2026-04-10 · 6 min read
Start with a clear comparison plan
MultiHR works best when you know which device is the reference and which devices are under test before the session starts.
A common setup is a trusted chest strap such as Polar H10 as the reference, with optical armbands or watch broadcast signals as test devices.
Connect every device before recording
- Wear every sensor correctly and confirm it is awake.
- Scan for nearby devices and connect the ones you want to compare.
- Keep the phone close to the sensors to reduce BLE instability during the session.
Record one synchronized session
After all required devices are connected, start one recording session and keep the activity continuous.
This gives you cleaner time alignment than stitching together separate recordings.
- Watch the live chart for sudden drops or drift.
- Record long enough to include both steady periods and changes in intensity.
- For meaningful analysis, longer sessions are usually better than very short spot checks.
Review the session after recording
- Export Excel if you want per-device columns and detailed timestamps.
- Export CSV if you plan to continue in R, Python, or another tool.
- Use in-app Pearson and Bland-Altman views for a quick first comparison.
Interpret the results carefully
High correlation is useful, but it is not the same thing as agreement. MultiHR includes Bland-Altman analysis because bias matters as much as trend matching.
Treat the results as device validation or training reference data, not as medical output.