You are reading a single comment by @andyp and its replies.
Click here to read the full conversation.
andyp-
As it's also PPG the base issues remain the same, but I quite like the location and form-factor. Plus the team behind it came out of Polar, and the algorithms aren't bad.
(Its probably the one I would go for, if I didn't already have an Apple Watch)If you want a proper deep dive into the limitations and the algorithmic basis for what either are doing:
https://peterhcharlton.github.ioIf you really want a deep dive there are two dev kits- one from Sony and one that is based of the ESP architecture and git has plenty of code to harness to even do cuffless Blood pressure.
Aldosterone
Don't follow Brian Johnson- the methodology of throw all the pasta at the wall only works if you can identify the strand that sticks.
Don't get a Whoop- capillary bed photoplethysmography is a poor method of measuring [multiple stacked confounders]. (hence why everyone reverts to chest HRMs, and there is huge debate about their models, overfitting, and the true utility of HRV outside of some pathological states.
I think 8sleep does a great job of targeted marketing, but nothing I've seen within it really makes me want to buy it (bar the 2 side heat differential). The reported noise of the unit puts paid to that for me.
Zoe's has many issues, but I like the aims- much more focused on direct measurement. Methodology isn't up to scratch yet (considering the elements at play within the biological systems studied, cell turnover, etc).
There is a new wave of direct biosensors coming out, with things like:
https://www.nature.com/articles/s41551-023-01095-1
https://www.impli.org
Amongst others, on the horizon. We're going to start stripping inferential layers, and get richer data to build the models underlying. Thats when things start getting interesting.
We're not quite there yet, but soon.
For the time being, whatever you currently use will probably suffice.