Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

AI Recognises Race in Medical Images

explainthispaper.com

Post reply on HN
Print view View original

261–270 of 357 posts

Re: AI Recognises Race in Medical Images

#261

Post by throwaway894345 »

JamesBarney wrote: post #209
andreyk wrote: post #156

Earlier quoted context omitted.

From what I understand, the definition of a social construct is a bit nuanced, and it's possible to both believe race is a social construct and that it correlates with things. As a metaphor, let's say we create two categories in the world for all people - tallers (people above six feet in height) and shorters (people below six feet in height). Human height objectively exists, but these categories are social construct…

So if you were going to create a test for "Is this a social construct?" What would it be? It seems like the definition from your comment would be "Could we take this labeling system, and define different labels or the labels differently?" I watched some of the youtube, and in the thought experiment she proposes she take a continuous trait(height) and arbitrarily splits it into two buckets. And talks about how this is…

> I watched some of the youtube, and in the thought experiment she proposes she take a continuous trait(height) and arbitrarily splits it into two buckets. And talks about how this is kinda silly.

Something can be continuous but still very clustered, and the extent to which it's "silly" depends on the uniformity of the distribution. It can still be useful to label the clusters in the distribution.

Another reason to subdivide a continuous dimension is that there could be a threshold after which some other dependent variable begins its inflection point (think of a hockeystick distribution). For example, there's probably some blood pressure value beyond which we begin to rapidly see serious adverse health effects--it's useful to call this "hypertension" or something. For another example, there's a threshold for the average temperature of the planet beyond which global warming "runs away". These are useful thresholds even though the dimensions are continuous.

Re: AI Recognises Race in Medical Images

#262

Post by lostlogin »

ttyprintk wrote: post #214
knicholes wrote: post #5

I wonder if it has anything to do with the machines that are being used to take the images. Maybe some groups have access to one type of imaging machine where other groups have access to some other type of imaging machine.

Or some scans are ordered for a diagnostic that’s more common in one group over another. I hope that the details show that treatment and diagnostic quality are indistinguishable between race classifications.

This is a good thought, but for all of said group to have the condition would be concerning.

Re: AI Recognises Race in Medical Images

#263

Post by oh_sigh »

runnerup wrote: post #227
oh_sigh wrote: post #187

Earlier quoted context omitted.

But with the taller/shorter, you would commonly have two tallers producing a shorter, or two shorters producing a taller. If we were talking about husbandry, we would say that tallers and shorters don't "breed true". But with race, you never have two white people produce a black person, or vice versa (and no, a black person with albinism is not a white person).

"never" is too strong here. Two black people occasionally produce a white baby without albinism. There's a lot of genes that code skin color and you can have a bunch of recessive "white skin" genes and still be very black-skinned yourself. Then sometimes your child gets all the recessive white-skin genes from both parents. I had a personal friend who was white with two black parents, not albino, and genetic tests pro…

Well, if they are black people with white ancestors(or white people with black ancestors), then they aren't the people I was thinking about. Almost certainly, the only reason your black friends produced a white baby, is because they had white ancestors at some point.

If two doberman pinschers had a puppy that looked like an English bulldog, it would be strange and newsworthy. But, if the two doberman's actually had grandparents that were English bulldogs, the mystery would be fairly easy to solve.

At least part of this confusion is associated with the culture of race(at least in the US) as opposed to the genetics of race. For example, we consider Barack Obama black, but he is equally as white as he is black. There's no genetic basis for making that kind of determination.

Re: AI Recognises Race in Medical Images

#264

Post by HideousKojima »

tpm wrote: post #38
happytoexplain wrote: post #19

Earlier quoted context omitted.

Yes, some anatomical differences are certainly affected by environment, but the quoted phrase is typically used in the context of treating races differently based on behavior ("X people are immoral", "Y people are stupid"). It's meant to encourage people to treat each other with equity and consider that such observed differences may be distorted by news, your like-minded cohorts, sociological conditions, etc, which a…

The idea that races have anatomical differences is very much contested, and rightly so.

You'd better hope your anesthetist doesn't believe that next time you need surgery: https://pubs.asahq.org/anesthesiology/article/107/1/4/8323/E...

Re: AI Recognises Race in Medical Images

#265

Post by throwaway894345 »

awild wrote: post #233
JamesBarney wrote: post #209

Earlier quoted context omitted.

So if you were going to create a test for "Is this a social construct?" What would it be? It seems like the definition from your comment would be "Could we take this labeling system, and define different labels or the labels differently?" I watched some of the youtube, and in the thought experiment she proposes she take a continuous trait(height) and arbitrarily splits it into two buckets. And talks about how this is…

> Maybe these things are social constructs, but if they are then we surely must come to the conclusion that almost everything we care about in the world is a social construct. That's the point really. And it's not inherently a value judgement to say something is a social construct. Using it as a value judgment usually is meant to convey that some secondary attributes are not inherent, or even more so, just historical…

In my experience, the implication is that things which are social constructs have no basis in reality and are thus not useful (and their use is harmful). I'm often inclined to respond, "money is also a social construct", but I'm afraid they would agree and proceed into some anti-capitalist, utopian diatribe.

