> Many unwittingly used a data set that contained chest scans of children who did not have covid as their examples of what non-covid cases looked like. But as a result, the AIs learned to identify kids, not covid.
> Driggs’s group trained its own model using a data set that contained a mix of scans taken when patients were lying down and standing up. Because patients scanned while lying down were more likely to be seriously ill, the AI learned wrongly to predict serious covid risk from a person’s position.
> Errors like these seem obvious in hindsight. They can also be fixed by adjusting the models, if researchers are aware of them.
I wouldn't trust any method of fixing errors in data collection. You need to redo data collection and perform a proper case-control study. You need to select your cases and controls such that the demographics look similar.
In general, your goal is not to pick cases out against a background of the general population. It's to pick cases out from an at risk group.
And if there's any difference in procedure used for data collection between cases and controls, your data is compromised. Researchers often suggest correcting the dataset (cropping out artifacts etc.). In my opinion this doesn't really work... artifacts (for example in image data) can often cause subtile global differences in illumination which ML will pick up on but not be obvious to someone inspecting the images.
The only solution when you have such artifacts (and only one that would be acceptable to me if I was doing DD, or evaluating the research) is to redo data collection in the context of a more reasonable case-control study.
If properly done, that approach would most likely generate a model that works really well and detects COVID for the right reasons. Put it out in the wild and it breaks down again because all radiologist would have to use the exact same method of scanning patients. Practice shows that scan parameters, position, annotations vastly differ across hospital, scanner and radiologist.
A client of mine wanted me to look into this dataset and help creating a model for detecting COVID. One of the doctors they where working with wanted to be able to submit pictures they took (from their point and shoot camera) of the CT scan. Good luck putting that variation in your dataset.
It's a first step I guess. At least if you can get the classification to work with controlled data collection you have reasonable evidence that the approach can work.
If all other parameters area randomized between case and control, this is also fine. I'd guess you can also add illumination and other artifacts to the datasets to make the training robust to this.
But ultimately in the extreme case (like wanting to submit images from a point and shoot camera) I suspect you'll have a hard time building a system that's robust to that... or robust enough to be used in a diagnostic context...
I personally don't think I'd be comfortable working without a standard configuration used by the radiologists and a controlled protocol.
>
“The models are so similar—they almost all use the same techniques with minor tweaks, the same inputs—and they all make the same mistakes,” says Wynants. “If all these people making new models instead tested models that were already available, maybe we’d have something that could really help in the clinic by now.”
This is similar to people wanting to volunteer after disasters, but get mad when they're told they'd be more in the way than helpful.
It's selfishness and narcissism masquerading as philanthropy. Instead of doing an actual useful task, they want credit for the exciting/pioneering front line work.
It was the same thing with the ventilator shortage. How many people publicly declared they were "inventing" a new, easy to make ventilator? Design was never the problem, manufacturing was. But few were interested in manufacturing extant (and already FDA approved) designs.
Very early on in the pandemic one research group opensourced "COVID-Net", using the described flawed dataset.
Given the fact that at that time there was still quite a lot of "if we all help we can beat the pandemic" positiveness going around. An easily run-able piece of small code that promised "results" was available and the fact that a lot of organizations wanted positive PR, you get these insultingly low quality "research" practices.
I looked at a lot of the initial papers and models and most were forks of COVID-Net or used an already established model like ImageNet and retrained it with the bad dataset. Presto! Paper out PR happy. Very few questioned the dataset or the approach in general. But they loved bragging about it on LinkedIn.
COVID exists, it causes physical changes and affects the world outside the host. 100% it can be modelled somehow in theory.
That being said, there are harsh rules against collecting data (for example, it would be extremely illegal to create a sample set by taking a random group of people and infecting half of them with COVID on purpose). So I expect actually getting training data is a huge hill to climb. If most of the models used the same training data I assume there is a reason like that.
The entire premise of modelling something rests on the fact that the behaviour of what you're trying to model in the future depends on its behaviour in the past.
This is not true for all systems, it's not even true for most systems. John Kay differentiates between 'resolvable uncertainty' and 'radical uncertainty'. Modelling a volcano eruption is resolvable because it's a reasonably predictable system and you can make confident stochastic claims. Complex human systems are chaotic, have 'unknown unknowns', are simply subject to chance or non-linear behaviour that completely throws any prediction off.
