Is ZEO accurate?

jnk wrote:I would nominate Max for the field trip, though, if it could be financed. He likes squigglies.

Is this in St. Barth's, Barbados or maybe over on the Amalfi Coast? We can see how the Zeo responds to tranquil slumber from ample umbrella drinks and a gentle well oiled back massage by a grass skirted young lady in Bora Bora...

I hear things are a little different from that at Uncle Dave's sleep facilities, especially during the winter up there in the almost-Canada. Slow-paced and luxurious, but oddly modern and formal. No major incidents, though, since that time a tech almost turned into a Zeo:

https://www.youtube.com/watch?v=xLg9Stk0-Ug

Everything is fine now, at least based on what people can recall.

Jay Aitchsee wrote:Jeff, why don't you take your Zeo over to Dave's and test it?

jnk wrote:Only if he promises to watch his language

Well this sounds like such a great opportunity, I'm certainly willing to give it a shot!



Oh well, the flesh is willing...

jnk wrote:...scoring ...

Speaking of scoring (can the Muffmeister make a smooth segue or what?) I have moderate befuddlement as to why DueceSleep.com is not abuzz about the new AASM Scoring Manual released yesterday.

Although the new manual looks a lot like the old manual.

OTOH, that would normally be the kind of P&Ming that would constitute "the buzz".

Eh? Stage scoring is all mostly rounded out, dice-rollin', voodoo stuff anyway, isn't it?

I mean, it can sure look and sound that way to uneducated people like me:

http://d21nsqb9vpwnq3.cloudfront.net/IS ... review.mp4

More than half the human experts got it "right"! Well done!

I hate squiggly lines.

(My apologies to the AASM for the deep link. Here is the overall page: http://www.aasmnet.org/isr/scoringeducation.aspx )

I love Q&A FAQs, though!:

Q: Is there a penalty if my technologists don’t reach their set criterion for agreement with the gold standard scorer?

A: There is no penalty. . . .

http://www.aasmnet.org/isr/Faqs.aspx

Yay! Zeo passes!

But wait! I've found a problem:

We [once a month] schedule a dinner meeting between the day and evening shifts to review the comparison on a video screen. As we munch on pizza or enchiladas, we analyze each epoch. When we disagree about the particular scoring of an event, we debate it and reach agreement - or at least compromise.

http://respiratory-care-sleep-medicine. ... ility.aspx

I can see how that's gonna be a BIG problem. Zeo can't eat pizza or enchiladas.

RATS!!!

(This post is brought to you by the makers of the exclamation mark.)

jnk wrote:Eh? Stage scoring is all mostly rounded out, dice-rollin', voodoo stuff anyway, isn't it?

I mean, it can sure look and sound that way to uneducated people like me:

http://d21nsqb9vpwnq3.cloudfront.net/IS ... review.mp4

More than half the human experts got it "right"! Well done!

Mmmm. Let's see, accuracy of scorers - 52.5%; accuracy of Zeo (Woodworker math) - 64%. Cost of 300 nights of Zeo analysis - $200 including replacement sensors. Cost of 300 nights of PSG analysis @ $2K per - $600,000. Mmmm...

J

Jay Aitchsee wrote:

jnk wrote:Eh? Stage scoring is all mostly rounded out, dice-rollin', voodoo stuff anyway, isn't it?

I mean, it can sure look and sound that way to uneducated people like me:

http://d21nsqb9vpwnq3.cloudfront.net/IS ... review.mp4

More than half the human experts got it "right"! Well done!

Mmmm. Let's see, accuracy of scorers - 52.5%; accuracy of Zeo (Woodworker math) - 64%.

That's great, but how can I be sure that I'll get a Zeo that's in the 64% group?

deltadave wrote:

Jay Aitchsee wrote:

jnk wrote:Eh? Stage scoring is all mostly rounded out, dice-rollin', voodoo stuff anyway, isn't it?

I mean, it can sure look and sound that way to uneducated people like me:

http://d21nsqb9vpwnq3.cloudfront.net/IS ... review.mp4

More than half the human experts got it "right"! Well done!

Mmmm. Let's see, accuracy of scorers - 52.5%; accuracy of Zeo (Woodworker math) - 64%.

That's great, but how can I be sure that I'll get a Zeo that's in the 64% group?

I think you use the same procedure to make sure that your live scorer is in the 52.5% group.

Drowsy Dancer wrote:

deltadave wrote:

Jay Aitchsee wrote:

jnk wrote:Eh? Stage scoring is all mostly rounded out, dice-rollin', voodoo stuff anyway, isn't it?

I mean, it can sure look and sound that way to uneducated people like me:

http://d21nsqb9vpwnq3.cloudfront.net/IS ... review.mp4

More than half the human experts got it "right"! Well done!

