HCI Bibliography Home | HCI Conferences | PETRA Archive | Detailed Records | RefWorks | EndNote | Hide Abstracts
PETRA Tables of Contents: 0809101112131415

Proceedings of the 2015 International Conference on PErvasive Technologies Related to Assistive Environments

Fullname:Proceedings of the 8th International Conference on PErvasive Technologies Related to Assistive Environments
Editors:Fillia Makedon
Location:Corfu, Greece
Dates:2015-Jul-01 to 2015-Jul-03
Publisher:ACM
Standard No:ISBN: 978-1-4503-3452-5; ACM DL: Table of Contents; hcibib: PETRA15
Papers:100
Links:Conference Website
  1. Usability and HCI issues
  2. Pervasive systems for the aged & smart health devices
  3. Robotic devices and multimodal interfaces
  4. Signal & image processing for ambient intelligence, pervasive computing
  5. Data modeling & information management for pervasive assistive environments
  6. Tools, infrastructures, architectures & techniques for deploying pervasive applications in assistive environments
  7. Wearable systems and monitoring devices
  8. Behavior modeling, analysis & prediction
  9. Signal and image processing for ambient intelligence & pervasive computing
  10. Virtual reality & robotic technologies for rehabilitation workshop (VRRTR)
  11. Tools, infrastructures, architectures & techniques for deploying pervasive applications in assistive environments
  12. Data modeling and information management for pervasive assistive environments
  13. Behavior modeling, analysis & prediction
  14. RAsEnv -- workshop on robotics for assistive environments
  15. Non-invasive monitoring technologies for disorder assessment workshop (SleepMon)
  16. Affective computing for biological activity recognition in assistive environments workshop (STHENOS)

Usability and HCI issues

Comparing projected in-situ feedback at the manual assembly workplace with impaired workers BIBAFull-Text 1
  Markus Funk; Andreas Bächler; Liane Bächler; Oliver Korn; Christoph Krieger; Thomas Heidenreich; Albrecht Schmidt
With projectors and depth cameras getting cheaper, assistive systems in industrial manufacturing are becoming increasingly ubiquitous. As these systems are able to continuously provide feedback using in-situ projection, they are perfectly suited for supporting impaired workers in assembling products. However, so far little research has been conducted to understand the effects of projected instructions on impaired workers. In this paper, we identify common visualizations used by assistive systems for impaired workers and introduce a simple contour visualization. Through a user study with 64 impaired participants we compare the different visualizations to a control group using no visual feedback in a real world assembly scenario, i.e. assembling a clamp. Furthermore, we introduce a simplified version of the NASA-TLX questionnaire designed for impaired participants. The results reveal that the contour visualization is significantly better in perceived mental load and perceived performance of the participants. Further, participants made fewer errors and were able to assemble the clamp faster using the contour visualization compared to a video visualization, a pictorial visualization and a control group using no visual feedback.
SmartTactMaps: a smartphone-based approach to support blind persons in exploring tactile maps BIBAFull-Text 2
  Timo Götzelmann; Klaus Winkler
Despite increasing digitalization of our society many blind persons still have very limited access to predominantly pictorial information such as maps. In this paper we introduce a novel approach to improve the accessibility of maps for blind users by utilizing the abilities of standard smartphones. A major issue of tactile maps is the limited discriminability of the humans' tactile sense. Textual annotation of maps is crucial, but adds much complexity to tactile maps. Additionally, only few Braille labels can be accommodated to maintain legibility. In our approach we link smartphones with adapted tactile maps which transforms the physical maps into interactive surfaces using both the tactile and the auditory modality. We integrate machine readable metadata into these maps which can be recognized by the smartphones' camera to immediately obtain detailed map descriptions from a free global database. During tactile exploration of the map, blind users can request auditory explanations by interacting with the mobile application. An experimental application and a user study demonstrate the feasibility of our approach.
Pot hunter: a virtual reality game for analyzing range of motion BIBAFull-Text 3
  Shawn N. Gieser; Peter Sassaman; Eric Becker; Fillia Makedon
Patients undergoing physical therapy go through a series of sessions performing exercises to help improve the range of motion (RoM) in affected regions of the body due to disease or injury. However, patients find these tasks repetitive and boring and end up not completing the prescribed therapy program. It has been shown that game based therapy exercises have led to increased rates of compliance. In this paper, we provide a continuation of previous work in VR-based therapy and present Pot Hunter, and one type of RoM analysis for when a person reaches above their head.
About the development of an interactive assistance system for impaired employees in manual order picking BIBAFull-Text 4
  Andreas Baechler; Peter Kurtz; Thomas Hoerz; Georg Kruell; Liane Baechler; Sven Autenrieth
For the user-centered development of an assistance system for manual order picking with impaired people, the interaction partners "human and machine" are interviewed and examined in a first analysis. A survey of user-tasks and requirements was carried out in form of an expert interview with a questionnaire in a sheltered workshop. Thereby 78 questionnaires about impaired employees are filled in by the pedagogical staff of the sheltered workshop. The results of the evaluation give information about the properties and capabilities, but also about the personal benefits of such a system for the impaired people. The evaluated results serve as a framework for the hardware- and software-design of the order picking system.
Accessible games for blind children, empowered by binaural sound BIBAFull-Text 5
  Konstantinos Drossos; Nikolaos Zormpas; George Giannakopoulos; Andreas Floros
Accessible games have been researched and developed for many years, however, blind people still have very limited access and knowledge of them. This can pose a serious limitation, especially for blind children, since in recent years electronic games have become one of the most common and wide spread means of entertainment and socialization. For our implementation we use binaural technology which allows the player to hear and navigate the game space by adding localization information to the game sounds. With our implementation and user studies we provide insight on what constitutes an accessible game for blind people as well as a functional game engine for such games. The game engine developed allows the quick development of games for the visually impaired. Our work provides a good starting point for future developments on the field and, as the user studies show, was very well perceived by the visually impaired children that tried it.
Design approaches for the gamification of production environments: a study focusing on acceptance BIBAFull-Text 6
  Oliver Korn; Markus Funk; Albrecht Schmidt
Gamification is an ever more popular method to increase motivation and user experience in real-world settings. It is widely used in the areas of marketing, health and education. However, in production environments, it is a new concept. To be accepted in the industrial domain, it has to be seamlessly integrated in the regular work processes.
   In this work we make the following contributions to the field of gamification in production: (1) we analyze the state of the art and introduce domain-specific requirements; (2) we present two implementations gamifying production based on alternative design approaches; (3) these are evaluated in a sheltered work organization. The comparative study focuses acceptance, motivation and perceived happiness.
   The results reveal that a pyramid design showing each work process as a step on the way towards a cup at the top is strongly preferred to a more abstract approach where the processes are represented by a single circle and two bars.

Pervasive systems for the aged & smart health devices

Test battery for assessment of cognitive function in older employees: performance, brain processes, and cardiovascular "costs" BIBAFull-Text 7
  Sergei A. Schapkin; Xenija Weißbecker-Klaus
Declines in cognitive function with advancing age may cause performance decrements and stress in older workers who have to do complex work. We aimed to develop a system for complex assessment of cognitive load in older employees with parallel registration of behavioural, neuronal and cardiovascular activity. We present the test battery that can provide reliable data on age-related cognitive deficits. Using this test battery, EEG and cardiovascular indexes were found which are stable over time, sensitive to ageing and cognitive load. The test battery and physiological indexes are suitable for the development of a computational model describing heart-brain interactions during work performance. The model may be further implemented into mobile devices using for the online assessment of cognitive load.
Maintaining good relationships in clinical setting: the bonus DOMUS project BIBAFull-Text 8
  Kevin Bouchard; Sylvain Giroux; Robert Radziszewski; Mathieu Gagnon; Quentin Szymanski; Stéphanie Pinard; Mélanie Levasseur; Nathalie Bier
The DOMUS laboratory has recently participated in the construction of a brand new residence for persons with Traumatic Brain Injury (TBI). This building which comprises six apartments, and four bedrooms is equipped with the latest smart home technology (sensors, effectors, etc.). It is a living lab where prototype software, algorithms and technologies can be deployed for long term evaluation. One of the challenges that we have to face in a living lab setting is the maintaining of the good relationships with both the professionals and the residents. In that regard, the DOMUS team worked toward implementing simple technological services that would rapidly and directly enhance social participation and the quality of life of the residents. The goal is also to motivate them into taking part of the various research projects and to establish a trust relationship. In this paper, we present the Bonus DOMUS, a project that was created toward these aims. It enables the residents to have customized alarms and motivational messages.
Automatic speed control for SmartWalker BIBAFull-Text 9
  Jiwon Shin; Ivo Steinmann; Bertrand Meyer
Ensuring mobility of the elderly is an important task in our aging society. To this end, this paper presents an automatic speed controller for the SmartWalker -- a high-tech extension of a regular walker. The walker locates its user by detecting the user's legs using a laser range scanner. The controller then determines the optimal speed for the walker using the user's location and other sensory data. We evaluated the walker and its speed controller with thirteen residents at three different retirement homes. Our analysis showed that the walker with the controller is slightly more comfortable and easier to maneuver than the walker without the controller and is more liked than traditional walkers.
A survey on an ingestible sensor for evaluating medication adherence in elderly people BIBAFull-Text 10
  Ioannis Kalpouzos; Kostas Giokas; Dimitrios Koutsouris
In this paper, we describe an ingestible sensor system to evaluate medication adherence in elderly people. After drug ingestion, an ingestible sensor, attached on the pill, is activated upon contact with gastric fluid and communicates with a wearable monitor that records ingestion data. The monitor records biometric data (e.g. heart rate, blood pressure) and communicates with a secured server, via cellular phone, for storage and further process. Summary reports can be generated back to the phone or potentially to anyone involved with the aged person caring. Clinical research trials proved the effectiveness of the Ingestible Sensor System in evaluating medication adherence.
Augmenting everyday artefacts to support social interaction among senior peers BIBAFull-Text 11
  Elena Nazzi; Tomas Sokoler
Novel technological possibilities emerge when tangible and social computing come together. This paper explores the potential of such technology when designing for seniors and their social interaction. Our research is guided by the concept of twitterIDo, which is to make seniors' everyday activities more visible by augmenting everyday artefacts to communicate the ongoing activity they are used for. We engaged a local community of seniors in a living lab to explore the possibilities of twitterIDo in real life situations. This paper presents a series of interactive prototypes of everyday artefacts and displays designed to start a dialogue with the seniors on how twitterIDo-technology may fit into their everyday situations. Our findings point out how augmented everyday artefacts can make a positive difference when designing technology in a domain such the one of seniors' and their social interaction. Presenting our experience we want to exemplify and open up new directions for designing tangible and social computing technology in this domain.
Cooking risk analysis to enhance safety of elderly people in smart kitchen BIBAFull-Text 12
  Rami Yared; Bessam Abdulrazak; Thomas Tessier; Philippe Mabilleau
Risk analysis during cooking activities enables to build a cooking-safe system in order to enhance safety of elderly people in smart kitchen. The Kitchen is the second place where majority of domestic accidents occur, and in particular the oven presents the most principal reason of fire accidents in the residence. The paper presents insightful cooking risk analysis that permits to determine the pertinent parameters to be monitored and measured during cooking in order to prevent risks. We investigated several cooking experiments, and analyzed the composition of heated cooking materials, the concentrations of gases in the cooking smoke, and humidity. The pertinent parameters determined in this paper are: concentrations of gases in the cooking smoke, ambient temperature, utensil temperature, burner temperature, relative humidity, and presence of an object on burner.