You cannot model something that is subject to forces you cannot even know at the point of making the model, and this is true for any complex human system pretty much. Modelling Covid as if we have access to Hari Seldon's psycho-history was always a terrible idea.
What? We have to model human health. Find good proxies for it (non invasive scans, bloodwork, PCR, rapid antigen test, whole genome sequencing, anamnesis [medical history]) and there's ample to predict. Of course there are better and worse signals to use.
no, the correct behavior under uncertainty isn't modelling, it's something akin to Taleb's notion of antifragility or robustness. The correct response to a pandemic isn't some sort of Hari Seldon psycho-history which, as the article points out, is futile. The right response is building systems so responsive they can crush a pandemic before it gets to that stage.
Antifragility depends on knowing how to respond so that next time you'll be less fragile for the "same" stimuli.
For that we need better public health measures, those depend on modeling, vaccine development, etc.
There's no escaping understanding human health. It underpins and informs our responses. (Which kind of viruses are likely to pose a real threat, monitoring those threats, evaluating them, models, predictions, data, etc.)
I'm not talking about one big Seldon-esque equation, but about the public/global health system we already have, the whole of medical research, and so on.
These toy classifiers trained on COVID scans are in the same class as Tesla's dangerous "autopilot", because the system is too limited, it doesn't understand context, it lacks a world model. (Tesla's lack epistemic convergence and stability for object detection, ie. it lacks the understanding that cars, roads, obstacles don't flicker in and out of existence, nor do they change from frame to frame. Similarly a "medical AI" needs to understand that medical imaging artifacts are unlikely to be good proxies for real diagnosis, so it needs to be able to abstract away from the image, process away the artifacts and recording environment differences, etc.)
You think AI is so primitive that we can't make any headway into the biggest problem we currently have? That seems like a fairly fringe view. Especially on a problem that has well defined data like medical imaging.
Medical diagnosis has been one of the core ai domains domains since the early 1960s so people have been breaking the picks at that coalface for almost 60 years.
And those older systems has more intelligence in them. Todays’s “AI” is a small number of tricks aimed at large amounts of data, with an unprecedented and enormous amount of marketing added.
> And those older systems has more intelligence in them. Todays’s “AI” is a small number of tricks aimed at large amounts of data, with an unprecedented and enormous amount of marketing added.
If you're talking about deep neural networks, I can understand this viewpoint. But generally the last decade has proven widely successful for high quality vision recognition models that just weren't available before then. And with something like transfer learning, you can take a powerful off-the-shelf model and specialize it without a huge dataset.
The real barriers IME are around legal and privacy implications. There's also a strong argument about if these models creating enough value in the first place, but they can work on a technical level.
Oh yeah. Just as AI has not made the stock market predictable. There is (maybe) progress in pockets but not in general. Systems that change strongly and react to forecasts are really difficult to handle (and also stuff happens, eg new variants).
Also on the wet side, there is very little progress on something like general antiviral drugs.
My guess would be that it's hard to get a clean well labeled dataset where case and control datasets are otherwise similar. Particularly, in a pandemic the people doing the data collection are otherwise occupied.
Another problem you have is when you do it right the results look worse than a biased dataset. So there's some incentive to continue working with a biased dataset.
There's almost certainly something to model, you don't hear stories about hospitals filling up with patients and nobody knows if they're covid patients or how to deal with them; or at least, not anymore.
100s of wrong models doesn't sound extremely unlikely to me. It's really easy to train a model and test that it works on some dataset and then find it doesn't work on a different one. Setting up the inputs for training is difficult because of data privacy, there's intense pressure to publish, models have no expectation of explanability, lots of reasons for a wrong model that looks promising.
I've worked adjacent to people trying to solve problems with machine learning and it's rough going. When things work, ok nice, but when they don't, it's much harder to tweak than something based on heuristics (which, of course, don't always apply either).
100s of DL groups with little previous experience in this area set out to train similar models at the outset of the pandemic. Many rushed out models. I’m not surprised that they ran into similar pitfalls.
As someone with a background in CS (but not AI/ML) and working in bioinformatics:
People with a background in methods like solving problems that are easy to solve with the methods they know. Those are rarely the problems people in biology/medicine want to solve. Some common pitfalls include:
* Asking the right questions.
