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

411–420 of 927 posts

Re: Midjourney Medical

#411

Post by rlt »

poilcn wrote: post #401
wkoszek wrote: post #382

Earlier quoted context omitted.

All doctors say this, and that sort of drove me away from healthtech. As if there were absolutely no way to take a step in a direction of fixing it. The faster and earlier we start to scan everyone regularly, as long as scanning methods aren't invasive, the more certainty we'll have what to warn people about and what not to tell them. Perhaps with the regular screening (imaging quarterly, if the scan is fast) you cou…

Healthcare resources are very limited, you'd overwhelm it with lots of "yeah that's a defect, but 40% have it", things that would go away on its own, false positives, things that do not require urgent intervention, 10x increase of hypochondriacs and health deterioration caused by anxiety You'd have a system where every resource is allocated for diagnostics, but no medical staff to treat it Also a significant part of…

Maybe it's not a coincidence an AI company is building this thing...

Re: Midjourney Medical

#412

Post by teekert »

wkoszek wrote: post #393
teekert wrote: post #381

A problem with large scale "screening" is the explosion of false positives (even at very high specificity) and the follow-ups that those generate will overwhelm our current healthcare systems. So any machine that does something medical must address this. Either that, or don't be medical. But then you might just as well tell people: "Move around a bit more. Talk to other people. Eat real food, not too much, mostly pla…

If we scan patient every 6mo starting from age 18 lets say, you could identify the masses in the patient body and track what stays the same, whats growing etc.

But what if most "masses" are cysts or other harmless structures that form during the 80 years we're walking around? I think that after about their 3rd useless biopsy people start to feel the problem with this.

Of course we can keep tuning and tuning the models, but in the limit it may well make more sense to wait for symptoms. At least that is the current experience.

Now maybe this machine will make sense in screening age 55+, 20 year+, 2 pack+ smokers for Lung lesions (where a much large portion of detected lesions are true positives). We do this currently with CT and this may be better or cheaper. But it doesn't look like it is, and it looks like far (very far) lower res than MRI (often the follow-up of a CT-scan).

Re: Midjourney Medical

#413

Post by KingMob »

wkoszek wrote: post #382
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…

All doctors say this, and that sort of drove me away from healthtech. As if there were absolutely no way to take a step in a direction of fixing it. The faster and earlier we start to scan everyone regularly, as long as scanning methods aren't invasive, the more certainty we'll have what to warn people about and what not to tell them. Perhaps with the regular screening (imaging quarterly, if the scan is fast) you cou…

It's more statisticians saying this, and not doctors per se. You run into issues of signal detection theory, false positives, and the lay confusion that Bayesian P(A|B) !== P(B|A).

You're right that we could take steps to fix it, but unfortunately, those steps involve mass education that every human body has anomalies, and many of those should just be ignored.

We'd get a wave of anxiety, lawsuits, and unnecessary interventions, until humanity collectively internalized this.

Re: Midjourney Medical

#414

Post by rlt »

icantevenhold wrote: post #400
user43928 wrote: post #394

Earlier quoted context omitted.

Are people really going to perform invasive procedures over mere concern if there are no symptoms and the doctor recommends against it?

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?

Re: Midjourney Medical

#415

Post by deanc »

Kristencline wrote: post #408

ER Nurse here: This produces images as good as an MRI- did I get that right? We already have those- they are relatively cheap ($2000 if you paid cash) and have already been scaled. The only difference seems to be the speed of the test. But how long does it take to be lowered in and out of the water, not to mention the fact that you are soaking wet afterward. An MRI of the brain takes 15 minutes, only requires you to…

It’s valuable to have input of healthcare professionals here. I don’t disagree with the majority of what you’re saying.

However, the value add here is it can do your whole body a lot faster than doing a full body MRI (which would take hours at least?)

Re: Midjourney Medical

#417

Post by Kristencline »

sberens wrote: post #65

I don't understand how people can hate on this. It's probably the most novel & ambitious consumer health device ever? Plus they're doing it fully bootstrapped. Let them cook!

Ultrasonic imaging is definitely not novel. And it requires you tolerate being fully submerged. And all you get is an image that is the SAME quality as an MRI. Except now you are soaking wet.

As a consumer health device, we haven't even gotten the population at large to wear biometrics and the CGM fad is over. Full body scans that cannot be used by a physician are not generally useful. If they aren't targeting FDA approval right off the bat, they are wasting their time. This is not solving any current problem in healthcare- you can get an MRI for $2K cash out of pocket and you get to keep your clothes dry

Re: Midjourney Medical

#418

Post by mommys_little »

poilcn wrote: post #401
wkoszek wrote: post #382

Earlier quoted context omitted.

All doctors say this, and that sort of drove me away from healthtech. As if there were absolutely no way to take a step in a direction of fixing it. The faster and earlier we start to scan everyone regularly, as long as scanning methods aren't invasive, the more certainty we'll have what to warn people about and what not to tell them. Perhaps with the regular screening (imaging quarterly, if the scan is fast) you cou…

Healthcare resources are very limited, you'd overwhelm it with lots of "yeah that's a defect, but 40% have it", things that would go away on its own, false positives, things that do not require urgent intervention, 10x increase of hypochondriacs and health deterioration caused by anxiety You'd have a system where every resource is allocated for diagnostics, but no medical staff to treat it Also a significant part of…

That's the real problem! That healthcare costs are a goldmine for Big Pharma instead of being a cheap and widely available service. And, as someone said before, the huge amount of data it produces, would decrease the rate of false positives to zero in no time! And your arguments about hypochondriacs are very similar to those that were once given against teaching reading to all people!

Re: Midjourney Medical

#419

Post by risyachka »

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…

>> It's a controversial and complicated idea

sure, and there will be downsides.

But that data will be valuable nonetheless.

Re: Midjourney Medical

#420

Post by KingMob »

munificent wrote: post #303
arcticbull wrote: post #283

Earlier quoted context omitted.

You’re dealing with populations here. Literally the odds of a positive being false would be over 90%. Much higher in the more rare conditions. I’m not exaggerating. That means every almost every follow up you do is a waste of time, money and limited resources, denying care to those who need it. Including you when you actually do need it. It also exposes you to the risks of unnecessary follow-ups like infection. Your…

It's scary in both directions. If you let it give out tons of false positives, then patients are trained to ignore it when it cries wolf. If you dial it back so that it gives out fewer positives, then now it starts giving out false negatives and not helping sick people.

Heh, Signal Detection Theory strikes again! This problem is as old as detecting whether a radar blip is a WW2 bomber on its way or not.

Sadly, there's no perfect threshold when the signal and noise distributions overlap substantially, just different trade-offs.

(Love CI, btw!)

Post reply on HN