Fall Detection Scheme based on Deep Learning Model for High-Quality Life

Elderly individuals worldwide have a significant risk of injury or death due to falls. An aged person's mobility, independence, and quality of life can be significantly impacted if he or she falls and does not receive timely assistance. This paper proposes the creation of a low-power network, smart devices, and cloud computing-based fall detection system for senior individuals in indoor spaces as part of this study. 10 volunteers volunteered to have their fall and non-fall movement data collected using a wireless telemetry surface electromyography capture system. Generative Adversarial Network (GAN) models use the effective signal segments as input. Notifications are automatically sent to caregivers of the elderly if a fall is detected, and an alert is triggered. As the final step, the system offers cloud-based services. There is a service that allows healthcare professionals to obtain the fall data for further examination from a medical standpoint. In experiments, fall detection has been found to be highly accurate, precise, and recallable.