* Framing the questions correctly.
* Finding computational problems that capture the essence of the questions and that can be solved efficiently.
* Interpreting the results.
* Avoiding systemic biases in rare but important edge cases.
Finding the right problems to solve takes time, and the first attempts will probably fail.
The first attempts made by any particular person are likely to fail. And when many people approach similar problems from similar perspectives, they are likely to make similar mistakes. Give them a few years, and someone will probably figure out what the mistakes were and how to avoid them.
Right- thousands of people making the same predictable learning curve mistakes isn't thousands of man-hours of new research. It's thousands of (wo)men all duplicating the same few hours of research, which isn't likely to result in any breakthrough.
If 10 people hike to Mt Everest's base camp, it's not the equivalent of one person climbing ~10 times further to the peak.
The silly thing here is that it’s very easy to train models nowadays, so the medical AI space is very noisy.
With the right data and the right application, this technology can definitely be helpful.
However, it is very hard to sort out signal from noise since any reasonably competent undergrad can throw a convnet at some images and get an AUC.
Actually, I have even seen a high school science fair where they did a medical computer vision application.
I think this speaks more to the accessibility and ease of use of the technology that is out there than it does to the inherent limitations of the tools we have.
I am reasonably confident that there is a group out there that was more careful and has something that might have been useful, but it’s just impossible for clinicians to know who to trust.
I feel that both uncritical, un-nuanced hype (Radiologists will be replaced in 5 years!) and uncritical, un-nuanced closed mindedness (AI will never be clinically useful) is harmful for tackling the biggest problem the field faces right now: trust.
> I am reasonably confident that there is a group out there that was more careful and has something that might have been useful
What is that confidence based on? It seems like the researchers here made a reasonably thorough effort to find "something that might have been useful" and only turned up two that were even worth further investigation:
"She and her colleagues have looked at 232 algorithms for diagnosing patients or predicting how sick those with the disease might get. They found that none of them were fit for clinical use. Just two have been singled out as being promising enough for future testing."
So are you reasonably confident that these researchers just didn't look hard enough to find the good stuff?
Why not? Plenty of researchers aren’t thorough. I’m not saying that’s the case here but it’s entirely plausible from what I know about academia and wouldn’t be a crazy claim to make.
If a model did particularly well while almost every other model did badly, it would have gained attention in the community (not to mention, the researchers who worked on it would be actively marketing this!). Even if the researchers who worked on the paper weren't aware of it, it would almost certainly be brought up during peer review. The only way that such a model would not end up on such an extensive list is deliberate omission, not lack of rigor.
I don’t think this is true. Medical practices change slowly and it takes time to build trust. It’s possible a covid model will become clinically useful, but it probably requires having enough buy in to run a clinical trial and then for the economics to line up so that using it leads to more money for hospitals somehow
I think the issue is not with the training step but with the validation step. The only way to build enough trust is to carefully run a prospective clinical trial, and before that you probably need enough retrospective evidence to get some sort of FDA approval.
These processes take a lot of time and it’s not surprising that a year and a half after COVID-19 became widespread in the US, we still don’t have a fully validated model.
In fact the claim the paper makes is that no model has been sufficiently clinically validated.
Problem is: how do you get enough buy in from clinicians to run a clinical trial with YOUR model, when there are hundreds of crappy models out there? It takes a lot of time to run these studies for this reason to start with and then depending on the size of the clinical trial, it takes time for the trial to run it’s course.
I'm not sure if this has been attempted, so please do correct me, but I would very much like to see AI models designed to assist rather than give coarse classifications like "X disease detected". The best example I can think of would be creating bounding boxes around specific tells, or listing specific markers that a clinician would be looking for themselves, leaving the final decision up to them.
As a biologist who started working in ML, I see exactly the same issues in the field I am working in. I am spending most of my time fixing bad while others just use the public datasets as is. Maybe I am wasting my time because they are in the top journals and I am not.
The hardest part about ML is understanding the domain. I agree with Jeremy Howard on this, that it's easier to teach a domain expert to do ML than to teach an ML expert the domain. I think you are on the right path.
