Toward Useful Services for Elderly and People with Disabilities

As the device ecosystem in the intelligent environments becomes more complex, the need for supporting a wider range of devices becomes crucial. To address this challenge, this paper describes a framework, Imhotep, which can be used to develop mobile applications, easily adapting them to the constraints imposed by the user and device capabilities through a set of preprocessor directives. These directives are enhanced with concepts that are automatically adjusted to the current trends of mobile devices by using a fuzzy knowledge-eliciting reasoner.

The growth in popularity of smart environments has been quite steep in the last decade and so has the demand for smart health assistance systems. A smart home-based prompting system can enhance these technologies to deliver in-home interventions to a user for timely reminders or a brief instruction describing the way a task should be done for successful completion. This technology is in high demand with the desire for people who have physical or cognitive limitations to live independently in their homes. In this paper, we take the approach to fully automating a prompting system without any predefined rule set or user feedback. Unlike other approaches, we use simple off-the-shelf sensors and learn the timing for prompts based on real data that is collected with volunteer participants in our smart home testbed.

Recently researchers have proposed activity recognition methods based on adapting activity knowledge obtained in previous spaces to a new space. Adapting activity knowledge allows us to quickly recognize activities in a new space using only small amounts of unlabeled data. However, adapting from dissimilar spaces not only does not help the recognition task, but might also lead to degraded recognition accuracy. We propose a method for automatically selecting the most promising source spaces among a number of available spaces. Our approach leads to a scalable and quick solution in real world, while minimizing the negative effects of adapting from dissimilar sources. To evaluate our algorithms, we tested our algorithms on eight real smart home datasets.

We present a human activity recognition (HAR) system for smart homes utilizing depth silhouettes and

$\mathcal{R}$

transformation. Previously,

$\mathcal{R}$

transformation has been applied only on binary silhouettes which provide only the shape information of human activities. In this work, we utilize

$\mathcal{R}$

transformation on depth silhouettes such that the depth information of human body parts can be used in HAR in addition to the shape information. In

$\mathcal{R}$

transformation, 2D directional projection maps are computed through Radon transform, and then 1D feature profiles, that are translation and scaling invariant, are computed through

$\mathcal{R}$

transform. Then, we apply Principle Component Analysis and Linear Discriminant Analysis to extract prominent activity features. Finally, Hidden Markov Models are used to train and recognize daily home activities. Our results show the mean recognition rate of 96.55% over ten typical home activities whereas the same system utilizing binary silhouettes achieves only 85.75%. Our system should be useful as a smart HAR system for smart homes.

Activity recognition in smart environments and healthcare systems is gaining increasing interest. Several approaches are proposed to recognize activities namely intrusive and non-intrusive approaches. This paper presents a new fully non-intrusive approach for recognition of Activities of Daily Living (ADLs) in smart environments. Our approach treats the activity recognition process as an information retrieval problem in which ADLs are represented as hierarchical models, and patterns associated to these ADLs models are generated. A search process for these patterns is applied on the sequences of activities recorded when users perform their daily activities. We show through experiments on real datasets recorded in real smart home how our approach accurately recognizes activities.

DANTE is an application, which supports the annotation of ADLs captured using a pair of stereo cameras. DANTE is able to interpret the position and orientation of any object that is tagged with a special marker. Offline, users navigate frame-by-frame through captured scenes to annotate onset/completion of object interactions. The main utility is supporting the development of large annotated datasets, which is essential for the development and evaluation of context-aware models to interpret and monitor occupant behaviour within smart environments. DANTE only records scenes during which ‘tagged’ objects are interacted with therefore significantly reducing the amount of redundant footage recorded. The current study has extended the concepts of DANTE and has used it to support the annotation of additional sensor platforms. Results demonstrated both the capability of DANTE to support annotation of other platforms along with reducing the amount of time previously required to manually annotate such data by more than 45%.

The increasing aging population has inspired many machine learning researchers to find innovative solutions for assisted living. A problem often encountered in assisted living settings is activity recognition. Although activity recognition has been vastly studied by many researchers, the temporal features that constitute an activity usually have been ignored by researchers. Temporal features can provide useful insights for building predictive activity models and for recognizing activities. In this paper, we explore the use of temporal features for activity recognition in assisted living settings. We discover temporal relations such as order of activities, as well as their corresponding start time and duration features. To validate our method, we used four months of real data collected from a smart home.

