‘The potential to improve the field of medicine is an incredible motivation for me.’ – Jessica Burgner-Kahrs

Computer scientist Jessica Burgner-Kahrs talks to us about continuum robots, revolutionary brain surgeries, and an afternoon at the zoo

Interview: Dirk Liesemer

JAM: The robots that you’re studying look rather innocuous at first glance. In any case, they don’t resemble humans or the robots that we’re familiar with from the automobile industry.

Jessica Burgner-Kahrs: That’s true, many people expect to see that kind of similarity. When it comes to our continuum robots, the similarities are rather indirect. To put it simply, they consist of two parts: a power unit and a type of tube. The movements of the tube call to mind the tentacles of an octopus, or, when the tube grasps an object, it’s a lot like the trunk of an elephant. But nonetheless, what we have here is a robot. According to the classic definition of the term, a robot is a steerable machine that can carry out positionings. There’s even a DIN standard for robots. The term refers to programmable machines in general, so this category contains a broad range of machines, even including the household vacuum cleaner.

JAM: What sets continuum robots apart from classic robots?

Jessica Burgner-Kahrs: Continuum robots don’t have joints or rigid couplings; instead their design tends to be inspired by nature and has elements that are reminiscent of an earthworm or an elephant trunk, as I mentioned earlier. The smallest continuum robot we’re studying has a diameter of just one millimetre. We want to find out how unarticulated mechanisms can be implemented, controlled, and mathematically described in engineering. In the end, we need to be able to steer and regulate the robots we create. That’s why we’re interested in seeing how humans and machines can interact with each other. In other words, how can we steer a robot that doesn’t have a shoulder joint, elbows, or a wrist joint?

JAM: Your robots are intended to be deployed during surgery; for example, to remove brain tumours. How will that work?

Jessica Burgner-Kahrs: Our robots need to be able to reach areas that have so far been inaccessible or difficult to access, including in the human body. That will fundamentally revolutionise surgery by making it much less invasive than it is nowadays. At the moment, if a tumour is located at the base of the skull, it’s necessary to cut a very wide opening into the head. Our robots will only need a small opening, such as the mouth, the nose, or a tiny hole in the skull. A surgeon will be able to insert the continuum robot through one of these openings and then steer it all the way to the tumour. We can already imagine how much the work, skills, and procedures in the medical field will change as a result. Right now, rigid instruments such as scalpels, scissors and grippers are still being used in surgeries in a very straight, direct way. Our instruments will make it possible to perform operations that no one today has even thought of yet.

JAM: How does one steer this type of robot during an operation?

Jessica Burgner-Kahrs: That will naturally depend on the type of operation. Tumours, for example, could be removed via laser ablation. For this type of procedure, the patient would be placed into a magnetic imaging tomograph. With the help of these images, it will be possible to monitor where in the body the robot is located and to confirm its movements. In the field of abdominal and urological surgery, for example during a partial kidney resection, doctors would even be able to obtain a direct real-time image, because cameras will be integrated into the front of each robot. Then, even more detail could be obtained by layering ultrasound images over the video image. For these types of procedures, the robot’s diameter would not be limited to a millimetre, in fact, it could even have a thickness of up to several millimetres.

JAM: By now our readers will certainly be impatient to know: how far along are you with your research?

Jessica Burgner-Kahrs: At the moment we’re still working with prototypes. Before we can put the robots to use during an operation, we need to answer a number of fundamental questions. How can we mathematically calculate how the tiny, elastic robots are going to behave? I’ll give you a visual example: if I’m pulling a garden hose along behind me, obviously I don’t want it to get caught on every obstacle in the way, rather ideally to wind itself around the obstacles elastically. I definitely don’t want it to harm either the obstacles or itself. A continuum robot needs to have a comparable degree of elasticity when it’s in the human body. But to ensure that, I need to have a sensor that can register and measure the form and size of the robot and, if necessary, the obstacle. And when you have a tiny robot, the sensors also need to be tiny. The smaller everything gets, the greater the metrological challenges become. On top of that, the mathematical modelling needed to predict the robot’s behaviour is complicated and requires designing complex algorithms to plan movements. When you tell a robot to do a certain movement, then that movement needs to be carried out exactly as you have described, and the Bowden cables or the tubules inside the robot need to move against each other. All of these challenges need to be solved before a robot can be safely deployed in an operation. That’s why at this point we are only working on pieces of dead animals or on very naturalistic replicas, such as silicon brains.

JAM: In the laboratory, you’re also working together with surgeons. How do they react to the robots?