We have good tools for diagnosing COVID-19, and have had them from almost the start. If your looking for medical AI problems, start with the 12-lead EKG which fits on one piece of 8x10 piece of paper, is 2D, and current formulaic algorithms misread about 50% of the time when compared to a paramedic or nurse. Much less a physician, cardiologist, or electrophysiologist.
If this algorithm was better it would absolutely save lives and make everyday life easier for clinicians.
Most people working in the AI field are completely ignorant to the underlying mechanisms. They just hack together some code from github and feed some data in. Just because you get an output doesn't mean its even remotely accurate.
Anyone who does understand the underlying mechanisms are probably still trying to clean the data sets.
I vaguely remember reading more about this elsewhere and I believe they only used COVID positive patients. It would be interesting if there are follow-up studies with mixed training data if true.
Ah, ok. I think this was the article I originally read which implies they only used 2,500 covid positive patients. But rereading it, maybe it’s just unclear reporting.
“Across around 2,500 captured cough recordings of people confirmed to have COVID-19, the AI correctly identified 97.1 percent of them – and 100 percent of the asymptomatic cases.”
Yeah, I guess it would be cool to do an X-ray or CT scan of every patient, preferably in repeatable and controlled conditions similar to the AI training set. On the other hand there supposedly were very good results with training dogs to identify covid patients by smell alone. Which sounds faster, cheaper and probably even less error prone (judging by this article). I guess it doesn't scale well and there's not much money to be made, so there's that.
I think it may be the case with AI researchers where if the only tool you have is a hammer then every problem starts looking like a nail /jk
X-rays and CT scans involve some expense and radiation exposure. It makes no sense to subject every patient to such imaging studies. We have cheaper and safer tests to diagnose most cases of COVID-19.
Should have built AI tools to educate the ignorant populace and encourage them to get a vaccine. I know there have to be tons of adboys and girls out there whose only job is to persuade people to do something that could have knocked this out of the park. I see we are still at about 50% vaccination and Walmart is begging people to come get a free vaccine. I’m sorry my tone is negative but this is really getting me frustrated and making me lose hope in the future of humanity really. The stupid is winning. It’s the Asimov quote coming true. :( We’ve got to take real action against this.
"Anti-intellectualism has been a constant thread winding its way through our political and cultural life, nurtured by the false notion that democracy means that 'my ignorance is just as good as your knowledge."
I understand your frustration, but the condescending remarks are not conducive to wholesome scientific discussion.
Diversity is the way nature builds robust and balanced ecosystems - AI tools that reinforce ideological homogeneity could be catastrophic, for reasons I consider self-evident.
Returning to the topic at hand, and as a counterpoint to your claims, there are legitimate second and third order effects to consider regarding vaccines and the evolutionary dynamics of SARS-CoV-2. Don't take my word for it - here is peer-reviewed research from the brightest minds and top institutions in the world, raising serious concerns about approaching eradication of the virus purely through mRNA vaccines in their current form [1].
I cannot emphasize enough that the spirit of science is to question assumptions and rigorously prove hypothesis.
Social diversity of opinion is also to be expected. No matter what the leadership decides, there will be people that criticize it and there will be people that are strongly against it.
The reasons almost don’t matter. If the leadership doesn’t have at least two plans to deal with this expected situation, then they have failed.
> I cannot emphasize enough that the spirit of science is to question assumptions and rigorously prove hypothesis.
But that's not what we're getting. Instead, we're getting a large number of people who haven't spent ten minutes bothering to learn anything about the science posting obvious bullshit.
How is anyone supposed to get actual science done if nearly all the "questioning" is repetitive worthless lies by people who have absolutely no interest in science at all and are simply screaming obvious lies at the top of their voices?
A question I ask a lot of the vaccine deniers on my Facebook page is, "Can you give a quick explanation in your own words why COVID-19 is so good at getting past the human immune system?"
>How is anyone supposed to get actual science done...
One would hope that scientists would not be diverted by questions by people they do not work with or know on Facebook and are able to get actual science done regardless of a minority thinks.
I wonder if it's rational or a good thing to defend scientists by potentially restricting the ignorance of lay people in public places?
I think many would say it's good to avoid Harm and protect lives. It's easy to be irrational to defend rationality.
