Aplicativo móvel integrado a anel e pulseira inteligente para monitoramento de idosos

Abstract

This work details the development of "RingCare,"a digital health ecosystem composed of a native Android mobile application integrated with a smart ring and a smart band, focused on the continuous monitoring of the elderly. This initiative is motivated by the growing aging population and its associated health risks, such as falls, sleep disorders, and conditions like nocturia, seeking to promote user safety and quality of life through Internet of Health Things (IoHT) and ubiquitous computing technologies. The central objective was to develop a tool that offers proactive and discreet monitoring, supported by theoretical concepts of self-adaptive systems and context-awareness to enhance user autonomy. The solution was implemented as a native application in Kotlin with Jetpack Compose, following the MVVM architectural pattern to ensure maintainability and scalability. The system connects via Bluetooth Low Energy (BLE) to low-cost devices, such as the Colmi R06 ring and the Mi Band 6, and integrates with Android’s Health Connect platform to collect and centralize vital data like heart rate, oxygen saturation (SpO2), steps, and sleep patterns. The solution’s architecture includes distinct "Patient"and "Caregiver"profiles, enabling remote data tracking and the dispatch of automatic alerts in risk situations, with all data management, profiles, and authentication handled by the Firebase platform. Critical functionalities, such as a 30-second confirmation dialog for the patient in case of heart rate spikes before notifying the caregiver, were implemented to optimize emergency response. The main outcome of the project is a functional prototype that validates the feasibility of an effective, accessible, and user-centric monitoring solution. "RingCare"establishes a solid foundation for the future of preventive care, although the study identified a vulnerability in the unencrypted communication with the smart ring, a point that requires future attention. For next steps, we suggest applying machine learning techniques to improve the detection of falls and physiological anomalies, adapting the interface for different levels of digital literacy, and conducting field validation studies with elderly users.