Stories
Slash Boxes
Comments
Search

News for nerds, stuff that matters

Slashdot Log In

Log In

Create Account  |  Retrieve Password

Doctors To Control Robot Surgeon With Their Eyes

Posted by kdawson on Sat Mar 22, 2008 01:50 PM
from the no-saccades-please-we're-operating dept.
Robotics Medicine
trogador writes "Researchers from Imperial College London are improving the Da Vinci surgical robot by installing an eye-tracker, which allows surgeons to control the robot's knife simply by looking at the patient's tissues on a screen. Tracking the eyes can generate a 3D map, which in turn can make moving organs — like a beating heart — appear to stand still for easier operation. Other features include 'see-through' tissues on the surgeon's screen (so tumors can be seen underneath tissues) and 'no-cut' zones, places where the robot won't allow the surgeon to cut by mistake. Says ICL Professor Guang Zhong Yang, 'We want to empower the robot and make it more autonomous.'"
+ -
story

Related Stories

Submission: Surgeons to control Da Vinci robot with their eyes by trogador (1170027)
[+] Bringing Surgical Robots Into the Mainsteam 73 comments
The New York Times is running a story about how using robots to perform surgical operations has been transformed from a controversial dream to reality. Dr. Frederic Moll abandoned his residency for Silicon Valley and helped to revolutionize the industry. The lengthy article also discusses some of his innovations. We've discussed various robot-assisted medical procedures in the past. From the Times: "'I was struck by the size of the incision and injury created just to get inside the body,' Dr. Moll says. 'It felt antiquated.' He took the idea to his employer, Guidant, a medical device company. Guidant decided that robotic surgery was too futuristic and too risky, so Dr. Moll rounded up backers, resigned, and in 1995, founded Intuitive Surgical. The company prospered by proving that robots could deftly handle rigid surgical tools like scalpels and sewing needles through small incisions in a patient's skin."
This discussion has been archived. No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.

Doctors To Control Robot Surgeon With Their Eyes 50 Comments More /

 Full
 Abbreviated
 Hidden
More
Loading... please wait.

  • Ah yes ... (Score:3, Funny)

    by ScrewMaster (602015) on Saturday March 22 2008, @01:53PM (#22830908)
    but is it a Robot Chicken? If so, it's important to remember that when you're playing chicken the first one to blink loses.

  • But the real question: (Score:4, Funny)

    by theaceoffire (1053556) on Saturday March 22 2008, @01:58PM (#22830930) Homepage
    ^_^ I foresee an issue with female clients.
    "AHHHH! NO, My eyes are UP HERE!"
  • !!! eye movement is jerky! why oh why would you use eye movement of all things to control a surgical instrament!? good thing that the robot will limit what areas the surgeon can cut

    • Re:jerky movement (Score:5, Informative)

      by ScrewMaster (602015) on Saturday March 22 2008, @02:18PM (#22831108)
      Those movements are called "sacchades" and they are important to preventing retinal fatigue. They're actually fairly predictable and it shouldn't be hard to average them out.
      • What if one of the surgeon's assistants is seriously hot and is in his peripheral vision? What gunna happen when the surgeons eyes keep darting to the nurses huge jugs?
      • You're thinking of microsaccades rather than saccades (which are larger scale jumps from one point of focus to a different point of focus). Microsaccades are probably actually actively helpful in precise eye tracking, because they end at the target of fixation, and therefore pick out the relevant spot within the area that the focus drifts within.
        • As I understand it, the primary function of saccades is to prevent retinal fatigue. Experiments have been performed where a grain-of-wheat lamp was literally glued to the eyeball. That way, the light source would move with the eye, and the corresponding image would be focused on the same area of the retina, regardless of eye motion. Within a few seconds, the image would fade and the light would become invisible as the retinal pigments were exhausted.