Edits. Oh, it could be that people are extending "follow the science" and seeing how others are not and thinking this as a kind of lack of faith. Perhaps the idea that actual scientists are being stopped doing science because of no faith in them by small numbers of the public is actually a religious thought. That actual science needs people to believe in it for it to work, and if an other doesn't believe it, they are anethema. That could have explain my feelings about the irrationality of this.
(also frustrated)
re: adboys and girls, the problem (at least in the US) is that vaccines are political, so persuading people to get them is seen as dangerously close to subverting democratic society. Same thing as taking action to counter foreign influence campaigns.
see also: the leaks to project veritas about Facebook's counter vaccine hesitancy program.
Well, back when Harris and Biden both said they were hesitant to get a vaccine produced under Trump, they kinda set the stage for the dumpster fire we have today.
That's evidence of media bias, not evidence that it wasn't said.
ETA: Here's Jezebel's pre election coverage of their statements, when everyone still claimed Trump was lying about the vaccine being close.
>On Sunday, during an interview on CNN’s State of the Union, Harris was asked whether she would trust a vaccine doled out by the Trump administration before Election Day.
>“I think that we have learned since this pandemic started, but really before that, that there’s very little that we can trust that comes out of Donald Trump’s mouth,” Harris said.
Several state governments including California stated that they wouldn't trust FDA vaccine approvals and wanted to do their own safety reviews. That didn't help with public confidence either.
The point was if he rushed it. The companies delayed their announcement and strengthened the trials so the current Trumpista favorite argument would in theory lose steam.
>The companies delayed their announcement and strengthened the trials so the current Trumpista favorite argument would in theory lose steam.
They [Pfizer] delayed the announcement to three days after the elections. I'm sure those three days were critical in testing and making the vaccine safe for the consumer./s
It was an obvious political move to deny back-then US president Trump a public success as a consequence from him enacting "Operation Warp Speed".[0][1][2]
Also, "Trumpista"? You must feel really clever and enlightened.
I live in New Orleans where unvaccinated cases are exploding and people in surrounding areas specifically cite Democrats as a reason not to take the vaccine. I’ll stand by my words. And I’m not fan of Pfizer trying to make vaccine policy by press release but didn’t one of the companies turn down Warp Speed money to avoid this perception?
All of the vaccines used in the US received funding through Operation Warp Speed. J&J and Moderna got R&D money and production orders. Pfizer only got production orders.
No. They didn't want Trump to have that "win." It was purely political. The vaccines they are pushing now are the very same ones that were developed and tested while Trump was in office.
It certainly wasn't purely political. It was based on the idea that the decision to approve the vaccine is the job of the FDA and that they have to pass their safety and effectiveness verdict.
The US FDA was among the first world-wide to approve (Europe was months later) which in my mind shows that there is not much to an argument that political forces delayed the announcement (and how could the opposition / democrats influence a government institution such as the FDA if the ruling party couldn't? Please no deep state bonkers).
> The vaccines they are pushing now are the very same ones that were developed and tested while Trump was in office.
I would challenge that Trump should receive any political credit for supporting the development of the vaccines, because when the development was happening at Moderna (US) and BionTech (Germany) in spring 2020 Trump was still rejecting SARS-COV-19 as irrelevant. Trump should receive credit for Warpspeed that made sure the production of the developed vaccines is happening and is happening in a way that benefits the American people first (to great frustration to many Europeans).
>It was based on the idea that the decision to approve the vaccine is the job of the FDA and that they have to pass their safety and effectiveness verdict.
You can't claim they were just following the rules, because they weren't. They bended and broke whatever rules they saw fit to get the vaccine done, including doing multiple phases simultaneously. If the FDA were following the rules, we still wouldn't have a vaccine available, because they still don't have full FDA approval.
>Trump was still rejecting SARS-COV-19 as irrelevant
Trump was called racist by Congressional D's for wanting to shut down the international travel to prevent the spread in February.
I got misled by parent. I knew that Biontech had received EU approval in December but from parents comment understood that in the US the approval came shortly after the elections. Sorry.
Still: don't you think the argument holds: US FDA is one of the first to approve is a sign that there can't be much delay.
Do you think if they had announced before the election and Trump had won all the current recalcitrants would be lining up in droves to take it, their concerns about rushing assuaged?
I've been telling people that the way to get people to take the vaccine would be to give Trump credit for operation lightspeed (which I think he deserves but that's irrelevant) and start referring to the vaccine as Trump's Vaccine.