Robotic devices and multimodal interfaces

An interactive framework for learning user-object associations through human-robot interaction BIBAFull-Text 13
  Michalis Papakostas; Konstantinos Tsiakas; Natalie Parde; Vangelis Karkaletsis; Fillia Makedon
A great deal of recent research has focused on social and assistive robots that can achieve a more natural and realistic interaction between the agent and its environment. Following this direction, this paper aims to establish a computational framework that can associate objects with their uses and their basic characteristics in an automated manner. The goal is to continually enrich the robot's knowledge regarding objects that are important to the user, through verbal interaction. We address the problem of learning correlations between object properties and human needs by associating visual with verbal information. Although the visual information can be acquired directly by the robot, the verbal information is acquired via interaction with a human user. Users provide descriptions of the objects for which the robot has captured visual information, and these two sources of information are combined automatically. We present a general model for learning these associations using Gaussian Mixture Models. Since learning is based on a probabilistic model, the approach handles uncertainty, redundancy, and irrelevant information. We illustrate the capabilities of our approach by presenting the results of an initial experiment run in a laboratory environment, and we describe the set of modules that support the proposed framework.
Towards user-centered design of a robotic prosthetic hand with EMG control interfaces BIBAFull-Text 14
  Velin Dimitrov; Nicholas Cebry; Çagdas Önal; Taskin Padir
We present the design evolution of robotic prosthetic hands at WPI over the last couple years. Following a user-centered design approach, we have identified a series of requirements that good prosthetic hand designs should have. We describe in detail the mechanical, electrical, and software subsystems of each hand, and the relative advantages and disadvantages of the design decisions made during the development phases. In addition, we describe, in depth, our work towards an intuitive and high bandwidth EMG interface for assistive products. We compare various amplifiers and their performance for our implementation of an EMG interface.
Nonverbal communication with a humanoid robot via head gestures BIBAFull-Text 15
  Salah Saleh; Karsten Berns
Social interactive robots require sophisticated perception and cognition abilities to behave and interact in a natural human-like way. The proper perception of behavior of interaction partner plays a crucial role in social robotics. The interpretation of these behaviors and mapping them to their exact meanings is also an important aspect that interactive robots should have. This paper proposes an interaction model for communicating verbally and nonverbally with human. Human behavior, during the interaction with the robot, is perceived and then interpreted depending on the situation in which the behavior has been detected. In this model, head gestures are used as a back channel (feedback) for the robot to adapt the interaction scenario. The back channel signals can be consciously or unconsciously generated by human. Simultaneously, the eye gazes are also detected to ensure right interpretation of head gestures. In order to recognize the human head gestures, head poses have been tracked over time. A stream of images with their corresponding depth information, acquired from a Kinect sensor, are used to find, track, and estimate the head poses of human. The proposed model has been tested in various experiments with different scenarios in interaction with human.
Position paper: accessible human-robot interaction (AHRI) BIBAFull-Text 16
  Claudia Loitsch; Michael Schmidt; Gerhard Weber
Assistive robots that address different impairments and continuously changing capabilities by providing manifold interactions based on profiles are not targeted in current research. Substantiated by a survey of human-robot interaction for assistive systems, we claim the need for establishing the innovative research topic of Accessible Human-Robot Interaction (AHRI). It is shown that limitations or loss of motoric, sensory, and mental capabilities as well as multiple impairments and resulting barriers regarding interaction, communication and perception are not sufficiently incorporated in designs of assistive robots. To achieve a design for all, fundamental research on varying needs, capabilities, and preferences along with derivation of profiles and developments of new adaptive and adaptable interaction concepts is necessary.
Development of MERCURY version 2.0 robotic arms for rehabilitation applications BIBAFull-Text 17
  N. Moustakas; P. Kartsidis; A. Athanasiou; A. Astaras; P. D. Bamidis
MERCURY is a robotic platform comprised of two mechatronic robotic arm manipulators, a body machine interface (BMI) in the form of a wearable hardware sensor sleeve and a brain computer interface (BCI). It is a prototype system primarily aimed at research in human-robotic interfaces, medical rehabilitation and assistive technologies for patients with Spinal Cord Injury. This paper discusses improvements implemented in the second generation of the system, following evaluation of results obtained from pilot testing the first generation robotic setup. The system now integrates two of the second generation MERCURY robotic arms. The main improvements are digitization of control signals, the addition of anthropomorphic hands in place of pincers, two additional degrees of freedom, improved telecommunications and BCI control.
Examples for a ubiquitous mobility assistant as outcome of the inDAgo project BIBAFull-Text 18
  Stefanie Müller; Antonija Mrsic Carl; Peter Klein; Henrik Rieß; Denise Bender
In the project inDAgo several demonstrators for helping elderly people in their mobility were developed. User research results showed an aversion to conspicuous assistive products. Thus the project focused on developing demonstrators with the appearance of invisible companions. The autonomous demonstrators were built in a ubiquitous, adaptive environment. This paper shows an extract of these products and describes their acceptance in the target group.

Signal & image processing for ambient intelligence, pervasive computing

Automatic soundscape quality estimation using audio analysis BIBAFull-Text 19
  Theodoros Giannakopoulos; Georgios Siantikos; Stavros Perantonis; Nefta-Eleftheria Votsi; John Pantis
The huge growth of population size along with all the accompanying impacts, like traffic flow, commercial and industrial activities have led to a respective increase of noise pollution in the urban environments. In most cases, noise pollution in big cities is characterized by low-frequency and continuous background sounds. This ever-growing environmental problem engages health risks and major complaints of annoyance on behalf of millions of citizens. Therefore, sustainable urban planning needs to seriously take into consideration the task of mitigating environmental noise. In addition, the quality of the acoustic environment plays an important role in urban as well as in rural and natural spaces, since it has been proven to affect biodiversity. In this paper, we demonstrate how efficiently assessing soundscape quality can be applied to real recordings from various sites. The evaluation of the qualitative attributes of the soundscape is carried out combining space-sound-human presence. The mapping of the extracted feature statistics to the perceived soundscape quality level is achieved through a Support Vector Machine Regression model. Extensive experiments have been carried out on a real-world dataset and the resulting performance evaluation proves that the proposed architecture can be applied to assess the soundscape quality of both natural and urban spaces.
An interactive and intuitive stem accessibility system for the blind and visually impaired BIBAFull-Text 20
  Rahul Kumar Namdev; Pattie Maes
We present an intuitive and interactive platform to make complex STEM (Science Technology Engineering and Mathematics) educational materials accessible to blind and visually impaired people using a mini-hyperbraille device with almost no loss of information as compared to printed materials.
   We have come up with a novel way to increase the effective resolution of the braille device by adding a mechanical XY gantry. Using this XY gantry enables us to create an ultrahigh resolution, larger surface braille device without paying the prohibitive price charged for the bigger hyper Braille displays available in the market. In addition to that, to further augment usability and intuitiveness of our system, we have integrated a nod-ring, which is a tiny finger-worn device for supporting additional hand gestures such as zoom in and out.
   Previous studies have shown that the use of zoom and pan can increase usability and improve the understanding of tactile graphics. Along with zooming and panning, our system uses vibrating patterns, rhythmic motions, synthetic voice and synchronized voiced-vibrations to convey information to blind users in an intuitive way. We also implemented a touch gesture recognition framework on our touch enabled braille device. Using these touch gestures and a high quality synthetic voice, we have developed a highly responsive system for providing voice annotations of the graphics content.
   An important contribution of this work is the implementation of a high-quality system for automatic transcription of STEM (including difficult math Nemeth translations) books into braille. Learning resources for blind people are quite sparse and transcription of STEM material is a very expensive and time consuming process. Using our automated transcription platform it is easy, fast and inexpensive for publishers to transcribe STEM books into braille books. The scope of this automated transcription platform is not only limited to STEM books but it can be used for transcription of any content/book/web-page available on-line.
Assistive positioning system based inertial techniques and wavelet denoising BIBAFull-Text 21
  Teodor Lucian Grigorie; Petre Negrea; Ioana Raluca Edu; Felix Constantin Adochiei
The here presented work refers to the development of a new bi-dimensional strap-down inertial navigator based on an algorithm that reduces their sensors noise by using a tuning method of the wavelet transform assisted by a GPS system. In this way is increased the INS standalone operation period with acceptable accuracy during the GPS outages. The tuning of the wavelet based filters for the inertial miniaturized sensor uses the Partial Directed Coherence method. Shown are: the navigator theory, the used wavelet tuning method basic principle, and the experimental validation of the improved navigator based wavelet denoising.
An intra-fraction markerless daily lung tumor localization algorithm for EPID images BIBAFull-Text 22
  Timothy Rozario; Sergey Bereg; Weihua Mao
In this work we report an intra-fractional markerless algorithm that accurately detects lung tumors on mV projections within the beam's eye view, while minimizing harmful effects such as poor soft tissue resolution, global image distortion, image blurring and scattering due to intrafraction target motion and radiation scatter.. First, we generate two sets of DRRs digitally reconstructed radiographs-background DRR without tumor and tumor only DRR from the 4D CT planning data after the tumor has been initially segmented out. Next, the composite DRR is generated by fusing the tumor DRR on the background. The composite DRR along with the matching mV projection are divided into a matrix of small tiles. The tile configuration is automatically set up such that the tumor always remains within the beam's-eye-view geometry on the composite DRR. In order to locate the tumor on the mV projection, the tumor DRR is fused at different locations on the background DRR while the tiles of the composite DRR are globally shifted. For each configuration, the composite DRR is matched with the corresponding mV projection. A simple NCC normalized cross correlation is used to compute the similarity between the composite DRR and corresponding mV projection tiles. Finally, the location of the lung tumor on the mV projection is identified based on the best match found.
   The algorithm was successfully tested on a dynamic chest phantom at our institution. Approximately 5700 raw images over 12 gantry angles were tested and the tumor was accurately located on every mV projection. Although, the chest phantom was created to mimic the human chest anatomy with neighboring organs, tissues and bony structures, which introduced strong signals, the maximum error reported was less that 1.6 mm while the average error reported was less than 0.7 mm.
Using dual camera smartphones as advanced driver assistance systems: NAVIEYES system architecture BIBAFull-Text 23
  Duguleana Mihai; Gîrbacia Florin; Mogan Gheorghe
Over the last years, automotive industry has shown a tremendous interest in Advanced Driver Assistance Systems (ADASs), especially the ones based on driver's bio-features. As most car producers strive to meet the increasing needs of high-end and even average consumers, more and more complex systems are being developed. The current trend is to maximize the synergy between humans and machines by designing better user interfaces (UIs) which can anticipate the behavior of the driver. Great research efforts are put into inferring car position, traffic environment, driver's condition (e.g. degree of drowsiness, driving skill, emotional state) and intentions (e.g. changing lanes, overtaking other cars). The goal of this study is to design and implement an ADAS based on a dual camera mobile phone. We address the mathematical models, the main software modules and the system architecture implied by such a system. Within this paper, we focus on developing the system architecture, as seen from different perspectives: User View, Logic Segmentation View and Process View.
An integrated RGB-D system for looking up the meaning of signs BIBAFull-Text 24
  Christopher Conly; Zhong Zhang; Vassilis Athitsos
Users of written languages have the ability to quickly and easily look up the meaning of an unknown word. Those who use sign languages, however, lack this advantage, and it can be a challenge to find the meaning of an unknown sign. While some sign-to-written language dictionaries do exist, they are cumbersome and slow to use. We present an improved American Sign Language video dictionary system that allows a user to perform an unknown sign in front of a sensor and quickly retrieve a ranked list of similar signs with a video example of each. Earlier variants of the system required the use of a separate piece of software to record the query sign, as well as user intervention to provide bounding boxes for the hands and face in the first frame of the sign. The system presented here integrates all functionality into one piece of software and automates head and hand detection with the use of an RGB-D sensor, eliminating some of the shortcomings of the previous system, while improving match accuracy and shortening the time required to perform a query.

