The WellnessWatch: A Wearable Device Integrating Fall Detection With Environmental Monitoring

Abstract

Falls are a major health risk for older adults and post-surgical patients, and exposure to hazardous environmental conditions such as gas leaks or extreme temperatures can further threaten personal safety. The WellnessWatch addresses these risks by integrating fall detection and environmental monitoring technologies into a compact, wearable device to improve personal safety and environmental awareness. The system is built around an embedded microcontroller that incorporates an inertial measurement unit (IMU) sensor to detect sudden motion changes, while an additional environmental sensor measures gas concentration, temperature, and humidity. Real-time sensor data from the WellnessWatch are continuously processed on the device to identify abnormal conditions that may indicate a fall or unsafe environment. Fall detection was implemented using a machine learning algorithm, and unsafe environments were defined in accordance with healthcare research recommendations. If a fall or hazardous condition is detected, the device alerts the user via an onboard buzzer and display, while simultaneously transmitting the event via Bluetooth Low Energy (BLE) to a custom iOS mobile application. To mitigate false positives and negatives, push buttons allow users to cancel alerts or manually trigger a fall event. Accuracy and power consumption were evaluated across multiple machine learning algorithms, demonstrating that an artificial neural network (ANN) achieved the best performance with an F1 score of 0.97 and power consumption of 0.5340 mJ. Experimental results show 0.9043 F1 score, a 0.026 false positive rate, and a 0.037 false negative rate. The system maintains reliable BLE communication with 14.5 hours of battery life. The WellnessWatch redefines health monitoring by combining fall detection and environmental sensing into a single wearable platform, delivering proactive support and enhanced safety for vulnerable populations.