Loading...
5 results
Search Results
Now showing 1 - 5 of 5
- Identification of Daily Activites and Environments Based on the AdaBoost Method Using Mobile Device DataPublication . Ferreira, José M.; Pires, Ivan; Marques, Gonçalo; Garcia, Nuno M.; Zdravevski, Eftim; Lameski, Petre; Flórez-Revuelta, Francisco; Spinsante, SusannaUsing the AdaBoost method may increase the accuracy and reliability of a framework for daily activities and environment recognition. Mobile devices have several types of sensors, including motion, magnetic, and location sensors, that allow accurate identification of daily activities and environment. This paper focuses on the review of the studies that use the AdaBoost method with the sensors available in mobile devices. This research identified the research works written in English about the recognition of daily activities and environment recognition using the AdaBoost method with the data obtained from the sensors available in mobile devices that were published between 2012 and 2018. Thus, 13 studies were selected and analysed from 151 identified records in the searched databases. The results proved the reliability of the method for daily activities and environment recognition, highlighting the use of several features, including the mean, standard deviation, pitch, roll, azimuth, and median absolute deviation of the signal of motion sensors, and the mean of the signal of magnetic sensors. When reported, the analysed studies presented an accuracy higher than 80% in recognition of daily activities and environments with the Adaboost method.
- Internet of Things Architectures, Technologies, Applications, Challenges, and Future Directions for Enhanced Living Environments and Healthcare Systems: A ReviewPublication . Marques, Gonçalo; Pitarma, R.; Garcia, Nuno M.; Pombo, NunoInternet of Things (IoT) is an evolution of the Internet and has been gaining increased attention from researchers in both academic and industrial environments. Successive technological enhancements make the development of intelligent systems with a high capacity for communication and data collection possible, providing several opportunities for numerous IoT applications, particularly healthcare systems. Despite all the advantages, there are still several open issues that represent the main challenges for IoT, e.g., accessibility, portability, interoperability, information security, and privacy. IoT provides important characteristics to healthcare systems, such as availability, mobility, and scalability, that o er an architectural basis for numerous high technological healthcare applications, such as real-time patient monitoring, environmental and indoor quality monitoring, and ubiquitous and pervasive information access that benefits health professionals and patients. The constant scientific innovations make it possible to develop IoT devices through countless services for sensing, data fusing, and logging capabilities that lead to several advancements for enhanced living environments (ELEs). This paper reviews the current state of the art on IoT architectures for ELEs and healthcare systems, with a focus on the technologies, applications, challenges, opportunities, open-source platforms, and operating systems. Furthermore, this document synthesizes the existing body of knowledge and identifies common threads and gaps that open up new significant and challenging future research directions.
- Activities of Daily Living and Environment Recognition Using Mobile DevicesPublication . Ferreira, José M.; Pires, Ivan; Marques, Gonçalo; Garcia, Nuno M.; Zdravevski, Eftim; Lameski, Petre; Flórez-Revuelta, Francisco; Spinsante, Susanna; Xu, LinaThe recognition of Activities of Daily Living (ADL) using the sensors available in off-the-shelf mobile devices with high accuracy is significant for the development of their framework. Previously, a framework that comprehends data acquisition, data processing, data cleaning, feature extraction, data fusion, and data classification was proposed. However, the results may be improved with the implementation of other methods. Similar to the initial proposal of the framework, this paper proposes the recognition of eight ADL, e.g., walking, running, standing, going upstairs, going downstairs, driving, sleeping, and watching television, and nine environments, e.g., bar, hall, kitchen, library, street, bedroom, living room, gym, and classroom, but using the Instance Based k-nearest neighbour (IBk) and AdaBoost methods as well. The primary purpose of this paper is to find the best machine learning method for ADL and environment recognition. The results obtained show that IBk and AdaBoost reported better results, with complex data than the deep neural network methods.