Data modeling & information management for pervasive assistive environments

A framework for self-managing database support and parallel computing for assistive systems BIBAFull-Text 25
  Dennis Marten; Andreas Heuer
There is no doubt that assistive systems are and will be a great part of our everyday lives. Thus, it is not suprising that in recent years researchers all over the world have been putting a lot of effort into their development. One of the most challenging problems usually is the handling of enormous amounts of data, which often has been collected by numerous sensors. This data is the basis of models, e.g. for prediction of movement, which has been derived by statistical methods, e.g. machine learning. However, due to the massive amounts of data, conventional statistical tools suffer from performance issues. In this paper, we would like to introduce and discuss a framework that combines the popular, statistical development tool R, database technology and the widely known MapReduce framework. Our main focus is placed on user-friendliness, meaning that the user does not have to change anything in his R-script, but still benefits from parallel computation and the in- and output power of databases.
Toward a platform for collecting, mining, and utilizing behavior data for detecting students with depression risks BIBAFull-Text 26
  Einoshin Suzuki; Yutaka Deguchi; Tetsu Matsukawa; Shin Ando; Hiroaki Ogata; Masanori Sugimoto
In this paper, we present our plan for constructing a platform for collecting, mining, and utilizing behavior data for detecting students with depression risks. Unipolar depression makes a large contribution to the burden of disease, being at the first place in middle- and high-income countries. We survey descriptors of depressions and then design a data collection platform in a classroom based on the assumption that such descriptors are also effective to students with depression risks. Visual, acoustic, and e-learning data are chosen for collection and various issues including devices, preprocessing, and consent agreements are investigated. We also show two kinds of utilization scenarios of the collected data and introduce several techniques and methods we developed for feature extraction and early detection.
Generating privacy constraints for assistive environments BIBAFull-Text 27
  Hannes Grunert; Andreas Heuer
Smart environments produce large amounts of data by a plurality of sensors, which constantly track our activities and desires. To support our daily life, assistive environments process these data to calculate our intentions and future actions. In many cases, more information than required are generated and processed by the assistive system. Thereby, the system can learn more about the user than intended. By this, the users' right to informational self-determination is injured, because they lose control how their data is used.
   In this paper, we present a model to let the user formulate requirements to protect his privacy in smart environments. These requirements are transformed into multiple integrity constraints, which ensure privacy.
An ontological approach towards psychological profiling of breast cancer patients in pervasive computing environments BIBAFull-Text 28
  Irini Genitsaridi; Kostas Marias; Manolis Tsiknakis
Pervasive computing has presented new unobtrusive patient monitoring solutions through the use of intelligent wearable devices and environmental sensors. In this work, we study the problem of psychological patient profiling in the context of pervasive computing environments. This paper is focused on the particular case of profiling breast cancer patients. Essentially, we present an ontological solution able to represent the psychological profile of breast cancer patients and we analyze the methodological approach to assess the patient's psychological state based on his profile. In addition, we distinguish the various types of information in the patient profile and study the possibility of collecting this information from unobtrusive monitoring solutions instead of the current, exhaustive process of filling clinical questionnaires.
Variation in oxygen saturation measurements in very low birth weight infants BIBAFull-Text 29
  Olli-Pekka Rinta-Koski; Jaakko Hollmén; Markus Leskinen; Sture Andersson
Low birth weight is heavily correlated with health issues. Very low birth weight (VLBW) infants, with a birth weight below 1500 g, are particularly at risk, and often subject to multiple developmental problems.
   The Neonatal Intensive Care Unit (NICU) at Helsinki University Central Hospital has been collecting patient data in a database since 1999. We studied data collected from 2059 VLBW infants admitted between 1999 and 2013. Our aim was to study the variance of oxygen saturation measurements and compliance with guidelines as an example of using statistical means to assess quality of care from vital trend measurements. As an example of quality control, we have studied the discrepancy between automatic measurements and manual readings taken from the same sensor output.

Tools, infrastructures, architectures & techniques for deploying pervasive applications in assistive environments

A prototype of a real-time solution on mobile devices for heart tele-auscultation BIBAFull-Text 30
  Julio Cesar Bellido; Giuseppe De Pietro; Giovanna Sannino
The accessibility of the health service is often criticized by patients whose clinical or geographical conditions make their transfer to health facilities problematic. This paper shows how the use of technology can help individual to exercise their right to healthcare by offering healthcare services that lead to an improved quality life for the patient. A prototype of a mobile tele-auscultation solution developed in this study could meet this need and relieve the patient from the burden of having to reach the appropriate health facilities to undergo a specialist examination. Patients equipped with an electronic stethoscope can self-monitor their health in any place and can start a live tele-auscultation session with a remote doctor.
Sole based tactile information display for visually impaired pedestrian navigation BIBAFull-Text 31
  Slim Kammoun; Wahiba Bouhani; Mohamed Jemni
Well understanding guidance instructions when using an electronic orientation aids is vital for people with visual impairments to reach destinations or avoid obstacles and hazards. In this paper we describes a study about how to make use of a tactile sole with 5 vibrators to present spatial information for visually impaired people while using an electronic orientation aid. We designed and evaluated a sole based tactile information display that delivers directional information in real time through shoes. A Wizard-of-Oz preliminary evaluation with 2 blindfolded subjects has been conducted. Result shows that designed haptic sole is a safer, non-intrusive and promising approach that can be used to simplify guidance instructions.
Adaptive music technology using the Kinect BIBAFull-Text 32
  Kimberlee Graham-Knight; George Tzanetakis
In this paper a new approach to music-making for people with disabilities is discussed. Until recently, the technology to enable people with disabilities to make music has been relatively limited, consisting primarily of mechanical approaches. With new developments in computing, including the Microsoft Kinect, touchless sensors are providing a new way for people with disabilities to interface with instruments in novel ways. There have been few papers that made empirical measurements of adaptive musical instruments, including latency. This paper will fill this gap by detailing an adaptive musical interface using the Microsoft Kinect. Then the overall latency, including response time of the musician, will be measured, and methods to decrease this latency will be proposed.
A multimodal adaptive session manager for physical rehabilitation exercising BIBAFull-Text 33
  Konstantinos Tsiakas; Manfred Huber; Fillia Makedon
Physical exercising is an essential part of any rehabilitation plan. The subject must be committed to a daily exercising routine, as well as to a frequent contact with the therapist. Rehabilitation plans can be quite expensive and time-consuming. On the other hand, tele-rehabilitation systems can be really helpful and efficient for both subjects and therapists. In this paper, we present ReAdapt, an adaptive module for a tele-rehabilitation system that takes into consideration the progress and performance of the exercising utilizing multisensing data and adjusts the session difficulty resulting to a personalized session. Multimodal data such as speech, facial expressions and body motion are being collected during the exercising and feed the system to decide on the exercise and session difficulty. We formulate the problem as a Markov Decision Process and apply a Reinforcement Learning algorithm to train and evaluate the system on simulated data.
Self-managed patient-game interaction using the Barrett WAM arm for motion analysis BIBAFull-Text 34
  Alexandros Lioulemes; Paul Sassaman; Shawn N. Gieser; Vangelis Karkaletsis; Fillia Makedon; Vangelis Metsis
In this paper, we present a framework for physical rehabilitation, that uses a combination of video gaming and robotic technology to allow the monitoring and progress tracking of a person during physical therapy. The system, called MAGNI, uses the advanced control capabilities of the Barrett WAM Arm robot and a custom-made video game. The MAGNI system helps the patient to complete a rehabilitation session through a user-system, game-based interaction program, involving exercises prescribed by a therapist. The system can control and supervise the rehabilitation sessions to ensure compliance and safe exercising. It uses motion analysis to provide an evaluation of the patient's progress over time. The MAGNI system records the position of the subject's hand during game interaction with the robotic arm and analyzes this data using pattern matching and machine learning algorithms, in order to guide self-managed physical therapy. Our experiments show that we can accurately classify user motion activity between a set of different exercises, and measure user compliance with the prescribed regimens.

