Slashdot Mirror
Can AI Replace Hospital Radiologists? (cnn.com)
An anonymous reader quotes CNN:
Radiologists, who receive years of training and are some of the highest paid doctors, are among the first physicians who will have to adapt as artificial intelligence expands into health care... Today radiologists face a deluge of data as they serve patients. When Jim Brink, radiologist in chief at Massachusetts General Hospital, entered the field in the late 1980s, radiologists had to examine 20 to 50 images for CT and PET scans. Now, there can be as many as 1,000 images for one scan. The work can be tedious, making it prone to error. The added imagery also makes it harder for radiologists to use their time efficiently... The remarkable power of today's computers has raised the question of whether humans should even act as radiologists. Geoffrey Hinton, a legend in the field of artificial intelligence, went so far as to suggest that schools should stop training radiologists.
X-rays, CT scans, MRIs, ultrasounds and PET scans do improve patient care -- but they also drive up costs. And now one medical imaging startup can read a heart MRI in 15 seconds, a procedure which takes a human 45 minutes. Massachusetts General Hospital is already assembling data to train algorithms to spot 25 common scenarios. But Brinks predicts that ultimately AI will become more of a sophisticated diagnostic aid, flagging images that humans should examine more closely, while leaving radiologists with more time for interacting with patients and medical staff.
X-rays, CT scans, MRIs, ultrasounds and PET scans do improve patient care -- but they also drive up costs. And now one medical imaging startup can read a heart MRI in 15 seconds, a procedure which takes a human 45 minutes. Massachusetts General Hospital is already assembling data to train algorithms to spot 25 common scenarios. But Brinks predicts that ultimately AI will become more of a sophisticated diagnostic aid, flagging images that humans should examine more closely, while leaving radiologists with more time for interacting with patients and medical staff.
And what's with the gap of several hours between stories yesterday, followed by several stories being backdated and posted in several minutes? And where is BeauHD? This site leaves a lot to be desired.
An AI is only as good as the people that have taught it. Of course it can accumulate experience and never forget, but humans also have a thing called intuition to see things in a different view and capture things that are completely new.
Humans and AI will however supplement each other for improved accuracy.
If builders built buildings the way programmers wrote programs, then the first woodpecker would destroy civilization.
And now one medical imaging startup can read a heart MRI in 15 seconds, a procedure which takes a human 45 minutes.
That's quite nice that you save 44.75 minutes, but isn't kind of more interesting how much money you save? Save for emergency situations, waiting an hour usually isn't a problem in US medicine. Even 45 minutes of CPU/GPU time is going to be cheaper than 45 minutes of human time.
Ezekiel 23:20
flagging images that humans should examine more closely
That's exactly how it's going to work. You could train a 5 year old to determine if an image showed scenario A or B. It's just fancy chicken sexing. Anything that is decisively A or B gets labeled as such. Images that are questionable get passed on to Level 2, the Human.
Then you re-train the network, rinse repeat.
Can AI Replace Hospital Radiologists?
No, AI can not. What AI can do is be an extremely valuable tool for radiologists and doctors, one that makes analyzing all the various forms of synthetic medical imagery more accurate, and most likely increasing their productivity. It can reduce oversights and errors, but it won't be able to fully replace expert human analysis for quite some time. Like most AI solutions it will most likely take far longer than AI experts predict. Perhaps we need an AI to predict timeframe for AI solutions since people seem to do that poorly. :-)
Does this perhaps lead to reducing the number of radiologists due to increased productivity? Probably not, more likely radiologists will be bombarded with more imagery to analyze as technology improves and costs lower and is more frequently used.
Dermatologists are looking towards extinction as well.
BTW, as a physician who spends half his day in the hospital and half in the office, I average speaking to a radiologist about once a year.
You could replace all the radiologists (excepting interventional radiologists) in the U.S. with IBM's Watson and a dozen humans for over-reads. You don't go into radiology if you're a people person.
Help! I'm a slashdot refugee.
.. while leaving radiologists with more time for interacting with patients and medical staff.
It will leave more time for radiologists to review additional flagged scans since that will probably make them the most money.
Obviously AI can replace radiologists. If new tools allow a person to handle twice or even ten times the case load, you've effectively replaced people with A.I.
Radiology can turn into a call center job as well, where a small team of people handle departments at multiple hospitals and medical centers. Imagine a world where a staff of 3-4 handles every case for 100 locations. Easily means 1000s of cases per day, but with A.I. it is conceivable.
Idiots think A.I. replacing a job means the original job no longer exists. That is a naÃve postion to take.
I'd at least want a second opinion.
Hi! I make Firefox Plug-ins. Check 'em out @ https://addons.mozilla.org/en-US/firefox/addon/youtube-mp3-podcaster/
Reading an X Ray is thousands of times harder than reading a captcha. Far, far, more subtle. Yes a broken bone is obvioius most of the time. But soft tissue problems are way, way, more difficult and subject to interpretation. The difference between a negative finding a postive one can be invisible to the untrained eye.