Jessica Burgner-Kahrs: Most of them are completely delighted when they see our robots and have the opportunity to steer them. Often we’ll get a whole flood of ideas from them about the kinds of ways that they could use the robots. This collaboration is fruitful for us because we’re suddenly made aware of demands on the robots that we hadn’t thought about ourselves. A robot needs to be more than just relatively flexible. A soon as it has reached its goal, it needs to be able to apply a lot of power to another task, such as sewing or cutting. So on the way to its goal, the robot needs to be able to move very elastically, but once at its destination, it needs to be able to make itself very rigid in order to provide the high amount of power required. We’re working with surgeons, but also with researchers in diverse fields such as computer science, electrical engineering, materials science, and biology. We receive new impulses from all around us. Research is not limited to the traditional structures anymore. I think that without these many perspectives on the same topic, continuum robotics wouldn’t exist as a research area. And this interdisciplinarity advances the revolution. Continuum robots can reach places in unconventional ways and make it possible for surgeons to obtain access that could not previously even have been thought of due to the limitations imposed by existing instruments. We’re constantly expanding the field of robotics.

JAM: In technical terms, how much longer will it take until one of your robots can be used in an operation?

Jessica Burgner-Kahrs: That should be possible within the next five to ten years. But we’re only working on prototypes. Before our research can help patients, continuum robots first need to be developed as products. Someone needs to take responsibility for that. Perhaps we as a team or someone from our group will create a start-up. It’s hard to predict when the first operation on a human being with the help of a continuum robot will take place. I hope to still experience it during my time as a researcher.

JAM: Will the robots also be used in other areas in the future?

Jessica Burgner-Kahrs: Continuum robots are always handy when a certain destination can’t be reached via a straight path, but rather one with curves or several angles. In addition to medical operations, the large field of technical inspections could benefit from these robots, such as checking airplane turbines, which have numerous compressor stages and turbine blades. And it’s likely that our robots will one day serve to provide humans with skills that they have lost, for example as arm prostheses with trunk-like structures for gripping diverse things. But also as exoskeletons: someone who has suffered a heart attack could use such an outer skeleton to learn how to lift his or her arm again. Or a moveable exoskeleton could enable workers to transport heavy objects from A to B.

JAM: It sounds like nature inspires your research.

Jessica Burgner-Kahrs: That is definitely the case. Last year my team and I went to the zoo. We wanted to see how unarticulated, flexible structures are used in nature. That’s why we observed snakes and worms, and flamingo necks, which contain more vertebrae than the human neck. And anteaters, who have the longest tongues in the animal kingdom. We asked ourselves: what are the characteristics of these elastic structures? What kinds of tasks can they carry out? Everyone felt very inspired. But recreating an exact replica of an elephant trunk is far beyond the scope of our mechanical ability, because the trunk consists of forty thousand individual muscle fibres, each of which would need to be individually steerable. An elephant can use its trunk to delicately grip a coin or lift a heavy load, and all the animal does is change the rigidity of individual groups of muscle fibres. We are attempting to get closer to this type of behaviour by building robots that can switch between being flexible and being rigid depending on the task.

JAM: What is the state of continuum robotics today? Still at the very beginning?

Jessica Burgner-Kahrs: The field is still new, although the very first robot of this kind was developed in the US in the 1960s. But for a long time scientists couldn’t get anywhere close to describing it mathematically. International research on that only began about 15 years ago. During my doctoral work, I attended a lecture by a pioneer in the field about a tiny continuum robot. I then went to work with that expert during my post-doc phase. It was incredibly exciting to learn how this completely different kind of robotics functions. Then with a reintegration scholarship I returned to Germany and received a five-year grant to conduct fundamental research in this field. That’s how I was able to set up the very first emerging scholars group on robotics within the DFG. My professorship is the first one in Germany to approach the topic holistically. I now work with six research assistants, an engineer, and a number of student researchers from different fields.

JAM: How did you come to choose the field of robotics?

Jessica Burgner-Kahrs: As a child I wanted to be a surgeon. But doctors bear an enormous responsibility for life and death. While taking a computer science class at school, I realised that I had a talent for mathematical thinking. And while studying computer science in Karlsruhe, I discovered that it was possible to use computer science and robotics to support the medical field. The fact that my work can be used to improve treatments for patients greatly motivates me.

The Computer Scientist Jessica Burgner-Kahrs joined the Junge Akademie in 2016. She is Head of the Laboratory for Continuum Robotics at Leibniz Universität Hannover.