          This effect is even more noticeable with people who suf
    • "Controlled with eye movement" Doesn't automatically mean that the robot is going to actually do anything to the place where the surgeon is looking. As the summary mentions, the eye-triggered commands could be as simple as switching a camera view, something that obviously isn't going to cut open the patient.

      I don't know why everyone on here seems to think that they're the first person to figure out that a robot shouldn't slice open a patient because the doctor blinked.
  • Back when I was working at the NASA Jet Propulsion Lab, I was stuck in a basement laboratory. One interesting thing was that there were old robots *everywhere*, including a big old surgical robot right beside my desk. It was really interesting, as it had two 6-Degrees of Freedom "scapel" controls, and a microscope-like magnifying viewer.

    I asked some folks about it, and they said that one of the main benefits was that they used Fourier transforms and other filtering to significantly reduce the "jitter" of the doctor's hands, without aversely affecting intentional motions. I thought that was really interesting, and might save a lot of complications. A former boss of mine has a surgeon for a father, and he said it was quite common fifty years ago for them to have an alcoholic beverage before a surgery to steady their nerves, and seems to think this was effective. I suppose the robotic version wouldn't have all the downsides.

    I also think an interesting possibility for this technology is plastic surgery - one of the biggest current problems with plastic surgery isn't expense (a lot of things people want aren't that expensive), it's the risk of complication, which can be quite significant. My brother did not get his face repaired after breaking a cheekbone due to this risk.

    This would especially be good for individuals who have recently lost a lot of weight - there's a lot of self consciousness about excess skin, and being able to safely remove it with much reduced risk of complications would go a long way towards helping these people feel better about themselves (which is one way to help keep the weight off).

    So yes, I for one welcome my robot surgeon. Some day it might save my life!

  • Precision? (Score:4, Informative)

    by ScrewMaster (602015) on Saturday March 22 2008, @02:04PM (#22830994)
    "It shows you the tumour in relation to its anatomical structure," Darzi said. That means the surgeon can be more precise and avoid cutting out large amounts of healthy tissue.

    Lack of precision isn't the only reason surgeons remove large amounts of apparently "healthy" tissue along with a tumor ... it's to make sure they get all of it.

    • Re: (Score:3, Insightful)

      by repvik (96666)
      They cut out a fair bit because they simply can't see the tumor... Cutting a safety margin comes on top of that again.
    • You would like a good margin around the entire tumor. If you were able to precisely get 1cm margins all around that would be far preferable to getting 1.7 cm margins in one place and 0.1 cm margins in another. As it stands even if the ideal is 1cm, you often shoot for more on average because you would rather have one place that is 3cm than one place that is 0.1 cm.

      So precision would definitely give you less tissue loss, if nothing else than by allowing a smaller target for margins.
  • We want to empower the robot and make it more autonomous.

    This won't end well.

  • Great research, but still only research! (Score:5, Interesting)

    by neapolitan (1100101) * on Saturday March 22 2008, @02:09PM (#22831042)
    I am a doctor that does surgical procedures (mainly pacemaker device implantations).

    These kind of devices are very much experimental, and pop up in the news every couple of months (a slashdot search can reveal similar ideas), but quite far away from any sort of typical use. Right now heart bypass surgery is decidedly low-tech, with a surgeon viewing the beating heart with loupe glasses and very skillfully lasso-ing the coronary arteries. It is a great fantasy in the hearts of all doctors to have a machine that offsets any heart movement -- it would make things much easier if reliable. The article doesn't mention that the movement would also have to be coupled with respiratory movement, and have some sort of fail-safe in the case of patient or external movement.

    However, standard surgery is still not done using these tools. There are way too many items required to make this feasible in the near future ( 10 years IMHO, although I hope I'm wrong!)

