Hacker News Reader

Comment by lambdadelirium | Hacker News Reader

"While the nursing team usually checked blood work around noon, the technology flagged incoming results several hours beforehand"

So they're understaffed and could just look into the results more, oh wow what an use of computational power.

View on HN

If it keeps me alive, I don't care if the hospital I'm in "should" have just hired a lot more people.

Well, maybe it would be even more efficient to have enough staff to keep you alive...

> So they're understaffed and could just look into the results more, oh wow what an use of computational power.

How is this not a good use of compute?

Continuous monitoring will always be better than manual checks. Also this is not an LLM and uses less power than your email software.

That is literally what AI means - the ability of a computer to perform a task that would otherwise require human reasoning.

What is an example of a computer that doesn't employ AI, by this definition?

The way I see it, AI is about the tasks a system handles, rather than the computer itself. I would say that AI encompasses the set of tasks where the computer system is in some way better than its user not just by having access to more computational resources, but by actually "reasoning" better. As a simple example, I'd argue that a basic spell-checker which works via dictionary lookup doesn't employ AI, but an extensive modern grammar checker does, as it can "reason" about the English language better than I (and most people) can.

Another way of thinking about it is that non-AI systems must always perform a task correctly, or we'd say that they have a bug. Conversely, an AI system performs tasks in situations where there is some measure of uncertainty or subjectivity, and they might arrive at a way of performing the task that is suboptimal, or even entirely inappropriate, without being buggy - for these systems we'd say that they did their best given the circumstances.

In the case of this hospital study, if they had used a simple "beep if measure goes above X" system, that wouldn't have been AI, but they used an ML model which integrates many interdependent factors over time [0] and while it has a significant ratio of false positive triggers (and as such is often wrong), it applies what would absolutely count as "reasoning" in trained human nurses.

[0] "The deterioration prediction model was a time-aware multivariate adaptive regression spline (MARS) model (Appendix, Sections 1–4). The model is made time-aware by incorporating risk score predictions from earlier in the encounter, the change in risk score since the previous assessment, and summaries of changes in the risk score over time." https://www.cmaj.ca/content/196/30/E1027

Muh AI shill

In Canada they're dangerously understaffed. And the staff are burnt out and lack qualities like: attention to detail, common sense and basic human empathy. Or at least I hope it's because they're burnt out and not because the hospital regularly hires amoral robots, which is also a possibility.

Either you go with America and get bankrupted by medical care if you don’t have excellent insurance, or you go to Canada or Europe where the average doctor is paid 1/3rd as much and there are significant waiting periods for non immediately necessary procedures. Heads I lose, tails you win.

People wonder why folks hate doctors or get “white coat” syndrome. Same shit from dentists wondering why everyone hates them.

> Europe where the average doctor is paid 1/3rd as much and there are significant waiting periods for non immediately necessary procedures

I'm not sure where exactly you evaluated this based on (personal experience I suppose?) but this hasn't been true for me in Spain with either public healthcare nor private. Don't remember it being like that in Sweden (public healthcare) either, and I'm sure there are plenty of other European countries where the waiting time isn't significant either, and you also get great care.

Some countries seems to just have figured out how to make healthcare costs manageable, with great care, well educated doctors/nurses and also relatively low waiting times. I'd probably still say they're underpaid, because they're literally saving people's lives, but I guess that's true for everywhere, even the US.

Canadian doctors earn decently well:

https://www.dr-bill.ca/blog/career-advice/doctor-salary-us-v...

Sure, there's the exchange rate, but it's still quite good. The disparity for tech workers is much greater.

I think they (doctors here) have other concerns more about regulation / paperwork and overhead that comes with it, less than total compensation. Family doctors anyways.

That and the schools simply won't graduate enough of them. Doctor shortage is a serious problem. But so is nurse shortage post-COVID.

System here seems to be in crisis. Combination of many factors.

But all my experiences in the last few years have been... very positive? Excellent recent care for my teen at McMaster Children's Hospital. Family doctor 5 minute drive away, can get appointments quite quickly. So, yeah, it's regional and situation dependent.

The english NHS is not Europe in general. Some western European nations still have relatively good quality care without ruining themselves. Although admittedly, this is getting less and less common.

We could view the patient as a process under control, with all the sensors we have, and simply apply process control technology to that without waiting for a human to interpret the data many hours after it's relevant.

Visit a hospital and see how measurement samples are taken. Your idea might be applicable with 10x the budget.

I've been there. They take the physical samples to minimize patient sleep.

Otherwise, you're hooked up to monitoring equipment...

... that beeps and boops nonstop. Yes, but on top of that the error rate in measurements is staggering. It's difficult even to get reliable vital signs. Also, much of the stuff we're measuring is of unclear purpose. We don't have a solid understanding of what the measurements actually mean for the patient.

I talked with some doctors. They said they already have alert fatigue, mostly from so many false alarms built into the systems they already have.

I mean... this is a perfectly legitimate use of computational power? What is the downside?

I suppose there is a risk they will downsize more. But this is like thinking cameras were bad because they reduced the number of security guards needed to secure an area. No?

Well, calling this AI seems like a long shot. What seems causal here is 'warn early', and indeed I'm sure it would work even better if you outputted a warning displayed full screen on the nurse's phone. It's quite possible you could have the same effect with trivial thresholds instead of a stat model. Still, I'd say it's indeed a good use of computers in general to produce targeted warnings.

Oh, fair. To an extent, at least. If they had said these were ML processed samples, would you balk as hard at it?

That is, I'm willing to chalk up use of "AI" as a descriptor being an editorial choice. Agreed that it isn't impressive just because it is AI, but it does still seem to be a good use of computational power.

Why 'balking hard'? Just saying that this is trivial use of statistics, but for once it's intelligent use. Still, if the false positive rate is too high, the effect won't last long.

Apologies, I thought you were the GP.

I took your tone to be a bit of push back on this being a good use of compute.

Reply on news.ycombinator.com