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

Proceedings of the 2014 International Conference on PErvasive Technologies Related to Assistive Environments

Fullname:Proceedings of the 7th International Conference on PErvasive Technologies Related to Assistive Environments
Editors:Fillia Makedon; Mark Clements; Catherine Pelachaud; Vana Kalogeraki; Ilias Maglogiannis
Location:Island of Rhodes, Greece
Dates:2014-May-27 to 2014-May-30
Publisher:ACM
Standard No:ISBN: 978-1-4503-2746-6; ACM DL: Table of Contents; hcibib: PETRA14
Papers:71
Links:Conference Website
  1. Reasoning systems and machine learning for assistive environments
  2. Wearable systems and monitoring devices
  3. Ambient assisted living and smart homes
  4. Workshop on Virtual Reality and Technologies for Health and Rehabilitation: Promises, Proofs, and Preferences (VRTHR)
  5. Data modeling and information management for pervasive assistive environments
  6. Signal and image processing for ambient intelligence and pervasive computing
  7. Workshop on Affective Computing for Biological Activity Recognition in Assistive Environments (STHENOS)
  8. Workshop on Robotics in Assistive Environments (RasEnv)
  9. Usability and HCI issues
  10. Workshop on Assistive Technologies for Safe Operation of Complex Technological Systems including Industrial Sites, Shipping, Off-Shore Platforms and Mining Activities (SafeOperation)
  11. Robotic devices and multimodal interfaces
  12. Tools, infrastructures, architectures and techniques for deploying pervasive applications in assistive environments I
  13. Accepted posters
  14. Healthcare and industrial innovations in assistive technologies
  15. Tools, infrastructures, architectures and techniques for deploying pervasive applications in assistive environments I
  16. Workshop on distributed sensor systems for assistive environments (Di-Sensa)
  17. Accepted posters

Reasoning systems and machine learning for assistive environments

Human activity recognition in smart homes based on passive RFID localization BIBAFull-Text 1
  Kevin Bouchard; Jean-Sébastien Bilodeau; Dany Fortin-Simard; Sebastien Gaboury; Bruno Bouchard; Abdenour Bouzouane
Modern societies are facing an important ageing of their population leading to arising economical and sociological challenges such as the pressure on health support services for semi-autonomous persons. Smart home technology is 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, with different kinds of effectors (light, sound, screen, etc.) to a resident suffering from cognitive deficits in order to foster their autonomy. To implement such technology, the first challenge we need to overcome is the recognition of the ongoing inhabitant activity of daily living (ADL). Moreover, to assist him correctly, we also need to be able to detect the cognitive errors he performs. Therefore, we present in this paper a new affordable activity recognition system, based on passive RFID technology, able to detect errors related to cognitive impairment in morning routines. The entire system relies on an innovative model of elliptical trilateration with several filters, and on an ingenious representation of activities with spatial zones. This system has been implemented and deployed in a real smart home prototype. We also present the promising results of a first experiment conducted on this new activity recognition system with real cases scenarios about morning routines.
Towards a classifier for predicting cardiovascular events using privileged information BIBAFull-Text 2
  G. Giannoulis; S. V. Yotov; M. Naghavi; M. Budoff; I. A. Kakadiaris
Learning Using Privileged Information (LUPI) is a learning paradigm that aims to improve supervised learning in the presence of additional (privileged) information available during training, but not during the testing phase. For example, the Multi-Ethnic Study of Atherosclerosis (MESA) used in epidemiological studies related to heart disease, contains data from 186 attributes, only eight of which are used in current risk prediction algorithms.
Towards intelligent decision making for risk screening BIBAFull-Text 3
  Panagiotis Moutafis; Ioannis A. Kakadiaris
Predicting the best next test for medical diagnosis is crucial as it can speed up diagnosis and reduce medical expenses. This determination should be made by fully utilizing the available information in a personalized manner for each patient. In this paper, we propose a method that uses synthesis to infer the best learning cohort for the patient under consideration. The constrained sample space is then used to select the best next test by maximizing the expected information gain.
Benchmarking dynamic time warping on nearest neighbor classification of electrocardiograms BIBAFull-Text 4
  Nikolaos Tselas; Panagiotis Papapetrou
The human cardiovascular system is a complicated structure that has been the focus of research in many different domains, such as medicine, biology, as well as computer science. Due to the complexity of the heart, even nowadays some of the most common disorders are still hard to identify. In this paper, we map each ECG to a time series or set of time series and explore the applicability of two common time series similarity matching methods, namely, DTW and cDTW, to the problem of ECG classification. We benchmark the two methods on four different datasets in terms of accuracy. In addition, we explore their predictive performance when various ECG channels are taken into account. The latter is performed using a dataset taken from Physiobank. Our findings suggest that different ECG channels are more appropriate for different cardiovascular malfunctions.
A supervised learning approach for fast object recognition from RGB-D data BIBAFull-Text 5
  David Paulk; Vangelis Metsis; Christopher McMurrough; Fillia Makedon
Object recognition serves obvious purposes in assisted living environments, where robotic devices can be used as companions to assist humans in need. The recent introduction of vision based sensors, which are able to extract depth sensing information about the environment, in addition to the traditional RGB video, presents new opportunities and challenges for more accurate object recognition.
   The current work, presents an object recognition approach that uses RGB-D point cloud data and a novel feature extraction methodology, in combination with well-known supervised learning algorithms, to achieve accurate, real-time recognition of a large number of objects. In our experiments, we use a dataset of household objects organized into 51 categories, and evaluate the recognition accuracy and time efficiency of a set of different supervised learning methods.

Wearable systems and monitoring devices

Is photoplethysmography-derived pulse shape useful for fall detection? BIBAFull-Text 6
  Jason Leake; Nigel Harris; Edmund Keogh; Christopher Eccleston
Falls are a common source of serious injury for elderly people. The harm can be mitigated by using a fall detector. However they are far from ideal and produce a large number of false alarms for every real fall detected. This paper describes an experiment to examine whether the reliability of fall detectors might be improved by using photoplethysmography, which evaluates heart rate and blood flow in microvascular tissue, to make inferences about the body position. The results show a correlation between body position and pulse shape. However the effect is small and is of similar size to the effect of the position of the arm on which the sensor is mounted. Further work is needed to better understand how arm position affects pulse shape and the extent to which these results may be applicable to elderly people.
Advent: a system architecture for advanced monitoring of elders with chronic conditions BIBAFull-Text 7
  Theodor C. Panagiotakopoulos; Christos P. Antonopoulos; George Koutalieris; Panagiotis Kalantzis; Costas Theodoropoulos; George Koumanakos; Achilles Kameas; Nikolaos Voros; Stavros Koubias
This paper introduces the ADVENT project that focuses on providing a comfortable, safe and secure environment, supporting daily living of elders, while empowering their mobility and independency. We present a generic system architecture that emerged from user and system requirements analysis, which consists of four parts: i) the home monitoring environment, ii) the mobile personal monitoring and support, iii) the service deployment platform and iv) the communication infrastructure. These system parts are further described and several challenging issues from the ADVENT perspective are discussed. In addition, some potential R&D directions that will be carefully examined and evaluated in the next phases of the project are highlighted.
Battery optimization in smartphones for remote health monitoring systems to enhance user adherence BIBAFull-Text 8
  Nabil Alshurafa; JoAnn Eastwood; Suneil Nyamathi; Wenyao Xu; Jason J. Liu; Majid Sarrafzadeh
Remote health monitoring (RHM) can help save the cost burden of unhealthy lifestyles. Of increased popularity is the use of smartphones to collect data, measure physical activity, and provide coaching and feedback to users. One challenge with this method is to improve adherence to prescribed medical regimens. In this paper we present a new battery optimization method that increases the battery lifetime of smartphones which monitor physical activity. We designed a system, WANDA-CVD, to test our battery optimization method. The focus of this report describes our in-lab pilot study and a study aimed at reducing cardiovascular disease (CVD) in young women, the Women's Heart Health study. Conclusively, our battery optimization technique improved battery lifetime by 300%. This method also increased participant adherence to the remote health monitoring system in the Women's Heart Health study by 53%.
WI-PATCH: stick-on wireless sensor platform for continuous monitoring of human physiology BIBAFull-Text 9
  Igor Paprotny; Varsha Evangeline Sudula
World's population is aging at an unprecedented rate due to increased average lifespan. It is estimated that by 2050, over 45% of human population will be above the age of 50. Consequently there is great need for monitoring technologies that can facilitate remote and automated monitoring of elderly, as well as other population groups with other potential health monitoring needs. Current advances in wireless systems and rapidly growing use of cellphones enable deployment of wireless monitoring devices that can automatically collect patient physiological information, process and transmit it to a central point-of-care physician. In this work, we present the WI-PATCH, which is a flexible low-power stick-on wireless sensing platform that can be used to monitor selected parameters of human physiology. With form-factor similar to a common Band-Aid, the WI-PATCH can be attached on the skin of the patient. Forming a body-sensor network, multiple WI-PATCH devices can coordinate distributed data collection and can infer patient position or orientation. Using active low power RF technology, the WI-PATCH can remain operational for days without the need for frequent re-charging or battery replacement.
Towards a low cost open architecture wearable sensor network for health care applications BIBAFull-Text 10
  P. M. Papazoglou; T. I. Laskari; G. K. Fourlas
Wireless sensor networks present a growing interest in health care applications since they can replace wired devices for detecting signals of physiological origin and continuously monitoring health parameters, offering a reliable and inexpensive solution. In this paper a low cost open architecture wearable sensor network for health care applications is presented. Through this study, an experimental wireless sensor network (WSN) architecture has been built from scratch in order to investigate and present the development procedure and the corresponding capabilities and limitations of such a system. Moreover, technological aspects regarding implementation are also presented.

