TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Deep learning algorithm diagnoses skin cancer as well as seasoned dermatologists

596 点作者 capocannoniere超过 8 年前

35 条评论

rscho超过 8 年前
As MDs, I think it is very clear that all of us who understand even the slightest about computers and tech see that machine learning is the way to go. Medicine is ideally suited to ML, and in time, it will absolutely shine in that domain.<p>Now for people eagerly awaiting the MDs downfall, I think you are precipitating things a bit. We all tend to believe in what we do, and I concur in saying that expert systems will replace doctor judgement in well-defined, selected applications in the decade to come. But thinking that the whole profession will be impacted as hard as factory workers, with lower wages and supervision-only roles, is not realistic. What will be lacking is the automation of data collection, because you seem to underestimate by far the technical, legal, and ethical difficulties in getting the appropriate feedback to make ML appliances efficient. I firmly believe in reinforcement learning, and as long as the feedback system will be insufficient, doctors will prevail, highly-paid jerks or not.<p>I myself am an anesthesiologist, a profession most people think of as a perfect use case for those techs (as I do), and wonder why we haven&#x27;t been replaced already. The reality is that the job is currently far beyond what an isolated system could do. We already have trouble in making cars follow the right lane in non-standard settings. I hope people realize that in each and every medical field, the number and complexity of factors to control is far greater than driving in the right lane.<p>People who drive the medical system have no sense of technology. They cannot even envision the requirements for machines to become efficient in medicine. That is why we are seeing quite a lot of efficient isolated systems pop up, but we won&#x27;t be seeing fully integrated, doctor-replacement systems for a long time. This will require a new generation of clinical practitioners, who will understand how to make the field truly available to machine efficiency.
评论 #13487037 未加载
评论 #13486999 未加载
评论 #13486286 未加载
评论 #13486537 未加载
评论 #13488136 未加载
评论 #13487258 未加载
评论 #13488278 未加载
评论 #13486187 未加载
评论 #13486127 未加载
评论 #13488414 未加载
评论 #13486578 未加载
评论 #13486113 未加载
评论 #13486292 未加载
评论 #13487113 未加载
评论 #13487444 未加载
评论 #13488097 未加载
评论 #13492012 未加载
评论 #13486793 未加载
评论 #13486935 未加载
评论 #13487015 未加载
评论 #13486591 未加载
评论 #13488492 未加载
评论 #13486121 未加载
brandonb超过 8 年前
This is the second major study applying deep learning to medicine, after Google Brain&#x27;s paper in JAMA in December, and there are several more in the pipeline.<p>If you&#x27;ve developed expertise in deep learning and want to apply your skills to healthcare in a startup... please email me: brandon@cardiogr.am. My co-founder and I are ex-Google machine learning engineers, and we&#x27;ve published work at a NIPS workshop showing you can detect abnormal heart rhythms, high blood pressure, and even diabetes from wearable data alone. We&#x27;re working on medical journal publications now based on an N=10,000 study with UCSF Cardiology.<p>Your skills can really make a difference in people&#x27;s lives. The time is now.
评论 #13485111 未加载
评论 #13487756 未加载
评论 #13485254 未加载
评论 #13487280 未加载
评论 #13485600 未加载
iamleppert超过 8 年前
Honestly, I can&#x27;t wait for deep learning and computational methods to dethrone doctors and upend the medical profession. In the next five years, expect a computer to be able to predict most diseases a lot better than doctors can -- and with none of the attitude, high cost, or inconvenience.<p>Mind you I&#x27;m not talking about researchers, who will always have a job. I&#x27;m talking about practitioners. I&#x27;ve had a medical condition from birth and I&#x27;ve had to deal with my share of doctors. Outside of the insurance system, they are easily the most unpleasant part of the whole ordeal to deal with. There are some gems, but most you will encounter are pompous, arrogant, and &quot;commanding&quot; -- when they enter a room, they are flanked by &quot;residents&quot;, &quot;assistants&quot; and generally give off this air of superiority which is really just because of their route experience. The whole thing comes off more as a performance than anything else. Worse, they often get mad when you question them or ask them to explain themselves, or how they arrived at a conclusion.<p>Good luck finding work when an algorithm can do your job better than you. It&#x27;s only a matter of time.