Wearable systems and monitoring devices

CapWalk: a capacitive recognition of walking-based activities as a wearable assistive technology BIBAFull-Text 35
  Marian Haescher; Denys J. C. Matthies; Gerald Bieber; Bodo Urban
In this research project, we present an alternative approach to recognize various walking-based activities based on the technology of capacitive sensing. While accelerometry-based walking detections suffer from reduced accuracy at low speeds, the technology of capacitive sensing uses physical distance parameters, which makes it invariant to the duration of step performance. Determining accurate levels of walking activity is a crucial factor for people who perform walking with tiny step lengths such as elderlies or patients with pathologic conditions. In contrast to other gait analysis solutions, CapWalk is mobile and less affected by external influences such as bad lighting conditions, while it is also invariant to external acceleration artifacts. Our approach enables a reliable recognition of very slow walking speeds, in which accelerometer-based implementations can fail or provide high deviations. In CapWalk we present three different capacitive sensing prototypes (Leg Band, Chest Band, Insole) in the setup of loading mode to demonstrate recognition of sneaking, normal walking, fast walking, jogging, and walking while carrying weight. Our designs are wearable and could easily be integrated into wearable objects, such as shoes, pants or jackets. We envision such gathered information to be used to assist certain user groups such as diabetics, whose optimal insulin dose is depending on bread units and physical activity or elderlies whose personalized dosage of medication can be better determined based on their physical activity.
Human gait classification using a tri-axial accelerometer BIBAFull-Text 36
  Ahmad Lotfi; Minh Nguyen; Caroline Langensiepen
Gait is the pattern of movement of limbs and can indicate people's health status. Most of the proposed systems for gait analysis either use complex facilities or only provide statistical data to describe gait parameters. An affordable system is required which is able to cover the complete process of gait analysis. To address this requirement, a machine-learning based system using a wearable accelerometer and an enhanced feature extraction algorithm to identify different types of human gaits is proposed. It also determines the optimum learning algorithm to be used. Experimental study proves that the system is reliable in gait classification and therefore has the potential to be used as a contributory tool in disease diagnosis.
Sensors activation time predictions in smart home BIBAFull-Text 37
  Mohamed Tarik Moutacalli; Abdenour Bouzouane; Bruno Bouchard
Activity recognition is the most challenging stage of technological assistance which offers automatic support, when needed, to elderly and disabled people such as Alzheimer's patients living in smart homes. Many approaches and techniques were proposed for activity recognition while other technological assistance stages were barely explored. In this paper, after presenting our activity recognition approach and explaining how the artificial agent will use it to decide when to intervene offering help, we use time series forecasting in order to better choose the intervention time.
Performance characterization of self-calibrating protocols for wearable EEG applications BIBAFull-Text 38
  Thrasyvoulos Karydis; Filipe Aguiar; Simmie L. Foster; Andreas Mershin
Therapeutic Neurofeedback (NFB) using real-time electroencephalography (EEG) data works by reinforcing desired brainwave patterns. Although EEG is a well-established diagnostic tool and EEG-NFB shows great promise for enhancing cognitive performance and treating neurological disorders, proof of its efficacy has been limited. Here we characterize a novel Self-Calibrating Protocol (SCP) method coupled to five standard machine learning algorithms to classify brain states corresponding to the experience of "pain" or "no pain".
   Our results indicate that commercially available, wearable EEG sensors provide sufficient data fidelity to robustly differentiate the two "perceptually opposite" brain states. Crucially, use of SCP allows us for the first time to bypass the pitfalls associated with trying to force an individual's brain wave patterns to match "normed" target patterns obtained over population averages. These are necessary steps towards personalized NFB therapies and bespoke Brain-Computer Interfaces and brain training suitable to a wide variety of individual needs.
Reliability assessment in everyday-objects based physical-activity sensing using personal information BIBAFull-Text 39
  Claas Richter
Mobile physical-activity sensing and recognition in daily life can support medical and physiotherapeutic therapies by providing additional objective information. In our future scenario the patient is not equipped with any specialized sensor devices but uses sensors that are embedded in everyday objects, such as clothes, machines for household and office, and personal transportation and communication means. Those devices benefit from their omnipresence and their unobtrusive usage, but sensor data may suffer from low and unpredictably changing reliability and quality, and also from ambiguity and uncertainty. In this paper, we propose a new method that can assess the reliability of physical-activity measurement data from unreliable sensors embedded in everyday objects and further increase the reliability by an error compensation method.

Behavior modeling, analysis & prediction

Online recognition of people's activities from raw GPS data: semantic trajectory data analysis BIBAFull-Text 40
  Mehdi Boukhechba; Abdenour Bouzouane; Bruno Bouchard; Charles Gouin-Vallerand; Sylvain Giroux
The widespread use of GPS devices leads to an increasing availability of people traces. The GPS trajectory of a moving object is a time stamped sequence of latitude and longitude coordinates. The analysis and extraction of knowledge from GPS trajectories is important for several applications domains, ranging from traffic management to advertisement and social studies. We present an approach capable of incrementally extracting semantic locations from people's trajectories and inferring the activity done by users. We associate the places visited by people during their movements to a meaningful human activities using a novel algorithm that cluster incrementally user's moves into different type of activities. Studies using GPS records from a confined spatio-temporal region demonstrate that the proposal is effective and is capable of inferring human activities without depleting the phone resources.
ADOxx based tool support for a behavior centered modeling approach BIBAFull-Text 41
  Judith Michael; Fadi Al Machot; Heinrich C. Mayr
Meta-modeling platforms that support the automatic generation of modeling tools open a new quality in information systems development for engineers: Emphasis can be put on the design and use of a modeling language that is customized to the particular needs and desired features. This may contribute to strengthen the information-system design phase as it helps to reduce the developers' aversion against overloaded modeling languages and inflexible or expensive modeling tools. Our demo paper introduces HCM-L Modeler, a modeling tool for the Human Cognitive Modeling Language (HCM-L), which has been implemented using the meta-modeling platform ADOxx. The modeler is component of an ambient assistance information system for supporting elder persons in mastering their daily life activities.
PupilWare: towards pervasive cognitive load measurement using commodity devices BIBAFull-Text 42
  Sohail Rafiqi; Chatchai Wangwiwattana; Jasmine Kim; Ephrem Fernandez; Suku Nair; Eric C. Larson
Cognitive load refers to the amount of effort required by an individual to process information. Dating back more than fifty years, the cognitive psychology community has conducted experiments showing that the cognitive load experienced by an individual can be measured using sub-millimeter fluctuations in their pupil size, assessed using medical grade infrared devices known as pupillometers, and more recently, infrared eye-trackers. However the cost and availability of these eye-trackers limits most pupil response measurement to laboratory settings. We argue that ubiquitously measuring pupillary response could transform the next generation of context aware computing applications -- enabling computational devices to understand a user's current ability to process information, especially for users with cognitive disabilities. To this end, we present PupilWare, a system that analyzes pupil size changes through commodity cameras like those in a laptop. We evaluate PupilWare's ability to measure changes in pupil dilation using classic cognitive psychology experiments and validate its performance compared to infrared gaze trackers and medical grade pupillometers. We conclude that, in controlled conditions, PupilWare is as accurate as infrared eye-tracking for assessing task evoked cognitive load, though has problems with dark eyed individuals and eyelid occlusion.
Multi-channel ECG classification using forests of randomized shapelet trees BIBAFull-Text 43
  Isak Karlsson; Panagiotis Papapetrou; Lars Asker
Data series of multiple channels occur at high rates and in massive quantities in several application domains, such as healthcare. In this paper, we study the problem of multi-channel ECG classification. We map this problem to multivariate data series classification and propose five methods for solving it, using a split-and-combine approach. The proposed framework is evaluated using three base-classifiers on real-world data for detecting Myocardial Infarction. Extensive experiments are performed on real ECG data extracted from the Physiobank data repository. Our findings emphasize the importance of selecting an appropriate base-classifier for multivariate data series classification, while demonstrating the superiority of the Random Shapelet Forest (0.825 accuracy) against competitor methods (0.664 accuracy for 1-NN under cDTW).
A multi: modal decision making system for an ambient assisted living environment BIBAFull-Text 44
  Christos Panagiotou; Theodor Panagiotakopoulos; Achilles Kameas
Modern ubiquitous services demand the fusion of data from multiple modalities, but so far few approaches have achieved to present such solutions. This challenge becomes even more demanding when the AAL environment comes to serve the requirements of elderly people who need continuous monitoring and care. In light of this, this paper presents a multi -- modal decision making system that consists of mixed knowledge and non-knowledge based subsystems that deliver the appropriate intelligence among three modalities (i.e. ambient, health, fall detection). The main effort has been given on the health status assessment module, where a SVM classifier has been trained using the Physionet's MIT-BIH Arrhythmia database in order to detect abnormal heart beats based on time-domain and statistical features. An initial study on the classification scheme showed satisfactory results for the purposes of a system that is responsible of early screening and dangerous event detection.

Signal and image processing for ambient intelligence & pervasive computing

A quantitative comparison of the effectiveness of instructional visual stimuli for therapy exercise BIBAFull-Text 45
  Paul Sassaman; Shawn Gieser
In the interest of determining if several visual instruction stimuli used in commercial exercise games/software are viable for use with rehabilitative exercise games/software, this study quantitatively compares a selection of instructional stimuli. We hypothesize that if a difference in the accuracy of the results produced by each stimuli type exists, this should indicate which instructional stimuli provide better instruction in comparison to the others.
A survey on vision-based fall detection BIBAFull-Text 46
  Zhong Zhang; Christopher Conly; Vassilis Athitsos
Falls are a major cause of fatal injury for the elderly population. To improve the quality of living for seniors, a wide range of monitoring systems with fall detection functionality have been proposed over recent years. This article is a survey of systems and algorithms which aim at automatically detecting cases where a human falls and may have been injured. Existing fall detection methods can be categorized as using sensors, or being exclusively vision-based. This literature review focuses on vision-based methods.
HMAGIC: head movement and gaze input cascaded pointing BIBAFull-Text 47
  Andrew Kurauchi; Wenxin Feng; Carlos Morimoto; Margrit Betke
Augmentative and alternative communication tools allow people with severe motor disabilities to interact with computers. Two commonly used tools are video-based interfaces and eye trackers. Video-based interfaces map head movements captured by a camera to mouse pointer movements. Alternatively, eye trackers place the mouse pointer at the estimated position of the user's gaze. Eye tracking based interfaces have been shown to even outperform traditional mice in terms of speed, however the accuracy of current eye trackers is not enough for fine mouse pointer placement. In this paper we propose the Head Movement And Gaze Input Cascaded (HMAGIC) pointing technique that combines head movement and gaze-based inputs in a fast and accurate mouse-replacement interface. The interface initially places the pointer at the estimated gaze position and then the user makes fine adjustments with their head movements. We conducted a user experiment to compare HMAGIC with a mouse-replacement interface that uses only head movements to control the pointer. Experimental results indicate that HMAGIC is significantly faster than the head-only interface while still providing accurate mouse pointer positioning.
A framework for ECG denoising for mobile devices BIBAFull-Text 48
  S. Cuomo; A. Galletti; R. Farina; G. De Pietro; G. Sannino
Wearable sensors are widely adopted for the provision of healthcare services. Unfortunately the noise always degrades the quality of the acquired signals. In this paper, we propose a framework for mobile ECG denoising, based on a novel numerical scheme with low computational requirements. The proposed system is able to store a signal from a wearable sensor and process it in a remote way or directly on the device.
Skeleton-based human action recognition using basis vectors BIBAFull-Text 49
  Stylianos Asteriadis; Petros Daras
Automatic human action recognition is a research topic that has attracted significant attention lately, mainly due to the advancements in sensing technologies and the improvements in computational systems' power. However, complexity in human movements, input devices' noise and person-specific pattern variability impose a series of challenges that still remain to be overcome. In the proposed work, a novel human action recognition method using Microsoft Kinect depth sensing technology is presented for handling the above mentioned issues. Each action is represented as a basis vector and spectral analysis is performed on an affinity matrix of new action feature vectors. Using simple kernel regressors for computing the affinity matrix, complexity is reduced and robust low-dimensional representations are achieved. The proposed scheme loosens action detection accuracy demands, while it can be extended for accommodating multiple modalities, in a dynamic fashion.
Micro-scale thermal behavioral analysis for active evacuation routes BIBAFull-Text 50
  Nikolaos Doulamis; Anastasios Doulamis; Konstantinos Makantasis; Konstantinos Karantzalos; Konstantinos Loupos
Evacuation is a complex process influenced by multiple parameters that have significant impact on the design and execution of an efficient Active Evacuation Route (AER). Computer vision algorithms are critical for an effective AER, since it indicates the current situation awareness of the environment. Thermal imaging is an alternative effective computer vision mechanisms for the analysis of the crowd behavior either at the micro or macro scale. Thermal imaging allows efficient determination of people from the background even if highly dynamic scenes, illumination, occlusions or content alterations. This allows micro-scale analysis of the crowds resulting in an efficient active evacuation design. Experiments on thermal data from Athens International Airport indicate the assistive performance of our method.

Virtual reality & robotic technologies for rehabilitation workshop (VRRTR)

Benchmarking lower limb wearable robots: emerging approaches and technologies BIBAFull-Text 51
  Diego Torricelli; Antonio J. del Ama; Jose Gonzalez; Juan Moreno; Angel Gil; Jose L. Pons
Lower limb wearable robots are entering an exciting era. An increasing number of solutions are moving out of the lab, approaching the everyday rehabilitation practice and home-based assistive scenarios. In this context, the quantitative assessment of the technology is crucial for its correct inclusion in the market. Nevertheless, the tool normally used to support this process, i.e. benchmarking, hasn't been formulated yet in the wearable robotics field. Within the European projects H2R and Biomot we are developing a scheme for the definition of benchmarks specifically designed for wearable robot devices. This scheme takes into account two main perspectives, named "functional" and "interaction". The functional perspective aims at evaluating the stability, efficiency and correctness of motion of the global system constituted by the patient wearing the robot. The interaction perspective aims at evaluating the symbiotic interplay between the user and the device, under the physical, cognitive, and psychophysiological standpoints. In addition, we also identify the critical role of the technology in the process of bringing benchmarks into the everyday laboratory practice, with a specific focus on the emerging measurement and perturbing solutions.
Analysis of human stepping dynamics using a wii balance board with a webcam: a comparison study BIBAFull-Text 52
  Ismet Handzic; Kyle B. Reed
This paper demonstrates the assembly and verification of an inexpensive and straightforward stepping dynamics assessment system capable of simultaneously recording human lower limb motions, vertical ground reaction forces (GRF), and two dimensional foot center of pressures (COP) during the gait stance phase. This proposed system uses a single webcam video camera for motion analysis in the sagittal (side) plane. A color detection image processing Python script enables the webcam to track distinctly colored marker tape placed on the ankle and knee joint while stepping on and over a Nintendo® Wii Balance Board (WBB). The WBB is used to measure vertical GRF and foot COP. Marker positions and COP are used to construct a foot roll-over shape (ROS), the effective rocker shape that a lower limb system conforms to during a step. The accuracy of our WBB-webcam system is evaluated by the comparing marker motion, GRF, COP, and derived ROS measurements to a commercial force plate (FP) and eight-camera infrared motion capture (IRMC) system.
Eye-hand coordination assessment method using a haptic virtual environment with a complex valued neural networks training algorithm BIBAFull-Text 53
  Norali Pernalete; Amar Raheja; Alex Knaack
Post-stroke patients usually undergo therapy that focuses on general gross-motor movements such as walking, balancing, involving lower limbs; or general arm movements, leaving them with fine-dexterity and eye-hand coordination problems at their chronic stage. In this paper, we discuss the possibility to come up with assessment metrics for eye-hand coordination therapy, using a mapping between a robotic haptic device to a virtual environment and a training algorithm based on Complex Valued Neural Networks that will determine how close a determined movement pattern is in relationship with that traced by a healthy individual. Most of the current robotic systems' therapy relies on the patient's performance on standardized clinical tests such as the functional independence measure (FIM), motor power score, and the upper limb subsection of the Fugl-Meyer (FM) scales. These systems don't have other standardized metrics for assessment purposes. There is a need to establish a more intelligent and tailored therapy that could be implemented for patients to use at home in between therapy sessions, or in the long term. This therapy should be based on performance data gathered by the robotic/computer system that will provide an assessment procedure with improved objectivity and precision. This paper presents a preliminary design and simulation results of virtual environment tasks that interface with a haptic robotic device, as well as the training of a complex valued based neural network using patterns traced by healthy individuals. The idea is to use this trained algorithm, to identify in the future, how close the patients' fine motor movements are to the healthy subject's fine coordination movements; and determine a form of metrics that can be used for assessment and future therapy decisions.
VR4VR: vocational rehabilitation of individuals with disabilities in immersive virtual reality environments BIBAFull-Text 54
  Lal Bozgeyikli; Evren Bozgeyikli; Matthew Clevenger; Andrew Raij; Redwan Alqasemi; Stephen Sundarrao; Rajiv Dubey
This paper presents a virtual reality for vocational rehabilitation system (VR4VR) that is currently in development at the University of South Florida's Center for Assistive, Rehabilitation, and Robotics Technologies (CARRT). VR4VR utilizes virtual reality to assess and train individuals with severe cognitive and physical disabilities. Using virtual reality offers several advantages such as being inexpensive, safer and easily adjustable to different user needs through customization of environments, content and real time interventions. The system is composed of the following components: a virtual reality training area surrounded by an optical motion tracking system, a curved screen with two projectors, a server computer, a remote control interface on a tablet computer for job coaches, and a virtual assistive robot. This paper focuses on virtual reality training for underserved individuals with cognitive disabilities, such as autism spectrum disorder (ASD) and traumatic brain injury (TBI). We describe six transferrable skill modules and corresponding design considerations. Future work focuses on people with severe mobility impairment, such as spinal cord injury (SCI).
Development and evaluation of a dynamic virtual reality driving simulator BIBAFull-Text 55
  Sarah Tudor; Stephanie Carey; Rajiv Dubey
This paper describes the driving simulator developed at University of South Florida using the Computer Assisted Rehabilitation Environment (CAREN). This driving simulator was developed to help train individuals with spinal cord injury learn how to drive in a safe and controlled environment. The simulator includes a 180 degree projection screen and a 6 degree of freedom motion base. Two types of control options were integrated into the system: regular driving controls, which mimic the controls found in regular cars, and adaptive driving controls, which are used in vehicles modified for individuals with spinal cord injury. Testing was done with healthy individuals in order to obtain feedback about the system. Subjects completed trials with and without motion feedback, two environments (a city and a highway), and the two control options. After these trials, subjects completed surveys which included ratings and questionnaires. Results from the surveys, determined design parameters that will be implemented in the future in order to improve the driving simulator.
Simulation algorithm for the upper limb for better training and prosthesis prescription for amputees BIBAFull-Text 56
  Dimitrios Menychtas; Stephanie L. Carey; Rajiv Dubey
Prosthesis prescription and training for the upper limb is based on the experience of the prosthetist and user feedback. As a result, many users may receive a less than optimal prosthetic device. Even though the need for a prosthesis does not change, failure of the prosthetic device to meet expectations can cause users to abandon their prosthesis. Further optimization of a prosthesis is often completed using trial and error. Currently, there is no universally accepted way to quantify the fitting and performance of the prosthesis. In order to address that issue a simulation algorithm is being developed at the University of South Florida (USF) in an effort to quantify and evaluate the performance of both the user and the prosthesis. The algorithm uses motion capture data to generate a statistical prediction of the expected performance of the user. Currently, the algorithm can be used for able-bodied subjects and is being optimized for left side transradial amputees. In this paper an update on the state of the algorithm is presented.

Tools, infrastructures, architectures & techniques for deploying pervasive applications in assistive environments

Programmatic availability of virtual classrooms for assistive technologies BIBAFull-Text 57
  Carla Schäkel; Wiebke Köhlmann
Blind learners access digital materials and platforms using assistive technologies such as screen readers and braille displays. Virtual classrooms are e-learning applications, which allow learners to interact e. g. using text chat, audio and video conference and dynamic whiteboard. However, their dynamic, interactive and visual character poses barriers to blind learners who access information text- and line-based using assistive technologies. This paper presents an analysis of programmatic availability of selected virtual classrooms -- i. e. which information can be retrieved via screen readers -- in the context of a requirements analysis for accessible virtual classrooms for blind learners. The goal is to determine to what extend assistive technologies can access the user interface elements and their properties through the virtual classrooms' accessibility APIs by using different inspection tools. The result helps to determine the virtual classroom with the best programmatic availability for assistive technologies.
Intelligent contextual data stream monitoring BIBAFull-Text 58
  Kostas Kolomvatsos; Christos Anagnostopoulos; Stathes Hadjiefhtymiades
Contextual data monitoring plays an important role in increasing the quality of life of humans. Sensors observing specific activities report contextual data to a central system capable of situational reasoning. The system responds to any event related to the observed phenomenon. We propose an intelligent mechanism that builds on top of sensors measurements and derives the appropriate decisions for immediate identification of events. The mechanism adopts multivariate data fusion, time-series prediction, and consensus theory for aggregating measurements. We adopt Fuzzy Logic for handling the induced uncertainty in the decision making on the derived alerts. Simulations over real contextual data showcase the advantages and disadvantages of our monitoring mechanism.
Cascaded multimodal analysis of alertness related features for drivers safety applications BIBAFull-Text 59
  Mohamed Abouelenien; Mihai Burzo; Rada Mihalcea
Drowsy driving has a strong influence on the road traffic safety. Relying on improvements of sensorial technologies, a multimodal approach can provide features that can be more effective in detecting the level of alertness of the drivers. In this paper, we analyze a multimodal alertness dataset that contains physiological, environmental, and vehicular features provided by Ford to determine the effect of following a multimodal approach compared to relying on single modalities. Moreover, we propose a cascaded system that uses sequential feature selection, time-series feature extraction, and decision fusion to capture discriminative patterns in the data. Our experimental results confirm the effectiveness of our system in improving alertness detection rates and provide guidelines of the specific modalities and approaches that can be used for improved alertness detection.
Instrumented insole for weight measurement of frail people BIBAFull-Text 60
  Eric Campo; Yoann Charlon; Damien Brulin
In this paper, we describe an instrumented shoe insole designed to collect the weight data of a person during his or her daily activities. This research is within a more global project about new wearable systems for monitoring human health status. The work presented compares two solutions: one based on a force-sensing resistor and another one on a pressure sensor. This paper describes the insole device that integrates specific sensors to get weight data and the test bench developed to characterize their performances. The first results show that a 1kg weight change can be obtained which is perfectly meets the needs of healthcare professionals.
Wireless sensor network deployment for remote elderly care monitoring BIBAFull-Text 61
  Athanasios Dasios; Damianos Gavalas; Grammati Pantziou; Charalampos Konstantopoulos
This paper reports hands-on experiences in designing, implementing and operating a wireless sensor network (WSN)-based prototype system for elderly care monitoring in home environments. The monitoring is based on the recording of environmental parameters like temperature, humidity and light intensity as well as micro-level incidents which allow to infer daily activities like moving, sitting, sleeping, usage of electricity appliances and plumbing components. The prototype is built upon inexpensive, of-the-shelf hardware (e.g. various sensors, Arduino microcontroller platforms, ZigBee-compatible wireless communication modules) and license-free software, thereby ensuring low system deployment cost. Upon detecting significant deviations from the ordinary activity pattern of individuals and/or sudden falls, the system issues automated alarms which may be forwarded to authorized persons via a variety of communication channels. Furthermore, measured environmental parameters and activity incidents may be monitored through web interfaces.

