Signs of Hope
| By: Tanya M. Anandan, Contributing Editor
It’s all about control. This is where robots excel. They have an amazing ability to control the trajectory, depth and speed of their movements. And to limit the forces they exert.
In the operating room, robots help guide surgical instruments to precise treatment locations. They can repeat the same movements over and over again without fatigue, or remain completely stationary for long periods of time. Robots go where traditional surgical tools can’t, and perform tasks unimaginable without computer assistance, sophisticated algorithms, and advanced motion control technology. They make the impossible – possible.
Robotic surgery is on the cutting edge with flexible, maneuverable robots and machine learning. The market is a hotbed of development.
According to an Allied Market Research report, the global surgical robotics market was valued at just over $56 million in 2017 and is projected to exceed $98 million by 2024. North America accounts for the largest share of the global market for surgical robotics. Orthopedic surgical robots, like the Mako™ robot from Stryker for hip and knee replacement surgery, could see the greatest potential.
Industry insiders are seeing an uptick in the number of medical device companies with limited technology platforms looking to expand into the robotics realm. Everybody is trying to grab their slice of the pie.
“They might only be providing surgical hand tools today and they are wondering if they should be building a full-on robotic system,” says Corey Ryan, Manager of Medical Robotics – North America for KUKA Robotics Corporation in Shelby Township, Michigan. “They want to be able to stay current and keep their customer base. As the market starts to expand, people want to be in on the next Intuitive.”
Ryan is referring to Intuitive Surgical, maker of the da Vinci® Surgical System, arguably the best known medical robot on the market and certainly a pioneer in robotic surgery. Since FDA approval in 2000, the da Vinci robot has seen thousands of OR installations around the world. The fourth generation, the da Vinci Xi, launched in the U.S. in 2014. Improvements make it easier to use, more versatile, and provide better access to the patient. Take a look.
A contender, the Senhance™ Surgical Robotic System from TransEnterix Surgical, Inc. received FDA clearance in October 2017, making it the first new market entrant to abdominal surgical robotics since 2000. For the first time in a robotic surgical platform, eye-tracking camera control enables surgeons to move the laparoscopic camera simply by moving their eyes. With haptic feedback, they can “feel” forces encountered by the robotic arms.
Healthcare, like many industries, is experiencing a labor shortage as physicians retire or leave the profession for other pursuits. At the same time, the 65-and-older population, the primary users of surgical services, is expected to double by 2060, further straining the healthcare system. Surgical robots like the da Vinci and Senhance systems enable less experienced surgeons to perform at higher levels and help prolong the careers of our aging population of highly experienced surgeons.
“If you have a doctor that only does minimally invasive surgery once every month or two, they’re not as fast or as skilled as somebody who does 4 to 10 treatments a day,” says Ryan. “Robotic technology allows a doctor that has limited experience or is new at a procedure to deliver treatment at a level that they wouldn’t be able to do for potentially years until they have that experience. If you can take a surgeon with limited experience and they can perform like a world-class surgeon, there’s a huge advantage to that.”
We also want our experienced surgeons in the game as long as possible, as long as they can safely perform surgery. Robots help diminish the effects of hand tremors and can constrain unintended or accidental movements.
“One of the big advantages of medical robotics is you can take the tremor out of older hands as well,” says Ryan. “Surgeons have a very advanced ability to control their hands, but over time that does degrade. If you have a robot that can take those tremors out, they can extend the life of surgeons, but also make the system much safer.”
Surgical robotics also increases throughput. If you can reduce the time it takes to treat patients and deliver the same quality of care, then you can treat more patients.
|Robotic hair transplant system harvests and implants thousands of tiny hair follicles for more precise, reliable and faster minimally invasive hair restoration procedures. (Courtesy of KUKA Robotics Corporation)|
Superhuman Motion Control
A robot’s superhuman repeatability and motion control make surgical devices like this hair transplant system a dream application for robots. The ARTAS® iX hair restoration system was developed by Restoration Robotics, Inc. in San Jose, California. The system uses a KUKA LBR Med robot.