Ambient assisted living and smart homes

A new system for assistance and guidance in smart homes based on electrical devices identification BIBAFull-Text 11
  Corinne Belley; Sebastien Gaboury; Bruno Bouchard; Abdenour Bouzouane
The increasing needs for support services offered to cognitively-impaired people have serious social and economic impact on our societies. Assistive technology is often see as a potential answer to this issue that may help giving more autonomy to these people. This paper presents a new assistive system for smart homes, which is based on the analysis of electrical load signatures at the steady-state, in order to provide supervision and assistance in carrying out activities of daily living for people with cognitive impairment. The proposed system exploits a new algorithmic approach to determine the erroneous behavior related to cognitive deficits and to guide the person through the completion of his ongoing task. We implemented and deployed our system in a real size smart home prototype where we used only a single power analyzer at the main electric panel which is invisible to end-users. Then, a complete experiment has been conducted on this new assistive system using breakfast sequences reproduced with electrical appliances. The simulated sequences included some cognitive errors modeled from real case scenarios coming from previous experiments with Alzheimer patients. The system showed very promising and robust results, both for activity recognition and guidance. It demonstrated that it is possible, using nonintrusive hardware like a simple power analyser, to compete with other assistive systems presented in the literature, which require intrusive equipment to properly monitor and guide.
Gesture recognition in smart home using passive RFID technology BIBAFull-Text 12
  Kevin Bouchard; Abdenour Bouzouane; Bruno Bouchard
Gesture recognition is a well-establish topic of research that is widely adopted for a broad range of applications. For instance, it can be exploited for the command of a smart environment without any remote control unit or even for the recognition of human activities from a set of video cameras deployed in strategic position. Many researchers working on assistive smart home, such as our team, believe that the intrusiveness of that technology will prevent the future adoption and commercialization of smart homes. In this paper, we propose a novel gesture recognition algorithm that is solely based on passive RFID technology. This technology enables the localization of small tags that can be embedded in everyday life objects (a cup or a book, for instance) while remaining non intrusive. However, until now, this technology has been largely ignored by researchers on gesture recognition, mostly because it is easily disturbed by noise (metal, human, etc.) and offer limited precision. Despite these issues, the localization algorithms have improved over the years, and our recent efforts resulted in a real-time tracking algorithm with a precision approaching 14cm. With this, we developed a gesture recognition algorithm able to perform segmentation of gestures and prediction on a spatio-temporal data series. Our new model, exploiting works on qualitative spatial reasoning, achieves recognition of 91%. Our goal is to ultimately use that knowledge for both human activity recognition and errors detection.
An evaluation of RGB-D skeleton tracking for use in large vocabulary complex gesture recognition BIBAFull-Text 13
  Christopher Conly; Zhong Zhang; Vassilis Athitsos
An essential component of any hand gesture recognition system is the hand detector and tracker. While a system with a small vocabulary of sufficiently dissimilar gestures may work well with approximate estimations of hand locations, more accurate hand position information is needed for the best results with a large vocabulary of complex two-handed gestures, such as those found in sign languages. In this paper we assess the feasibility of using a popular commercial skeleton tracking software solution in a large vocabulary gesture recognition system using an RGB-D gesture dataset. We also provide a discussion of where improvements in existing methods utilizing the advantages of depth-sensing technology can be made in order to achieve the best possible results in complex gesture recognition.
Live evaluation within ambient assisted living scenarios BIBAFull-Text 14
  Carlos Pereira; António Teixeira; Miguel Oliveira e Silva
Ambient Assisted Living is a new challenge within evaluation processes. It represents high levels of data and a strong focus on the user itself by encompassing dynamical environments with contextual data and by offering several interaction modalities. In such scenarios, traditional question-answer processes fail short to retrieve the necessary amount of data for more comprehensive evaluations.
   In this paper, we propose an enhanced evaluation approach adapted to these new demands. The proposal follows a user-centric approach and provides an integrative platform for evaluators to gather more information. This platform is characterized by a reusable enquiry module, its ability for real time monitoring and a event-driven dynamic module for adaptable evaluation processes. As a proof of concept, a concrete scenario based on a telerehabilitation application is also presented.
Low sampling rate for physical activity recognition BIBAFull-Text 15
  Gerald Bieber; Thomas Kirste; Michael Gaede
The monitoring of physical activity by acceleration sensors is very common. Smartphones and it's accessories (Smartwatch, wrist bands) are equipped with sensors and provide enough calculation power for data processing. Body worn mobile devices are recognizing various types of physical activities. The current concept consists of a very high sampling rate, the higher the sampling rate, the better the accuracy of classification. This strategy reduces the battery lifetime, especially for devices with limited physical dimensions, e.g. Smartwatches. Since sampling rate is a relevant factor for energy consumption, this work is analyzing the possibilities and performance of a very low sampling rate for physical activity recognition on Smartwatches. This work proposes the new concept of extremely low sampling rate for physical activity recognition.

Workshop on Virtual Reality and Technologies for Health and Rehabilitation: Promises, Proofs, and Preferences (VRTHR)

Virtual reality for gait rehabilitation -- promises, proofs and preferences BIBAFull-Text 16
  Wendy Powell; Vaughan Powell; Maureen Simmonds
Improving walking speed and quality after illness or injury presents a number of challenges, not least of which is keeping patients engaged with therapy which they may find boring or painful. The rapidly developing area of virtual reality offers technology which can track users movements and use them to drive interactions in virtual worlds.
   This paper examines the potential of virtual reality to ameliorate pain and to improve rehabilitation adherence and outcomes. The role of hardware and software in mediating movement is discussed, and key elements identified which may have a significant effect on optimising VR systems for rehabilitation outcomes.
Considerations for virtual environments for upper limb rehabilitation tasks BIBAFull-Text 17
  V. Powell; W. Powell; M. Simmonds
Shoulder pain and dysfunction are relatively common, but traditional treatment approaches and their efficacy are beset with problems, due in part, to the lack of adherence to therapeutic exercise programmes. The use of novel technology to support rehabilitation is increasing in popularity, with some evidence that it may be able to decrease the perception of pain, and improve movement quality, engagement and compliance with therapy. However, indiscriminate use of off-the-shelf games technology or haphazard approaches to the design of rehabilitation applications and virtual environments may not only reduce the potential benefits, but may even exacerbate the underlying problems. We consider the potential and the pitfalls associated with the use of novel technology in virtual reality rehabilitation, and suggest some strategies for reducing the risks and optimising therapeutic outcomes in virtual environments.
Innovative medical applications and beyond of 3D techniques in a responsive virtual reality lab: experience report BIBAFull-Text 18
  Miao Song; Peter Grogono; Serguei A. Mokhov; Wang Song; Maureen J. Simmonds
In this interdisciplinary work we present a brief report on our experience with virtual reality (VR) for rehabilitation research and present exemplar research projects on integrated pain, mind, and movement research. We specifically discuss our interdisciplinary experience with the use of 3D (stereo and non-stereo) computer graphics techniques for applications in VR rehabilitation research and include documentary production. Finally we present data that shows how virtual environments and movement can alter the pain experience and improve both physical and cognitive function.