A radiologist doesn't just rely upon the image but also upon the patient's history and symptoms. And a doctor has the accumulated resovoir of "gut feeling" gained from experience. He combines all these parts to reach his determination.
AI could be useful suppliment to the radiologist's toolbox, but it won't replace the human diagnostician.
Image recognition is not AI. Sorry.
No, AI can not.
Most of the world didn't think that the automobile would replace horse + carriage when first introduced.
vos nescitis quicquam, nec cogitatis quia expedit nobis ut unus moriatur homo pro populo et non tota gens pereat.
Radiologists are constantly misreading x-rays.
Anyway, radiology is one of those bullshit specialties, like anesthesiologist (that used to be a nurses job and it was just as safe), that does nothing but jack up the costs of healthcare.
Much of our healthcare is over priced because we have over qualified people ($$$$) doing jobs that really only need a nurse or even a nurse practitioner.
But, the AMA will lobby and legislate their existence like they do for pharmacists - there is absolutely no reason for that profession to exist.
cease fire stand down,, that's the spirit.. sing along.. https://www.youtube.com/results?q=stem+cell+therapy&sp=EgIIBQ%253D%253D
I think most routine radiology could end up at least being assisted by AI, given that the entire practice revolves around using imaging techniques that return incomplete data and making a judgement call. This is the sort of thing machine learning is good at -- reading billions of images and determining what something looks like. Real radiologists in training do the same thing -- except they have a much smaller data set to fall back on.
The real question is how we're going to deal with the sudden flip in what is considered a highly-skilled job:
- Doctors in general are a perfect example - because the supply of medical school slots is kept low, only the people with perfect grades and photographic memories, _and_ who can ace the MCAT get into med school (in the US.) If machine learning becomes a thing, then having a photographic memory is not going to be as important as it once was...it's already less important.
- The Bar Association didn't limit the amount of law school slots the way the AMA does, and the result today is that law school grads can't find work. Just 20 years ago, having a law degree would definitely get you a job, and having a Top 14 law degree would set you up for life permanently. Law is a profession that relies on interpreting vast amounts of data, and computers are really good at the routine parts of the job that junior associates used to make $180K a year doing.
- From the non-professions, another example is air traffic controllers. Even with computers aiding them, humans who have the unique ability to think natively in 3 dimensions and keep an entire sector of airspace's inhabitants in their brains along with their speed, altitude and heading have been doing it for ages. It takes years of training to understand and definitely qualifies in my book as a highly skilled job. They make a lot of money because few people have the ability to do it and keep their stress levels non-lethal. But, it also sounds like something computers could take over eventually.
And nobody honest claims that they can. For example, the Watson people from IBM say "will not replace experts in the next 50 years". The state of the art in AI gets massively overestimated all the time. Actual fact is that the only thing available is "weak AI", and that one does not have any actual intelligence and is restricted to library look-up and statistical classificators. Very useful, but not even remotely doing what a smart and experienced human being can do. These algorithms do not have insight or understanding and they cannot get it. It is not a question of computing power either.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Can computers exceed the best human ability when the training data is comprised of human opinion? I can see it exceeding the performance of the average radiologist and maybe even the best radiologist when he is having a bad day. But given how present day AI technology works, I don't see how AI radiologists can be better than a human radiologist. They can be faster for sure. But I am not seeing how it can exceed the best human radiologists.
Geoff Hinton said that machine learning and particularly deep learning is so good now that anything that amounts to classification will soon be done better by machines than humans. The except is here. So logically medical schools should stop churning out radiologists now.
I would like to agree with him, but note that there is still a large difference between even large synthetic tests and the real world. At the end of the day a trained medical specialist will have to make the diagnosis, machines are not there yet, and companies are not willing to risk this level of commitment. Compare it to self-driving cars if you want: it may be technically doable, but the practical and legal challenges are still enormous.
I should say that I have been working in automating medical imaging for more than 20 years now. 15 years ago, with others I demonstrated a system that diagnosed skin cancer better than humans, and won many awards for it. Nonetheless it was not a commercial success and our company folded. Maybe things will be different because larger companies are behind similar efforts now (like IBM) and they have the time, resources and money to make it happen. However make no mistake, getting these systems accepted is going to take a long time.
So I would suggest that rather than saying inflammatory things that sound like "we CS people do things better than you petty doctors", AI people should propose to work with medical schools to help train specialists to understand, use these systems correctly and benefit from them. Any other attitude is probably doomed to fail.
No, who are you going to sue. Will never happen.