评论 #13485503 未加载
评论 #13484846 未加载
评论 #13485694 未加载
评论 #13484951 未加载
评论 #13485811 未加载
评论 #13487137 未加载
评论 #13485330 未加载
评论 #13484960 未加载
评论 #13485124 未加载
评论 #13485712 未加载
romaniv超过 8 年前
Systems that outperform doctors in some specific area of diagnostics aren&#x27;t new. One of the earliest examples of such systems is Mycin [1], which also was developed at Stanford, but around forty-something years ago. Never went to production because of practical issues that have nothing to do with its accuracy. It&#x27;s interesting that all of those &quot;practical issues&quot; are no longer relevant, and yet we don&#x27;t see a widespread use of similar software.<p>[1] - <a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Mycin" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Mycin</a>
评论 #13485013 未加载
评论 #13484817 未加载
评论 #13485137 未加载
btilly超过 8 年前
This reminds me of a talk that I saw about wavelet based algorithms in the 1990s for detecting tumors in mammograms.<p>The algorithms found most of the tumors that humans had missed, with similar false positive rates. BUT humans refused to work with the software!<p>The problem was that the software was very, very good at catching tumors in the easy to read areas of the breast, and had lots of false positives in more complicated areas. Humans spent most of their effort on the more complicated areas. Every tumor that the software found that the human didn&#x27;t simply felt like the human hadn&#x27;t paid attention - it was obvious once you looked at it. The mistakes felt like stupid typos do to a programmer. But the software constantly screwed up where you needed skill. The result is that humans learned quickly to not trust the software.
评论 #13485561 未加载
transcranial超过 8 年前
Unfortunately the paper is in Nature, paywalled, instead of Arxiv, and data&#x2F;code&#x2F;model&#x2F;weights inaccessible. While publishing in Science&#x2F;Nature&#x2F;NEJM&#x2F;JAMA is definitely the right approach for deep learning to gain validity in the medical community, faster progress could be gained by having a more open platform, with constant and real-time validation with more data, more medical centers and clinics. The reason progress in DL has been so breathtaking is in no small part due to the culture of openness and sharing.
评论 #13487622 未加载
nbmh超过 8 年前
This is interesting and impressive work, however, I noticed that they compared the algorithm&#x27;s performance to dermatologists <i>looking at a photo</i> of a skin lesion. This seems like a straw man comparison because any dermatologist would presumably be looking directly at a patient and would benefit from a 3D view, touch, pain reception etc. I realize that this was the only feasible way to conduct this study, but still suggests that an algorithm looking at a photo cannot match the performance of a dermatologist looking directly at a patient.
评论 #13487747 未加载
doesnotexist超过 8 年前
Eric Topol puts this up there as the most impressive AI&#x2F;medicine publication to date. <a href="https:&#x2F;&#x2F;twitter.com&#x2F;EricTopol&#x2F;status&#x2F;824318469873111040" rel="nofollow">https:&#x2F;&#x2F;twitter.com&#x2F;EricTopol&#x2F;status&#x2F;824318469873111040</a><p>The paper ends with &quot;deep learning is agnostic to the type of image data used and could be adapted to other specialties, including ophthalmology, otolaryngology, radiology and pathology.&quot;
ThomPete超过 8 年前
As someone with two melanomas under my belt (and more than a 1000 moles) what I really want is the ability to do a mass scan of my body also further down at the cellular level not just looking at the moles on the surface.<p>I am lucky enough to have Sloan Memorial as my hospital and no other than Dr. Marghoob one of the leading experts and I actually have a scan of my body made with 50 or so High Definition Cameras (I am litterally a 3d model in blue speedos and with a white net on my head).<p>They have a new system where they can look at the cell level without doing a biopsy and actually found my melanoma before they did the biopsy (i.e. they knew it was melanoma before they did biopsy) but it&#x27;s really a cumbersome process and I had 6 experts studying and working to position that laser properly.<p>So the real challenge today is how do we get the data into the system.
lucidrains超过 8 年前
This is why we need a platform for these models asap. I would totally download this app today and use it regardless of what the FDA thinks.
评论 #13484916 未加载
评论 #13484816 未加载
calebgilbert超过 8 年前
This is not hard to imagine at all. I know that there must be some absolutely excellent doctors out there, but I don&#x27;t trust the bottom 80% of doctors much at all, and honestly would rather have an algorithm most of the time, especially starting off. The lack of robust consumer level &#x27;medical doctor apps&#x27; is one of the biggest mysteries to me.