Data modeling and information management for pervasive assistive environments

Tell me your apps and I will tell you your mood: correlation of apps usage with bipolar disorder state BIBAFull-Text 19
  Jorge Alvarez-Lozano; Venet Osmani; Oscar Mayora; Mads Frost; Jakob Bardram; Maria Faurholt-Jepsen; Lars Vedel Kessing
Bipolar Disorder is a disease that is manifested with cycling periods of polar episodes, namely mania and depression. Depressive episodes are manifested through disturbed mood, psychomotor retardation, behaviour change, decrease in energy levels and length of sleep. Manic episodes are manifested through elevated mood, psychomotor acceleration and increase in intensity of social interactions. In this paper we report results of a clinical trial with bipolar patients that amongst other aspects, investigated whether changes in general behaviour of patients due to onset of a bipolar episode, can be captured through the analysis of smartphone usage. We have analysed changes in smartphone usage, specifically app usage and how these changes correlate with the self-reported patient state. We also used psychiatric evaluation scores provided by the clinic to understand correlation of the patient smartphone behaviour before the psychiatric evaluation and after the psychiatric evaluation. The results show that patients have strong correlation of patterns of app usage with different aspects of their self-reported state including mood, sleep and irritability. While, on the other hand, the patients' application usage shows discernable changes in the period before and after psychiatric evaluation.
A goal-oriented requirements engineering approach for the ambient assisted living domain BIBAFull-Text 20
  Davide Calvaresi; Andrea Claudi; Aldo Franco Dragoni; Eric Yu; Daniele Accattoli; Paolo Sernani
The Ambient Assisted Living (AAL) domain is associated with a large number of stakeholders such as patients, their relatives, caregivers and physicians. This variety introduces a great heterogeneity in system requirements, which sometimes results in conflicting needs that must be considered when developing effective AAL systems. In this work we adopt a Goal-Oriented Requirements Engineering (GORE) approach to map out needs and requirements for the AAL domain. Following the requirements mapping, we also propose a preliminary architecture for a home care system (named e-Ward) to assist patients in their domestic environment as if they are in a hospital room.
Infrastructure for data management and user centered rehabilitation in Rehab@Home project BIBAFull-Text 21
  Elisa Ferrara; Sonia Nardotto; Serena Ponte; Silvana G. Dellepiane
In this paper, we describe the Rehab@Home Operational Infrastructure which functioning essentially relies on the acquisition, processing, exchange and interpretation of a large set of heterogeneous data and information.
   These data are coming from existing clinical data records, rehabilitation workflow structure, user-system interaction, and explicit user feedback, basic information about expected and actual rehabilitation progress, biophysical sensors, ambient and contextual sensors. What in a more precise and detailed way has been described and analyzed is the specification and development of data protocol and data integration devoted to the acquisition, processing, exchange and interpretation of a large set of heterogeneous data and information coming from biophysical sensors, ambient and contextual sensors, existing clinical data records.
   It has been carried a study of user profiling and personalization, which will be exploited to adapt process and services with the aim of enhancing user satisfaction. Thanks to personalization of the user-system interaction, the explicit user feedback, the basic information about expected and actual rehabilitation progress are made available in the best way. Case-based reasoning further improves the extraction of useful information from a single patient and from compared analysis. Identification of the most relevant risk factors related to the rehabilitation process and the monitoring of the whole rehabilitation process was another field of study.
Mining candidates for adverse drug interactions in electronic patient records BIBAFull-Text 22
  Lars Asker; Henrik Boström; Isak Karlsson; Panagiotis Papapetrou; Jing Zhao
Electronic patient records provide a valuable source of information for detecting adverse drug events. In this paper, we explore two different but complementary approaches to extracting useful information from electronic patient records with the goal of identifying candidate drugs, or combinations of drugs, to be further investigated for suspected adverse drug events. We propose a novel filter-and-refine approach that combines sequential pattern mining and disproportionality analysis. The proposed method is expected to identify groups of possibly interacting drugs suspected for causing certain adverse drug events. We perform an empirical investigation of the proposed method using a subset of the Stockholm electronic patient record corpus. The data used in this study consists of all diagnoses and medications for a group of patients diagnoses with at least one heart related diagnosis during the period 2008-2010. The study shows that the method indeed is able to detect combinations of drugs that occur more frequently for patients with cardiovascular diseases than for patients in a control group, providing opportunities for finding candidate drugs that cause adverse drug effects through interaction.
Cognitive and context-aware applications BIBAFull-Text 23
  Sohail Rafiqi; Suku Nair; Ephrem Fernandez
Diminished vigilance caused by fatigue, stress, and excessive mental load have been cited as a major cause for most aviation accidents. The job of an Air Traffic Controller (ATC) includes helping pilots maintain safe distances between planes, avoid all obstacles, as well as safe landing and takeoff. Unfortunately, the ATC work environment along with extended hours and overtime consistently results in widespread fatigued Air Traffic Controllers. Similarly, distracted, and fatigued drivers are also the cause of an increasing number of traffic related accidents and deaths. Dynamic working conditions along with the user's physical and emotional state have significant impact upon the usability and security of the system. In order to address this problem we present a Cognitive and Context-Aware Framework (CCF) that determines user's affective state, cognitive load, and context information. This enables applications to make real-time decisions to improve usability, security, or simply enhance user experience. The CCF framework constantly captures and analyzes user's biometric data (e.g. pupillometric indices of cognitive load), environmental analysis, location and time. The CCF can be used to develop a variety of applications that want to make decisions based upon user's current affective states and context. The application's decisions can be as simple as alerting a supervisor in the event an ATC is found to be fatigued/sleepy or disable critical functionality. The CCF also has applications in the gaming industry, healthcare, and advertisement. In this paper, we present the overview of the framework along with a dynamically adaptable user interface as an example of CCF application.