Activity Recognition in Smart Environments presents a difficult learning problem. The focus of this paper is a 10-class activity recognition problem using motion sensor events over time involving multiple residents and non-scripted activities. This paper presents the results of using three different graph-based approaches to this problem, and compares them to a non-graph SVM approach. The graph-based approaches are generating feature vectors using frequent subgraphs for classification by an SVM, an SVM using a graph kernel and nearest neighbor approach using a graph comparison measure. None demonstrate significantly superior accuracy compared to the non-graph SVM, but all demonstrate strongly uncorrelated error both against the base SVM and each other. An ensemble is created using the non-graph SVM, Frequent Subgraph SVM, Graph Kernel SVM, and Nearest Neighbor. Error is shown to be highly uncorrelated between these four. This ensemble substantially outperforms all of the approaches alone. Results are shown for a 10-class problem arising from smart environments, and a 2-class one-vs-all version of the same problem.

Our research aim is that even caregivers who are musical novices perform part of a music therapy activity. In this paper, we present an accompaniment system for calming the symptoms of patients with dementia with mental instability who repeat stereotypical behaviors and utterances. This system converts patients’ utterances into pitches in response to an operator’s key entry and automatically plays a cadence based on those pitches. The cadence begins on a chord that resonates with a patient’s emotions and finishes on a chord that calms his symptoms. Because the use of this system is simple, even musical novices can use it to calm dementia patients’ continuous stereotypical utterances.

Bed-related situation monitoring is a crucial task in geriatric healthcare

.

Due to shortage of qualified caregivers, automatic detection of the situation in a bedroom is desirable since risks may arise when an elder gets up from the bed alone. In addition, analyzing the total amount of care time of an elder is helpful when the social effect on the improvement of the elder’s health needs to be evaluated. This paper presents a context-aware healthcare system which makes use of multi-modal and un-obtrusive sensing technology meanwhile taking human feeling into account. Specifically, we choose ambient sensors such as pressure straps and a laser scanner to monitor both the activity of the elder and his/her surroundings. Moreover, a context fusion is further proposed to infer the situation of the elder. Experimental results demonstrate the high promise of our proposed methods for bed-related situation awareness.

Location information can be useful to construct a profile of a person’s activities of daily living. This paper proposes an approach with the aim of improving the accuracy and robustness of a location recognition approach based on RFID technology. A method was introduced for the optimal deployment of an RFID reader network, which aims to minimize the hardware cost whilst achieving high localization accuracy. A functional subarea mapping approach was proposed based on both a coarse-grained and a fine-grained method. Experimental results indicated that the coarse-grained mapping provided a higher overall accuracy of location detection. The average subarea location accuracy achieved based on coarse-grained and fine-grained data was 85.4% and 68.7%, respectively. Nevertheless it was found that the fine-grained mapping approach was capable of providing more information in relation to the details of the functional subareas. This implies that we need to balance the requirements between the size of the subareas and the location accuracy.

Wireless Body Sensor Network (WBSN) consists of a large number of distributed sensor nodes. Wireless sensor networks are offering the next evolution in biometrics and healthcare monitoring applications. The present paper proposes a congestion control protocol based on the learning automata which prevents the congestion by controlling the source rate. Furthermore, a new active queue management mechanism is developed. The main objective of the proposed active queue management mechanism is to control and manage the entry of each packet to sensor nodes based on learning automata. The proposed system is able to discriminate different physiological signals and assign them different priorities. Thus, it would be possible to provide better quality of service for transmitting highly important vital signs. The simulation results confirm that the proposed protocol improves system throughput and reduces delay and packet dropping.

For providing real-time data collecting in a telehealthcare system composed of wireless sensor network and home automation network, a middleware called CodaQ is designed. It provides context data and takes into account QoS requirements of the applications. In this paper, we show how the data are modeled for including context information and how the QoS requirements are handled within a middleware. First measurements on a test bed have been carried out showing the good performance of our design.

The rich sensing ability of smart mobile phones brings an unique opportunity to detect and long-term monitor people’s physical activities. However, with mobile phone the application has to comply with people’s usage habit of it and thus capture the right moment to recognize activities, which will potentially cause great in-class variances. As a result, the model potentially becomes complex and costs much computing resources in mobile phone. This paper recognize people’s physical activities when they place the mobile phone in the pockets near the pelvic region. Experiment results show that the accuracy could reach 97.7%. To reduce the model size, evaluation of each feature attribution contribution for the accuracy is performed. And the result shows that we can cut the feature dimension from 22 to 8 while obtaining the smallest model.

