Last Updated on January 10, 2023 by pg@petergamma.org
– we made an evaluation of near to 800 pages with is about meditation and physiological sensors
– the closer we look at different options, we can see, that a lot of them are not very contemporary
– We have this instruction here from Robshala in India:
– the DFRobot PPG sensor is quiet new
– we have an ESP 32 example for it
– we have a blink app for it
– we can get sensor raw data from it
– we know that we can reduce noise from it with HeartPy
– we know after studying the papers about consumer grade sports sensors, that there is now winner (exept the Apple watch 3, which we doubt that it is the most accurate)
– following the example of the Polar OH1, we can build a PPG device which is non-wrist based
– following the example of the Polar OH1, we can combine multiple DFRobot modules to increase accuracy
– Gold standard for HR measurement are 3 lead ECG devices. With 3 lead ECG devices, we have the best chance that our device will be accepted by cardiologists and reviewers of scientific papers
– we can take this instruction and modify it:
– we don t have to take the DF Robot ECG module, we can also take Scott Hardens module. There is 10 years of experience ins Scott Hardens Circuit
– is it worth to take .csv files from Andoid Apps which are quite old and invest time for data analysis in Python?
– the closer we look at the circuits, we can see that everything we wish is already there. But we have to one level deeper at the level of the code and at the level of high quality instructions.
– A lot of the instructions and the devices we introduced on this site here are superficial
– We can add a 16 or 24 bit A/D converter
-we already have Python examples for PPG and ECG
– Is OpenBCI used by high quality scientists in high quality journals?
– None of the instructions we presented here could convince us completely.
– We eventually have to go to a deeper level to instructions as those presented by Talha Iqbal, but he uses as far as we know a simple pulse sensor, we can take the higher quality DF Robot instead
- from 3 lead ECG, we know, that it is accepted as gold standard for HR measurement
- for PPG we have the papers of Milind Desai and the plots of Rob ter Hort from where we know that for instance Apple watches are highly accurate. But do we have other validation data for PPG devices?
Comments are closed.