Signal and image processing for ambient intelligence and pervasive computing

A comparative study on 3-D stereo reconstruction from endoscopic images BIBAFull-Text 24
  Mostafa Parchami; Gian-Luca Mariottini
Advances in robotic surgery and growing use of robots in minimal access surgery (MAS), has increased the need of adapting computer-vision algorithms for surgical-vision applications. While methods for 3-D reconstruction are extensively investigated on man-made environments, the surgical-vision lacks such studies on 3-D reconstruction methods and their pros and cons on endoscopic images. In this paper we extensively compared several dense stereo reconstruction methods on a mock-up model using videos acquired from the daVinci endoscope. Also, the advantages and disadvantages of each method for different stages of stereo reconstruction are mentioned and supported by exhaustive experiments on endoscopic images.
Hand detection on sign language videos BIBAFull-Text 25
  Zhong Zhang; Christopher Conly; Vassilis Athitsos
For gesture and sign language recognition, hand shape and hand motion are the primary sources of information that differentiate one sign from another. Building an efficient and reliable hand detector is therefore an important step in recognizing signs and gestures. In this paper we evaluate three hand detection methods on three sign language data sets: a skin and motion detector [1], hand detection using multiple proposals [12], and chains model [9].
A markerless tiling method for tracking daily lung tumor motion on imperfectly matched images BIBAFull-Text 26
  Timothy Rozario; Sergey Bereg; Weihua Mao
In order to locate lung tumors on projection images without internal markers, digitally reconstructed radiograph (DRR) is created and compared with projection images. Since lung tumors always move and their locations change on projection images while they are static on DRRs, a special DRR (background DRR) is generated based on modified anatomy from which lung tumors are removed. In addition, global discrepancies exist between DRRs and projections due to their different image orientations, scattering, and noises. This adversely affects comparison accuracy. A simple but efficient comparison algorithm is reported to match imperfectly matched projection images and DRRs.
Evaluation of time and frequency domain features for seizure detection from combined EEG and ECG signals BIBAFull-Text 27
  Iosif Mporas; Vasiliki Tsirka; Evangelia Zacharaki; Michalis Koutroumanidis; Vasileios Megalooikonomou
In this paper, a large scale evaluation of time-domain and frequency domain features of electroencephalographic and electrocardiographic signals for seizure detection was performed. For the classification we relied on the support vector machines algorithm. The seizure detection architecture was evaluated on three subjects and the achieved detection accuracy was more than 90% for two of them and slightly lower than 90% for the third subject.

Workshop on Affective Computing for Biological Activity Recognition in Assistive Environments (STHENOS)

Human centered computing for the development of assistive environments: the STHENOS project BIBAFull-Text 28
  I. Maglogiannis
The paper presents the research conducted within the framework of the STHENOS project (www.sthenos.gr), which aims at the development of methodologies and systems for assistive environments. The research within the project aims at the development of methodologies and tools to compose pervasive human-centered systems, which will be able to understand the human state (identity, emotions and behavior) in assistive environments using audiovisual and biological signals. The proposed systems and applications are capable of offering services such as support for the aged/disabled/chronic patients, detection of critical situations from audiovisual content, biosignal and neurophysiology analysis for the detection of pathology (e.g. Alzheimer's disease), as well as for treatment follow-up. The paper includes an overview of the three perspectives of human centered computing studied in the STHENOS project, namely: audiovisual activity and status recognition, affective computing and neurophysiological analysis.
Human movement detection using attitude and heading reference system BIBAFull-Text 29
  George K. Fourlas; Ilias Maglogiannis
Among different types of human movement, falls are the most important since they related with high social and economic costs. Falls can cause various unintentional injuries such as fractures or in the worst-case scenario even lead to death, elderly citizen. Wearable devices present a growing interest in health care applications since they can detect signals of human activity and continuously monitoring critical parameters, offering a reliable and inexpensive solution. In this paper, an attitude and heading reference system -- inertial measurement unit (IMU) is used in order to detect human movement and especially different type of falls.
Multicamera fusion for online analysis of structured processes BIBAFull-Text 30
  Dimitrios Kosmopoulos; Ilias Maglogiannis
We propose a novel framework for online analysis of visual structured processes, using fusion from multiple cameras. Online recognition is performed through particle filters supported by hidden Markov models. We evaluate three fusion methods, an early fusion, a simple multiplication of the observation probabilities and a multi-stream one implying cross-stream coupling of observations and states. The performance is thoroughly evaluated under two complex visual behavior understanding scenarios: a visual process for table preparation in a kitchen and a real life manufacturing process in an industrial plant. The obtained results are compared and discussed.
Employing affection in elderly healthcare serious games interventions BIBAFull-Text 31
  Charis Styliadis; Evdokimos Konstantinidis; Antonis Billis; Panagiotis Bamidis
Serious games for elderly healthcare provide a promising and novel way to promote the well-being of senior citizens. The gaming environment, originally designed for a younger target population, benefits from the increasing power of personal computers, mobiles devices (phones, tablets) and SmartTVs, as well as the recent emergence of motion capture technology developed for videogame consoles via worn physical sensors and controller free sensors. Measuring self-efficacy by elderly individuals on such gaming environments has characterized them in terms of their effectiveness to motivate an audience reluctant to undertake more conventional forms of activities. However, the relationship between the features making the game challenging and the senior user's behaviour in becoming motivated so as to interact with the system independently and effectively is still elusive. For instance, failure in understanding the game's instructions or adapting to the speed and complexity may be thought of as barriers that affect the seniors' satisfactory interaction within the gaming environment. A step towards this direction would require using emotions in the midst of the gaming environment so as to allow for the embodiment of real-time mental (cognitive and emotional), and physical data. The main goal is to focus on assessing the features that encourage the elderly individuals to interact with the gaming environment on a daily basis. We propose to further enhance the gaming environment through the use of personal biosensors (i.e. wireless EEG) and cameras (i.e. fisheye camera) so as to collect the user's mental and physical changes over time and fuse them in a decision support system. This information will eventually provide feedback on the gaming experience so as to modify it according to the user's affective state.

Workshop on Robotics in Assistive Environments (RasEnv)

Safety challenges in using AR.Drone to collaborate with humans in indoor environments BIBAFull-Text 32
  Alexandros Lioulemes; Georgios Galatas; Vangelis Metsis; Gian Luca Mariottini; Fillia Makedon
This paper presents an Unmanned Aerial Vehicle (UAV), based on the AR.Drone platform, which can perform an autonomous navigation in indoor (e.g. corridor, hallway) and industrial environments (e.g. production line). It also has the ability to avoid pedestrians while they are working or walking in the vicinity of the robot. The only sensor in our system is the front camera. For the navigation part our system rely on the vanishing point algorithm, the Hough transform for the wall detection and avoidance, and the HOG descriptors for pedestrian detection using SVM classifier. Our experiments show that our vision navigation procedures are reliable and enable the aerial vehicle to fly without humans intervention and coordinate together in the same workspace. We are able to detect human motion with high confidence of 85% in a corridor and to confirm our algorithm in 80% successful flight experiments.
Indoor quadrotor state estimation using visual markers BIBAFull-Text 33
  Ghassan M. Atmeh; Isura Ranatunga; Dan O. Popa; Kamesh Subbarao
This paper discusses the problem of estimating the full state-vector (position/orientation) of an AR.Drone quadrotor using measurements from an inertial measurement unit (IMU) and an on-board camera taking images of predefined markers. The platform used is an inexpensive commercial quadrotor. The open-source Robot Operating System (ROS) is used to manage communication with the quadrotor. To estimate the AR.Drone states, an extended Kalman filter is used. The state estimates are propagated using a nonlinear dynamic model of the AR.Drone available in the literature. The estimation error covariance is propagated through the continuous-time Riccati equation using the model Jacobian. The estimated states are updated based on measurements of angular velocity from the IMU along with position and orientation from the camera. Convincing experimental results are presented. The work introduced here allows for an overall inexpensive setup for estimating the states of a quadrotor for flight in GPS denied environments using visual markers.
Service robotics for the home: a state of the art review BIBAFull-Text 34
  Kris Doelling; Jeongsik Shin; Dan O. Popa
This paper provides a review of the state of the art of service robotics for the home. We were primarily interested in major features of commercially available robots and prototype research robots, as well as remaining research challenges and gaps for the future. The review was undertaken in support of an initiative to build advanced smart homes for disabled military service members and their families. These homes will be constructed with the aim to include advanced home automation, robotics, and other assistive technologies.

