Integração de sistemas de saúde digital para detecção de quedas em sistemas hospitalares

Abstract

Falls in the elderly can lead to serious injuries and even fatalities, considering the vulnerability of this age group. Thus, it is important to develop systems that assist in monitoring and caring for elderly individuals in cases of falls. In healthcare, the Internet of Things (IoT) has been growing, with applications that offer everything from emergency care to long-term interventions, driving the creation of new systems. An example is the Health Risk system, which detects falls in patients using wearable devices and machine learning, with an accuracy of 94.4%. This work describes the development of an API for the integration of two healthcare systems: a mobile fall detection system based on the Internet of Things, Health Risk, and a hospital system designed for a university hospital, the Smart Hospital. To integrate these systems, a literature review was conducted on various topics related to the Internet of Things to carry out an in-depth study of standards in the development process and examples of successful applications involving IoT-based systems integrated with other healthcare systems. Subsequently, the API proposed in this work, named FallReportAPI, which is responsible for the integration of Health Risk and Smart Hospital, was built using Java as the programming language, Spring Boot for creating the routes, RabbitMQ for data exchange, and PostgreSQL as the database. To evaluate the complete integration of the systems, API tests were conducted focusing on functional tests to ensure the expected results were achieved. With the integration of these systems facilitated by the API developed in this work, communication between the systems results in immediate updates in case of falls by users of the Health Risk app to the Smart Hospital system. With this agile communication between healthcare systems, it is expected to reduce response time for healthcare teams in case of falls that require urgent medical attention.