Re: AI Recognises Race in Medical Images

#266

Post by naasking »

wwalexander wrote: post #248
naasking wrote: post #180

Earlier quoted context omitted.

> This is the definition I’ve always heard It's not as precise as you think, because fertility isn't transitive. Consider members of a species, M1, M2, F1, F2. M1 might be fertile with F1, M2 with F2, but M1 may not be fertile with F2. Are they all really members of the same species? Read up on the species problem for more information (there are now 26, not 24): https://en.wikipedia.org/wiki/Species_concept

I can’t admit I have any qualifications to speak about biology, and I know HN is known for proposing reductive “solutions” based on oversimplified models of reality. That being said, it seems to me that it makes sense that a species would include the entire connected graph whose nodes are members and whose edges represent the ability to have fertile offspring. I found [1] which is an interesting example of a very lar…

> They don’t inject arbitrary social categories into the definition. Definitions of “race” have no empirical basis on which to be proved or disproved in the first place.

That's simply not correct. The literature abounds with all sorts of correlates with race, like propensity for sickle cell anemia, vitamin D deficiencies, susceptibility to alcohol, and morphological differences. These are just as empirically justified as any classification of species, and just as with species, not all of those properties need apply to every single member in that category.

So to the extent that we find "species" a meaningful category when applied correctly, then we should also find "race" a meaningful category when applied correctly. The key in both scenarios is to apply them correctly, and we should abandon them when we find more precise metrics.

That said, you are correct that there are also numerous cultural and social properties that are sometimes lumped in with race in a manner that we don't see with species, mainly because "species" hasn't been politicized. That doesn't imply that there's nothing "there" once you tune out that baggage.

Re: AI Recognises Race in Medical Images

#267

Post by lostlogin »

We were marvelling at a surface shaded render made on a new Siemens MR from an T1 MPRage on a very still and compliant patient. It basically looked like a black and white photo (though with the tools at hand we could cut the image in half and look at the brain). You could see the facial hair and you could identify the patient if you knew them. Medical imaging is moving along at pace and it would be interesting to see what could be inferred from a dataset of images of this quality.

Re: AI Recognises Race in Medical Images

#268

Post by kovek »

CWuestefeld wrote: post #81
cubano wrote: post #31

Earlier quoted context omitted.

Huh? What do you mean, not working? That the AI was randomly choosing the correct race 82% of the time by luck? I'm confused by what your implying because it would seem to me that the authors went through many steps to try to pinpoint how the AI was doing this identification and how baffling it was to everyone that even with a lot of x-ray information removed (8x8 pixels compared to say 4k), it somehow was still corr…

I'm just making this up, but... Perhaps hospitals that treat a disproportionate share of poor people (which themselves are disproportionately not white), tend to use a different brand of X-ray film, and that brand has different contrast ratios than that of the brand preferred by rich hospitals. Thus, they'd be detecting the different brand of X-ray film rather than anything about the patients themselves. Of course, a…

That sounds like a great idea and they can test for it! Classify the scans on the “type of film” and then alter the scan and see if the model recognizes it

Re: AI Recognises Race in Medical Images

#269

Post by throwaway894345 »

heywherelogingo wrote: post #241
throwaway894345 wrote: post #215

Earlier quoted context omitted.

I don’t know what point you’re making, but my “definitional distinction” remark meant that “sex” is more or less defined by the ability of one group to bear children (fertility issues aside, of course). Not sure if that addresses your concern.

There is no "concern". Think about the other part of your comment rather than the gender bit. Good luck.

If you insist on being unclear I'm not going to work to tease out your meaning. Have a good day.

Re: AI Recognises Race in Medical Images

#270

Post by throwaway894345 »

ceejayoz wrote: post #154
throwaway894345 wrote: post #151

Earlier quoted context omitted.

You're misunderstanding the context of the thread. At this point in the thread, we're talking about whether or not there is any biological basis for race, not about the AI example in particular. The premise here is that there is a biological basis--that race is a label applied to a cluster of genetic traits (various facial shape features, skin color, etc) which are almost certainly genetically constructed.

The comment I replied to stated "clearly there must be some biological basis for race, if an AI can determine race from an x-ray". I'm pointing out that's not necessarily true. First, we have to prove that the AI is determining race from an x-ray. It's effectiveness at doing it via 64 pixels makes me skeptical that it is. After that, we get into sticky territory determining whether any biological differences we can i…

Ah, my mistake.
Post reply on HN