Usability and HCI issues

A marker detection method using hysteresis thresholding for human posture tracking: a head tracking system BIBAFull-Text 35
  Ahmet Cengizhan Dirican
This paper presents an easy to implement and fast marker detection method suitable for real-time marker-based human posture tracking. The method works on a color segmentation algorithm based on hysteresis thresholding conducted on meaningful pixels in an image. After the segmentation algorithm is described, the experimental results for an artificial scene are given. Then, the applicability of the method is examined by means of a postural tracking application. The change in elbow angle between upper arm and forearm of a person is tracked during continuous flexion and extension movements with 21.5 Hz frequency for 640×480 resolution. The paper finally introduces the head tracking system developed using the proposed marker detection method.
An audiovisual patient identification system for rehabilitation BIBAFull-Text 36
  Konstantinos Tsiakas; Dimitrios Zikos; Fillia Makedon
Exercising is an essential part of the rehabilitation. Patients that suffer from chronic diseases, such as Rheumatoid Arthritis (RA) or fibromyalgia, must be committed to a daily exercise routine that helps them to eventually return to their normal daily activities. However, rehabilitation programs can be expensive and time-consuming. The patient must be in contact with their therapist on a frequent basis and follow a really demanding routine. Therapist must always be able to track each patient's progress and activity and make personalized decisions. This work describes a tele-rehabilitation system which facilitates the initiation and monitoring of rehabilitation exercises. We propose a patient identification system that logs the user in this rehabilitation system, based on audiovisual feature extraction. The context of our system assumes that patients are able to perform their scheduled activities prescribed by the therapist either with their therapist or even alone at home. The therapist can remotely keep track of the progress of each patient. The above scenario requires that our system guarantees safety, accuracy and private data security.
Context-aware assistive systems at the workplace: analyzing the effects of projection and gamification BIBAFull-Text 37
  Oliver Korn; Markus Funk; Stephan Abele; Thomas Hörz; Albrecht Schmidt
Context-aware assistive systems (CAAS) have become ubiquitous in cars or smartphones but not in industrial work contexts: while there are systems controlling work results, context-specific assistance during the processes is hardly offered. As a result production workers still have to rely on their skills and expertise. While un-impaired workers may cope well with this situation, elderly or impaired persons in production environments need context-sensitive assistance.
   The contribution of the research presented here is three-fold: (1) We provide a framework for context-aware assistive systems in production environments. These systems are based on motion recognition and use projection and elements from game design (gamification) to augment work. (2) Based on this framework we describe a prototype with respect to both the physical and the software implementation. (3) We present the results of a study with impaired workers and quantifying the effects of the augmentations on work speed and quality.
Target reverse crossing: a selection method for camera-based mouse-replacement systems BIBAFull-Text 38
  Wenxin Feng; Ming Chen; Margrit Betke
We propose a selection method, "target reverse crossing," for use with camera-based mouse-replacement for people with motion impairments. We assessed the method by comparing it to the selection mechanism "dwell-time clicking," which is widely used by camera-based mouse-replacement systems. Our results show that target reverse crossing is more efficient than dwell-time clicking, while its one-time success accuracy is lower. We found that target directions have effects on the accuracy of reverse crossing. We also show that increasing the target size improves the performance of reverse crossing significantly, which provides future interface design implications for this selection method.
Development of a smart insole tracking system for physical therapy and athletics BIBAFull-Text 39
  Timothy E. Roden; Rob LeGrand; Raul Fernandez; Jacqueline Brown; James (Ed) Deaton; Johnny Ross
Development of a smart insole tracking system is described. Originally designed for healthcare applications, the system has found applications in both physical therapy and athletic training. The entire system is distributed between insole hardware, mobile device applications that interface with the insoles and a central Internet server for data warehousing and analysis. We describe the development of these components so far including a discussion of custom algorithm development required for the system. The athletic version has been commercialized while the more complex healthcare version is still under development.

Workshop on Assistive Technologies for Safe Operation of Complex Technological Systems including Industrial Sites, Shipping, Off-Shore Platforms and Mining Activities (SafeOperation)

Multi-sensor target detection and tracking system for sea ground borders surveillance BIBAFull-Text 40
  Panagiotis Agrafiotis; Anastasios Doulamis; Nikolaos Doulamis; Andreas Georgopoulos
Border safety is a critical part of national and European security. This paper presents a vision-based system for ground and maritime surveillance using fixed and moving PTZ cameras. This system is intended to be used as an early warning system by local authorities. For the ground surveillance scenario, we introduce a stable human tracker able to efficiently cope with the trade-off between model stability and adaptability. More specifically, we adopt probabilistic mixture models like the Gaussian Mixture Models (GMMs) which exploit geometric properties for background modelling. Then, we integrate iterative motion information methods, concerned by shape and time properties, to estimate image regions of high confidence for updating the background model. For the maritime surveillance scenario for ship detecting and tracking, the system incorporates a visual attention method exploiting low-level image features with an online adaptable neural network tracker. No assumptions about environmental or visual conditions are made. System performance was evaluated in real time for robustness compared to dynamically changing visual conditions with videos from cameras placed at a test area near Athens for the ground scenario and at Venetian port of Chania.
Investigation of coinciding shipping accident factors with the use of partitional clustering methods BIBAFull-Text 41
  Eva Lema; Dimitris Papaioannou; George P. Vlachos
Aim of this paper is to investigate how a series of different factors are coexisting in shipping accidents. We analyzed 355 shipping accident reports from the European Maritime Safety Agency (EMSA), which are publicly available from the official EMSA website. For this purpose we used the K-means clustering method with 15 a priori defined clusters. Our results indicated that human factors often coexist with parameters related to the condition of the ship and other external factors (i.e. bad weather). Our investigation aims to contribute to the better understanding of underlying factors so that more targeted staff training, manning and shipping maintenance measures can be taken to prevent future events.
Monitoring and evaluating failure-sensitive strategies in air traffic control simulator training BIBAFull-Text 42
  Stathis Malakis; Tom Kontogiannis; Panos Psaros
The introduction of new air traffic management systems changes the demands on the provision of Air Traffic Control (ATC) services. At the core of the system air traffic controllers are responsible for the safe, expeditious and orderly flow of the air traffic. The growth of air traffic requires an increase in the capacity of the airspace, controller tools, and operating procedures. In the case of modern ATC operations, we are dealing with a complex; information rich and dynamic environments that require air traffic controllers to attend to multiple events anticipate aircraft conflicts, implement new concepts and make sense of evolving scenarios. Major system wide interventions; the Single European Sky Air Traffic Management Research (SESAR) and Next Generation Air Transportation System (NextGen) present significant changes to the delegation of authority between pilots and controllers, which requires further research on how controllers employ failure sensitive strategies in complex scenarios. Failure sensitive strategies in the context of Cognitive Systems Engineering refer to higher level strategies related to macrocognitive processes (e.g., problem solving), which are supported by the same automation functions as normal operations. High-level strategies are essential in keeping the system safe, and include strategies that forestall possibilities for failure. Normal day-today operations require conflict resolution strategies and abnormal situations call for high-level strategies. Studying how controllers employ failure sensitive strategies to cope with traffic complexity is very important if we are to understand how modern information technology and new operational demands may affect system performance. To this end, an observational field study was performed during the annual refresher training of Air Traffic Controllers in a medium complexity European airport with seasonal traffic. This was later complemented with a follow up small scale trail on simulator training to corroborate initial findings. The purpose of the study was to explore the capabilities of existing simulator based training in monitoring and evaluating failure-sensitive strategies in normal operations where new concepts are employed. Initial results indicated that current simulators can be a useful tool in monitoring and evaluating failure sensitive strategies without substantial alterations of their characteristics and in line with training curricula. In a first level and using cognitive task analysis failure sensitive strategies were elicited and documented. In a second level, a set of user centred requirements was compiled towards the development of a low cost debriefing tool based on existing simulators and training practices. In a third and final level a complexity metrics was developed in order to define thresholds of cognitive complexity that prevent controllers from full implementation of new concepts on day to day operations. Practical benefits can be derived especially in the areas of decision support systems, safety management systems (e.g. supporting safety assessments of new concepts) and training in the context of SESAR and NextGen.
Bayesian network to predict environmental risk of a possible ship accident BIBAFull-Text 43
  Ioanna Koromila; Zoe Nivolianitou; Theodoros Giannakopoulos
The goal of the AMINESS project is to promote shipping safety in the Aegean Sea though a web portal offering different levels of access to relevant stakeholders such as ship owners, policy makers, the scientific community and the general public. The portal will have three principle uses. The first is to suggest both vessel and environmentally optimal safe route planning for ships. The second is to produce alerts for ships in real time with respect to potential hazards associated to other ships, as a function of its location and planned route, its cargo and the meteorological/sea conditions. Finally, the third is to support policy recommendations, through analysis of historical data in short and long term periods that correlate safety with ship trajectories. To that end, the risk of a possible accident occurrence in the Aegean Sea is being calculated using Bayesian networks (BN). Two types of accident scenarios (collision and grounding) have been studied. A simplified Bayesian model has been developed to predict the risk of an accident given the main characteristics of the vessel, namely the ship type, size, age and flag, which are inputted to the present model. The appropriate input data has been provided by the AIS (Automatic Identification System) sign us with. Training of the developed Bayesian network was performed using the data of both the historical accident database of Marine Rescue Coordination Center and the AIS and some use cases in the area of Aegean Sea is presented in this paper.
AMINESS: a platform for environmentally safe shipping BIBAFull-Text 44
  Theodoros Giannakopoulos; Ioannis A. Vetsikas; Ioanna Koromila; Vangelis Karkaletsis; Stavros Perantonis
Reducing the possibility of ship accidents in the Aegean Sea is important to all economic, environmental, and cultural sectors of Greece. Despite the increased traffic and the related obvious risk, there are currently no national-level monitoring policies in Greece. To this end, we develop the AMINESS platform that will integrate information from multiple sources (e.g. real-time vessel movements, weather data, traffic patterns, information on type and cargo of vessels, environmental data etc.) as well as historical maritime data. This platform offers a web portal accessible by ship owners, policy makers and the scientific community, which can be used to (a) suggest vessel and environmentally optimal safe route planning, considering both the risk as well as the cost (in fuel and time) of the alternative routes, (b) deliver real-time alerts for ships (e.g. danger of collision or capsizing, suspicious/illegal behaviour of vessels), and (c) support policy recommendations both by analyzing statistics and patterns as well as by simulating the impact of different policies. Through these services, the platform aims directly to reduce the risk of a ship accident and consequently to contribute in the safety, management and monitoring of the sea environment and the Aegean Sea in particular.
   In this paper, we present our ongoing research and development of the AMINESS platform. In particular, we start by presenting the project's general architectural scheme and then the specifications and characteristics of the individual components and modules that make up this system. We particularly focus on some components, especially on the tools that fuse and process the information providing the optimal routing, the alerting and the policy recommendation outputs of the system.