      - Testing: This is literally a life-and-death situation, and any robot "error" in a real person is likely to set the technology back 5 years after it hits the press and hospital review boards. Very, very damaging PR.
      - Education: Surgeons would have to essentially be re-trained to use such a system.
      - Feedback: It is really, really difficult to give an operator feedback on how something "feels." Part of during surgery (no kidding) a surgeon will often run his finger along cardiac arteries -- you can almost "feel" the calcified plaque in a diseased vessel. It would be really hard to approximate anything like this with a virtual robot.
      - Cost: Labor is relatively cheap compared to the capital expenditure to R&D something like this... of course, this will change as time goes by.

    Exciting news, but incremental technology.
    • Da Vinci is FDA approved and used clinically in a number of hospitals. Robotic surgery for EVERYTHING is a long way off, and I doubt eye controlled surgery will ever be useful, but robotic surgery for some procedures is here now.
      • I guess it depends what you mean by "used clinically." These generally are technology expos as part of a clinical trial -- FDA approval allows you to use it on a person, but you can bet that the patient signed a very long form indicating he is part of a research study / nonstandard procedure. Reading a few articles about this indicates the most high volume centers have done ~100 of these "robotic procedures," and mine (a major Harvard teaching hospital) has done... zero.

        I'm all for it too -- but again, I
        • In the first paragraph you say that nobody in your centre has done any robotic surgery, then in the last paragraph you mention that the GI doctors have. A "remote device", unless I've completely misunderstood you, is a robot, and may well actually BE a Da Vinci robot. It is not an autonomous robot, but it is a robot.

          I agree, autonomous surgery, where the procedure is under the control of a computer is a long way off. Non-autonomous robotic surgery is not.

          I notice that Massachusetts has nine Da Vinci robo
          • Your attitude is why I don't post here much.

            Dude, stop being the contrarian, it does not make you sound "smart," only childish and argumentative. People will treat you better in life if you ask questions for clarification instead of trying to trap somebody into silly fallacies to "prove your point," as the first sentence in your response. Occasionally we have trainees with your style, and gradually they socialize. :)

            Again, I'm not sure the name of the GI device -- I saw a tech demo of it in one of our vir
            • I wasn't trying to trap you, but by your response maybe I did?

              Since we're trading silly insults, I've heard some surgeons don't really like any hint they might be wrong, about anything, even if its just a silly mistake or outside their field of expertise.

              For your information (or rather for someone else reading this who is actually interested), Da Vinci appears to be fully FDA certified, NOT as a research device, for some procedures since 2000 or 2001. I believe they have been certified for some endoscopic
              • Well, thanks guys. I am happy to provide the perspective of an MD in a sub-specialty. I actually paged a few of my friends that do CT surgery to ask them if they used the da Vinci system because I was curious. All of them had heard of it, and a few had been in cases with it, and again, unfortunately, all of them said it was a sideshow currently. At another major teaching hospital, the most experienced surgeon had about 100 cases with the device: far, far fewer than standard (they do 2-3 operations *per
    • These kind of devices are very much experimental, and pop up in the news every couple of months...but quite far away from any sort of typical use.

      They are used somewhat routinely in many hospitals across the US, Canada and Europe. It's been several years since they were research-only. The only reason they aren't used more frequently is simply the cost and size, even the places that use them regularly only have one OR set up with the equipment, while they may be running two dozen ORs on a daily basis.

      You're

  • Oh. My. God. (Score:4, Funny)

    by ChaoticLimbs (597275) on Saturday March 22 2008, @02:17PM (#22831100) Journal
    I am so glad I am not a woman with really nice tits.
    • Here's a curious anecdote, that I feel is appropriate to the joke.

      I was looking for some clinical experience between the first and second years of med school, so I shadowed a surgeon, who specialized in breast tumors, and wound up spending lots of time in a "breast clinic" (screening and followup) for about a month and a half. While I probably saw some great breasts, I never felt aroused. Given that I am a very sexually-active heterosexual man, I was amazed at how my mindset was able to adjust itself, and h
  • I wouldn't want to be under the knife at that moment when a gaze is diverted.