I'm okay with handing Trump a win if it can help save lives and end this pandemic.
Doesn't that confirm what I said? That's, in Biden counties, 46.7% of people are vaccinated and in Trump counties, 35% of people are vaccinated.
Those numbers are vaccination rates in red vs. blue counties where the blue counties have higher rates than the red counties.
That interpretation is supported by the way the article is framing the data in the conclusion.
> Although there has been an overall significant slow-down in COVID-19 vaccination rates in the U.S., these findings show a widening divide of communities at risk for COVID-19 along partisan lines. A key component of any effort to boost vaccination rates among Republicans will be...
That's also a correlation of a correlation, which can be extremely misleading. It might be, for example, that Biden supporters in Trump-dominated districts are less likely to get vaccines.
Your idea wouldn't work. MAGAs hate liberals, and whatever liberals say appears to them like the voices of the grown-ups in the Peanuts cartoons. Maybe they pull out a few phrases to parody.
We've had over a year of near-continuous propaganda from the Republicans. Masks threaten your freedom. COVID is just the flu. No one dies from it, they put it down on every death certificate. Vaccinations kill you, not COVID. Vaccinations spread COVID, that's the new one.
Giving Trump credit for it would change nothing, even if the MAGAs heard it, and they wouldn't.
> which I think he deserves
It astonishes me that after years of watching Trump lie and lie and lie and lie, rambling and incoherent, often contradicting himself from day to day, watched eighteen months of him lying about COVID including that first key year where he did everything he could as President to prevent an effective response to COVID, that people somehow believe that he's secretly a competent administrator who "gets things done".
Is Trump not responsible for Operation Warp speed? The program that spent billions of dollars to stockpile untested vaccines while they were going through testing so that we'd have an ample supply to distribute in the event they were approved (on an accelerated time scale)?
Operation Warp speed is why, while everyone else is still struggling with supply, everyone the US who wants a vaccine can get a vaccine.
Unsurprisingly, the Biden administration did the exact opposite of what I suggested. Further entrenching COVID as a partisan issue.
> In January 2021, White House press secretary Jen Psaki announced that the program was expected to undergo a restructure and renaming under the Biden administration. Also in January 2021, Dr. Moncef Slaoui, former Operation Warp Speed lead, was told not to use the name Operation Warp Speed anymore.
So maybe the real mistake was making it free. IF a company could make a few bucks for every vaccine administered, we would have seen the ad industry being its full might to bear to harvest as many people getting it as possible.
I don't think there is an absence of software with the purpose of convincing the public to consume the right pharmaceutical products. I would say every owner of major internet property has adopted some kind of feature or process to promote them. Indeed, it has become quite the fashion statement to signal one's zeal for pop-science and one's contempt for the "deniers." I'm sure there's a market for 'Wear. The. Mask.' bumper stickers.
I would agree that stupid is winning, but I probably agree more with Kant on what Enlightenment means.
"About three-fourths of people infected in a Massachusetts Covid-19 outbreak were fully vaccinated, according to new data published Friday by the CDC.
The new data, published in the U.S. agency’s Morbidity and Mortality Weekly Report, also found that fully vaccinated people who get infected carry as much of the virus in their nose as unvaccinated people.
"
And another point related to education. Couple days ago a White House official stated :"We're dealing with an evolving virus. There is no precedent, and there is no playbook". Note, it isn't the White House of the previous buffoon, it is supposedly educated current White House. No wonder that being led by the people like this the current pandemic response isn't that much better than the ones of the previous centuries.
Despite virtually everyone in the town being vaccinated, 25% of the cases in the outbreak were in the remaining handful of unvaccinated people! As more areas reach close to 100% vaccination, we will start to see weird quirks of statistics like this more often, but if your takeaway here is that it's really easy for vaccinated people to catch and spread COVID, you're very wrong. Breakthrough infections are prevented by vaccines like Pfizer with about 99% efficacy, and without a breakthrough infection there is no transmission. If pretty much everyone were vaccinated, the odds of transmission would go to zero very rapidly.
Unless I'm reading your source wrong it seems to make the opposite point. Your source says that Provincetown has a 114% vaccination rate because the number of people who have used a Provincetown address to get vaccinated is greater than the number last counted in the census - 2k. However, currently, Provincetown, a summer vacation spot, has a population of 60k people.
