On the Cutting Edge
A virtual operating theatre could train hundreds of surgeons at once for every possible eventuality, in any operation. “There’s no point building anything where you presume there are limits,” argues David Broadbent, whose firm Cyberjenn are working with the NHS to develop just that. Nick Carson dissects their proposal.
“A mate of mine is an eye surgeon; he says surgery is a bit like plumbing – physical dexterity can be as important as intelligence,” suggests David Proud, who has since left Cyberjenn. To cut waiting lists, the NHS may try moving minor operations down the chain to allow nurses to conduct them, calling for highly-effective task-specific training.
“The government are trying to reduce cataract queues, for example, by setting up mobile surgeries. But perhaps the solution is to train more people on the NHS. Rather than having one person watch a trained surgeon perform the operation, have five hundred people performing a virtual operation.”
“We were asked to create characters to demonstrate the human body functionally, that could literally be dissected on-screen,” reveals Broadbent. “Characters that are one-hundred-percent correct in all of their bodily functions and totally interactive, so you can use a mouse and keyboard to conduct a full operation.”
“Take something simple like a cataract operation – what can go wrong? Let’s have all the scenarios, map them out; have a go,” Proud continues. “Before you actually get live on a body you’ve run it through, and can revisit it as many times as you need. In a live operating theatre you can hardly say, ‘I didn’t catch that. Can you do it again?'”
Training is not only quicker; it’s also more clearly defined. “You know exactly what they’ve all gone through,” he reasons. “Instead of an individual surgeon teaching it their way, you can have the approved way of doing it – best practice. And the graphics, the what-if scenarios, they can all be updated. You don’t just do it and leave it; it’s not like making a film – you know, tick – and then having to go back and do the whole thing again in two months’ time, when there’s something else. Maintenance is critical to make it valuable.”
If the glove fits
Dissecting a kidney with a keyboard is all very well, but in a profession where the dexterity of the surgeon can make the difference between life and death, does knowing what will happen if the scalpel slips really stop you doing it?
“The actual art of cutting, they can all do,” reasons Proud. “Sewing, fine work, that’s all there; you couldn’t simulate that properly, but initially you wouldn’t have to. This is a quick way of simulating a particular operation; it’s about exploring. What if I did this; what if I did that; what happens if bleeding starts there; or bruising occurs there; one of the stitches leaks.”
Nevertheless with human interface devices progressing at a rate of knots, a pin-point accurate simulation needn’t be out of reach if demand is there. “Virtual glove systems can easily be built into the software,” Broadbent points out. “With the technology growing side-by-side, you would be able to bring in smaller modules within a classroom environment, and then use it again in a larger training environment using virtual gloves.”
“That’s the key factor: There’s no point building anything where you presume there are limits. A client gives me the basics that they want, and then I add something. I think that’s necessary; longevity is crucial. In [military training package] America’s Army, shooting ranges are calculated using data sets collected over the last forty-odd years. When you introduce accurate data sets into surgery, the tool becomes even more invaluable – and it’ll only extend.”
Anything that uses technology for technology’s sake is purely self-serving, and Cyberjenn are not oblivious to the arguments against using CGI to train high-level professionals. “In most situations, if you put a virtual character trying to explain something in front of a bloke in his thirties with a modicum of intelligence he’ll think you’re taking the piss, to be honest,” admits Proud. “In the right place it’s a really powerful addition, but it’s foolish to say that we’ll use this technology in every possible application. It’s about finding that hotspot.”
Patronising is the last thing you can afford to be when training a future practitioner, so don’t expect a Mickey Mouse spin-off. “A little squeaky character talking to the surgeon? Forget it. Obviously it has to be someone credible describing the operation.” He stresses the importance of keeping the end-user in mind. “Why would you want a virtual character if you could have a real person? Well, one of the problems in colleges of further education is that they can’t get the resources; they can’t all go and dissect a body. There’s your application.”
Unlimited variables also allows for unlimited trial and error, and there’s no ‘undo’ option when you have a real patient under the knife. “It’s not a straightforward ‘let’s dissect a body’ scenario; the model reacts to the way you do the operation,” explains Broadbent. “If they cut in the wrong place it bleeds, or haemorrhages; they have to deal with that. What makes this technology so rich is the AI being developed alongside it.”
Strength in numbers
With high-end software so easily available, small companies and innovative individuals are raising the bar for industry heavyweights, which Broadbent argues can only be a positive thing. “Because of the widespread availability of software like 3D Max, Lightwave and Maya you can now produce broadcast-quality material yourself. You don’t need to be Pixar studios to output a half-hour film, or even a thirty-second animation, at the standard that would win you an Oscar.”
“Because small firms are constantly pushing the boundaries, it forces people like Pixar to keep bringing in new innovations,” he goes on. “I guarantee you that the techniques used in The Incredibles, you will be able to do at home by Christmas. The industry is constantly driving that way – attempting to be cutting-edge by making the technology available to smaller, up-and-coming studios and individual artists experimenting.”
“Pixar released RenderMan deliberately for that purpose. They realise that eventually their work will become stale, and they need to inject new innovation. So they make certain tools available to the general public and small businesses to develop new innovations, which they can then take up. Many businesses no longer innovate, and instead farm out to small units like Cyberjenn.”
“Bigger companies are exploiting that R&D opportunity,” concludes Proud. “If there are five companies offering a solution, they don’t care which one they take so long as it suits them, and then they make the money. They’ve got the infrastructure, the clients and the contacts – a network of people who can deliver, knitted together to suit their client base. If we can portray that we have enough people on our network to deliver what they want, then of course we’ll get a slice of the action.”
© Nick Carson 2005. First published on Channel 4’s IDEASFACTORY West Midlands