Fall Detection in Elderly People Using Compact Wearable Sensors with AI
Venkateswaran Radhakrishnan, B. Suresh Babu, M. Parameswari, Purushothapatnapu Naga Venkata VamsiLala, S. Ambigaipriya, N. V. Keerthana · 2024
The most crucial public concern regarding health is elderly people's falls which tend to undesirable injuries leading to an increase in healthcare expenses for them. This research paper describes a method that uses AI technology with a barometer sensor to detect falls in elderly people. In addition to this, the AI model is trained on a diverse dataset comprising various activities of daily living, enabling it to distinguish between normal movements and those that may lead to falls. This method probably gives some notifications or missed calls to the caretakers as well as to the hospitals. When compared to many more sensors such as gyroscopes and some fitness trackers, the Apple Watch 4, and Samsung Gear S3 barometer sensor perform in a better better way. Key performance metrics such as absolute and altitude are calculated and compared based on a natural language processing (NLP) algorithm.