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

Midjourney Medical

midjourney.com

Post reply on HN
Print view View original

471–480 of 927 posts

Re: Midjourney Medical

#471

Post by disgruntledphd2 »

RobotToaster wrote: post #467
convnet wrote: post #287

Earlier quoted context omitted.

It's a controversial and complicated idea. The downside, and the reason why most doctors do not recommend full body scans, is that every human body is a bit weird and there will almost always be something "wrong" that will be visible in a full body scan. This can lead to unnecessary testing, anxiety, and even unnecessary procedures. Many of these oddities flagged by the scan would never have caused any actual issues…

This always feels like a thinly veiled excuse to ration healthcare. Would these same doctors refuse a full body MRI to a billionaire paying out of pocket? Anything found can be monitored with focused follow up scans. It doesn't have to be immediately biopsied if it's in a location where that would pose a risk of iatrogenic harm.

At a population level, this would be both extremely time-consuming, and rather expensive.

More generally, no test is perfectly accurate, and for low base rate conditions the vast majority of positive tests will be false positives.

Like, again, as a data person I adore this idea in principle, but there would be a lot of details that we'd need to figure out to make it a reality.

Re: Midjourney Medical

#472

Post by MagicMoonlight »

convnet wrote: post #287
tgsovlerkhgsel wrote: post #233

I think a lot of medical diagnosis could be solved with mass data collection if it was cheap enough. Right now, blood draws are somewhat routinely done because they provide a lot of human-interpretable indicators from a small number of values, and there is some evidence that e.g. "dogs can smell cancer" etc. (i.e. some diseases cause detectable odors). With a big enough data set of [all kinds of bio values, including…

It's a controversial and complicated idea. The downside, and the reason why most doctors do not recommend full body scans, is that every human body is a bit weird and there will almost always be something "wrong" that will be visible in a full body scan. This can lead to unnecessary testing, anxiety, and even unnecessary procedures. Many of these oddities flagged by the scan would never have caused any actual issues…

[dead]

Re: Midjourney Medical

#473

Post by multjoy »

ACCount37 wrote: post #468
moffkalast wrote: post #437

Earlier quoted context omitted.

If the current state is anything to go by, an automated test would not only flag your out of distribution results but try to gaslight everyone reading its output with additional false indicators to map you into an area that's in distribution. Statistical models cannot accept the existence of extremely rare edge cases.

Modern LLMs routinely beat human doctors at diagnosing "extremely rare edge cases". They have unmatched breadth of knowledge by default, and can maintain attention across entire medical histories.

Citation very much needed.

Re: Midjourney Medical

#474

Post by multjoy »

rlt wrote: post #414
icantevenhold wrote: post #400

Earlier quoted context omitted.

People take horse dewormer against COVID so yes they will do all kinds of irrational things

Oh we're still doing the "horse dewormer" thing despite 250 million humans taking it each year?

Yes, because it's nonsense and those 250m humans need to get off Twitter.

Re: Midjourney Medical

#475

Post by ahtihn »

bialpio wrote: post #463
blensor wrote: post #428

Earlier quoted context omitted.

How do you measure the body regularly without potentially introducing problems just by measuring it?

My understanding is that both MRIs and ultrasounds do not introduce problems.

MRIs by themselves no, but depending on what you want to actually see you need to inject a contrast agent which is probably not something you want to do too frequently.

Re: Midjourney Medical

#476

Post by fn-mote »

ACCount37 wrote: post #468
moffkalast wrote: post #437

Earlier quoted context omitted.

If the current state is anything to go by, an automated test would not only flag your out of distribution results but try to gaslight everyone reading its output with additional false indicators to map you into an area that's in distribution. Statistical models cannot accept the existence of extremely rare edge cases.

Modern LLMs routinely beat human doctors at diagnosing "extremely rare edge cases". They have unmatched breadth of knowledge by default, and can maintain attention across entire medical histories.

