Robot performs first realistic surgery without human help: System trained on videos of surgeries performs like an expert surgeon
Technology
113
Beiträge
72
Kommentatoren
1.8k
Aufrufe
-
And then you‘re lying on the table. Unfortunately, your case is a little different than the standard surgery. Good luck.
I assume my insides are pretty much like everyone else's. I feel like if there was that much of a complication it would have been pretty obvious before the procedure started.
"Hey this guy had two heads, I'm sure the AI will work it out."
-
I'd bet on at least twenty years before it's in general use, since this is a radical change and it makes sense to be cautious about new technology in medicine. Initial clinical trials for some common, simple surgeries within ten years, though.
This is one of those cases where an algorithm carefully trained on only relevant data can have value. It isn't the same as feeding an LLM the unfiltered Internet and then expecting it to learn only from the non-crazy parts.
The idea that a carefully curated data set may yield better results seems to be something that even the likes of Google engineers can't get their heads around.
-
Oh yeah, I've been successfully propagandized into thinking rich people became rich through merit, I forgot how many of them are complete morons XD
Thanks for reminding me
Maybe we could install a murder mode switch.
Perhaps an algorithm where its effectiveness is inversely proportionate to your bank account.
-
See the part that I dont like is that this is a learning algorithm trained on videos of surgeries.
That's such a fucking stupid idea. Thats literally so much worse than letting surgeons use robot arms to do surgeries as your primary source of data and making fine tuned adjustments based on visual data in addition to other electromagnetic readings
That's such a fucking stupid idea.
Care to elaborate why?
From my point of view I don't see a problem with that. Or let's say: the potential risks highly depend on the specific setup.
-
you could not pay me enough to have my surgery done by a robot
I trust a good machine much more than any human.
-
You underestimate the demands on a surgeon’s body to perform surgery. This makes it much less prone to tiredness, mistakes, or even if the surgeon is physically incapable in any way of continuing life saving surgery
That's absolutely not the point. I was criticizing the journalism, not technology.
-
They obviously don't feel comfortable with the robot doing surgery on humans just yet either which is why they're not actually suggesting doing that yet. It will have to go through years and years of certification before that's even considered.
I'm sure most surgeries will still be conducted by humans but there are situations where one of these would be extremely helpful. Any situation where a surgeon isn't currently accessible and can't quickly get there. Remote communities, Disaster relief, Arctic research facilities, Starships trapped in the Delta quadrant, War zones, Ships at sea.
Do you think a 5 bed hospital will have the money to afford a robotic surgeon?
-
That's such a fucking stupid idea.
Care to elaborate why?
From my point of view I don't see a problem with that. Or let's say: the potential risks highly depend on the specific setup.
Being trained on videos means it has no ability to adapt, improvise, or use knowledge during the surgery.
Edit: However, in the context of this particular robot, it does seem that additional input was given and other training was added in order for it to expand beyond what it was taught through the videos. As the study noted, the surgeries were performed with 100% accuracy. So in this case, I personally don't have any problems.
-
I trust a good machine much more than any human.
Have you considered that the machine is made by a collection of humans?
-
That's such a fucking stupid idea.
Care to elaborate why?
From my point of view I don't see a problem with that. Or let's say: the potential risks highly depend on the specific setup.
Imagine if the Tesla autopilot without lidar that crashed into things and drove on the sidewalk was actually a scalpel navigating your spleen.
-
Yeah but the training set of videos is probably infinitely larger, and the thing about AI is that if the training set is too small they don't really work at all. Once you get above a certain data set size they start to become competent.
After all I assume the people doing this research have already considered that. I doubt they're reading your comment right now and slapping their foreheads and going damn this random guy on the internet is right, he's so much more intelligent than us scientists.
Theres no evidence they will ever reach quality output with infinite data, either. In that case, quality matters.
-
Being trained on videos means it has no ability to adapt, improvise, or use knowledge during the surgery.
Edit: However, in the context of this particular robot, it does seem that additional input was given and other training was added in order for it to expand beyond what it was taught through the videos. As the study noted, the surgeries were performed with 100% accuracy. So in this case, I personally don't have any problems.
I actually don't think that's the problem, the problem is that the AI only factors for visible surface level information.
AI don't have object permanence, once something is out of sight it does not exist.
-
I actually don't think that's the problem, the problem is that the AI only factors for visible surface level information.
AI don't have object permanence, once something is out of sight it does not exist.
If you read how they programmed this robot, it seems that it can anticipate things like that. Also keep in mind that this is only designed to do one type of surgery.
I'm cautiously optimist.
I'd still expect human supervision, though.
-
Theres no evidence they will ever reach quality output with infinite data, either. In that case, quality matters.
No we don't know. We are not AI researchers after all. Nonetheless I'm more inclined to defer to experts then you. No offence, (I mean there is some offence, because this is a stupid conversation) but you have no qualifications.
-
Imagine if the Tesla autopilot without lidar that crashed into things and drove on the sidewalk was actually a scalpel navigating your spleen.
Absolutely stupid example because that kind of assumes medical professionals have the same standard as Elon Musk.
-
Do you think a 5 bed hospital will have the money to afford a robotic surgeon?
You assume an Antarctic research facility lacks funds?
-
Absolutely stupid example because that kind of assumes medical professionals have the same standard as Elon Musk.
Elon Musk literally owns a medical equipment company that puts chips in peoples brains, nothing is sacred unless we protect it.
-
Have you considered that the machine is made by a collection of humans?
Yes. But if the machine has proven to work reliably it will usually do so for its lifetime, while humans are prone to e multitude of errors. Especially in the medical field.
-
No we don't know. We are not AI researchers after all. Nonetheless I'm more inclined to defer to experts then you. No offence, (I mean there is some offence, because this is a stupid conversation) but you have no qualifications.
It's less of an unknown and more of a "it has never demonstrated any such capability."
Btw both OpenAI and Deepmind wrote papers proving their then models would never approach human error rate with infinite training. It correctly predicted performance of ChatGPT4.
-
That's such a fucking stupid idea.
Care to elaborate why?
From my point of view I don't see a problem with that. Or let's say: the potential risks highly depend on the specific setup.
Unless the videos have proper depth maps and identifiers for objects and actions they're not going to be as effective as, say, robot arm surgery data, or vr captured movement and tracking. You're basically adding a layer to the learning to first process the video correctly into something usable and then learn from that. Not very efficient and highly dependant on cameras and angles.
-
-
-
-
YouTube relaxes moderation rules to allow more controversial content. Videos are allowed if "freedom of expression value may outweigh harm risk"
Technology
1
-
Reddit sues Anthropic, alleging its bots accessed Reddit more than 100,000 times since last July
Technology
1
-
-
‘Alexa, what do you know about us?’ What I discovered when I asked Amazon to tell me everything my family’s smart speaker had heard
Technology
1
-