  • Eye control? (Score:3, Informative)

    by ceoyoyo (59147) on Saturday March 22 2008, @02:26PM (#22831176)
    Why? I did some work on development of a surgical robot. Surgeons use their tactile senses a lot, and its important to have 3D haptic feedback. Why in the world would you want to control a robot with your eyes? You can't even tell it how hard or deep to cut.

    • Why in the world would you want to control a robot with your eyes? You can't even tell it how hard or deep to cut.

      Because most surgeons only have two hands. This represents a new means of input and control. Think of it as using both your hands and feet for driving. You can certainly drive with only your hands and a setup with the brake and gas on the steering wheel. However, having the break and gas operated by the feet gives the driver greater ability to use his hands for steering or drinking his coffee while shaving and eating a muffin.

      • This is the example you use to demonstrate that eye controlled surgery is a GOOD idea??

        To use a slightly different example, you could give a guy in a tank a bunch of video cameras, eye controlled aiming (already exists) and a voice command fire control, plus the regular driving controls. You don't, you build tanks to use a crew of typically three. How come? Because when job performance counts it's best to use multiple people, not overload one. Particularly not with flakey, taxing things like eye control
      • A group in my lab is building one (with full haptic feedback) that can scale down the surgeon's movements to micrometre accuracy. So it can move a scalpel pretty softly indeed. Superhumanly softly.
      • And they'd record that you're focused on the surface of your 2D monitor....

        Da Vinci is mostly used for endoscopic surgery where you can't directly see the surgical field, robot or no. Even if you use some kind of 3D, do you really want your surgery to depend on where your surgeon looks?

        Okay, I want to cut here, here and here, but NOT there... oh crap.
      • Might work. Surgeons would then be in good company. I have more than once written a paper that contained the phrase "we performed a study in which blinded radiologists...."
  • "...it ain't a robot"? Aren't robots autonomous? If not, my car is a robot.
  • I, for one, welcome our new robot controlling overlords.
  • We want to empower the robot and make it more autonomous.
    • [Leela]: Well at least here you'll be treated with dignity. Now strip naked and get on the probulator.
    • [Fry]: Ooo!
  • On Friday, I had a balloon angioplasty & stent put in -- the procedure was done at St. Joseph's [slashdot.org] in Atlanta, which is one of the main training centers for the daVinci system. If my case had required bypass surgery, it might well have been done by the daVinci.
  • 2 beered up surgeons in theatre:-

    "Check it out - I even closed!"
    "OK smart arse - so you can do an appendectomy with your eyes huh? - let's see you do one using your tongue then!"

  • HAL: "I'm sorry Dave, but I can't let you do that"

    DAVE: "But HAL, it's a tumor, it's got to come out!"

    HAL: "My records show this patient hasn't paid their insurance premium this month. I'm very sorry Dave, I can't let you operate."

    • Re: (Score:2, Insightful)

      by Aglassis (10161)

      The human body is not a machine, and we cannot even come close to mimicking one, let alone thousands to millions.

      Why isn't the human body a machine? I'll grant that we can't mimic it but that doesn't mean that the systems of the body aren't mechanical in nature. Would you argue against using titanium rods to help with fractures, the use of plastic joints to replace failed joints, or other technologies that attempt to repair some mechanical failure in the body (including the use of micro-robots to perform surgery)?

      On a more ethical standpoint, being interested in Medicine and Machines myself, I feel this is going in the opposite direction of what I hope for medicine. Distancing doctors from patients, and life from reality may prove a nasty combo.

      Ethical? I'm not sure that word means what you think it means. You are advocating an almost supersti

        • "The human body is too complex for a machine to encompass and predict all possible outcomes."

          It's too complex for a physician to predict all possible outcomes. Given the minimal time allowed each patient and the extreme pressure to see many patients while coping with all his/her other duties, technology as a medical "force multiplier" will become ever more necessary to empower the doctor as a "care supervisor".

    • Re: (Score:3, Insightful)

      by vertinox (846076)
      The human body is not a machine, and we cannot even come close to mimicking one, let alone thousands to millions.