In this latest iteration of the hair transplant system, the robot not only harvests hair follicles from the back of a patient’s head, but it also implants the newly harvested follicular units into targeted areas of the scalp. In previous iterations of the system, the robot only handled harvesting. Then a technician would need to manually insert the harvested follicles. Now, integrated force sensing ensures the robot maintains desired force when making contact with the patient during harvesting and implantation.
See the ARTAS Robotic System in an actual procedure (warning, some graphic content) as CEO Ryan Rhodes and COO Gabe Zingaretti describe the follicular unit extraction (FUE) method for hair transplantation and the role of the LBR Med in the advanced process.
The LBR Med robot is an adapted version of the LBR iiwa used in industrial applications. Just like the industrial version, the LBR Med is lightweight and safe for humans to work around right out of the box. Integrated force torque sensors in all of its seven joints provide built-in sensitivity and collision avoidance. A fully redundant safety system, including redundant software and hardware, meet the RIA safety standards for human-robot collaborative systems.
The main difference from the industrial version is that the LBR Med robot is certified to meet the medical device standard. This means all of the mounting areas are hidden, so you don’t have exposed mounting bolts where fluids could accumulate and potentially grow bacteria. All of the cables run under the base, so it can be sealed. All of the electrical connections run inside the flange, so there’s no exposed wiring on the outside. The device meets requirements for overload testing. Plus the power ports inside the Med have to meet leakage current requirements.
Ryan explains that every electrical device leaks current onto the ground line. Typically, it’s very minimal. But even minimal leakage current, if it goes onto the ground line and through a patient, could be deadly. The controller and robot must be designed to ensure that you’re not accidently driving current onto the ground line.
The LBR Med is undergoing development with several OEMs trying to bring robotic medical products to market. FDA clearance, especially for robotic surgical devices where actual excision occurs, can take years to come to fruition. A robot like the LBR Med that is already certified to the medical device standard and designed for the human-robot collaborative workspace can make the development process more efficient.
A major area of development for KUKA’s medical robotics team is trocar kinematics.
“When you’re working inside the human body, you are working through a port (a small incision through the layers of tissue). So that port becomes this pivot point,” explains Ryan. “You have to maneuver all the surgical instruments about this pivot point. We’re working on software to handle robot motion directly out of the box by allowing you to set a trocar pivot point. That’s the next big thing we see coming. Customers are looking for that kind of functionality.”
This functionality is already available for hand guiding the LBR Med robot. You can see it in action at 46 seconds into this video demo. Now the robot manufacturer is working on making it available for automated motion.
|Robotic intraoperative radiotherapy device may someday provide focused cancer treatment during surgical procedures. (Courtesy of Sensus Healthcare, Inc.)|
Ryan says there is a lot of research taking place in robotics for biopsy procedures, medical screw insertion, and even suturing. Surgeons at Children’s National Health System in Washington DC demonstrated the ability to robotically suture live animal tissues with more precision than a human hand. Check out STAR, the suturing robot. The technology spinoff, Activ Surgical, Inc., formerly Omniboros, is further developing surgical technology using robotics and machine learning.
The Sculptura™ system made by Sensus Healthcare in Boca Raton, Florida, uses the LBR Med for intraoperative radiotherapy (IORT). This kind of radiation therapy occurs during surgical removal of a cancerous tumor. By performing the procedure during surgery, this delivers a high dose of radiation precisely to the targeted area with minimal exposure to surrounding healthy tissues. IORT also avoids the usual delay between the time the tumor is surgically removed and when conventional external beam radiotherapy is typically administered.
While the Sensus works-in-progress product is still awaiting FDA clearance, applications like IORT demonstrate another advantage of robotics. These types of systems reduce the risk of human exposure to radiation, both for the patient and the healthcare professional.
Only Possible with Robots
Robotic surgery also delivers a level of treatment that is nearly impossible without robots. Take the CARLO® system developed for craniomaxillofacial surgery by Swiss-based AOT AG. This robot-guided laser bone cutting device uses KUKA’s LBR Med to guide the laser beam to the precise ablation locations.