Mmmm. Let's see, accuracy of scorers - 52.5%; accuracy of Zeo (Woodworker math) - 64%.

That's great, but how can I be sure that I'll get a Zeo that's in the 64% group?

I think you use the same procedure to make sure that your live scorer is in the 52.5% group.

OK, I'll give it a shot.

"Hey baby, what's your sign?"

jnk wrote:Stage scoring is all mostly rounded out, dice-rollin', voodoo stuff anyway, isn't it?

For Zeo? Absolutely!

However, it attempts to overcome its inaccuracy by avoiding a major source of scoring ambiguity in combining N1 and N2. However2, the obvious problem with this is that N1 is basically worthless, so Zeo scoring can be very deceiving in poor sleepers. However3, if, in removing the N1-N2 ambiguity, it's IR is still only 68%, then this should move out of the 149$ price range and be sold at the Dollar Store.

Howevermostimportant, since the Zeo bars show 5 minute blocks instead of 30 second epochs (unless you're running a program like Max or tetragon), you undoubtedly stand a much better chance of getting a passing IRS by doing "heads or tails" flipping a baseball.

But what do I know? I'm just a pizza-munchin', enchilada-slingin' IRSMF.

Jay Aitchsee wrote:Mmmm. Let's see, accuracy of scorers - 52.5%; accuracy of Zeo (Woodworker math) - 64%.

Actually, since Zeo clumps N1 and N2 together, it would have totalled 16 seconds of LS, making it the majority of sleep, calling that epoch LS and attaining 0.0% accuracy against GS.

deltadave wrote: . . . Zeo clumps N1 and N2 together, it would have totalled 16 seconds of LS, making it the majority of sleep, calling that epoch LS and attaining 0.0% accuracy against GS.

Excellent point! That illustrates a basic difference, I think, and a significant one.

Personally, I would want the medical people scoring a medical test to do so according to AASM rules, which although somewhat arbitrarily set, are vital to follow for consistency in order for breathing events to be properly noted and measured. In that context, I would perfer the epoch to be scored wake.

On the other hand, that said, I personally would prefer my little electronic sleep coach, who isn't interested in the medical significance of my breathing at that moment, to score that epoch light sleep in its tally of whether to report the 5-minute chunk of time as wake or sleep in the chart, since there was more N1/N2 than wake. I wouldn't want my sleep coach's hands to be tied by the medical rules for scoring medical tests.

To illustrate, I use a $20 watch with a stopwatch that reports my pulse when I do stationary bike, stair-climbing, treadmill, etc. That way I can time my workout and can keep my heart rate between 140 and 150 as I do it. I don't need to be wired each time with EKG or to have a full stress test each time I exercise to constantly monitor other things about my heart rhythm. The $20 watch does fine for what I am using it for. The information is very valuable to me for what it is. A runner might use such a device for a race, even though he knows the device will not give his official time for a race and will not let him know the exact health condition of his heart.

So it is with Zeo. Don't expect your sleep coach to give you medically valid information like a medical test run by medical people for medical reasons. But for ballpark estimates useful for ideas of how to improve your sleep hygiene, I am unaware of anything more useful or reasonably priced than Zeo for that. Keep it on your nightstand next to your heart-rate watch that you will use during exercise the next morning!

Just my opinion.

deltadave wrote: Howevermostimportant, since the Zeo bars show 5 minute blocks instead of 30 second epochs (unless you're running a program like Max or tetragon), you undoubtedly stand a much better chance of getting a passing IRS by doing "heads or tails" flipping a baseball.

Small point, but... the 5 minute blocks are only used for graphic presentations. All the tallies are based on 30 sec blocks, which are easily downloaded from the site in CSV format.

J

jnk wrote:So it is with Zeo. Don't expect your sleep coach to give you medically valid information like a medical test run by medical people for medical reasons. But for ballpark estimates useful for ideas of how to improve your sleep hygiene, I am unaware of anything more useful or reasonably priced than Zeo for that. Keep it on your nightstand next to your the heart-rate watch that you will use during exercise the next morning!Just my opinion.

Well said, Jeff. I agree.

J

deltadave wrote:

Jay Aitchsee wrote:

jnk wrote:Eh? Stage scoring is all mostly rounded out, dice-rollin', voodoo stuff anyway, isn't it?

I mean, it can sure look and sound that way to uneducated people like me:

http://d21nsqb9vpwnq3.cloudfront.net/IS ... review.mp4

More than half the human experts got it "right"! Well done!

Mmmm. Let's see, accuracy of scorers - 52.5%; accuracy of Zeo (Woodworker math) - 64%.

That's great, but how can I be sure that I'll get a Zeo that's in the 64% group?

But just think... If you don't get a Zeo, you won't even be in the 64%, 50%, 10%, 5% of knowing anything.

So any percent rate of accuracy has to count for something!

/flame away.