مقاله انگلیسی رایگان در مورد اینترنت اشیا برای مراقبت از بیماری های دمانس – IEEE 2018

This section describes the design and implementation of our IoT architecture, which is capable of combining different, yet complementary, technologies to develop novel functionalities. Our system collects real-time data from multiple publishers and delivers the data to the TIHM back-end system. At this stage, an advanced analytics tool is used to process the integrated data and generate insights and notifications based on the status of patients. Note that clinical knowledge and experience has been used to set the parameters for alerting and notifications. The high-level overview of the interactions in TIHM architecture is presented in Figure 1. As shown, it is composed of four main parts: • sensors installed at homes; • SMEs’ back-end servers; • the TIHM back-end system, including the storage and analysis servers; and • the user interface for data visualization and management. At a finer level of detail, several passive sensors are embedded in patients’ homes to collect data from the environment and surroundings, such as humidity and temperature conditions, appliance usage, etc. In addition, medical devices and wearable technologies are used to measure important physiological parameters, such as blood pressure, pulse, etc. All the sensor and medical devices record and send the data to their corresponding gateways over Wi-Fi or Bluetooth or, in exceptional cases, via auxiliary interfaces of such a device. Gateways relay these data to the companies’ back-end systems over GPRS (General Packet Radio Service), SigFox, or home broadband. At this stage, all participating companies comply with a common JSON (JavaScript Object Notation) data model for the TIHM project, which is defined in the next section. This is followed by the communication with the TIHM back-end system through a publish/subscribe (pub/sub) or message queue (MQ) model; specifically, the advanced message queuing protocol (AMQP) is employed. These data are then validated and persisted to a NoSQL Mongo database for further analysis. To make the analyzed data more readily accessible and consumable by remote users (mainly the clinical-monitoring team), a web-based graphical interface has been implemented with defined privilege levels to access both real-time and historical data12—comprehensive details are included in the section “User Interface.” Observe that the MQ is the core exchange for TIHM messages and is able to receive, process, and reply to requests coming from the companies’ back-ends, TIHM back-end system, and user interface.