Teemu Henrikki Laine

Activity Recognition on Smartphones via Sensor-Fusion and KDA-Based SVMs

Although human activity recognition (HAR) has been studied extensively in the past decade, HAR on smartphones is a relatively new area. Smartphones are equipped with a variety of sensors. Fusing the data of these sensors could enable applications to recognize a large number of activities. Realizing this goal is challenging, however. Firstly, these devices are low on resources, which limits the number of sensors that can be utilized. Secondly, to achieve optimum performance efficient feature extraction, feature selection and classification methods are required. This work implements a smartphone-based HAR scheme in accordance with these requirements. Time domain features are extracted from only three smartphone sensors, and a nonlinear discriminatory approach is employed to recognize 15 activities with a high accuracy. This approach not only selects the most relevant features from each sensor for each activity but it also takes into account the differences resulting from carrying a phone at different positions. Evaluations are performed in both offline and online settings. Our comparison results show that the proposed system outperforms some previous mobile phone-based HAR systems.

Using Unity 3D to facilitate mobile augmented reality game development

Mobile augmented reality (mobile AR) enables virtual content such as 3D models, animations and annotations to be placed on top of a real world objects in any context. We applied mobile AR to develop the Calory Battle AR exergame to tackle worldwide childhood obesity. In this game the player finds and defuses virtual calory bombs in a real world environment. Specifically, we present the development of two game versions. First prototype was created without a third party game engine and it led to many challenges. To explore solutions to these challenges, we created a new version of game with the Unity 3D game engine. Using the Unity 3D, the game development process was simplified. A mixed-method usability evaluation on children and university students indicated that especially interaction with AR content and user interface clarity were improved in the Unity 3D version. This study produced three important contributions: 1) a novel mobile AR exergame to motivate children to move; 2) reimplementation of the game using the Unity 3D; and 3) results of a usability evaluation comparing two game versions. We expect that game engines such as the Unity 3D will become essential for AR game development in the future.

Designing Mobile Augmented Reality Exergames

Exergames aim to make exercise more enjoyable, especially for children and young adults who are accustomed to digital technologies. Calory Battle augmented reality (AR) is a mobile exergame that utilizes context awareness and AR to enable interaction with virtual content. Designing mobile exergames and AR interaction has received little scholarly attention. This article has several contributions to the design discussion: (1) implementation of a mobile AR exergame, (2) discourse on the game design process, (3) evaluation with 29 South Korean elementary school children and university students who suggested a good reception of the game and generated ideas for improvements of usability and AR interaction, (4) analysis of the game with respect to established game motivators and the Immersion, Scientificalness, Competitiveness, Adaptability, and Learning (ISCAL) exergame design model, (5) design principles and lessons learned, and (6) discussion of the flow experience in exergames. These results can be used by designers to create motivating and interactive mobile AR games.

Mobile Gateway for Ubiquitous Health Care System Using ZigBee and Bluetooth

A ubiquitous health care system takes the advantage of portability and small size of wireless sensor nodes to provide remote health care services and real-time health monitoring. A gateway is needed to mediate communication between a local sensor network and remote data consumers. In current implementations of ubiquitous health care systems ZigBee-based sensors are often used to gather vital sign data such as ECG and heart rate. Transferring large quantities of vital signs from a ZigBee-based sensor sink node to a gateway requires a bandwidth that ZigBee alone is not able to provide. In this paper, we present a technical design of a Bluetooth-based mobile gateway that bridges the connection between a sensor network and the Internet. Our system enables ubiquitous health care experience while providing a platform for additional services such as alarms, notifications and analysis of medical data. Controlling a sensor network from the mobile gateway is also possible. The flexible design of the system does not restrict its usage only to health care services - the gateway can be configured to work with any kind of sensor network having a sink node with Bluetooth capability.

ManySense: An Extensible and Accessible Middleware for Consumer-Oriented Heterogeneous Body Sensor Networks

Consumer-oriented wearable sensors such as smart watches are becoming popular, but each manufacturer uses its own data access mechanism. At the same time, the need for inferred context data is increasing in context-aware applications. A system is needed to provide a unified access to heterogeneous wearable devices for context-aware application developers. We propose ManySense—an Android-based middleware for heterogeneous consumer-oriented BSNs. Extensibility is achieved through adapter interfaces which allow sensors and context inferencing algorithms to be coupled with the middleware. Accessibility of the middleware allows third party applications to access raw sensor data and inferred context data uniformly. This paper provides two main contributions which are divided into several outcomes: (1) design and implementation of the ManySense BSN middleware that allows low-effort addition of new sensors and context inferencing algorithms through adapter interfaces, provides unified access to optionally filtered sensor data and inferred context data for third party applications, mediates control queries to sensor adapters and context inferencing adapters, and facilitates adapter development through an SDK and (2) evaluation of ManySense by comparing its performance with manual sensor data acquisition, analysis of ManySenses extensibility through adapter interfaces, and analysis of ManySenses accessibility from third party applications.

