IoT is a field within Computer Science that has grown quickly in recent years. Kevin Ashton introduced the term “Internet of Things” in 1999 [3]. One can define IoT as a global network of smart devices that can sense and interact with their environment for communication with users and other things (smart devices) and systems. In this context, things could be identified solely by using radio-frequency identification (RFID) [4] tags in order to be connected to the Internet and publish their information. Things are physical objects such as refrigerators, cars, walking sticks, dog collars, and whatever object comes to mind.

Thus, using sensors, actuators, and RFID-like technology, objects in the environment could be viewed, identified, and controlled more autonomously. In this case, things themselves could specify when they needed to be replaced, fixed, or report if they could provide data [3].

To develop the IoT patient-monitoring application described in this paper, 3 main IoT technologies have been used: RFID, micro-controllers, and sensors.

RFID is an automatic identification method that utilizes radio signals, recovering and storing data remotely through devices called RFID tags. These devices are used for identification, sensing, and communication [5].

Arduino [6] microcontrollers, which are open source platforms for electronic prototyping, are also used: Uno R3 [7] and Yún [8] models (Figure 1). Microcontrollers can be programmed to process inputs and outputs of connected external components (Figure 1). One can use embedded computing to allow the construction of systems that interact with the environment using hardware and software [9].

A variety of sensors can be used to collect data for IoT applications such as temperature, humidity, light level, oxygen level, and sensor presence, among others.

In eHealth, it is common for some devices to contain a number of sensors linked together, such as in the HealthPatch MD [10] Vital Connect health-monitoring sensor (Figure 2). The sensor is a small adhesive patch with a module that measures heart rate, breathing frequency, body temperature, posture, detection of falls, and also has Internet connectivity. Another example is the eHealth Sensor Platform Complete Kit [11]. It contains an eHealth Sensor Shield compatible with Arduino and Raspberry Pi [12] microcontrollers (Figure 2), plus 10 sensors to collect biometric data (Figure 2): pulse, oxygen levels in blood, airflow (breathing), body temperature, electrocardiogram (ECG), glucometer, galvanic skin response, blood pressure, patient position (accelerometer), and muscle or electromyography sensor (EMG).

A software agent [13] is an element of a computational system that is situated in an environment where it can perform autonomous actions in order to reach its assigned goals. An agent is both autonomous and capable of learning from its experience. Autonomy has been acknowledged as a key characteristic of an agent in satisfying its goals [14]. In this context, autonomy means operating without the intervention of humans or other systems, although the set of possible actions should be previously defined.

Although agents control their behavior and internal states, they do not have full control of the environment in which they operate. Agents contain a set of actions that can carry out tasks, the execution of which can result in changes in the environments. For this reason, one can consider that an agent can have partial control and influence over its environment depending on the action performed [15].

In general, the use of software agents is justified by the fact that asynchronous software systems require autonomous operation, a general argument that can be applied to our solution.

Frameworks are tools used to generate applications related to a specific domain; that is, to cope with a family of related problems [16]. The choice of using existing frameworks or developing new application generators is based on whether the framework can offer design and code reuse. Thus, frameworks can usually increase software development productivity and shorter time-to-market, compared with traditional approaches.

Frameworks contain fixed and flexible points known as frozen spots and hot spots, respectively. Hot spots are extension points that allow developers to create a new application from the framework instantiation process. In this case, developers should create specific application code for each hot spot, through the implementation of abstract classes and methods defined in the framework. Frozen spots consist of the framework’s kernel, corresponding to its fixed parts, previously implemented and hard to change. A frozen spot calls one or more of the application’s hot spots and is present in each framework’s instance [16].

Creating a new instance of a framework consists of 3 main steps: (1) Domain analysis, (2) Design, and (3) Instantiation. The domain analysis step includes requirements elicitation including definitions of hot and frozen spots. The design step is responsible for specifying the hot and frozen spots through a modeling language such as UML [17] diagrams. Design patterns [18] are also used in this phase. The instantiation phase corresponds to the application generation phase through hot spot implementation [16].

Our proposal takes a similar approach to that in [19]. This paper shows the implementation of a distributed information infrastructure that uses the intelligent agent paradigm for: (1) automatically notifying the patient’s medical team regarding the abnormalities in his or her health status; (2) offering medical advice from a distance; and (3) enabling continuous monitoring of a patient’s health status. In addition, the authors have promoted the adoption of ubiquitous computing systems [20] and apps that allow immediate analysis of a patient’s physiological data such as a personalized feedback of their condition in real time, by using an alarm-and-remember mechanism. In this solution, patients can be evaluated, diagnosed, and cared for through a mode that is both remote and ubiquitous. In the case of rapid deterioration of a patient’s condition, the system automatically notifies the medical team through voice calls or SMS messages, providing a first-level medical response. This proposal differs from ours, in that the resulting application is closed, as opposed to our broader eHealth application generator.

The approach in [21] focuses on design and development of a distributed information system based on mobile agents to allow automatic and real-time fetal monitoring. Devices such as a PDA, mobile phone, laptop, and personal computer are used to capture and display the monitored data.

In [22], mobile health apps are proposed as solutions for (1) overcoming personalized health service barriers; (2) providing opportune access to critical information on a patient’s health status; (3) avoiding duplication of exams, delays and errors in patient treatment.

In this step, problems that health professionals currently deal with in their patient monitoring routines are considered. As mentioned earlier in this paper, the decision was made to build a software framework instead of one or more apps. The choice to use framework design techniques was motivated by the fact that the construction of a more generic and complex system would provide a cost-benefit, in that frameworks can usually increase software development productivity and shorter time-to-market.

This layer works as a remote database of a patient’s vital signs.

Through this layer, the L3 can communicate with the L1 layer’s remote database.

The L3 Layer is responsible for the entire information management of the instantiated applications. It is composed of 10 modules with well-defined responsibilities. These modules are shown next, along with a description of their purpose and examples of how our framework can be extended by specific application code implemented in IoT4Health’s instances.

To provide the data distribution service to the application, a remote data storage service called Parse [26] was utilized. However, because the Parse hosted service will be retired in early 2017, we are moving our database to another platform called MongoDB [27].

The Agents4Health communicates with Parse (L1) through the REST API [28]. The application sends and retrieves data to and from the cloud through HTTP requests.

The data management layer comprises the IoT application, with its 8 modules (M1-M8) as follows: