Smooth operator – the making of i-Snake®

People’s dedication to their work comes out in all kinds of ways. Today, there are six of us and a camera crammed into the mechatronics lab at the Hamlyn Centre, Imperial College London. As we chat about how some of the machines are used in the jewellery trade, engineer and Co-Director of the Centre Professor Guang-Zhong Yang slips off his wedding ring and passes it around.

It seems Professor Yang’s passion for engineering extends even to the symbolism of love, as he explains that the ornate ring has been made with one piece in three dimensions using a technique called electric discharge machining. I note this down on a growing list of things I’m learning this afternoon.

Our visit here – to a department that develops technology to minimise the physical and psychological impact of surgery – began in much the same vein, with my colleague and me struck dumb by the lab’s rapid prototyper. Rapid is right: in front of our gaping eyes, it produced a fully formed mechanical component, three-dimensionally printed in layers just 16 microns deep. “You could even scan your head and get a model of that printed,” one of the project’s engineers tells us.

In the next office, a surgeon talks to us as he assembles an endoscope from a large case. In an engineering lab across campus sits a large plastic box with small holes in the lid, used to train surgeons in keyhole surgery. Inside, garish latex organs wobble ominously. It’s becoming clear that this is no ordinary engineering project.

The project – funded by a Strategic Translation Award made by the Wellcome Trust’s Technology Transfer division – aims to transform keyhole surgery. Such ‘minimally invasive’ techniques have already improved the care of many patients. Instead of having to cut open the body to expose the organs, surgeons can slide cameras and surgical instruments into the body, through tiny incisions, on the end of flexible rods.

Meet members of the team and find out more about making i-Snake® in this short film:

Running time: 3 min 24 s
Read the transcript [PDF 84KB]

Led by Guang-Zhong and his colleague, surgeon (and former health minister) Professor Lord Ara Darzi, the team of engineers and surgeons are developing a robotic snake-like device called i-Snake®. It will be self-propelled, able to move through the body to its target, where the surgeon can operate it by remote control. “This project is about advancing the current generation of keyhole surgery into something that’s more accessible and more flexible in the long term,” says Guang-Zhong.

Robots in surgery
Using robots in the operating theatre is not new. “At St Mary’s Hospital we were the first group in the UK to use robots to help us do some surgical procedures,” says Ara. “The robotically assisted platform was itself quite a big step in allowing us to do surgery much more precisely, even more than through a keyhole or open approach.”

St Mary’s Hospital, part of Imperial College, acquired in 2001 a da Vinci surgical robot, made by Intuitive Surgical. Used originally in cardiac surgery, the robot was adopted by surgeons removing the prostate gland – a procedure that carries the risk of leaving the patient incontinent and impotent, with huge implications for their future quality of life.

“There are many proposed benefits of the da Vinci, but there are costs too,” explains Mr James Clark, a surgeon and PhD student on the i-Snake® team. “In the UK, the robot is predominantly used for removing the prostate for prostate cancer, although some groups are exploring its potential in other fields including bowel and thyroid surgery. It enhances the surgeon’s precision, and reduces the chance that vital nerves close to the operation site will be damaged. For many operations, only specific parts of the procedure require such precision.

“With i-Snake® we’re trying to exploit this and see if we can develop a device that is not only compact but that can be used on an on-demand basis. We are always looking for new surgical applications that could be derived from this sort of technology.”

Keyhole surgery is not the only potential use of i-Snake® – it may also be useful in an emerging field of surgery called natural orifice transluminal endoscopic surgery (NOTES), in which surgeons operate through body orifices such as the mouth or vagina. This means that no incision is required – so there will be no scar (see ‘In brief: What is NOTES?’, below).

The robotic snake is not just about improved access to distant tissues, but also about augmenting the surgeon’s own senses – acting as an extension of their eyes and hands. “‘i-Snake®’ stands for ‘imaging-sensing navigated and kinematically enhanced’,” says Guang-Zhong. “We use imaging and sensing so that it can view the tissue it’s touching and can see where it is going.”

Imaging will be a major benefit of the robot. When removing a tumour, for example, clinicians could scan the tissue before surgery to differentiate between cancerous and healthy tissue. Then, during the operation, the tissue could be overlaid with the scans, to help ensure that all the cancer is removed and that healthy tissue is spared.

Surgeons will also be able to use real-time imaging during the operation to ‘see’ beyond the tissue they’re working on. This will help to stop them damaging tissues, such as blood vessels, that they can’t see, and will obviate the need for other imaging, such as histopathology, to be performed before the operation.

Professor Guang-Zhong Yang (far right) and other members of the i-Snake® team.

Taking control

Visions of slithering reptiles are put to rest as we are shown a prototype – a long, thin metal device made of narrow jointed segments connected in a line. The robot is just over 1 cm wide, and hollow, so that instruments can be put inside, emerging from the tip when required. The length of the robot will vary from 20 cm to two or three times this, depending on the type of operation and how the organ is accessed.

Although the word ‘robot’ implies automation, with this device, the surgeon is always in control. It is, essentially, a telemanipulator – something that you control manually but that works remotely. Balancing what the surgeon can and can’t do with the robot, the so-called human-robot interface, is a major activity for some of the engineers working on the project.

“i-Snake® consists of many articulated parts. It’s not efficient to control the joints one by one, so my work is to design a control scheme for the operator to manipulate those joints simultaneously,” says medical robotics engineer and PhD student Ka-Wai Kwok. His colleague and fellow PhD student Valentina Vitiello adds: “We consider the head of the robot to be controlled by the surgeon – we don’t want to make the whole thing automatic. However, while the surgeon is carrying out the procedure we need to make sure that the rest of the articulated body is not causing tissue damage.”

To do this, the engineers can set a limited workspace, which means that certain body areas are made physically out of bounds to the snake. “Safety is always our concern,” says Ka-Wai. “Thanks to advanced medical imaging we can frequently update the ‘allowable workspace’ of the surgical robot inside the body – to provide appropriate guidance to the robot. We’re doing more than making the robot avoid obstacles, we’re preventing tissue damage and controlling the force the robot applies to the tissue.”

A testing environment
While some engineering projects can run for years before the product is tested in a real-life environment, i-Snake® couldn’t be more different. In the three years of the project so far, six different iterations of the snake robot have existed, each tested in surgical trials using pigs.

Each trial sees the team test the previous and the new model of robot. “The reality is that we’re constantly refining and improving,” says engineer and research assistant David Noonan. “We’ll improve the actual robot, the articulated sections, but also the software, the driver electronics to control it, the user interface. It’s almost a complete overhaul of the system each time.”

It’s also the time that the surgeons get to be hands-on with the device, which can be nerve-wracking for the engineers. “We had one trial where the system was damaged by the surgeon testing it. It’s not their fault – although we do joke about it with them within the team.”

The team hopes to complete the design and engineering in the next 18 months, and the next stage is to seek regulatory approval to test the device in humans. Ultimately, it could have multiple applications: gastrointestinal, gynaecological and, eventually, cardiothoracic surgery. But will surgeons adapt to the new technology?

“Clinicians are a fairly conservative bunch – they don’t like change,” says Ara. “When laparoscopic surgery first emerged, people thought I was mad.” He tells of several other occasions in his career when new technologies received a frosty welcome from some, including the introduction of surgical robots.

He sees i-Snake® as the next advance in line, but is adamant about the need to innovate: “If Henry Ford had taken the advice to invest in a cart, not in a car, we’d be in a very different place now.”

• For more on the Hamlyn Centre’s research, see the Hamlyn Centre website.

In brief: What is NOTES?

What is it?
NOTES (natural orifice transluminal endoscopic surgery) is an emerging surgical technique in which surgeons go through a natural orifice (such as the mouth, rectum or vagina) instead of making an incision on the outside of the body.

What benefits does it offer?
No incision means no incision-related complications and no visible scars.

What about the risks?
It can be difficult to navigate precisely to the target organ with current instruments. There is still an access hole (albeit inside the body) that needs to be closed safely.

What‘s happening at St Mary‘s?
There is a clinical trial of NOTES underway. To date, two gallbladder removals have been completed, with access via the vagina.

References

Vitiello V et al. DOF minimization for optimized shape control under active constraints for a hyper-redundant flexible robot. In: P Jannin et al (eds). Information Processing in Computer-Assisted Interventions 2011 (Lecture Notes in Computer Science). Springer: Berlin/Heidelberg; 2011

Kwok KW et al. Control of articulated snake robot under dynamic active constraints. Med Image Comput Comput Assist Interv 2010;13(pt 3):229-36

Noonan DP et al. Gaze contingent control for an articulated mechatronic laparoscope. IEEE International Conference on Biomedical Robotics and Biomechatronics, Tokyo, Japan. IEEE; 2010

Shang J et al. An articulated universal joint based flexible access robot for minimally invasive surgery. IEEE International Conference on Robotics and Automation (accepted)