Vandermi J. Silva

An experience to use robotics to improve Computer Science learning

In the teaching of engineering and computing, one of the major difficulties in modern education is to create an environment to motivate the student to learn, allowing them to assimilate the abstractions of computer science with playful and practical actions. The use of robotics in educational environments has proven to be an appropriate tool for the development of activities that involve creating, designing and planning, thereby facilitating the teaching-learning process and further expanding the integration between different areas of knowledge. Robotics in the educational context is called educational robotic and can be defined as an environment acquisition of concepts through mechanical devices, allowing development of logical reasoning and creativity, besides being a multidisciplinary tool for learning. Our goal in this work is to provide an environment for open source programming by using robotics in education and help students to improve their learning of programming languages and software engineering disciplines through a study case where we used the Lego Mindstorms Educational Kit.

UbMed: A ubiquitous system for monitoring medication adherence

One of the biggest problems with chronical patients in treatment is medication adherence. Studies indicate that taking drugs out of time influence the patients treatment decreasing the drugs effect. To minimize this problem, we developed a ubiquitous and intelligent system able to monitor the taking of medicines and to identify whether the patient is meeting the requirements prescribed by the doctor. An architecture provided with a decision system based on rules and trees to evaluate data collected from an intelligent medicine cabinet, sensors and electronic devices available in the patients home was designed. The system classified the drugs taken pattern and released messages on social networks, SMS, and consumer electronic devices such as TV, smartphone and tablets, without human interference. Its goal is to help keeping the medication on time and helping to decide what to do in case of missing the right time. The algorithms J48, Rep and Random tree, were tested to classify the taking medicine patterns and to chose the right services available. The obtained results are very promising and reached an acceptable accuracy rate.

Monitoring of hypertensive patients through conventional medical devices integrated to the Brazilian digital TV

This paper presents an implementation of a healthcare monitoring system integrated to the model of Digital TV adapted in Brazil (SBTVD), which provides the convergence among wireless sensor network, mobile devices, the World Wide Web (WEB) and embedded systems, through the connection of these devices to a central computer called healthcare gateway. This gateway manages the service requests and access via Digital TV (DTV) and mobile devices like cellular phones. Conventional sensors capture the patients body temperature, pressure, heartbeat and room temperature which are sent to the gateway. Then a reactive software agent based on the behavior, verifies the status of the sensors and dispatch a message to a DTV and/or a cellular phone previously registered in the system via WEB interface. The impact of this new concept will be very high on the Brazilian population due to the great number of people suffering from hypertension and the high level usage of digital TV and mobile devices can help health teams in monitoring the patient.

On the use of inertial sensors and machine learning for automatic recognition of fainting and epileptic seizure

This paper depicts a machine learning method for fainting and epileptic seizures automatic recognition. We evaluated five machine learning techniques in order to find out which classification method maximizes the accuracy level and, at the same time, minimizes the computational complexity since the experimental environment has very limited computational resources (processing power). We prototype such method in a wearable device, taking into account F-Score and Accuracy metrics. The experimental evaluation shows that there are no significant difference between KNN, PART, and C4.5. However, KNN has high computational cost when compared to PART and C4.5. PART has low computational cost when compared to C4.5 since it identified less rules.

Using Brazilian Digital TV to Integrate Health Care Services Embedded in Medical Commercial Devices

The Digital TV (DTV) System is getting a stimulus within industry and in Brazil that stimulus is based on the belief that this new system will be successful. Several proposals for extending the potential of interactivity, for providing innovative services are available today. In this context, health care is a concept being studied. The health care concept is relative to care about health, in preventing diseases, quality of life and with applications to hospitals, elderly care centres and so on. Integrating health care technology with home devices is a new trend. Thus this chapter shows how the software programmer can integrating devices and sensors with the architecture based in filters presented here.