Robotic devices and multimodal interfaces

Implementation of advanced manipulation tasks on humanoids through kinesthetic teaching BIBAFull-Text 45
  Sven Cremer; Matt Middleton; Dan O. Popa
In this paper, we describe a software framework for programming by demonstration (PbD) using kinesthetic teaching. A Personal Robot 2 (PR2) robot platform was used to demonstrate teaching effectiveness and conduct dual arm manipulation operations during a wine pouring demonstration. During the teaching process, an operator directs the PR2 arms to execute complex joint trajectories leading to desired handling of objects in joint or Cartesian space. The user can efficiently store and playback programmed motions using Extensible Markup Language (XML) parsing. We present repeatable wine pouring motion results that were executed by the PR2 during a demo for several thousand guests at a public event.
Multiple-robot monitoring system based on a service-oriented DBMS BIBAFull-Text 46
  Yutaka Deguchi; Daisuke Takayama; Shigeru Takano; Vasile-Marian Scuturici; Jean-Marc Petit; Einoshin Suzuki
In this paper, we present a human-targeted monitoring system composed of two autonomous mobile robots based on a service-oriented DBMS, mainly from the viewpoint of positioning control. Each robot is equipped with a Kinect and monitors the target human from appropriate angles and distances. The service-oriented DBMS, which manages the monitoring system and enables a rapid development and extension of the system, views each robot as a data source which generates a data stream to be stored and processed in the DBMS. The results of the experiments conducted in a real office are promising.
Quantitative evaluation of the Kinect skeleton tracker for physical rehabilitation exercises BIBAFull-Text 47
  Shawn N. Gieser; Vangelis Metsis; Fillia Makedon
Using video game technology in physical rehabilitation has shown many positive results in the past few years. The release of the Microsoft Kinect has presented many new opportunities for development in physical rehabilitation technologies. However, there have been questions about the Kinect's accuracy in actual experimentation. In this paper, we compare skeleton data obtained by a Kinect to that obtained by a VICON system in order to determine the accuracy of the Kinect while a tracked subject is moving their arm around. This is the first steps towards a much larger physical rehabilitation system.

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

Mobile bandwidth prediction in the context of emergency medical service BIBAFull-Text 48
  Stefanie Judith Opitz; Nicole Todtenberg; Hartmut König
For the emergency medical service a reliable sharing and transmitting of medical data and electronic patient records between the ambulance and the hospital is of great importance for the quality of patient care. During the last years new mobile technologies have evolved with much higher bandwidths than before. Nevertheless, the available network bandwidth strongly varies from area to area. In this paper we present an approach to investigate the predictability of the mobile network performance based on the statistical evaluation of the gathered transmission data on frequently used routes. This allows one to pro-actively adapt the available bandwidth to improve the utilization of the mobile network capacity and achieve a more reliable transmission of the medical data. We performed data rate measurements on different routes and derived predictions of the available mobile resources. As a first approach, the predictions are based on the average of former measurements, but taking into account the current position and temporal variation of the mobile resources. Compared with the results obtained through the straightforward reference approach our predictions are more accurate and precise. Consequently, our empirical studies confirm that despite high variations of the available wireless bandwidth reasonable predictions for known routes are possible.
Sign language recognition using dynamic time warping and hand shape distance based on histogram of oriented gradient features BIBAFull-Text 49
  Pat Jangyodsuk; Christopher Conly; Vassilis Athitsos
Recognizing sign language is a very challenging task in computer vision. One of the more popular approaches, Dynamic Time Warping (DTW), utilizes hand trajectory information to compare a query sign with those in a database of examples. In this work, we conducted an American Sign Language (ASL) recognition experiment on Kinect sign data using DTW for sign trajectory similarity and Histogram of Oriented Gradient (HoG) [5] for hand shape representation. Our results show an improvement over the original work of [14], achieving an 82% accuracy in ranking signs in the 10 matches. In addition to our method that improves sign recognition accuracy, we propose a simple RGB-D alignment tool that can help roughly approximate alignment parameters between the color (RGB) and depth frames.
Mobility behavior assessment using a smart-monitoring system to care for the elderly in a hospital environment BIBAFull-Text 50
  M. Chan; E. Campo; W. Bourennane; F. Bettahar; Y. Charlon
The number of elderly people in developed countries is increasing, and the number of individuals suffering from age-related diseases is also growing. Elderly people prefer to live as independently as possible at home, but living independently has its risks, such as falling, weakening bodies, memory loss, and wandering that limit mobility and activities. Basic functions like walking and sleeping are key indicators in determining the performance of an elderly person's home activities. For this reason a low cost, smart and real-time home monitoring system has been developed to collect individual mobility data (walking, sleeping, and going to the washroom) and warns formal or informal caregivers immediately when there is a dangerous event or motor behavior impairment. This paper reports on our current trials of a monitoring system in an Alzheimer's unit to determine the mobility behavior of some patients.
Java framework for dialysis monitoring using a distributed network architecture BIBAFull-Text 51
  Hiren Sangani; Andreas Fink; Christian Peter
Monitoring of dialysis patients is a typical scenario for multi-sensor systems, measuring vital parameters like heart rate or blood pressure. The main issue is to display and monitor the patients' vital parameters anytime and anywhere, using so-called pervasive environments. The vital parameters shall be available at the remote devices and at a central monitoring station, e.g. for targeted alerting and simultaneous monitoring of all patients from a central location. The proposed system contains different sensors to measure vital parameters, cell phones as remote devices and a web application as central monitor. A web server is implemented to handle the HTTP requests and responses from cell phones and web application. On the cell phone, the Android platform is used to implement the functionality for sensor and network communication interfaces using the integrated WLAN and classic Bluetooth protocols. The hierarchical architecture of the monitoring network allows the extension of the system by additional patients, wearing additional sensors. The functionality of the monitoring system is validated in an exemplary use case for two patients showing valuable results in terms of response time and availability.

