Healthcare for the Elderly Using the IoT
summary
A prototype system combines Bluetooth LE iBeacons for room changes, accelerometers and Google Fit for step counts, GPS for outdoor geofences, a plug‑in power meter for TV use, and a simple BLE “flic” button for alarms; data flows via MQTT through an ARM gateway to a Spark pipeline that is tested on 130 million events and uses anomaly rules and machine learning to raise alerts. Fall detection runs at about 85% hit rate; phones + Android Wear (Sony) prove the most reliable, earlier custom hardware fails on battery and usability.
AI-written summary · about
chapters (22)
- 0:00Introduction
- 1:19Depopulation on the countryside
- 3:08Rural access to healthcare
- 4:53People are alone
- 6:00Health indicators
- 8:41Alerts
- 10:36Data Processing
- 10:55Hardware
- 12:22Beacons
- 16:13First attempt
- 17:38Second attempt
- 18:53Third attempt
- 25:05Mobile App
- 28:21Inference
- 30:00Buttons
- 34:48Aggregate data
- 37:08Data storage
- 38:21Data analysis
- 39:14Spark
- 44:53Desktop client
- 49:01Setup
- 50:42Conclusion
related talks
50mHealthcare for the Elderly using the IoT
21mManaging an IoT landscape in healthcare with FHIR
29mBuilding a mobile location aware system with beacons
35mLeveraging IoT Biometrics and Zephyr RTOS for Neonatal Nursing in Uganda
37mIoT, JavaScript and Beer Brewing
06mEvery Byte Matters: Traffic Analysis of BluetoothWearable Devices
01h 00mBeeIoT - Enter the World of Bees
43mIoT with Bluetooth and Qt: Wirelessly Connected