- Internet of Things Architectures for Enhanced Living EnvironmentsPublication . Marques, Gonçalo Miguel Santos; Santos, Nuno Manuel Garcia dos; Babić, ZdenkaAmbient Assisted Living (AAL) is an emerging multidisciplinary research area that aims to create an ecosystem of different types of sensors, computers, mobile devices, wireless networks, and software applications for enhanced living environments and occupational health. There are several challenges in the development and implementation of an effective AAL system, such as system architecture, human-computer interaction, ergonomics, usability, and accessibility. There are also social and ethical challenges, such as acceptance by seniors and the privacy and confidentiality that must be a requirement of AAL devices. It is also essential to ensure that technology does not replace human care and is used as a relevant complement. The Internet of Things (IoT) is a paradigm where objects are connected to the Internet and support sensing capabilities. IoT devices should be ubiquitous, recognize the context, and support intelligence capabilities closely related to AAL. Technological advances allow defining new advanced tools and platforms for real-time health monitoring and decision making in the treatment of various diseases. IoT is a suitable approach to building healthcare systems, and it provides a suitable platform for ubiquitous health services, using, for example, portable sensors to carry data to servers and smartphones for communication. Despite the potential of the IoT paradigm and technologies for healthcare systems, several challenges to be overcome still exist. The direction and impact of IoT in the economy are not clearly defined, and there are barriers to the immediate and ubiquitous adoption of IoT products, services, and solutions. Several sources of pollutants have a high impact on indoor living environments. Consequently, indoor air quality is recognized as a fundamental variable to be controlled for enhanced health and well-being. It is critical to note that typically most people occupy more than 90% of their time inside buildings, and poor indoor air quality negatively affects performance and productivity. Research initiatives are required to address air quality issues to adopt legislation and real-time inspection mechanisms to improve public health, not only to monitor public places, schools, and hospitals but also to increase the rigor of building rules. Therefore, it is necessary to use real-time monitoring systems for correct analysis of indoor air quality to ensure a healthy environment in at least public spaces. In most cases, simple interventions provided by homeowners can produce substantial positive impacts on indoor air quality, such as avoiding indoor smoking and the correct use of natural ventilation. An indoor air quality monitoring system helps the detection and improvement of air quality conditions. Local and distributed assessment of chemical concentrations is significant for safety (e.g., detection of gas leaks and monitoring of pollutants) as well as to control heating, ventilation, and HVAC systems to improve energy efficiency. Real-time indoor air quality monitoring provides reliable data for the correct control of building automation systems and should be assumed as a decision support platform on planning interventions for enhanced living environments. However, the monitoring systems currently available are expensive and only allow the collection of random samples that are not provided with time information. Most solutions on the market only allow data consulting limited to device memory and require procedures for downloading and manipulating data with specific software. In this way, the development of innovative environmental monitoring systems based on ubiquitous technologies that allow real-time analysis becomes essential. This thesis resulted in the design and development of IoT architectures using modular and scalable structures for air quality monitoring based on data collected from cost-effective sensors for enhanced living environments. The proposed architectures address several concepts, including acquisition, processing, storage, analysis, and visualization of data. These systems incorporate an alert management Framework that notifies the user in real-time in poor indoor air quality scenarios. The software Framework supports multiple alert methods, such as push notifications, SMS, and e-mail. The real-time notification system offers several advantages when the goal is to achieve effective changes for enhanced living environments. On the one hand, notification messages promote behavioral changes. These alerts allow the building manager to identify air quality problems and plan interventions to avoid unhealthy air quality scenarios. The proposed architectures incorporate mobile computing technologies such as mobile applications that provide ubiquitous air quality data consulting methods s. Also, the data is stored and can be shared with medical teams to support the diagnosis. The state-of-the-art analysis has resulted in a review article on technologies, applications, challenges, opportunities, open-source IoT platforms, and operating systems. This review was significant to define the IoT-based Framework for indoor air quality supervision. The research leads to the development and design of cost-effective solutions based on open-source technologies that support Wi-Fi communication and incorporate several advantages such as modularity, scalability, and easy installation. The results obtained are auspicious, representing a significant contribution to enhanced living environments and occupational health. Particulate matter (PM) is a complex mixture of solid and liquid particles of organic and inorganic substances suspended in the