Accepted posters

TigerBites: an assistive notification system for local dining BIBAFull-Text 52
  David Paulk; Lisa Kim; Diogo Adrados
This work proposes an automated notification system that allows the user to select their favorite dining items and sends a message to the user via email when a 'favorited' item is being served at a nearby dining location [and/or when a similar item is available]. The system is tested in Princeton, NJ and designed for student users, where Princeton's four residential college dining halls, the graduate college dining hall, and the Center for Jewish Life, are considered the nearby dining locations. Under CAS authentication, a user can add past and presently displayed dining items to the user list of favorites by clicking a plus sign beside the item when logged in. There will be several interfaces from which users can select favorites. The default presents a list of locations. By selecting a location, a user can expand the breakfast, lunch, and dinner menus for the current day. The search interface presents a search bar with features including search filters and similarity suggestions. In addition to modifying favorited dining items, a user can submit suggestion letters to the dining locations under their CAS login. A user will be able to access the suggestion box by clicking on a link to it from the main page.

Healthcare and industrial innovations in assistive technologies

Evaluation of classification methods for the prediction of hospital length of stay using Medicare claims data BIBAFull-Text 53
  Dimitrios Zikos; Konstantinos Tsiakas; Fadiah Qudah; Vassilis Athitsos; Fillia Makedon
In this paper, we investigate the performance of a series of classification methods for the prediction of the hospital Length of Stay (LOS), based on two temporally sequential clinical scenarios. We used a 2012 Medicare Provider Analysis and Review (MedPar) dataset, which contains records of Medicare beneficiaries who used inpatient hospital services. Our subset included 300,000 randomly selected cases. During the prepossessing we added new features and linked our data with external datasets, using common key identifiers. In the first scenario our goal was to predict the LOS using a subset of information which is readily available to the clinician upon the patient admission, while the second scenario assumes that there is available additional data (information on the patient diagnosis and clinical procedures). For our experiments we used three different classifiers: Naïve Bayes, AdaBoost and C4.5 Decision tree, for two different LOS cut-off points (4 day and 12 day hospital stay). The overall performance of our classifiers was ranging from fair to very good. On the other hand the true positive rate, that is the correct classification of the long hospital stays, was low, with an exception of Naïve Bayes, which demonstrated significantly better performance in the second scenario. Our results indicate that Naïve Bayes may be used for the prediction of the in-hospital LOS. Our analysis also indicates that the MedPar data combined with other data resources has the potential to provide a good basis for robust prediction analytics in hospitals.
Improving healthcare cost transparency through mobile application development BIBAFull-Text 54
  David Allen; Ovidiu Daescu
Managing the rising costs of healthcare is of critical importance to both the public and private sectors. Those who need healthcare face a daunting challenge in determining the projected price of care as there is currently a lack of transparency with respect to obtaining the price of procedures from healthcare providers such as hospitals and labs. The price of a given medical procedure can vary by thousands of dollars depending on where it is performed, with patients often unable to compare prices and select the best one. Patients are often at a disadvantage when it comes to negotiating healthcare costs as the care may be of life or death importance to the patient. Higher healthcare costs put a greater financial burden on patients, insurers and employers. In order to increase healthcare price transparency and support patients in the search for quality healthcare, we have developed a mobile device application that compares the price of procedures at local hospitals based on publically available price data.
Computer games to decrease pain and improve mood and movement BIBAFull-Text 55
  Maureen J. Simmonds; Dimitrios Zikos
Pain is a pervasive problem that can compromise mood and movement leading to depression and disability. Computer games can enhance self-esteem, mood, and movement in healthy individuals. To what extent such games can improve mood and movement and decrease pain in individuals with chronic pain is not known. This study compared the effects of two computer games on pain, mood and movement in patients with fibromyalgia (FM) compared to a pain free cohort. Twenty-nine people with (FM) and 19 healthy controls were randomized to play a game to enhance mood or a game with no emotional salience. Standardized measures of clinical pain, thermal pain thresholds, self-efficacy, mood, self-esteem and physical performance were obtained before and after game play. Both games improved pain threshold, mood and physical performance (p≤.019). There was no differential effect of games suggesting that for these subjects and after one game play, attention to the game rather than the game itself is the likely explanation.

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

eFRIEND: an ethical framework for intelligent environment development BIBAFull-Text 56
  Simon Jones; Sukhvinder Hara; Juan Augusto
Intelligent Environments bring technology closer to daily life and aim to provide context-sensitive services to humans in the physical spaces in which they work and live. Some developments have considered the ethical dimension of these systems; however this is an aspect, which requires further analysis. A literature review shows that these approaches are rather disconnected from each other, and that they are not making an impact on real systems being built. This paper summarises the ethical concerns addressed by previous work, highlights other important concerns, which have been overlooked so far, and proposes a more holistic approach. It explains how these concerns can be used to guide part of the development process in such a way that Intelligent Environments being engineered in the future will consider the ethical dimension in practice, not just in theory.
Interdisciplinary partnerships between rehabilitation therapists and computer scientists: a proposed model BIBAFull-Text 57
  Angie Boisselle; Maureen Simmonds
The concept of inter-professional collaboration to optimize solutions for complex rehabilitation problems is not novel. However, the processes involved in successful and optimal collaboration between rehabilitation therapists and computer scientists is not well studied. In this paper, we examine strategies to connect technology driven problems and solutions in the lab to clinically driven problems and constraints for usable solutions in the field. We highlight gaps in collaboration such as differences in discipline language, hypothesis driven vs. function driven outcomes and understanding of the 'end-user'. We also discuss future ideas for successful collaboration to optimize usability of rehabilitation technology through creative problem solving.
Text summarization as an assistive technology BIBAFull-Text 58
  Fahmida Hamid; Paul Tarau
Automated text summarization can be applied as an assistive tool for people with vision deficiency as well as with language understanding or attention deficit disorders. In this paper, we introduce an unsupervised graph based ranking model for text summarization. Our model builds a graph by collecting words, and their lexical relationships from the document. We apply a handful of available semantic information (definition, sentimental polarity) of words to enhance edge-weights (interconnectivity) between nodes (words). After applying a polarity based ranking algorithm over the graph we collect a subset of high-ranked and low-ranked words, name those as keywords. We, then, extract sentences that possess a higher rank defined by the rank vector of keywords. Sentences extracted in this manner correlate with each other and express the summary of the document quite successfully. Summaries formed by our model can appease readers with vision difficulties while keep them updated.