Enhancing Physical Education with Exergames and Wearable Technology

Increases in the numbers of obese and overweight children are a major issue in post-industrial societies because obesity can lead to severe health-related problems. In addition, many challenges affect the quantity and quality of physical education (PE) provided by schools. Exergames that combine exercise with gaming have been recognized as a possible method for motivating children to become physically active and to make PE more fun. Furthermore, exergames that utilize wearable sensors devices allow players movements to be tracked for estimating the efficiency of exercise. In this study, we developed the Running Othello 2 (RO2) exergame, where players wear a smartphone and a smart wrist band to compete in a board game enhanced with physical and pedagogical missions. In physical missions, the game uses inertial sensors and a heart rate meter to detect the physical activities of players. The pedagogical part of the game is based on the South Korean PE curriculum. We evaluated RO2 with 61 South Korean third grade elementary school students, 32 of whom learned curriculum topics by playing the game. The other 29 students comprised a control group who studied the pedagogical content using handouts. The results indicated that learning with RO2 was more efficient, the players were engaged, and their heart rates increased. Based on the evaluation, we identified several issues to be addressed in future research. Finally, we discussed how RO2 supports the educational affordances of wearables and we explained how exergames using wearables can overcome some of the challenges faced by PE.

Step counting on smartphones using advanced zero-crossing and linear regression

In this paper, we describe the design and implementation of a step counting algorithm for android-based smartphones. It uses an advanced zero crossing scheme to count the number of zero crossings (ZCs) and then employs linear regression to estimate the number of steps using ZCs. The system was tested for three subjects using both subject-dependent and subject-independent settings. During experiments, not only a high accuracy of estimation is recorded for both settings, the battery consumption is also shown to be very low. This proves the feasibility of employing our algorithm for step counting on smartphones for longer durations.

User-centered design of a context-aware nurse assistant (CANA) at Finnish elderly houses

In the age of mobile and ubiquitous computing, many nurses working in the healthcare sector still utilize traditional tools (e.g., paper calendars and clipboards) and pre-installed mobile applications (e.g., web browser, calendar) in their work activities. There exists a variety of mHealth applications, but none of them combines essential professional tools for nursing. We tackled this problem in the Finnish elderly house context through the User Centered Design (UCD) method whereby the target users actively participate in the design process. Together with 12 nurses, we first identified their profiles and their expectations on work-related mHealth application functionalities. The results were utilized in the conceptual design of Context-Aware Nurse Assistant (CANA), which combines the identified functionalities and provides context-sensitive services to consolidate nurses work activities. This paper contributes to the field of ubiquitous healthcare as follows: 1) initial user survey that shows the need for CANA; 2) UCD design process of CANA and discussion of UCDs suitability for designing ubiquitous healthcare systems, 3) technical architecture based on modular web services and 4) evaluation results of a low-fidelity CANA prototype. These results are of interest to software designers, healthcare professionals and context-aware application developers.

Machine learning and dynamic user interfaces in a context aware nurse application environment

The increasing usage of smartphones in daily life has received considerable attention in academic and industry driven research to be utilized in the health sector. There has been development of a variety of health-related smartphone applications. Currently, however, there are few to none applications based on nurses’ historical or behavioral preferences. Mobile application development for the health care sector requires extensive attention to security, reliability, and accuracy. In nursing applications, the users are often required to navigate in hospital environments, select patients to support, read the patient history and set action points to assist the patient during their shift. Finally, they have to report their performance on patient related activities and other relevant information before they leave for the day. In a working day, a nurse often visits different locations such as the patient’s room, different laboratories, and offices for filling reports. There is still a limited capability to access context relevant information on a smartphone with minimal recourse such as Wi-Fi triangulation. The Wi-Fi triangulation signals fluctuate significantly for indoor location positioning. Therefore, providing relevant location based services to a mobile subscriber has become challenging. This paper addresses this gap by applying machine learning and behavior analysis to anticipate the potential location of the nurse and provide the required services. The application concept was already presented at the IMCOM 2015 conference. This paper focuses on the process to ascertain a user’s context, the process of analyzing and predicting user behavior, and finally, the process of displaying the information through a dynamically generated UI.

Sustainable usage through emotional engagement: a user experience analysis of an adaptive driving school application

This paper explores the factors affecting sustainable usage of digital services, such as mobile learning (m-learning) applications. We define a conceptual model of digital service sustainability and its measurement indicators and criteria. We assess the conceptual model by applying it to an adaptive m-learning application, which provides theory and assessment for driving school students based on their competence and learning progress. Additionally, the application provides mandatory self-evaluation reports to instructors after each practical driving session. The application assessment focused on how students’ performed the predefined tasks at a usability laboratory. The assessment was based on educational components in conjunction with user experience factors of delightfulness, adjustability, satisfaction, and reliability. The key qualitative performance aspects proved to be user-centered (e.g. individual differences in cognitive and learning capabilities), UI-focused (e.g. interaction and adaptability to screen resolution), and content-related (e.g. contextual delivery, emotional appeal, adaptability). These elements resulted in overall delightful and effortless experience. The evaluation results indicate that the user experience factors impact sustainable usage because students are emotionally attached to the application. Therefore, we suggest that emotional attachment assessment results may serve as an indicator for sustainability of m-learning applications. These results help mobile application designers to design sustainable applications.