AI Detects Heart Failure from One Heartbeat: Study • Hacker News Board

Re: AI Detects Heart Failure from One Heartbeat: Study

#71

Post by et2o » Sat, Oct 19, 2019, 3:23 AM UTC

RosanaAnaDana wrote: post #34

et2o wrote: post #5

I'm going to sound like a skeptical jerk here, but 490,000 heartbeats is how many patients? From what I recall these public ECG datasets are like 20 patients who underwent longitudinal ECGs. 500k heart beats is like 5 person-days of ECG recordings. Ninja Edit: N=~30 patients. For something like ECGs which are readily available, they really should have tried to get more patients. A single clinic anywhere does than 30…

I mean, the issue would be in the structure of the cross validation approach. Say, set training = 29, test = 1, build models etc: how well did you do on the one? Rinse, wash hands, repeat 30x. This is your cross validation error rate. Its not that difficult.

That would not solve this

Re: AI Detects Heart Failure from One Heartbeat: Study

#72

Post by paulcole » Sat, Oct 19, 2019, 4:51 AM UTC

et2o wrote: post #54

paulcole wrote: post #21

Earlier quoted context omitted.

> I'm going to sound like a skeptical jerk here Well you're on the right site for it! At least you didn't claim that you could come up with something better by tinkering on a rainy Sunday afternoon.

Probably could though

there it is

Re: AI Detects Heart Failure from One Heartbeat: Study

#73

Post by soVeryTired » Sat, Oct 19, 2019, 10:23 AM UTC

Cass wrote: post #64

As a doctor as opposed to an AI researcher, so many of the choices this study makes are baffling to me. First of all, why just one heartbeat? You never capture just one heartbeat on an ECG anyway, and "Is the next heartbeat identical to the first one?" is such an important source of information, it seems completely irrational to exclude it. At least pick TWO heartbeats. If you're gonna pick one random heartbeat, how…

> First of all, why just one heartbeat?

I think it's a sort of academic machismo. "Look what we can do - isn't it amazing?"

I saw the same thing in Robotics recently. An academic came to give a talk on localisation using computer vision: they cross-referenced shop signs that were seen by a robotic camera with the shop's location on a map to get a rough estimate of where the robot was. My first question was "what is the incremental benefit of this approach when it was combined with GPS?". It turned out that the researchers just hadn't used GPS at all - almost like they considered it to be "cheating".

I feel like many academic disciplines have unwritten 'rules' that you need to follow if you want to be included in the conversation. Not all of those rules are sensible.

Re: AI Detects Heart Failure from One Heartbeat: Study

#74

Post by manmal » Sat, Oct 19, 2019, 8:42 PM UTC

jacquesm wrote: post #25

manmal wrote: post #18

Earlier quoted context omitted.

Care to explain?

It sort of spoils the joke but if there are heartbeats it hasn't failed yet. So no heartbeats = failure.

Oh I really didn't get it. Thanks for explaining. I still don't find it funny.