S.O.S. - My Grandparents: Using the Concepts of IoT, AI and ML for the Detection of Falls in the Elderly

The increase in life expectancy and aging of the population have highlighted the need for innovative solutions to support and protect older people, who become more vulnerable to household accidents, particularly falls. This scientific work focuses on the development of an intelligent monitoring and alert system using Arduino Nano 33 IoT and TinyML technology to differentiate between falls and ordinary bending among the elderly. The system aims to promptly intervene in case of an unexpected event, ensuring access to medical assistance. The study explores the implementation of the system, including the use of advanced sensors and TinyML technology for accurate fall detection. Specialized literature on falls in the elderly and the potential of force platform measurements to predict falls is also reviewed. Ethical considerations and data privacy are emphasized throughout the research. The results show that the implemented system can accurately differentiate between sitting and falling movements, enabling rapid responses in emergency situations, thus contributing to improved elderly care and reducing the risk of complications associated with accidental falls.