> Modern LLMs routinely beat human doctors at diagnosing "extremely rare edge cases".

There is a selection bias here. Not saying it wouldn’t work, but right now you hear about exceptional cases, not when the LLM wants to amputate for a wart.

We all work with LLMs, right? It hasn’t been long at all since an LLM gaslit me while attempting to recover an unbootable laptop. I should have been recommended a few simple steps to try; instead, it was unable to ignore the irrelevant details and led me on an hours-long chase. To me that means the LLM will also struggle to ignore irrelevant medical information.

Re: Midjourney Medical

#477

Post by RobertoG »

arcticbull wrote: post #273
tgsovlerkhgsel wrote: post #233

I think a lot of medical diagnosis could be solved with mass data collection if it was cheap enough. Right now, blood draws are somewhat routinely done because they provide a lot of human-interpretable indicators from a small number of values, and there is some evidence that e.g. "dogs can smell cancer" etc. (i.e. some diseases cause detectable odors). With a big enough data set of [all kinds of bio values, including…

Don’t make me tap the sign. Bayes Theorem: https://en.wikipedia.org/wiki/Bayes'_theorem There’s a very good reason we don’t test asymptomatic people in low incidence populations. Basically all positives are false positives when you do that, no matter how accurate the test is. When you’re testing healthy randos for everything the odds of a positive being false have so many 9s it would make an SRE weep. Unless this is…

The argument has some merits, but we should remember that, from the point of view of Bayes, you could apply the same argument to symptoms, which is only evidence. High odds of a false positive, means that you have not enough evidence, not that you have not useful information.

Re: Midjourney Medical

#478

Post by Dathuil »

seer wrote: post #423
convnet wrote: post #287

Earlier quoted context omitted.

It's a controversial and complicated idea. The downside, and the reason why most doctors do not recommend full body scans, is that every human body is a bit weird and there will almost always be something "wrong" that will be visible in a full body scan. This can lead to unnecessary testing, anxiety, and even unnecessary procedures. Many of these oddities flagged by the scan would never have caused any actual issues…

Exactly - I had switched to a one meal per day setup and have been mostly following it for a few years. Then after a routine “heart health” check all my indicators were super out of whack - the doctors thought I was on my deathbed - but I am perfectly happy pain free, in shape, physically active person… Then _i myself_ had to dig into all these tests and figure out that they were measuring the wrong thing - since the…

Reminds me that a few years ago my wifes grandfather (80+) was wondering if he should cut back on the amount of exercise he was doing. He would regularly be knocking on our door at 6am to see if one of us would want to got for a quick 10k run or to hit the gym.

He was a firefighter in NY in his youth and had never stopped exercising even after retirement.

He went to his GP explained his workout routine and was basically told there is no precedent for it as people his age tend to not be running 10km a day. In short he was told if you're not in pain or fatigued keep at it.

I think he's nearly 90 now and has cut back the running to only a day or so a week, but last time we went to visit he was in his garage bench pressing 50kg

Re: Midjourney Medical

#480

Post by srdjanr »

jibal wrote: post #331
appplication wrote: post #290

Earlier quoted context omitted.

So you are certainly correct but you can also tighten up your definitions for true positives as you have more information on your false positives. There may exist additional signal as well. To your point though I think there is a difference between collecting and evaluating additional data sources and using them as diagnostic tools. I suppose I fundamentally disagree with the implication of your post that there is no…

Positive for what, exactly? Quoting convnet, above: > The downside, and the reason why most doctors do not recommend full body scans, is that every human body is a bit weird and there will almost always be something "wrong" that will be visible in a full body scan. This can lead to unnecessary testing, anxiety, and even unnecessary procedures. Many of these oddities flagged by the scan would never have caused any act…

The root comment is talking about adding blood, breath, urin, spit... analysis. For body imaging only I agree with you. But if we add all this, I guess we'd be able to rule out many false positives
Post reply on HN