Smart homes provide a technologically enhanced environment that helps user’s complete activities of daily living, thus increasing their autonomy. However, predominant use of verbal prompts in smart homes, with little knowledge of their effectiveness, affects significantly their efficiency by providing prompts that are not optimized with the profiles of the users and the characteristics of the tasks. In order for prompts to be effective, they have to compensate the deficits of its users by exploiting their remaining strengths. To contribute solving the issue, we present, in this paper, basic guidelines that are useful for increasing prompt efficiency in smart homes. We identify relevant significant individual profiles and task characteristics that affect prompt efficiency and how to use prompts accordingly. In addition, we illustrate our efforts to validate the proposed guidelines by giving preliminary results from our ongoing experimentations and by presenting the experimental protocol and software application being used.

Falls are one of the major risks for seniors living alone at home. Computer vision systems, which do not require to wear sensors, offer a new and promising solution for fall detection. In this work, an occlusion robust method is presented based on two features: human centroid height relative to the ground and body velocity. Indeed, the first feature is an efficient solution to detect falls as the vast majority of falls ends on the ground or near the ground. However, this method can fail if the end of the fall is completely occluded behind furniture. Fortunately, these cases can be managed by using the 3D person velocity computed just before the occlusion.

An increasingly aging population, information technology developments and shift to high-rise residential neighborhoods combined with lack of aged facilities have led recently to a specific type of residential development in Korea. They appear in the form of silver towns integrated with smart technologies and targeted specifically at the aging population. The silver towns are increasingly being conceived, designed and marketed as smart living environments for the elderly. This paper analyses selected silver towns to obtain insights from the elderly perspective about these developments and understand design challenges for future silver towns and smart technologies.

Because of cognitive problems, it is very difficult for individuals with Alzheimer’s disease (AD) to manage their time. Consequently, they are dependent on their caregivers or use pen-and-paper organizers, both of which have limitations. A more interesting alternative could be an electronic organizer that optimizes the functional autonomy of persons with AD. This article describes the development of an electronic organizer called AP@LZ by an interdisciplinary team of computer specialists and clinicians. The results show how knowledge of technology and knowledge of the cognitive capacities of persons with AD can be combined. A pre-experimental phase confirmed that AP@LZ was easy to use by elderly participants.

One of the biggest challenges we currently face is to keep elderly people immersed in their social environment when they leave their home and enter a retirement home. Many of them feel isolated. The TV stands as their favorite media, and our first experiments showed that listening to vocalized local news and receiving TV messages and photos from family helped in fighting these feelings of isolation. With online social networks, we wish to involve the elderly in new types of interactions, more various and frequent. They will be more active and included in micro-conversations around multimedia contents. The retirement homes will benefit also from social networking capabilities. They will participate to the local news dedicated to the elderly people. In addition, the remote family will be informed of activities through an agenda and various publications.

One of the main features associated with those suffering from dementia is memory loss. Recent studies have shown that the deployment of technological reminder based solutions, have to a certain extent, been capable of alleviating some of the issues associated with memory loss. Nevertheless, technical and usability challenges are still major hindering factors to large scale uptake. This work presents the details of the evaluation of a mobile phone based video reminding system with 4 persons suffering from mild dementia. We gathered 101 days worth of usability data during an evaluation period of 5 weeks along with a qualitative collection of pre and post evaluation questionnaires. The results from our evaluation have shown that out of 216 reminders delivered during the evaluation period only 18 of them were not acknowledged by the persons with dementia. Upon examining the post-evaluations, it was found that the carers played a significant role in terms of the success of the solution and that an initial settling in period of on average 14 days was required before users felt comfortable using the technology.

Ambient Assisted Living (AAL) systems integrate stand-alone assistive technologies with smart homes and telehealth. This paper reports on a study that focused on the envisioned impact of AAL systems on the lives of end-users using an alternate research approach. A qualitative design was used and semi-structured interviews were conducted with 12 older adults in Southwestern British Columbia, Canada. While a high degree of acceptance regarding AAL was found in the present study, the research also discovered that such technologies have the potential to profoundly affect, both positively and negatively, participants meaning and experience of the home environment. These findings suggest that research and development paradigms need to be expanded if our intention is to produce a product that will be accepted and helpful to the end-user.

Representing knowledge via building an ontology for healthcare messages is important to achieve semantic interoperability among healthcare information systems and to better execute decision-support systems. HL7, an ANSI-accredited developing organization, developed the most widely-used messaging standard in healthcare information systems. This messaging standard, in its version 3, is based on an object-oriented model called the Reference Information Model (RIM). In this paper, a description logic-based ontology in Web Ontology Language for HL7-RIM is proposed. This proposed ontology is used to instantiate medical records written in HL7, and also has been used to reason a sample medical rule. The evaluation results show that the ontology can capture necessary clinical information elements and can be used successfully by decision-support systems.