rawnlq超过 8 年前
There&#x27;s an app used by over a million doctors called &quot;Figure 1&quot; that allows them to share medical images for crowdsourced diagnosis and treatment of rare cases.<p>I wonder when we will get to a point where machine learning can help there?<p>[1]<a href="https:&#x2F;&#x2F;figure1.com&#x2F;medical-cases" rel="nofollow">https:&#x2F;&#x2F;figure1.com&#x2F;medical-cases</a>
ChuckMcM超过 8 年前
I read the headline and wondered how ml could train the difference between a new dermatologist and a seasoned one. Cancer I get, looks totally different than non-cancerous skin :)<p>That said, pulling this is one of the best ML applications to date. Recognizing cats or scenery doesn&#x27;t seem nearly as useful
lscholten超过 8 年前
Great results! Deep learning has been gaining track in other medicine areas as well.<p>One such task is lung cancer nodule detection from CT scans. A paper I recently co-authored applied many different architectures to this detection and achieved very good results. (<a href="https:&#x2F;&#x2F;arxiv.org&#x2F;pdf&#x2F;1612.08012.pdf" rel="nofollow">https:&#x2F;&#x2F;arxiv.org&#x2F;pdf&#x2F;1612.08012.pdf</a>)<p>The best combination of systems detected cancer nodules which were not even found by four experienced thoracic radiologists.
评论 #13485864 未加载
sungam超过 8 年前
Dermatologist here. Most skin cancer diagnosis is relatively straightforward and if the lesion is suspicious will require a biopsy to establish the subtype of the cancer and plan further treatment. There is no reason why this initial visual diagnosis cannot be performed at the same level as a dermatologist by a machine or indeed by a non-doctor trained intensively for a relative short period to interpret photographs.<p>The difficulty is two-fold. Firstly liability - a dermatologist aims not to miss a single case of melanoma in the tens of thousands of patients seen over their career, if this algorithm is used widely in millions of patients then either the sensitivity will have to be higher and more biopsies performed or there will have to be an acceptable rate of missed diagnosis for melanoma.<p>Secondly, in edge cases such as moles that are slightly atypical. In these scenarios there is no way that I would be comfortable making an assessment from a photograph. Now of course, a machine could also gather further information via methods such as in vivo confocal microscopy but in this case the cost savings are likely to be negligible.
hughdbrown超过 8 年前
Can someone clarify for me how the training and testing sets were constructed? One problem is that cancerous and benign skin are unbalanced in a representative population. How was this imbalance handled in testing? How was the testing set constructed? And so on.
评论 #13489822 未加载
kevinalexbrown超过 8 年前
One major, major advantage that medical imaging has for deep learning is the similarity of each data point, especially the &#x27;background data.&#x27; For instance, human brains typically look very similar across individuals (up to scanning parameter differences), except in the abnormalities - which are often precisely what you want to highlight.<p>As an example, I recently trained a neural neural network to perform a useful task for our lab using 3 (!) hand-labeled brains.
评论 #13485564 未加载
habosa超过 8 年前
Diagnosis based on image recognition is something machines are already very good at, even without recent deep learning techniques (although I am sure they will help).<p>For instance in college I worked with a radiologist to write an image-recognition program to identify osteoporosis from 3D MRI data. We used some super-basic image segmentation algorithms to identify the bounds of the bone layer that we cared about. From there a model was able to determine mechanical properties of the bone and therefore make an assessment with much more granularity than the human eye.<p>This was a first-year grad student class and I was coming at this totally naive with some Matlab scripts, and we managed to get usable results in weeks.<p>Here&#x27;s a sample of that professor&#x27;s research: <a href="https:&#x2F;&#x2F;www.ncbi.nlm.nih.gov&#x2F;pmc&#x2F;articles&#x2F;PMC2926228&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.ncbi.nlm.nih.gov&#x2F;pmc&#x2F;articles&#x2F;PMC2926228&#x2F;</a><p>While I am not in the camp of &quot;machines will replace doctors&quot;, I think radiology and other similar fields are in for a sea-change in technique and a large reduction in the use of human judgement.
drfritznunkie超过 8 年前
Coming from a family of people in the medical professions, they&#x27;ve all seen reports of how _everything_ is going change in their fields because some new computer program can do X...<p>To which my father usually mutters something like: &quot;Why fuck are they wasting their time with that? Can&#x27;t they fix the fucking medical billing system instead?&quot;<p>Most of the medical professionals I know echo similar sentiments.
