within a millimeter.

Collaboration, not eureka moment

The origin of the device’s promise rests not in medicine’s rare eureka moment, but in the kind of gesture that many people experience – an introductory email or a serendipitous meeting. For Alterovitz, assistant professor of computer science and head of the Computational Robotics Research Group in UNC’s Department of Computer Science, it began with an email he sent shortly after joining UNC in 2009.

Photo fo Ron Alterovitz.

Ron Alterovitz, assistant professor of computer science and head of UNC’s Computational Robotics Research Group in the Department of Computer Science. (Photo by Donn Martin, Endeavors)

The message went to School of Medicine faculty member Richard Feins, professor of surgery in the division of cardiothoracic surgery. Feins, an early user of the superDimension lung navigation system, was immediately interested in hearing what Alterovitz had to say.

“As surgeons, most of what we do in terms of treatment involves getting where we need to go, so when Ron came to me with the concept of a steerable catheter that could get anywhere he preprogrammed it to go, it was exciting,” says Dr. Feins.

Surgical precision is also an interest of Brent Senior, Nathaniel and Sheila Harris Distinguished Professor of Otolaryngology. But Senior’s work with Alterovitz didn’t start with an email; it began at church. Senior was mentoring a student of Alterovitz’s at church when the two got to talking about their work at UNC. It soon became clear to Senior that he should meet professor Alterovitz.

“I was excited to hear what he was doing because it really is exactly what we need to be thinking about for the future of what we do in skull-base surgery,” says Dr. Senior.

The airway as a tree

“Think about the airway as a tree,” says Dr. Feins. “The limbs get smaller and smaller as you fan out. You’d like to climb out to grab an apple off one of the branches, but you can only go so far before you have to stop because the limbs are too small and will break. That’s sort of what we’re talking about with reaching peripheral lung lesions.”

Diagnosing peripheral-zone lung cancer is difficult. The nodules can be reached with a CT scan and biopsied, but doing so runs the risk of collapsing the lung. Furthermore, even when the lesion is reached, the amount of tissue that can be gathered may be limited, and therefore the sample is potentially inaccurate.

Image of probe entering lung. Snake-like robotic needles, attached to and deployed by the bronchoscope, may provide more accurate diagnoses. As curvilinear devices that can be programmed, the needles can snake their way through the parenchyma of the lung to access nodules in the peripheral zone, striking their target with precision.

Tentacle-like robots have the potential to extend the range of bronchoscopes, enabling safe biopsy of tissue in the peripheral region of the lung for early-stage lung cancer diagnosis without risk of lung collapse. (Image courtesy of Ron Alterovitz)

Snake-like robotic needles, attached to and deployed by the bronchoscope, may provide more accurate diagnoses. As curvilinear devices that can be programmed, the needles can snake their way through the parenchyma of the lung to access nodules in the peripheral zone, striking their target with precision.

“If the nodule identified in the CT scan is on the peripheral zone, and the bronchial tubes are so small that you can’t use existing devices like the superDimension, then that’s where these robots can work,” says Alterovitz.

Early-stage diagnosis through biopsies of peripheral nodules has the potential to save lives. And in the immediate term, that’s the focus of Alterovitz and Dr. Feins. The long-term goal, however, is to actually treat the tumors. Dr. Feins uses the analogy of the early days of cardiology.

“Cardiology was primarily a diagnostic specialty,” says Dr. Feins. “They did angiograms and saw what was wrong with the patient, and then they’d have to send the patient to the surgeon. Eventually they found therapeutic options. They could dilate the arteries or put stents in the arteries. Those therapies changed the whole dynamic. I think it’s possible that if we can get the technology to precisely where we need it to go, not only can we make a diagnosis, but we can add therapeutics like localized radiation, localized chemotherapy, or even localized freezing or radio-frequency ablation.”

Read more from the article by Zach Read, UNC Health Care and School of Medicine.

Published Feb. 24, 2014.