Even if all of Provincetown residents were vaccinated that would not imply that anywhere near the majority of the people currently in Provincetown were vaccinated. I'm not sure how we can usefully estimate the vaccination rate of the people who have been in Provincetown recently. Google says the vaccination rate for Massachusetts is about 72%. That seems really worrying to me because if that vaccination rate describes the people in Provincetown then the infection rate implies that the vaccine is offering no real protection - i.e. 75% of population vaccinated and 75% of infected were vaccinated.
>What's neglected in that article is that essentially the whole town is vaccinated:
i don't think it was neglected, as it is more like obvious thing wasn't explicitly stated - as vaccine obviously not increasing infection chances, 3/4 of infected being vaccinated means that vaccination rate >= 3/4, and that pretty much means "essentially the whole town is vaccinated"
I guess you are trolling. And is really sad that such an one-sided argument as yours even exists.
First, you are missing that this is a Delta out-break. Delta has been found to be able to evade the vaccines more than other variants. This is something new and something we didn't know just 2-3 months ago.
Next, without stating how many percent of the population where vaccinated in MA what does three-fourths imply?
Governments world-wide are struggling with how to respond to Delta. Because it is more contagious and even for vaccinated people there is a risk of hospitalizations, there is a struggle what to do. Lift all restrictions (as England has done)? Stay rather strict (as in France/Germany where there are still public mask mandates)?
of course. Posting CDC data and WH quotes. While somehow nobody mentioned trolling wrt. GGP calling opponents "stupid" and "anti-intellectuals" without any data to back his position. Not surprising. Just a typical for these days debate with the "intellectuals".
>This is something new and something we didn't know just 2-3 months ago.
You are repeating the politicians' talking point what i was talking about . The very high chances that the "improved" 2.0 version would come were obvious pretty much from the start for anybody who paid even slightest attention to the precedents. Yet politicians have as always been playing ideologically charged short game all along right into populace ignorance while completely disregarding the long game.
The vaccine doesn't stop you from getting covid. It reduces the effects so you're less likely to need hospitalization. So saying x% of people with covid are vaccinated doesn't mean much. There's some education for you.
lets that sink. Especially considering that according to the above mentioned CDC data infected vaccinated people spread virus as infected unvaccinated.
>It reduces the effects so you're less likely to need hospitalization.
So, as current vaccines don't stop infection and spread, why would the most of the population whose risk of hospitalization is very low need (and even forced) to vaccinate?
>So, as current vaccines don't stop infection and spread, why would the most of the population whose risk of hospitalization is very low need (and even forced) to vaccinate?
>Superspreaders (antimaskers, antilockdown, antidistancing, antivaxxers) are the people keeping COVID alive.
It's not as black and white as you try to paint it.
I'm not living in the US but I've encountered a sizable number of people who dropped all caution and safety measures as soon as they got vaccinated even though they still can catch the virus and spread the virus with the added bonus of favoring the formation of immune escaping mutations.
SARS-CoV-2 is now endemic in the worldwide human population, just like several other coronaviruses. It will never go away regardless of what people do. Fortunately the vaccines can limit the death toll.
> Driggs’s group trained its own model using a data set that contained a mix of scans taken when patients were lying down and standing up. Because patients scanned while lying down were more likely to be seriously ill, the AI learned wrongly to predict serious covid risk from a person’s position.
> Errors like these seem obvious in hindsight. They can also be fixed by adjusting the models, if researchers are aware of them.
I wouldn't trust any method of fixing errors in data collection. You need to redo data collection and perform a proper case-control study. You need to select your cases and controls such that the demographics look similar.
In general, your goal is not to pick cases out against a background of the general population. It's to pick cases out from an at risk group.
And if there's any difference in procedure used for data collection between cases and controls, your data is compromised. Researchers often suggest correcting the dataset (cropping out artifacts etc.). In my opinion this doesn't really work... artifacts (for example in image data) can often cause subtile global differences in illumination which ML will pick up on but not be obvious to someone inspecting the images.
The only solution when you have such artifacts (and only one that would be acceptable to me if I was doing DD, or evaluating the research) is to redo data collection in the context of a more reasonable case-control study.