Data modeling and information management for pervasive assistive environments

Empath2: a flexible web and cloud-based home health care monitoring system BIBAFull-Text 62
  Robert F. Dickerson; Enamul Hoque; Ifat Afrin Emi; John A. Stankovic
Home health care sensing systems are projected to streamline the efficiency of the practice of medicine by decreasing the costs of senior care and by providing preventative care to keep people out of hospitals and nursing homes. Many current sensing systems are not yet flexible enough to easily handle widely different medical applications. Empath2 provides a flexible three layer architecture that uses the Cloud and can easily be instantiated for different home health care applications. To demonstrate the flexibility of the architecture, Empath2 was instantiated for three widely different purposes. We present the design of Empath2 stressing properties that support flexibility and discuss its differences from other flexible home monitoring architectures. Evaluations for three sets of real home deployments (two of which with actual patients, and one with healthy people) are presented showing the short deployment times, short software development times, and its effectiveness for the applications at hand. Lessons learned are also presented.
Privacy-preserving intelligent networked video surveillance for patient monitoring and alarm detection BIBAFull-Text 63
  Apostolos Meliones; Simos Kokkovos
An adaptive intelligent video surveillance and motion detection system employing a network of IP cameras for patient monitoring and alarm generation is presented which ensures patients privacy through the use and processing of motion information instead of the real image of a monitored person. Minimization or even complete avoidance of false alarms is achieved, since the system proceeds to the announcement of alarms only when a series of user-defined conditions is met. The proposed system is highly configurable to adapt both to different areas of video surveillance and to several categories of monitored persons as well as to generate those alarms that concern the health carer. The system demonstrates a noteworthy performance enabling efficient system deployments involving large camera networks, as well as the packaging of a full-featured application in tiny cost and size computing devices.
Frequent pattern clustering for ADLs recognition in smart environments BIBAFull-Text 64
  Dany Fortin-Simard; Sebastien Gaboury; Bruno Bouchard; Abdenour Bouzouane
Smart habitats are considered by many researchers as a promising potential solution to help supporting the needs of elders. It aims to provide cognitive assistance by taking decisions, such as giving hints, suggestions and reminders to a resident in order to increase their autonomy. Smart homes can be seen as a huge data warehouse on the person's lifestyle. However, one of the major issues which emerge from this context of big data is learning. So it is essential to develop techniques to learn from patients before being able to assist them. In fact, each person makes a number of recurring activities, but not necessarily the same, not in the same way, not at the same time, etc. It is difficult for an expert to establish a knowledge library of activities as is often the case in the literature. A promising solution that is beginning to be explored seriously by many scientists concerning the application of data mining techniques to learn behaviors, habits and routines of people. About it, we present in this paper an affordable activity recognition system, based on frequent sensor clustering, able to recognize the patterns of the daily routine activities.
Rating quality in metadata harvesting BIBAFull-Text 65
  Sarantos Kapidakis
The quality of the data and metadata affects the interoperability of the collections and the quality of all processing. Our metadata quality metric helps the metadata harvester collection administrators detecting and improving the weaknesses of their metadata, and harvesters locating the most problematic collections, in terms of metadata quality, and prompt their administrators to improve their metadata. We extended and used an adaptive quantitative metadata quality metric and a tool to implement it. In controlled values, their value distribution is considered, and in free text values the length of their description. Moreover, we also consider additional information in the OAI-PMH XML responses, that is not normally mapped in metadata elements, but still contains metadata information, such as XML attributes. We used the tool to make quality observations, to examine collections for patterns and irregularities and to produce the appropriate advice for the collection administrators. Some of these observations are demonstrated here. We compared the reported quality over a 3-year period, to get a general quantitative and qualitative feeling of the diversity in the record descriptions, and the changes in their quality during their lifetime. We verified the assumption that the quality increases over time: usually by a tiny amount, in every collection, and by a lot on a small number of collections. Also, the lower quality collections are the ones that stop responding and vanish.
Smart homes and the challenges of data BIBAFull-Text 66
  Kevin Bouchard; Sylvain Giroux
This paper discusses research on smart home in the advent of Big Data. It justifies the interests of data driven research and the emergence of large data warehouses in smart home research. The paper describes the main applications of Big Data for smart homes. Then, few of the most important challenges related to large data warehouse are described. In particular, the difficulties related to data storage from the sensors are reviewed and many of the data mining problems in the context of Big Data are discussed.

Behavior modeling, analysis & prediction

Modelling and simulation of activities of daily living representing an older adult's behaviour BIBAFull-Text 67
  Abubaker Elbayoudi; Ahmad Lotfi; Caroline Langensiepen; Kofi Appiah
The availability of datasets for monitoring the activities of daily living is limited by difficulties associated with the collection of such data. There have been many suggested software solutions to overcome this issue. In this paper, a new technique to generate realistic data is proposed. The new method provides virtual data to the researchers with the ability to rapidly generate a large simulated dataset with different factors that could be used to represent different behaviour of a user. This paper describes the use of Hidden Markov Model (HMM) and Direct Simulation Monte Carlo (DSMC) to generate data for Activities of Daily Living (ADL) representing an older adult's behaviour. The combination of HMM and DSMC facilitates the generation of datasets capturing behaviour in terms of occupancy and movement activity performance in the environment. Simulated data is validated against data collected from a real environment.
Extracting news text from web pages: an application for the visually impaired BIBAFull-Text 68
  Erik Lundgren; Panagiotis Papapetrou; Lars Asker
Apart from the actual content, web pages contain several other components (referred to as boilerplate text) that describes how, and in what context the content should be displayed. We show how content bearing text can be efficiently separated from boilerplate text using a random forest classifier. We compare the performance with another state-of-the-art method for boilerplate detection that uses a decision tree classifier and shallow features extracted from the text. The result is a general improvement using the random forest classifier for both classifying problems analyzed, significantly so for the more complex problem. We also show that a small increase in feature set range can lead to even further improved accuracy. The conclusion is that random forest classification can achieve significantly higher accuracy rates than at least one of the current state-of-the-art methods for content extraction. The results can improve content extraction methods for a variety of applications, including search engine optimization and making the web more accessible for the blind or visually impaired.
Abnormal reaching behaviour in virtual environments: preliminary observations BIBAFull-Text 69
  Vaughan Powell; Wendy Powell
In this paper we present the preliminary findings of 3 patterns of abnormal movement observed with some individuals during VR reaching tasks in a shoulder rehabilitation application. These motions are consistent with models of rigid system behavior and appear similar to behaviours indicated in the VR literature. Such motion strategies may be in response to diminished visual perception cues in the VE and are typically inefficient in their recruitment and use of appropriate musculature and suggest that some individuals may require instruction or a much longer period of acclimatization to avoid potentially undermining treatment outcomes.
An automated framework for predicting obstructive sleep apnea using a brief, daytime, non-intrusive test procedure BIBAFull-Text 70
  Lauren Samy; Paul M. Macey; Nabil Alshurafa; Majid Sarrafzadeh
Sleep constitutes a big portion of our lives and is a major part of health and well-being. The vital repair and regeneration tasks carried out during sleep are essential for our physical, mental and emotional health. Obstructive sleep apnea (OSA) is a sleep disorder that is characterized by repeated pauses in breathing during sleep. These pauses, or apneas, deplete the brain and the rest of the body of oxygen and disrupt the normal sleep cycle. OSA is associated with a number of adverse safety and health consequences, including excessive daytime sleepiness and fatigue, which increase the risk for motor vehicle and work-related accidents. OSA also results in an increased risk for hypertension, cardiovascular disease, the development of diabetes and even premature death. The gold standard method for diagnosing OSA patients is polysomnography (PSG). PSG is an overnight sleep test that monitors a participant's biophysical changes (EEG, ECG, etc.) that occur during sleep. Despite its wide use and multi-parametric nature, there are multiple complications associated with that test that make it ineffective as an early-stage diagnosis tool. In this paper, we propose a daytime OSA screening tool that addresses the shortcomings of PSG. The framework consists of a data collection component that acquires information about the subject being tested, and a prediction component that analyzes the collected data and makes a diagnosis.
   We identify patients' key physiological, psychological and contextual features and apply advanced machine learning algorithms to build effective prediction models that help identify OSA patients in the comfort of their own home. The system was deployed in a pilot sleep apnea study of 16 patients. Results demonstrate the proposed system's great potential in helping sleep specialists in the initial assessment of patients with suspected OSA.

RAsEnv -- workshop on robotics for assistive environments

A two-step identification method for human-robot interaction in assistive environments BIBAFull-Text 71
  Alexandros Lioulemes; Nikolaos Sarafianos; Theodoros Giannakopoulos; Vangelis Karkaletsis
Integrating robotic platforms in smart home environments can improve the monitoring quality of daily activities. In this study, we explore a scenario where a robot provides a service to the users, which in our case is delivering a cup of coffee. The users place their order via an application, which at the same time captures a short video from their upper-body and their face to obtain information about their identity and to recognize them during the delivery phase. The proposed approach comprises three distinct steps. At a first step the robot detects groups of people, then it captures information from their faces and their upper body and measures the distance with the probe and identifies the person with the higher probability. Finally it approaches this person, performs an additional identification and delivers the cup of coffee. Through real-time preliminary tests under different illumination conditions, we verified that the robot can execute the task with high accuracy.
A novel EMG-free prosthetic interface system using intra-socket force measurement and pinch gestures BIBAFull-Text 72
  Joe Sanford; Oguz Yetkin; Sven Cremer; Dan O. Popa
In this paper, we present a novel system to drive a robotic prosthetic hand through the measurement of intra-socket pressure, and gesture selection from the healthy hand. A prototype HRI interface was implemented and used to compare the proposed method with standard state of practice. Grip-selection was made using finger pinch-gestures, was shown to have adequate functionality to provide a user with on-the-fly grip determination and functionality consistent with commercial systems. A moving average filter acting as a signal classifier was created to determine "open" vs "close" patterns sensed by the socket mounted piezo-resistive sensors. Sample windows were user defined as were thresholds used to determine the subject's intent. The subject was able to successfully switch between three predetermined grip configurations of a Touch Bionics i-Limb robotic hand and choose appropriate opening and closing actions.
User adaptable tasks for differential teaching with applications to robotic autism therapy BIBAFull-Text 73
  Isura Ranatunga; Namrata Balakrishnan; Indika Wijayasinghe; Dan O. Popa
Autism is characterized by limitations in social interaction, speech, imitation and motor coordination. Recent studies suggest a link between imitation and Autism. In this paper, a robot is utilized to practice imitation learning in individuals with Autism. A system is proposed to gradually improve the imitation and social interaction of patients by adaptive training, which involves repetition of tasks close to the individual capacity limit and gradually increasing the capacity by improving the difficulty level. An ideal robot motion such as a hand wave performed by a trainer is recorded using a Kinect sensor and is replayed on the small humanoid robot Zeno. The motion is encoded on the robot using the Dynamic Movement Primitives (DMP) architecture, which is a set of non-linear differential equations used to generalize a motion by just changing the time, frequency and amplitude of the trajectory. The robot can adapt its motion to match that of the subject by analyzing the subject's movements by changing the DMP trajectory parameters accordingly. In the future, it is expected that this system will help subjects of different capabilities learn in a consistent manner by adaptively adjusting to their progress.