air. Moreover, it is considered the pollutant that affects more people. The most damaging particles to health are ≤PM10 (diameter 10 microns or less), which can penetrate and lodge deep within the lungs, contributing to the risk of developing cardiovascular and respiratory diseases as well as lung cancer. Taking into account the adverse health effects of PM exposure, an IoT architecture for automatic PM monitoring was proposed. The proposed architecture is a PM real-time monitoring system and a decision-making tool. The solution consists of a hardware prototype for data acquisition and a Web Framework developed in .NET for data consulting. This system is based on open-source and technologies, with several advantages compared to existing systems, such as modularity, scalability, low-cost and easy installation. The data is stored in a database developed in SQL SERVER using .NET Web services. The results show the ability of the system to analyze the indoor air quality in real-time and the potential of the Web Framework for the planning of interventions to ensure safe, healthy, and comfortable conditions. Associations of high concentrations of carbon dioxide (CO2) with low productivity at work and increased health problems are well documented. There is also a clear correlation between high levels of CO2 and high concentrations of pollutants in indoor air. There are sufficient reasons to monitor CO2 and provide real-time notifications to improve occupational health and provide a safe and healthy indoor living environment. Taking into account the significant influence of CO2 for enhanced living environments, a real-time IoT architecture for CO2 monitoring was proposed. CO2 was selected because it is easy to measure and is produced in quantity (by people and combustion equipment). It can be used as an indicator of other pollutants and, therefore, of air quality in general. The solution consists of a hardware prototype for data acquisition environment, a Web software, and a smartphone application for data consulting. The proposed architecture is based on open-source technologies, and the data is stored in a SQL SERVER database. The mobile Framework allows the user not only to consult the latest data collected but also to receive real-time notifications in poor indoor air quality scenarios, and to configure the alerts threshold levels. The results show that the mobile application not only provides easy access to real-time air quality data, but also allows the user to maintain parameter history and provide a history of changes. Consequently, this system allows the user to analyze in a precise and detailed manner the behavior of air quality. Finally, an air quality monitoring solution was implemented, consisting of a hardware prototype that incorporates only the MICS-6814 sensor as the detection unit. This system monitors various air quality parameters such as NH3 (ammonia), CO (carbon monoxide), NO2 (nitrogen dioxide), C3H8 (propane), C4H10 (butane), CH4 (methane), H2 (hydrogen) and C2H5OH (ethanol). The monitoring of the concentrations of these pollutants is essential to provide enhanced living environments. This solution is based on Cloud, and the collected data is sent to the ThingSpeak platform. The proposed Framework combines sensitivity, flexibility, and measurement accuracy in real-time, allowing a significant evolution of current air quality controls. The results show that this system provides easy, intuitive, and fast access to air quality data as well as relevant notifications in poor air quality situations to provide real-time intervention and improve occupational health. These data can be accessed by physicians to support diagnoses and correlate the symptoms and health problems of patients with the environment in which they live. As future work, the results reported in this thesis can be considered as a starting point for the development of a secure system sharing data with health professionals in order to serve as decision support in diagnosis.
- Recognition of Activities of Daily Living and Environments Using Acoustic Sensors Embedded on Mobile DevicesPublication . Pires, Ivan; Marques, Gonçalo; Garcia, Nuno M.; Pombo, Nuno; Flórez-Revuelta, Francisco; Spinsante, Susanna; Teixeira, Maria Canavarro; Zdravevski, EftimThe identification of Activities of Daily Living (ADL) is intrinsic with the user’s environment recognition. This detection can be executed through standard sensors present in every-day mobile devices. On the one hand, the main proposal is to recognize users’ environment and standing activities. On the other hand, these features are included in a framework for the ADL and environment identification. Therefore, this paper is divided into two parts—firstly, acoustic sensors are used for the collection of data towards the recognition of the environment and, secondly, the information of the environment recognized is fused with the information gathered by motion and magnetic sensors. The environment and ADL recognition are performed by pattern recognition techniques that aim for the development of a system, including data collection, processing, fusion and classification procedures. These classification techniques include distinctive types of Artificial Neural Networks (ANN), analyzing various implementations of ANN and choosing the most suitable for further inclusion in the following different stages of the developed system. The results present 85.89% accuracy using Deep Neural Networks (DNN) with normalized data for the ADL recognition and 86.50% accuracy using Feedforward Neural Networks (FNN) with non-normalized data for environment recognition. Furthermore, the tests conducted present 100% accuracy for standing activities recognition using DNN with normalized data, which is the most suited for the intended purpose.