Hacker News new | past | comments | ask | show | jobs | submit login

Out of curiosity : how do you know it reliably detects it?

I have a polar watch that also tracks my sleep and presents it in a nice plot. I have no way to check whether that plot is a good representation of my night, or whether they just randomly put in a few blocks and mark them as REM.

I don't mean to imply that I don't trust the device or anything. I'm just saying I have no way to verify how reliable it really is.




Because every time I wake up after having a dream, I see it being properly registered. Conversely, if I wake up without having dreamt right before waking, then I see that the device didn't register REM.


> I don't mean to imply that I don't trust the device or anything. I'm just saying I have no way to verify how reliable it really is.

Dreem 2, for example, has a few studies to back it up:

> The aim of this study was to assess the signal acquisition and the performance of the automatic sleep staging algorithms of a reduced-montage dry-electroencephalographic (EEG) device (Dreem headband, DH) compared to the gold-standard polysomnography (PSG) scored by five sleep experts.

And the relevant results:

> The algorithm achieved an overall accuracy comparable to human-level performance of 85.76% (N1: 56%, N2: 88%, N3: 85%, REM: 92%, and Wake: 85%).

https://academic.oup.com/sleep/article/43/11/zsaa097/5841249


I am less skeptical about an EEG-based device like the Dreem than I am about a wristband.

Most wristbands have a heart rate monitor and an accelerometer. Is there a pattern in the heart rate that shows REM sleep versus deep or light sleep? Or do they use the accelerometer to detect twitches you make during REM sleep? How do they discriminate that from my cat jumping onto the bed?

Compare that to EEG-based devices : I think it is well known that sleep phases can be read in an EEG. The Dreem has fewer channels than a medical EEG, and maybe a little less accuracy, but that is beside the point : it is measuring the relevant signals for sleep phase detection.




Consider applying for YC's Summer 2025 batch! Applications are open till May 13

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: