Eldercare Monitoring System

The problem

The existing options for monitoring an elderly parent living alone are bad in different ways. Cameras solve the problem by filming someone in their own home, which most families rightly refuse. Panic buttons assume the wearer is conscious and able to press one, which is exactly the assumption a serious fall breaks. The gap in the market was a system that notices on its own, without watching.

The engineering risk sat in one trade-off: miss a real fall and the consequences are unthinkable; flood relatives with false alarms and they switch the system off within a month. Either failure kills the product.

The build

A monitoring system built around a wearable motion and audio sensor that reports a heartbeat every minute. The heart of the system is a five-stage verification pipeline that a suspected fall must pass before anyone is alerted: audio classification flags the event, motion data corroborates it, the system asks the wearer to respond by voice, a distress analysis model evaluates the reply, and speaker verification checks the reply is the wearer's own voice, so a television in the background cannot fool the system.

Alongside incident detection, the system builds a daily picture of activity and compares it against age-normed baselines, so relatives see a quiet, readable trend line of how their parent is doing.

The workflow

Sensor events flow through RabbitMQ across four isolated queues, so a slow analysis job can never delay an urgent one. The AI stack runs PyTorch for audio classification, Whisper for speech recognition, and a speaker verification model for identity confirmation. Data is encrypted in transit and at rest, and the system keeps raw audio off the servers wherever possible, storing derived signals in its place.

The outcome

Incident confirmation runs end to end in under 30 seconds, and the multi-stage pipeline cut false positives drastically compared to single-signal detection. Families get an autonomous safety net that respects the dignity of the person wearing it. The system runs in production today.