Project Title: Home Management System for Frontotemporal Dementia (FTD)
Problem Statement:
Individuals with
Frontotemporal Dementia (FTD) often experience difficulties with
spatial awareness, which puts them at risk of unintentionally
leaving safe areas or encountering hazardous objects in their
homes. Current assistive devices typically require constant
caregiver supervision, smartphone applications, or internet
connectivity, limiting their practicality and ease of use in
everyday settings.
Design Approach:
To address this
challenge, our team developed a wearable wristband system that
provides real-time sensory alerts to assist individuals with FTD
in maintaining safe boundaries at home. The system operates
independently without needing a smartphone or internet connection,
enhancing usability and reliability. We employed Bluetooth Low
Energy (BLE) proximity sensing to detect when the wearer
approaches predefined hazard zones, such as doors or kitchen
stoves.
The stationary component consists of a Raspberry Pi 5 microcontroller placed at these key hazard zones. The wearable device, worn as a wristband, uses a FireBeetle ESP32-S3 development board with integrated BLE to communicate with the Raspberry Pi. When the wristband detects proximity to a hazard zone through BLE signal strength, it triggers an auditory alert via a small speaker embedded in the wristband, warning the user to avoid the area.
Prototype Description:
The wrist
wearable device integrates the FireBeetle ESP32-S3 board, chosen
for its low power consumption, compact size, and BLE capabilities.
It receives signals from the Raspberry Pi stationed near hazardous
locations. When the Bluetooth signal strength (RSSI) exceeds a
preset threshold (approximately -70 dBm), the device activates a
speaker using an IRF520 MOSFET Driver Module controlled through
digital output (pin IO7). The speaker emits a pulse tone alert,
providing immediate sensory feedback to the wearer.
Power is supplied by a 3.7-volt lithium-potassium battery connected via a PH 2.0 interface, ensuring sustained operation for both BLE communication and audio output while maintaining comfort and wearability. The ESP32 functions as a BLE GATT server with a custom service, allowing the Raspberry Pi to send “ON” or “OFF” commands to control the alert sound based on proximity detection.
The Raspberry Pi runs a Python script using the bleak library to continuously scan for the wearable device’s BLE broadcasts. It reads signal strength and sends commands to activate or deactivate the wristband’s speaker accordingly. The system minimizes unnecessary commands by tracking the current state and handles automatic reconnection in case of signal interruptions.
This prototype effectively demonstrates an independent, wearable assistive technology solution designed to enhance safety and autonomy for individuals with Frontotemporal Dementia.
To the left of this description are images of the Raspberry Pi in its custom 3D-printed container, the wearable wrist component with its speaker visible through the lid, and our team’s project poster presented at the Assistive Technology fair.