Non-invasive monitoring technologies for disorder assessment workshop (SleepMon)

Monitoring obstructive sleep apnea with electrocardiography and 3-axis acceleration sensor BIBAFull-Text 74
  Jong-Ha Lee; Hee-Jun Park; Yoon-Nyun Kim
Obstructive sleep apnea syndrome is a sleep-related breathing disorder that is caused by obstruction of the upper airway. This condition may be related with many clinical sequelae such as cardiovascular disease, high blood pressure, stroke, diabetes, and clinical depression. To diagnosis obstructive sleep apnea, in-laboratory full polysomnography is considered as a standard test to determine the severity of respiratory disturbance. However, polysomnography is expensive and complicated to perform. In this research, we explore a computer-aided diagnosis system with portable ECG equipment and tri-accelerometer (x, y, and z-axes) that can automatically analyze biosignals and test for OSA. Traditional approaches to sleep apnea data analysis have been criticized; however, there are not enough suggestions to resolve the existing problems. As an effort to resolve this issue, we developed an approach to record ECG signals and abdominal movements induced by breathing by affixing ECG-enabled electrodes onto a triaxial accelerometer. With the two signals simultaneously measured, the apnea data obtained would be more accurate, relative to cases where a single signal is measured. This would be helpful in diagnosing OSA. Moreover, a useful feature point can be extracted from the two signals after applying a signal processing algorithm, and the extracted feature point can be applied in designing a computer-aided diagnosis algorithm using a machine learning technique.
Non-invasive sleep-environment monitoring system BIBAFull-Text 75
  Fabio Lobato; Bruno Silva; Rodrigo de Bem; Diane Miranda
Sleep disorders affect approximately 30% of the adult population, due to this fact, it is considered an important public health issue. Some medical conditions are correlated with sleep disturbances, including: obesity, diabetes, cardiovascular disease, hyperactivity disorder and early mortality. The current mainstream sleep disorder detection and assessment method, the laboratory polysomnography, is very expensive and inconvenient for patients who are extracted from their own sleep-environment. Aiming to avoid the high costs and to perform an assessment in loco, we present in this paper a non-invasive sleep-environment monitoring system in order to aid the detection of environmental factors that may be contributing to poor sleep. The stand-alone device was designed in order to provide robustness, scalability and usability to a completely built-in sleep assessment system. We highlight that the main goal of this in-home device is to give more accurate information to physicians and technical staff, assisting in the screening process, reducing costs and helping to improve the well-being of people with sleep disorders.
Non-intrusive infant monitoring, sensor data fusion and tele-alerting prototype system (asmart cot MAIA2) BIBAFull-Text 76
  Miltiadis Yfantis; Lilia Raducan; Alexander Astaras
Sleep monitoring is an increasingly popular practice, both for medical and lifestyle purposes. In the case of infant safety monitoring, however, most of the devices used are inapplicable due to the utilisation of wires, cords, obtrusive sensors, constant radio wave transmission, low sensitivity and specificity.
   We proposed and are currently developing the second generation of a portable, unobtrusive infant safety system that can be fitted to most existing cots and can wirelessly tele-alert the infant's carers in case of emergency or other pre-defined circumstances. The MAIA system is based on the real-time algorithmic fusion of data obtained from multiple sensors distributed around the infant's cot, as part of a reasonably priced system which is quick to install, requires no alteration of existing infant care routines and demonstrates a high level of sensitivity and specificity.
Real-time subspace denoising of polysomnographic data BIBAFull-Text 77
  Vangelis Metsis; Ioannis D. Schizas; Gregg Marshall
Analysis of polysomnographic (PSG) biosignals, collected during sleep studies, is the current gold-standard for sleep disorder assessment. Motion and imperfect contact of the wired sensors attached to the human body, to acquire the data, can introduce noise and artifacts that can diminish the quality of the collected data. In this work we present a subspace denoising method that exploits the low-dimensionality of the acquired data, and is able to reduce the noise and increase the SNR ratio in real-time, resulting in improved data quality.
Monitoring breathing activity and sleep patterns using multimodal non-invasive technologies BIBAFull-Text 78
  Michalis Papakostas; James Staud; Fillia Makedon; Vangelis Metsis
The monitoring of sleeping behavioral patterns is of major importance for various reasons such as the detection and treatment of sleep disorders, the assessment of the effect of different medical conditions or medications on the sleep quality, and the assessment of mortality risks associated with sleeping patterns in adults and children. Sleep monitoring by itself is a difficult problem due to both privacy and technical considerations.
   The proposed system uses a combination of non-invasive sensors to assess and report sleep patterns and breathing activity: a contact-based pressure mattress and a non-contact 2D image acquisition device. To evaluate our system, we used real data collected in Heracleia Lab's assistive living apartment. Our system uses Machine Learning and Computer Vision techniques to automatically analyze the collected data, recognize sleep patterns and track the breathing behavior. It is non-invasive, as it does not disrupt the user's usual sleeping behavior and it can be used both at the clinic and at home with minimal cost. Going one step beyond, we developed a mobile application for visualizing the analyzed data and monitor the patient's sleep status remotely.

Affective computing for biological activity recognition in assistive environments workshop (STHENOS)