CARLO (for Cold Ablation Robot-guided Laser Osteotome) does not yet have FDA clearance. You can see CARLO in action at 3:40 into this video.
The robotic laser osteotomy system cuts the jaw bone in a three-dimensional, interlocking tongue-and-groove pattern. When taken apart and reassembled, the pieces fit together like a puzzle and hold the jaw in place without a lot of hardware.
“No surgeon could cut that in three dimensions,” says Ryan. “Robots enable treatments that are impossible without them.”
He says this is the same with pedicle screw insertion for spinal surgery.
“It’s very hard to line up a screw based on looking at a CT model, and try to plan to place it at 23 degrees off center and 35 degrees from the angle of the bone. Whereas a robot is a 3D motion control and positioning system. I don’t think any first-generation system will actually insert the screw. It will just align them and the surgeon will do the insertion. But the amount of time it will save is enormous. Nobody has released a product yet, but researchers have proven that you can do it. There will be robotic pedicle systems, no doubt. The math is just too compelling.”
The ExcelsiusGPS robotic guidance and navigation system from Globus Medical Inc. does the math and assists with spine surgery.
Machine Learning for OR Automation
For some, the math is very compelling. Researchers at University of California San Diego are developing algorithms to help guide and eventually automate surgery. Michael Yip, a professor of electrical and computer engineering at UC San Diego and director of the university’s Advanced Robotics and Controls Lab (ARCLab), is spearheading the effort.
We first met Yip when we covered rehabilitation robotics. His lab is conducting fascinating research into adaptive orthotics for vets and artificial muscles for robots.
With his work in artificial intelligence for surgical automation, the idea is to use techniques from AI and machine learning to help improve surgical robots’ control accuracy to submillimeter precision and expand their dexterity so they can reach more areas of the body.
“We want to make it easier for surgeons to perform their job, or to create tools or software to allow surgeons to do new procedures,” says Yip. “Automation isn’t here to take away jobs. We look at those tasks that can improve the throughput, consistencies, and speed at which they are done, and to do procedures that you couldn’t do without a robot.”
Yip envisions an OR where robots perform the mundane secondary tasks such as suction, irrigation and tissue retraction, which are typically done by an assisting surgeon resident or a second surgeon. He says this underutilizes the surgeon’s expertise. They could be treating other people. That’s why they want to automate.
|Researchers study advanced robotic surgical devices and machine learning to improve precision and dexterity, and automate routine surgical tasks so surgeons can focus on more complex procedures and treat more patients. (Courtesy of University of California San Diego)|
ARCLab researchers work with their own customized da Vinci Surgical System, for which they develop hardware and software to explore levels of shared autonomy. The system has four robotic arms. One arm holds the endoscope with a camera on the end. The other three arms can hold surgical instruments, but a surgeon can only control two arms at a time. Yip says the extra arm could be used for these secondary tasks.
Yip wants to impart the robotic arms with intelligence so they can follow the surgeon and do the secondary tasks with more autonomy. He provides an example.
“With tissue retraction, the instrument may be peeling back tissue a layer at a time as the surgeon cuts. Right now the surgeon has to stop dissection and grab the third arm, pull back the tissue a bit, and then switch back to the primary arm. That can be tedious and keep somebody in the OR longer than they need to be. These secondary or assistive tasks are low risk but high reward for robotic automation.”
UC San Diego is also a training ground for new surgical residents. The Center for the Future of Surgery offers training on the da Vinci system, surgical simulators, and laparoscopic surgery. Yip’s lab works closely with the residents practicing robotic surgery. They make sure his lab is addressing real problems in the surgical suite.
He thinks AI and machine learning is the right approach. “We want to use data to describe how to automate things rather than have an engineer handcraft behavior that might or might not be applicable in all scenarios. We’re interested in getting the data from watching surgeons perform. (This is where the surgical resident training center comes in handy. It’s an environment ripe for data collection.) With only a few examples, the robot has to be able to learn a new skill or task. This is called imitation learning, or learning from demonstration. The challenge is to learn using as little data as possible, but be able to generalize beyond what it just learned.”
