Hi Charles -

This is an interesting idea!

One thing I like about your approach is that bases the alerts on physiological measures, not purely on activity/motion.

Some fall detectors on the market now include "low activity alerts" that are intended to alert caregivers if their loved one is idle for a long time, based on activity/motion data. This aims to detect if the older adult is sick or having some other health issue.

More often, these alerts end up being triggered just because the older adult took off the wearable for an extended period of time, not because they are actually experiencing low activity or a health issue.

These false alarms cause a lot of stress for the caregiver, ultimately undermining their trust in the technology to some degree.

Your approach sounds like it will improve upon the status quo of this kind of alerting by basing the alerts on more rigorous data. A few things to bear in mind are whether the sensors would be comfortable and acceptable for the user to wear, and if there are any considerations that you should take in building your algorithms to prevent false alarms.