"There are no gods, no devils, no angels, no heaven or hell. There is only our natural world. Religion is but myth and
...from what I hear. They basically outsourced the analysis of X-Ray and other scans to actually check problems.
This IS all going to collapse at some point. According to Silicon Valley, we should by now: all have robot cars, have colonized mars, be at the mercy of superior computer intelligence, be replicating everything we need with our own personal 3D printing replicators, have somehow cured cancer with computers, have virtual reality contact lenses glued to our eyes, be able to download data directly into our brains, somehow miraculously be able to survive exclusively on a nutrition shake, be having sex with robots that are indistinguishable from people, etc., etc., ad nauseum.
That list spans many years and is incomplete, there's plenty more crapola where all that came from. Are investors really this stupid? Or is it possible, as with the housing bubble, they really are just this greedy?
More and more, as with Jawbone, we are going to see formerly billion dollar companies simply evaporating. History will refer to the early 21st century as the 'Vapor Era'. At some point resources will be funneled back into reality.
You may or may not believe this, but in most of your major hospitals there is a software that determines the amount of live tissue left after a patient has a stroke. This software determines the amount of live tissue and course of action for the stroke team, long before the Radiologist has a chance to look at it. The images from the CT are sent first to the Stroke software then to the PACS archive for the Radiologist to read, the software determines the live tissue percentage and course of action in under 3min most times from my experience 1-2min. an e-mail is then sent with images and verbiage to the Stroke team and they start treatment, if the Radiologist disagrees with the software's findings in most cases the stroke team has already started on that course of action. :) ohh if its late at night and the ED is busy, usually there are only 2-3 Radiologist actually working thus the time it takes for them to get to that stroke patient's study takes a bit more time depending on the size of the hospital. where I work we perform a about 950,000 studies a year.
There will always be things that the AI will miss.
Twitter supports and protects racists - by smearing their critics with the "Hate Speech" label.
Siri, do you concur with Watson?
A 21st century issue: the irony of technologies of abundance in the hands of those still thinking in terms of scarcity.
Programmers should have a duty of social responsibility
to not develop AI which will eventually replace any worker.
The easiest person to replace is an AI developer.
Go well
In medicine Radiology appears to be the hi tech thing everyone except the patient or radiologist want to automate/eliminate. Sure radiologists are expensive, and hospital admins hate them for their costs, but it's still going to be awhile before they are replaced. It's not about identifying the problems, but preventing false positives. AI algos are good at the former, not the latter.
But of course, radiology is heavy image based, so all our computer vision and machine learning CS folks want to completely solve "the problem", aka automation, instead of complimenting the actual user (radiologist).
The last thing a radiologist wants is to ever meet or have to interact with a patient. Seriously, the folks in med school with the worst social skills go into radiology.
I'm a radiologist in practice for 20 years. I also earned a PhD in medical imaging and wrote software to do image analysis for my PhD thesis. I typically don't post to any newsgroups but feel compelled to respond as I have extensive experience in this particular area.
My experience with AI in my field has been poor, at best. We use an industry standard CAD system to help us detect abnormalities in mammograms. The software is so poor that it overdetects findings virtually 99% of the time and flags things of no importance. This decreases sensitivity to the point of being nearly worthless, and is often distracting. If you want another opinion, read this: http://www.diagnosticimaging.com/cad/state-cad-mammography
I also had some experience with software which attempted to detect polyps in the colon. It too overcalled findings (mainly stool collections which it thought were polyps) so frequently that we no longer use it.
A third CAD system I've used tries to detect pulmonary nodules that could represent a small cancer. It is so unreliable that the techs no longer use it as preprocessing to help me in my work. They can't make it work, and I don't have the time to troubleshoot problematic software for a vendor who doesn't support their product.
I have yet to see any CAD or AI that has positively impacted my ability to do my job. Am I interested and hopeful that someday it will? Of course. However, we can't even get the obvious or easy solutions working. I see no threat in having my job disappear during the remaining part of my career. I am far, far more concerned that continuing declines in reimbursement rates will damage my profession to such a degree that nobody will want to spend a minimum of 9 years of post college education and be paid the same rate as a plumber or an electrician who can earn the same amount without a college degree. I have already advised my children NOT to enter my profession because it simply isn't worth the massive debt and lost time one takes on in medical school for reimbursement rates similar to that of skilled contractors.
I hope so, they miss things too often. I have a weird congenital defect in my leg that can lead to blood clots later on in life. Only one radiologist ever picked up on the defect,even though it is really easy to see.
"Geoffrey Hinton, a legend in the field of artificial intelligence..."
Dude I've never heard of is a "legend"? Some legend! I think the OP has a different definition of legend than I do...
Maybe try to dial back the rhetoric a bit. "Respected, long-time figure" would be one example. "Experienced researcher" would be another. The OP has to figure out their own phrasing but undermines their message when using excessively fawning descriptors.