      Wasn't that the whole point with using a machine?

      A human can only hold a knife so still and accurate where a machine could cut on an accuracy scales below a millimeter.

      I mean do we build planes with wings that flap like a bird? It would make sense to build a machine that does surgery without the flaws the inherit instability of the human hand.
    • I feel this is going in the opposite direction of what I hope for medicine. Distancing doctors from patients, and life from reality may prove a nasty combo.

      Well, it's a good thing most of us aren't hoping for the same things you are. Most sophisticated medical technology "distances" a doctor from his patient to some degree ... like, for example, a stethoscope. Should a doctor be required to press his ears against a patient's chest to listen the heart and lungs?

      The issue here is that the human body has
      • Well, it's a good thing most of us aren't hoping for the same things you are. Most sophisticated medical technology "distances" a doctor from his patient to some degree ... like, for example, a stethoscope. Should a doctor be required to press his ears against a patient's chest to listen the heart and lungs?

        Interesting that you should mention the stethoscope. I'm a few years out of my work as a nurse, but when I was working in that field there was considerable resistance to the use of electronically assist

    • Re: (Score:3, Insightful)

      by NIckGorton (974753) *

      On a more ethical standpoint, being interested in Medicine and Machines myself, I feel this is going in the opposite direction of what I hope for medicine. Distancing doctors from patients, and life from reality may prove a nasty combo.

      First, you need some distance. If you are in respiratory failure, I have precious little time to get you intubated and ventilated. That is not the time you want active listening and human touch. You want technical skill and someone who does not dick around. You may even want someone who has toys like this hella sexy video laryngoscope: http://medgadget.com/archives/2006/10/mcgrath_series.html [medgadget.com]

      Want.

      Of course with that I am less in touch with you physically and less likely to end up spattered in the hal

        • and that quick thinking decision kills me?

          The you're dead, just like you'll be if that doctor doesn't have what it takes to help you. Put it this way: if you have a competent ER physician on hand with the right equipment, you have a chance. Personally, I want that doctor to have the best medical technology available, because I want to live.
    • "Distancing doctors from patients, and life from reality may prove a nasty combo."

      I'd rather have a stable "machine tool" used to cut me for surgery than a less-precise human hand. Let the surgeons brain have a better interface with my repair process than just his imprecise fingers.

      This has nothing to do with "distancing" and everything to do with precision and control. If you want to use the tools they used on Mel Gibson in "Braveheart", have at it, but the point of advancing medical tech is to improve eff
    • Now that's a good, traditional British name. Wonder what his family crest looks like.

      You've never heard the phrase 'a good Englishman is 99% foreign'?

      My Family are English, and have lived in the UK for centuries, yet my Surname is Nordic. Extremely so in fact, It hasn't been englicised at all, in all this time, which I find quite odd. I've had no end of problems over the years with it being spelled wrong.

      • I've had no end of problems over the years with it being *spelled* wrong.
        The way you spelt that word suggests that you is Americaine?

        Cheers!
        • The way you spelt that word suggests that you is Americaine?

          Nope, but since the overwhelming majority of my exposure to the written word has been early american SF (can't help it, I worship at the alter of Gernsback) , I do find that my spelling is some sort of mutant hybrid between those of our two countries.

          Add to that a fascination for pre eighteenth century European dictionaries, and you end up with a vocabulary which is positively weird. I gave up trying to conform to the language norms of my country a
    • The last thing I'd need is the robot carving a 'star' shape into my aorta in an attempt to slow down time :P
    • 'We want to empower the robot and make it more autonomous.' When the robot uprising happens I will not help humanity precisely because of statements like these

      I wouldn't worry too much until you start hearing statements like "We want to empower the super-smart AI and make it more autonomous."

All trademarks and copyrights on this page are owned by their respective owners. Comments are owned by the Poster. The Rest © 1997-2009 Geeknet, Inc.