A novel hybrid approach for human silhouette segmentation BIBAFull-Text 79
  Konstantinos K. Delibasis; Theodosis Goudas; Ilias Maglogiannis
In this work we propose a novel algorithm for human silhouette segmentation, which combines characteristics from a number of well established and state of the art algorithms, such as the Gaussian mixture models, the Self Organizing Maps and the Illumination Sensitive method. The proposed algorithm is evaluated against user-defined ground truth segmentation for two different types of indoor video sequences, one of which was obtained by a hemispheric camera. The behavior of the algorithm with respect to its controlling parameters is investigated and its computational burden is studied.
Design of a lifestyle recommender system for the elderly: requirement gatherings in Germany and Greece BIBAFull-Text 80
  Stephan Hammer; Andreas Seiderer; Elisabeth André; Thomas Rist; Sofia Kastrinaki; Charline Hondrou; Amaryllis Raouzaiou; Kostas Karpouzis; Stefanos Kollias
As overaging is becoming a main societal challenge, the development of AAL (Ambient Assisted Living) systems has become the centre of many research projects the last years. Our own work is targeted towards the development of an AAL system -- called CARE -- that provides assistance in form of recommendations helping its users overcome typical difficulties of everyday life, and contributes positively to their well-being. To inform the design of the envisioned CARE system we recruited two peer groups of potential users, a group of 20 Greek seniors, and a group of 27 German seniors, and conducted structured interviews which were focused on the seniors' life-style, medical needs, attitude towards AAL technologies, and, more specifically, on desired functions and system configurations of a recommendation-giving CARE system. We discuss outcomes of the conducted interviews and sketch a first CARE prototype which appears as an augmented digital picture frame that interleaves the display of photos with recommendations and interventions to improve the seniors' life-style and well-being.
Fall detection using history triple features BIBAFull-Text 81
  Georgios Goudelis; Georgios Tsatiris; Kostas Karpouzis; Stefanos Kollias
Accurate identification and timely handling of involuntary events, such as falls, plays a crucial part in effective assistive environment systems. Fall detection, in particular, is quite critical, especially in households of lonely elderly people. However, the task of visually identifying a fall is challenging as there is a variety of daily activities that can be mistakenly characterized as falls. To tackle this issue, various feature extraction methods that aim to effectively distinguish unintentional falls from other everyday activities have been proposed. In this study, we examine the capability of the History Triple Features technique, based on the Trace transform, to provide noise robust and invariant to different variations features for the spatiotemporal representation of fall occurrences. The aim is to effectively detect falls among other household-related activities that usually take place indoors. For the evaluation of the algorithm the video sequences from two realistic fall detection datasets of different nature have been used. One is constructed using a ceiling mounted depth camera and the other is constructed using an RGB camera placed on arbitrary positions in different rooms. After forming the feature vectors, we train a support vector machine using a radial basis function kernel. Results show a very good response of the algorithm achieving 100% on both datasets indicating the suitability of the technique to the specific task.
A multimodal adaptive dialogue manager for depressive and anxiety disorder screening: a Wizard-of-Oz experiment BIBAFull-Text 82
  Konstantinos Tsiakas; Lynette Watts; Cyril Lutterodt; Theodoros Giannakopoulos; Alexandros Papangelis; Robert Gatchel; Vangelis Karkaletsis; Fillia Makedon
In this paper, we present an Adaptive Multimodal Dialogue System for Depressive and Anxiety Disorders Screening (DADS). The system interacts with the user through verbal and non-verbal communication to elicit the information needed to make referrals and recommendations for depressive and anxiety disorders while encouraging the user and keeping them calm. We designed the problem using interconnected Markov Decision Processes using sub-goals to deal with the large state space. We present the problem formulation and the experimental procedure for the training data collection and the system training following the methodology of Wizard-of-Oz experiments.
Joint segmentation and classification of actions using a conditional random field BIBAFull-Text 83
  Dimitrios Kosmopoulos; Ilias Maglogiannis
In this paper, we present results of joint segmentation and classification of sequences in the framework of conditional random field (CRF) models. We use a recently proposed dual-functionality CRF model: on the first level, the proposed model conducts sequence segmentation, while, on the second level, the whole observed sequences are classified into one of the available learned classes. We evaluate the efficacy of our approach considering a real-world application, and we compare its performance to popular alternatives.
Interpretation of behaviour evolution in activities of daily living BIBAFull-Text 84
  Ahmad Lotfi; Caroline Langensiepen; Abubaker Elbayoudi
Human behaviour can be difficult to interpret even with the sophistication of modern smart homes, yet an understanding of the way people conduct their activities of daily living is essential for any attempts to detect problems. We discuss the key indicators for various activities that can be relatively robustly measured, and how these indicators can lead to a holistic measure for the activity. Combining indicators to give a metric for an activity evolution such as sleep can assist in extracting trends which may indicate some change in well-being.
Exploiting future internet technologies: the smart room case BIBAFull-Text 85
  Charilaos Akasiadis; Evaggelos Spyrou; Georgios Pierris; Dimitris Sgouropoulos; Giorgos Siantikos; Alexandros Mavrommatis; Costas Vrakopoulos; Theodoros Giannakopoulos
In this paper we present SYNAISTHISI, i.e., a cloud-based platform, that provides the necessary infrastructure in order to interconnect heterogeneous devices and services over heterogeneous networks. SYNAISTHISI facilitates the orchestration of a collective functionality allowing several services to be managed through agents that dynamically allocate the available resources.. In a smart room use-case, multiple sensing and actuation units have been developed and deployed in a lecture room. Maintaining comfort levels for room occupants is achieved through automatic decision making that exploits information from a complex event recognition engine. Our goal is to improve the overall working environment, while minimizing energy losses.
Flexible metadata mapping using OAI-PMH BIBAFull-Text 86
  Sarantos Kapidakis; Nikos Houssos; Kostas Stamatis; Panagiotis Koutsourakis
We designed, implemented and tested enhancements allowing the transmission of richer information from sources, during the metadata transfer, to cater for the case that the desired harvested metadata format evolves. The new functionality will be useful in applications like disconnected health assistive environments that occasionally go online so that their information is polled, especially when the harvested data format is occasionally revised, like when replacing their sensors or their software with different versions. We provide the mechanisms to transfer the raw metadata of each source, while still converting it to the desired format locally at the OAI-PMH client, based on the provided XSLT mapping. The proposed solution ensures full interoperability with non enhanced OAI-PMH clients/servers, so that all OAI-PMH agents can be freely mixed, although the new functionality will only be available to client/data provider pairs that both incorporate the proposed enhanced features.
ifOnly: a smart phone app to capture unmet need and support the development of new products for people living with long term conditions BIBAFull-Text 87
  Lisa Austin; Nigel Harris; Christine Nagle
Given the rise of long term conditions, and focus on living independently in the community, we launched a free crowd sourcing community app called 'ifOnly' to encourage people with disabilities to share the problems they encounter in everyday life. The app allows people to record, upload and share videos and audios that demonstrate everyday problems they face at home. These were then shared on the 'ifOnly' website -- www.ifonlyitworked.com. Designers, recruited nationally, were asked to come up with innovative design solutions via a competition hosted by the University of Bath. The designs were evaluated for novelty and commercial potential by a panel of stakeholders.
Development and evaluation of a unity-based, Kinect-controlled avatar for physical rehabilitation BIBAFull-Text 88
  Dylan Ebert; Vangelis Metsis; Fillia Makedon
This paper presents our work in developing a 3D Avatar representation of a physical therapist, to guide the rehabilitation process of patient, while the therapist is not physically present. We describe our development approach, and assess the motion accuracy of an avatar that moves according to joint tracking input coming from Microsoft's Kinect, while the therapist showcases the exercises. It is found that there is a strong correlation between the velocities of the Kinect and avatar joints, enough to make a system with high potential for real-world application.
A novel, non-invasive dermatological diagnostic instrument for skin resistance scanning (DermaSense) BIBAFull-Text 89
  Inessa Kirsanidou; Alexander Zogkas; Christina Kemanetzi; Chrysovalantis Korfitis; Elizabeth Lazaridou; Alexander Astaras
Initial physical examination in dermatology is often based on visual inspection and clinical experience, supported by technologically simple devices such as the dermatoscope. While such devices can be rather effective in trained hands, they cannot provide objective measurement data and error margins. We propose a specific adaptation of electrical impedance tomography scanning, a technology which has proven its value in other medical disciplines, as a support tool for dermatological physical examination and -- eventually -- diagnosis. A proof of concept prototype of the DermaSense device has been constructed and is currently being evaluated, while clinical pilot measurements related to melanoma are being planned.
Monitoring of compliance on an individual treatment through mobile innovations BIBAFull-Text 90
  Athanasios Anastasiou; Kostas Giokas; Dimitra Iliopoulou; Dimitris Koutsouris
The present work examines the potential of the usage of smartphones in order to offer health services to elderly patients. The purpose of this work is the design, development, and implementation of a telemedicine application. This application aims to improve the monitoring mode and increase patient adherence to the instructions assigned by the medical staff. It consists of three parts: the doctor's application (Web Application), the patient's application (Android Application) and the Web Server of the platform, where the database is stored necessary for the smooth operation of the platform. Also the Web Server hosts the doctor's Web Application. The Web Application is based on web front-end technologies, providing the medical personnel with a variety of features and useful actions. These actions and capabilities are mainly relevant to the assignment of instructions to patients and the monitoring of the progress of their health. The Android Application has been implemented for the operating system of Android-based mobile devices and consists of a handy and user-friendly environment, equipped with the right tools so that the patient has the ability to update the system on the progress of his/her health by storing the appropriate medical measurements. Both applications also provide customization capabilities to the profile of patients and doctors.
Mobile-cloud platform for raising hearing loss awareness BIBAFull-Text 91
  Panagiotis Katrakazas; Kostas Giokas; Dimitrios Koutsouris
According to W.H.O., 360 million people worldwide have disabling hearing loss [1]. However, a number of patients are afraid to consider hearing loss a problem and subsequently are afraid to seek medical help for the hearing loss. This may potentially lead hearing impaired individuals to a further level of disability and handicap. [2]
   In the concepts of raising hearing loss awareness, a mobile platform is presented, allowing individuals at risk to conduct a day-to-day hearing test. The data gathered from these tests are uploaded to a cloud database, where health care professionals, e.g. audiologists and ENT clinicians, have access and assess the data provided by users in order to inform them about their condition and raise their awareness levels by providing information addressed to each individual separately.
Exploring assistive technology for assistance dog owners in emergency situations BIBAFull-Text 92
  Charlotte Robinson; Clara Mancini; Janet van der Linden; Claire Guest; Lydia Swanson
Many vulnerable individuals own an assistance dog. Previous work has shown that a domestic alarm, Ringsel, allows assistance dogs to "call for help" via a canine interface that they interact with by pulling a detachment off with their mouths. Here we discuss the potential for systems like the Ringsel to leverage distinct behavioral patterns exhibited by the canine users to aid the automatic detection of emergencies by being used in coordination with existing assistive technologies for emergency detection and response.
A framework for the assessment of wandering behavior BIBAFull-Text 93
  Theodora Toutountzi; Scott Phan; Fillia Makedon
The purpose of this study is to develop a framework for the assessment of wandering behavior by detecting certain wandering behavioral characteristics on elderly with symptoms of a dementia disease. In our research we attempt to propose a framework in order to classify someone as a potential wanderer. We propose a system to track human activity by using various motion sensors to enable us to detect patterns and abnormalities in a subject's behavior.
Large scale temperature monitoring system for detection of potential ebola patient BIBAFull-Text 94
  Shyuan Yang; Barry Fine; Ioannis Kymissis
Ebola crisis has taken its human and economic toll. It is an illustrative example of how inadequate healthcare infrastructure at the community level can have global implications. We are developing a low-cost continuous temperature acquisition system with embedded functionality for monitoring at the local, national and global level. Our system aims to not only track the health of patients in a range of resource constrained environments (e.g. small villages), empowering communities to contain potential outbreaks before they become unmanageable, but also to monitor healthcare workers during the course of their contact with patients and upon return to their local communities minimizing the probability of further disease transmission. We are confident that our system can be deployed in a number of use scenarios for Ebola tracking and management to improve the efficiency of healthcare resource allocation, control the spread of disease, and ultimately save lives.
3D mapping of visual attention for smart rehabilitation BIBAFull-Text 95
  Christopher D. McMurrough; Alexandros Lioulemes; Scott Phan; Fillia Makedon
The estimation of human attention as input modality has been suggested as a method for an advanced human-computer interaction. With an increasing interest and development of augmented reality tools, the advent of Microsoft HoloLens glasses and increasingly affordable wearable eye-tracking devices, monitoring the human attention will soon become ubiquitous. Also visual heat-maps have become very popular and simpler to create in the 2D space over the last few years. They are very compelling and can be effective in summarizing and communicating data. The innovation in our work is the implementation of visual 3D heat-maps of the real world combined with advanced Computer Vision libraries. Finally, we have incorporated the visual 3D heat-maps for rehabilitation purposes that deal with the loss of concentration in children with learning disabilities, or disabled patients to select items of interest for them across a room.
Self-calibrating protocols enhance wearable EEG diagnostics and consumer applications BIBAFull-Text 96
  Thrasyvoulos Karydis; Filipe Aguiar; Simmie L. Foster; Andreas Mershin
Advances in electroencephalography (EEG) sensors and embedded signal processing modules are driving interest in wearable EEG devices. EEG based Neurofeedback (NFB) works by reinforcing desired brainwave patterns and shows great promise for enhancing performance and treating mental disorders. Yet, both clinical and at-home efficacies remain low. We propose a novel Self-Calibrating Protocol (SCP) which can robustly classify brain states when coupled with five standard machine learning algorithms. Our results indicate that commercially available, wearable EEG sensors provide sufficient data fidelity at fast enough rates to differentiate between arbitrary, user-defined states in real time. We conclude that SCP allows for the first time to completely bypass the pitfalls of using "normed" neurophysiological states and leads off to individualized therapeutics and bespoke performance enhancement applications.
Robot-aided rehabilitation using force analysis BIBAFull-Text 97
  Maher Abujelala; Alexandros Lioulemes; Paul Sassaman; Fillia Makedon
The estimation of human arm forces is an important factor in physical therapy, especially in robotic-aided physical therapy. Force measurements reveal the rehabilitation progress of patients with poor upper extremity motor function. In this work, we managed to record and analyse the upper arm forces of the patient while executing upper extremity exercises. Our analysis of these forces allows the robot to identify the patient's motion capability and apply active motion control to the patient's arm when their arm's forces deviate from the desired motion set-up by the therapist.
RF powered sleep apnea monitoring system BIBAFull-Text 98
  Vishnoukumaar Sivaji; Dinesh K. Bhatia; Shalini Prasad
Sleep Apnea is a chronic and widespread problem that is characterized by periods of pauses in breathing during sleep. It can lead to several life threatening conditions if left undiagnosed. Conventional methods of diagnosing sleep apnea involve connecting the patient to various sensors to monitor several physiological parameters overnight in what is called a sleep study. Problems associated with this method have led to several undiagnosed cases. To overcome this, we have proposed a low cost, RF powered, portable sleep apnea monitor that simplifies the process of screening for sleep apnea that can be done at the comfort of home. Tests of our system proved that the device was successful in detecting extended periods of apnea.
An online learning approach for trend recognition BIBAFull-Text 99
  David Paulk
This work assesses the performance of the Gradient Descent and Exponentiated Gradient online learning algorithms on the Wikipedia Page Traffic Statistics Dataset. The two algorithms are trained to predict future Wikipedia page traffic during a 7-month period. Predictions are a weighted combination of feature attributes, which are various measurements of page traffic change. The algorithms improve their predictions as they learn patterns from the data sequence by updating a weight distribution over the features. Cumulative loss for predictions during the 7-month sequence is used as an evaluation metric. The cumulative loss measurements demonstrate that the Exponentiated Gradient algorithm predicts future Wikipedia page traffic more accurately Gradient Descent does with the computed features. The cumulative loss measurements also suggest that the Exponentiated Gradient algorithm becomes less confused than the Gradient Descent algorithm becomes when irrelevant features are present among relevant features.
Using binary logistic regression coefficients for the dynamic quantification of comorbidities BIBAFull-Text 100
  Dimitrios Zikos; Ismail Vandeliwala
Comorbidities are multiple co-occurring disorders related with a series of inpatient effects which affect the overall quality of care. We describe a methodology for the dynamic quantification of the effect of comorbidities on important health outcomes, such as the in-hospital mortality and patient complications. Using a comprehensive Medicare dataset, our algorithm utilizes the coefficients of binary logistic regression models to predict the impact of a comorbidity to a binary outcome and the effect of any new disease to this comorbidity profile. To demonstrate the functionality of the algorithm, we developed a pilot java based web application. The system can be useful upon the first patient encounter as well as during the entire service episode of the care.