Yip’s lab is working on a new project with the U.S. Army where they are trying to automate lifesaving procedures. These would be vital procedures such as performing a tracheotomy before someone suffocates or stopping hemorrhaging within minutes to prevent fatal blood loss. If military personnel were in a remote location, perhaps the desert or aboard a ship, and such an emergency arose, a small portable robot would automatically perform lifesaving procedures until the medics arrive. The goal is to develop a proof of concept by the end of the four-year project.
Flexible Robotic Catheters
ARCLab researchers are also responding to a big push toward flexible robotic catheters for surgery. It’s actually a need on the hardware side that is driving their focus on the software side.
“We’re building our own robot systems in my lab, partly because they might fill missing needs in the current landscape of available devices, but also as a way to explore machine learning and AI as a way to control these devices,” says Yip. “Surgeons currently use these flexible catheters and endoscopes. As they become thinner and more flexible, your ability to control them drastically decreases to where it’s almost impossible to manually control these devices. That’s where robotics comes in. It can figure out these complicated control mappings and recover that intuitiveness and controllability.”
The lab’s machine learning approach for controlling flexible robotic catheters has already been patented and licensed.
“That says something about the importance of AI and machine learning in the next class of robotic instruments,” says Yip. “These robotic catheters are really where all the device manufacturers are pushing.”
He cites a new flexible robot developed by Intuitive Surgical that is awaiting FDA clearance. According to the manufacturer’s new website, the Ion ™ robotic platform is indicated for minimally invasive peripheral lung biopsy. The system features a flexible, maneuverable 3.5 mm diameter catheter that can pass through small and difficult-to-reach airways deep inside the peripheral lung to perform transbronchial needle biopsies. This video preview teases the product release.
|Supporting the future of minimally invasive surgery, researchers are developing millimeter-sized, flexible robotic catheters to access the far reaches of the human body. (Courtesy of University of California San Diego)|
Yip says that’s the future, these flexible endoluminal devices. His lab is working on their own robotic catheters.
“Our robotic catheters are 1 to 3 mm diameter, which is arguably the smallest you would find anywhere,” says Yip. “This is still research so you won’t see anything that small in production – yet. But that’s the future. Smaller and smaller, so you as a patient will have less or no observable trauma from surgery.”
These robotic catheters could have many uses. In the U.S. Army project noted previously, they could act as balloon catheters to stop bleeding. Other uses include heart catheters for treating arrhythmias; transurethral approaches to break up kidney stones; transoral approaches for throat or lung cancer screening and biopsies; transoral gastrointestinal procedures to treat conditions like acid reflux; and for colonoscopy screenings to excise polyps on the spot rather than performing a second procedure at a later date.
“In general, everything is going more towards minimally invasive, flexible, endoscopic catheter-like devices,” says Yip. “It’s to the point where they are so hard to manually control by clinicians, you need the robots there to help out. The question then is how can the robots be programmed to know how best to control these instruments. That’s where I think machine learning has a huge advantage in terms of using real data rather than hand engineering these controllers.”
Yip’s catheter is about a meter and a half long. In fact, in this video of the ARCLab’s flexible robotic catheter, you can see Yip holding the undulating tip and the robot motor is so far away from the end of the catheter that it appears blurry in the background.
These snake-like robots are going where others can’t. Flexible surgical robots are advancing minimally invasive surgery with scar-free procedures.
Steering Toward Scar-Free
Flexible, snake-like robots are already in the OR and, surprisingly, on the corporate espionage hit list.