评论 #13485939 未加载
the_watcher超过 8 年前
Telemedicine has a lot of regulatory hurdles to get to market, but initiatives like this are extremely exciting, since they can likely be taken to market in a way that explicitly clarifies that it&#x27;s not a diagnostic, it&#x27;s simply a low barrier way to actually get that mole you&#x27;ve got looked at. If you don&#x27;t have health insurance, you could actually get an idea of how critical it is to get in to see a doctor. That said, the obvious concern would be the extreme cost of a false negative (although the evidence suggests that the algorithm is no more likely to provide one than a doctor, the concern over single accidents caused by self-driving cars, even when the overall rates are far lower makes it pretty clear that the bar for success to the public for non-humans is substantially higher than it is for humans)
评论 #13485216 未加载
评论 #13485030 未加载
jwtadvice超过 8 年前
In my opinion the way to stage these technologies is not to blitz toward a fully cyborg doctor replacement, but to bolster the capabilities of the doctor with new technology - similar to how calculators did not replace mathematicians (despite historical headlines suggesting this would happen).<p>Giving a doctor the ability to get a &quot;second opinion&quot; fast and cheaply to a patient is a large boon to medicine, and shouldn&#x27;t be underestimated. It allows the doctor to deal with all the nuance limited automated tools can not, and gives the MD the ability to check themselves against the computer. If the MD finds themselves disagreeing on something like a skin condition, the feedback can both improve the doctor&#x27;s service and provide bug information for the code and databases used to train the AI.
caycep超过 8 年前
I wouldn&#x27;t be surprised at tasks that involve image recognition - these include dermatology (visual inspection) and pathology. In fact, I wouldn&#x27;t be surprised if CNN&#x27;s were better at pathology as every time I looked at microscope slides, there is so much &quot;visual clutter&quot; in a typical tissue specimen that I&#x27;m sure I was missing a ton of information on the slide.
EternalData超过 8 年前
This is going to be part of a greater trend of automation starting to affect fields considered to be white collar and paths to prosperity. I think the same is going to happen with financial analysts, entry-level lawyers etc. It&#x27;ll be interesting to see the political response, especially given how charged the atmosphere has become around &quot;preserving&quot; jobs.
kafkaesq超过 8 年前
A significant finding, to be sure. But like the paper itself says:<p><i>Here we demonstrate classification of skin lesions using a single CNN, trained end-to-end from images directly, using only pixels and disease labels as inputs. </i><p>What they achieved was algorithm to <i>classify skin lesions</i> - not perform a &quot;diagnosis&quot; of the overarching pathology, i.e. skin cancer.
kumarski超过 8 年前
Skin conditions are one of the few modalities where ML make deep sense as a diagnostic.<p>I think pharmacovigilance is the other area based on my interaction with the industry of folks at pharma and healthcare provider companies who work in ML.<p>Disclaimer: i run mlweekly.com and help at semantic.md
bluenose69超过 8 年前
What about 3D aspects? The word &quot;bump&quot; is used in most descriptions I&#x27;ve seen online, although I don&#x27;t know if that is something the doctors consider or just something that&#x27;s enough to suggest a visit to the doctor.
yalogin超过 8 年前
These new methods appear to best suited to be used in the pet world sooner as the ethical and legal issues will be a bit less stringent than in a human context. May be that is where things will start to change.
WalterBright超过 8 年前
My (old) dermatologist could spot skin cancer from across the room. I asked him how he could do that, he said he&#x27;s seen a million of them. It&#x27;s the same idea as &quot;deep learning&quot;.
james_niro超过 8 年前
I truly believe with smart algorithms and big data we can change the way we live. Smart medicine, proper diagnoses and early detection of disease we can improve our lives
arikrak超过 8 年前
On a related note, does anyone know how IBM&#x27;s Watson health is doing? They&#x27;ve been developing it for years but I haven&#x27;t heard much about their results.
sekou超过 8 年前
Even though diagnosis is only one piece of the puzzle, what I would hope is that this becomes part of the answer to the high cost of healthcare.
trhway超过 8 年前
so, basically while i&#x27;m taking shower the HAL ... err ... Google Home cameras in the shower would check for moles development, blood O2 from the color of skin, vascular health from the reaction to the water temperature, pulse from visible pulsations, mental and other conditions from the eyes movements, etc...
kazinator超过 8 年前
Only as well? Not faster and cheaper?
adamnemecek超过 8 年前
Basic income can&#x27;t come soon enough.
评论 #13492733 未加载
评论 #13485130 未加载
zxcvvcxz超过 8 年前
Quick, someone tell me why doctors won&#x27;t be obsolete in 20 years!<p>Geoffrey Hinton believe that we should stop training radiologists <i>now:</i><p><a href="https:&#x2F;&#x2F;twitter.com&#x2F;withfries2&#x2F;status&#x2F;791720748624797697?lang=en" rel="nofollow">https:&#x2F;&#x2F;twitter.com&#x2F;withfries2&#x2F;status&#x2F;791720748624797697?lan...</a>
评论 #13487688 未加载