Workshop on distributed sensor systems for assistive environments (Di-Sensa)

Sink controlled reliable transport for disaster recovery BIBAFull-Text 59
  Charilaos Stais; George Xylomenos; Giannis F. Marias
We present a reliable transport layer protocol for sensor networks, targeting disaster recovery applications where human or robotic rescuers try to gather information from a possibly fragmented sensor network by moving through the disaster area. The mobility of the information sink means that the protocol must quickly adapt to a constantly changing view of the network, where connections and disconnections are the norm. Our protocol is purely sink driven, that is, the sink controls congestion by rate limiting the sensors, choosing how to assign the available bandwidth to different sensor types and deciding on the level of reliability to be achieved. In addition, our protocol operates at the application layer with minimal requirements from lower layers, allowing its integration with a disaster recovery application that will set its parameters depending on the disaster scenario. As a result, our protocol allows simple and inexpensive fixed sensors to be combined with expensive but reusable mobile equipment for disaster recovery purposes.
Towards real-time emergency response using crowdsourcing BIBAFull-Text 60
  Ioannis Boutsis; Dimitrios Tomaras; Vana Kalogeraki
Crowdsourcing has emerged as an attractive paradigm in recent years for information collection for disaster response, which utilizes data received from the human crowd, to provide critical information collection and dissemination during emergency situations and visualize this data to generate emergency maps for the human crowd. In this paper we investigate the use of crowdsourcing mechanisms for real-time emergency response and describe our approach for developing a crowdsourcing tool that can be effectively used to formulate questions and seek answers from the human crowd using a MapReduce programming model, and integrate this information into a novel spatiotemporal data structure and create a visual emergency map. Our experimental evaluation shows that our approach is practical, efficient and can be used for applications with real-time demands.
Tracking persons using a network of RGBD cameras BIBAFull-Text 61
  George Galanakis; Xenophon Zabulis; Panagiotis Koutlemanis; Spiros Paparoulis; Vassilis Kouroumalis
A computer vision system that employs an RGBD camera network to track multiple humans is presented. The acquired views are used to volumetrically and photometrically reconstruct and track the humans robustly and in real time. Given the frequent and accurate monitoring of humans in space and time, their locations and walk-through trajectory can be robustly tracked in real-time.
Defining a mobile architecture for structural health monitoring BIBAFull-Text 62
  Panos Sakkos; Dimitrios Kotsakos; Vana Kalogeraki; Dimitrios Gunopulos; Jaakko Hollmén
The proliferation of powerful, programmable mobile devices along with the availability of wide-area connectivity has provided a powerful platform to sense and share location, motion, acoustic and visual data. The new generation of smart devices feature a variety of sensors that can be used towards building scalable and extendable monitoring systems with hundreds of nodes. In this paper, we report on work in progress to develop a distributed mobile system for Structural Health Monitoring utilizing smart handheld devices. We describe a distributed clustering algorithm that clusters nodes that produce similar power spectra and thus enables the implementation of decentralized heavy computation monitoring algorithms.
The impact of resource heterogeneity on the timeliness of hard real-time complex jobs BIBAFull-Text 63
  Georgios L. Stavrinides; Helen D. Karatza
The rapid technological advances have made the use of domestic assistive environments vital in the daily life of elderly and disabled people. A major challenge in an assistive environment is the effective coordination and scheduling of the interdependent tasks required for the analysis and processing of the collected sensory data, in order to timely response to a life-threatening emergency event. These complex jobs are usually scheduled on distributed heterogeneous resources that may be part of a computational grid or cloud. We investigate by simulation the impact of the processor and network heterogeneity on the timeliness of such critical jobs.

Accepted posters

Wireless infrastructure setup strategies for healthcare BIBAFull-Text 64
  Nikunj Agarwal; M. P. Sebastian
The investments in wireless entities should be planned and implemented in such a way that they deliver the expected care to the patients. There should be maximum utilization of the infrastructure to meet the patient needs. In this paper we have proposed a methodology to measure the overall utility of the wireless capabilities based on the connectivity, applications, security practices, and the locations. The novelty of the paper is a department wise analysis of the different wireless entities to gauge the importance of each of them. The importance of different wireless entities will help streamline the investments.
Opportunistic sensing and detection of mild cognitive impairment BIBAFull-Text 65
  Bilgin Kosucu; Netzahualcóyotl Hernández; Temitayo Olugbade
Dementia is a growing healthcare problem, and it has become necessary to find a way to reduce the prevalence of this disease. Mild cognitive impairment is a risk factor for dementia, and being able to detect the onset of mild cognitive impairment gives healthcare professionals the chance to reduce the effect of dementia on the society. In this paper, we propose a system that seeks to detect mild cognitive impairment in otherwise healthy persons. We are particularly interested in being able to gather data for monitoring without interfering with the subject's daily living.
Talos: assistive robotic platform BIBAFull-Text 66
  Harris Enotiades; Fillia Makedon; Scott Phan; Christopher McMurrough; Panos Shiakolas
In this paper, we describe the approaches used to develop the TALOS robotic platform for use with assistive living environments and related work using an alternative custom robotic arm.
Greek-language verbal and non-verbal interaction with a philosopher humanoid robot BIBAFull-Text 67
  Michalis Papakostas
We envision a world where humanoid robots can be used as exciting museum guides, shopping mall robots, or interactive theatre actors, impersonating characters such as ancient philosophers who talk about their theories and lives with the public. Towards that goal, in this paper we present four basic applications regarding human-robot interaction. These applications can be considered as part of a robust dialogue system whose development is ongoing. All the implementations are applied on a humanoid robot named "Demokritos". Humanoid Demokritos has the ability of performing humanlike facial expressions and have verbal interaction in Greek with its interlocutor. Additionally a face recognition and tracking system has been established and an implementation which gives our robot the capability of recognizing and imitating the facial expression of the person interacting with it. All the applications are being described and analyzed in detail and we conclude by presenting some of our next goals and by proposing some interesting aspects related to human-robot interaction.
A probabilistic algorithm with user feedback loop for decision making during the hospital triage process BIBAFull-Text 68
  Dimitrios Zikos; Ismail Vandeliwala; Philip Makedon
In this paper, we describe a probabilistic algorithm with user feedback loop, which can be used for decision making during the patient triage process. Given an R{x, y} the method relies on the user defining a set of x values (i.e. symptoms) and the algorithm returns a collection of y values as a hidden layer (possible diseases), taking into consideration a possible false negative user reporting, by looking into candidate values of y and identifying x values (symptoms) which have not been initially provided by the user. The user can specify parameters such as the minimum probability ratio of the final output, the minimum probability ratio of the y values for which the non-user given x values will be re-evaluated, and the maximum number of user feedback loops. In order to validate the method, we use a comprehensive 2012 Medicare Claims dataset with 15 million cases.
Experimental testing of a prototype wireless tele-alerting system for monitoring sleeping infants (smart cot MAIA) BIBAFull-Text 69
  Evangelia Spiridonou; Grigoris Matralis; Panagiotis Kartsidis; Lilia Raducan; Miltiadis Yfantis; Alexander Astaras
Real time in situ medical monitoring has become increasingly common in the past few decades, driven by the development of a wide variety of affordable sensors and power autonomous, energy efficient microcontroller platforms. Portable and unobtrusive multi-sensor data acquisition is considered routine, including real-time data processing and alerting based on measurements within or around the human body.
   Project MAIA has developed an intelligent baby cot system capable of monitoring sleeping infants and remotely notifying their carers under pre-programmed circumstances. It aims to add a layer of protection for infants during the first 6 months of their life, primarily targeting medical emergencies such as choking, suffocation and the elusive Sudden Infant Death Syndrome (SIDS). The MAIA system also aims to contribute to current paediatric medical knowledge by providing rare in situ research data.
Creating custom fitted point of goggles through facial metrics BIBAFull-Text 70
  Cyril Lutterodt; Scott Phan; Filia Makedon
This paper displays the possible uses of facial metrics to create custom fitted POG goggles. This custom device will improve usage and enhance conformability for the end user.
A framework for the development of an assessment tool for children with disabilities BIBAFull-Text 71
  Philip Makedon; Lynette Watts; Dimitrios Zikos
We describe a framework for the development of a gaming environment for the assessment of children with a spectrum of disabilities (GE-CDA tool). A typical case example is that of Cerebral Palsy (CP). We follow an evidence-based approach to identify the required domains which need to be addressed in the GE-CDA tool. Three different axes are considered, namely the domain of disability (cognitive, physical, psychological and behavioral), the function to be assessed for those domains (i.e. hand-eye coordination, working memory capacity) and the assessment method for each one of the functions. We plan to use Minecraft to develop the assessment tool, since it supports activity based mini-games (fishing, archery, maze, puzzles, etc.) and also allows for modifications, serving as an ideal platform for our GE-CDA tool.