The Flex® Robotic System advances transoral and transanal surgery to perform procedures that are difficult or impossible to visualize and access with conventional straight, line-of-sight approaches. With the aid of an integrated 3D high-definition camera, the flexible robotic scope enables physicians to see and traverse anatomical curves and maneuver around complex internal structures en route to the target surgical site. Access is through natural orifices in the body, such as the mouth and anus, without the need for external incisions. The system is developed and manufactured by Medrobotics Corporation, a privately held medical device company located 35 miles south of downtown Boston in Raynham, Massachusetts.
|Robotic surgical system uses a joystick-like controller, advanced 3D vision for navigation, and a flexible, steerable scope for minimally invasive transoral and transanal surgery. (Courtesy of Medrobotics Corporation)|
This is a robot-assisted platform, meaning a physician is always at the helm. Using a joystick-like controller and watching the robot’s progress on a display screen, the physician steers the flexible robotic scope to the target site. Then flexible surgical instruments are inserted into guide tubes that run along the length of the scope. Once triangulated at the treatment site, the surgical instruments can be used for grasping, extracting or cauterizing polyps and lesions, and also suturing.
Watch the Flex Robotic System in the OR demonstrating its use for ENT procedures.
The flexible, steerable robotic scope allows for deeper access to hard-to-reach areas of the anatomy such as the vocal cords in transoral procedures and the base of the Sigmoid colon in colorectal procedures.
“In the past, these types of procedures had to be performed with more invasive procedures which required incisions through the skin,” says Medrobotics CEO Samuel Straface. “Those could either be laparoscopic or open surgery. We are enabling these scar-free procedures to be performed by using natural orifice access to the anatomy.”
With natural orifice procedures using conventional straight-line tools, access is often limited.
“With our system, it allows you to get right over a lesion in an angled approach that would not be possible with a straight instrument through the natural orifice,” says Straface. “In some cases it allows the procedures to be performed significantly less invasively and in other procedures it can facilitate a potentially easier or faster procedure for the surgeon. There is a benefit for the patient as well if it’s a shorter procedure.” (Patients are usually under general anesthesia.)
The Medrobotics system marries the best of two worlds, the flexibility of endoscopy and the stability of laparoscopy in a shapeable and steerable robotic scope.
“Unlike a flexible endoscope which remains floppy once you get it to where you want it, ours becomes rigid, so you have a stable surgical platform from which to operate,” says Straface. “You can use two instruments at a time, whereas with an endoscope, they usually only have one instrument. By having precise control of two instruments, you can perform common surgical procedures that require traction on the tissue and two instruments working in concert.”
|Flexible and steerable robotic scope equipped with a 3D high-definition camera and specially designed flexible surgical instruments helps surgeons visualize and treat hard-to-reach areas of the body. (Courtesy of Medrobotics Corporation)|
The core of the snake-like robotic system is its interconnected linkage design that takes advantage of a proprietary method for managing friction, allowing it to be flexible to conform to anatomical curves but rigid when needed. Outer and inner mechanisms move in tandem as the robotic scope advances into the body.
“Our system is really a set of links that have cables running through them,” explains Straface. “The links have very specific shapes, such that when you pull on the cables to tighten them together, the friction from one link to another forms a rigid system in whatever shape you have the links in at that time. Then we have this inner spine that advances through it. That adds more structure, support and rigidity.”
The roboticized endoscope was born in the Biorobotics Lab at Carnegie Mellon University, where the lab’s director, Howie Choset, develops various kinds of snake robots for search and rescue applications, manufacturing and nuclear inspection, among other applications. He cofounded Medrobotics in 2005. Choset and his cofounders created the proof of concept, which the company developed into the Flex system and commercialized in 2014. Having received CE mark certification, it was initially marketed in Europe. The system received FDA clearance in 2015 and was released in the U.S. Straface says the robotic system is also cleared in Australia and Singapore.
Just this past September, physicians at NYU Langone Health successfully performed surgery on a patient using the Flex robotic system to remove a large pre-malignant colorectal lesion without a traditional incision. The procedure marks the first time a robot was used in gastrointestinal endoscopy. In the future, the robotic system may be used for transvaginal gynecologic procedures and minimally invasive abdominal surgery.
Robotic minimally invasive procedures often mean less blood loss, shorter recovery times, and less trauma to surrounding tissues. Robots provide unprecedented control, access and flexibility to surgical procedures. The impossible is now possible. Someday, robots could automate routine surgical tasks so surgeons can concentrate on more complexity and more patients.