The “Care Delivery Frame (CDF)” presented here integrates two distinctly different applications, the home telehealth system and the remote photo sharing service of digital photo frame, to create a unique information channel for senior users who are not familiar with the operation of computers and Internet. In addition to health data monitoring, children or caregivers can “deliver care” to their seniors not living together by warm messages and thoughtful reminders on the CDF, as well as sharing their feelings, joy, and life experience through photos and video clips. CDF is positioned as the software for providing care and positive emotion to senior users, with the emerging new class of computers – the “Pad PC”, as its major target platform. CDF is the smallest possible home telehealth system that can be established between a single user and his/her care giver. It is our answer to the current difficulties and challenges of home telehealth systems.

Advances in sensor technologies have provided the opportunity to perform continuous and unobtrusive capturing of physiological signals. One particular application that has benefitted from this technology is the remote monitoring and management of cardiovascular conditions. In this paper, details of an investigation considering the impact of everyday activities on heart rate are presented. ECG and accelerometer signals collected from wearable wireless sensors have been utilized to investigate the underlying relationships between physiological and activity-related profile information. The impact of activities on heart rate has been captured through analysis of the patterns of heart rate using the CUSUM algorithm. Subsequently, results have shown that a change in the pattern of heart rate is detected shortly after an activity commences. Further extensions of the research are also proposed, including integration of a range of ECG features and intelligent data analysis techniques, thereby facilitating the future development of context aware health monitoring mechanisms.

Technological innovations, varying from ubiquitous computing, intelligent appliances, telecommunication, robotics, to wearable sensors, enable new Smart Home (SH) applications. More and more academic publications reporting on experiments on SH can be found. A comprehensive clustering of concepts and approaches is largely missing. Based on an extensive review of SH literature, this paper proposes a framework that decomposes the SH research into four domains and 15 sub-domains. The framework is applied to visualize the state of the art of SH research, and to outline future challenges. The framework helps researchers to identify gaps in SH research.

The appropriation process of the environmental space parameters by elders can be done through the use of color codes that are meaningful to the persons with loss of autonomy. This approach is based upon the natural interpretation of communicating signs and signals by the brain that helps to answer more quickly to critical situation at different degrees. We illustrate first with the transposition of the color codes used in the automobile domain to the housing environment. By extension, we show that intelligent furniture can be designed through ICT and home automation to assist persons with cognitive troubles or motor disability. Finally, we describe the application done in a dare care center for persons with cognitive troubles. Colors of the rooms have been chosen to allow the patients to naturally identify the role of each room.

The number of smart appliances and devices in the home and office has grown dramatically in recent years. Unfortunately, these devices rarely interact with each other or the environment. In order to move from environments filled with smart devices to smart environments, there must be a framework for devices to communicate with each other and with the environment. This enables reasoners and automated decision makers to understand the environment and the data collected from it. Semantic web technologies provide this framework in a well-documented and flexible package. In this paper we present the Casas Ontology for Smart Environments (COSE) and accompanying data from a test smart environment and discuss the current and future challenges associated with a Smart Environment on the Semantic Web.

A fully ambulatory urodynamics monitoring (AUM) system was developed in this study. Conventional cystometry (CMG) and AUM were performed for 28 patients with neurogenic bladders caused by spinal cord injury (24 males and 4 females, age: 49.4±13.9 years, BMI: 23.5 ±2.4). As a result, 10 of the patients were diagnosed as having different reflexibility of the bladder between conventional CMG and AUM (

p

<0.05), and in the patients with areflexic bladders the number of patients with detrusor overactivity was higher in AUM and leakage was observed more frequently. These results demonstrated that our system could be a useful additional tool in the clinical assessment of patients in which CMG failed to explain their symptoms.

The phenomenon of an aging society has derived problems such as shortage of medical resources, rising healthcare costs and reduction of quality in healthcare services. Pervasive healthcare aims to alleviate these problems, but many issues remain to be resolved. This paper presents a system called CARA (Context Aware Real-time Assistant) whose design goals are to address these issues in a pervasive long-term healthcare solution. CARA aims to provide efficient healthcare services by adapting the healthcare technology to fit in with normal activities of the elderly and working practices of the caregivers. This system can continuously measure physiological signals, and either store the data on the server or stream the data to a remote location in real-time. A design goal of ubiquitous access has prompted the design of the system as a rich internet application. The only tool required is a web browser with the commonly-available Adobe Flash plug-in installed. Thus the system enables access and analysis on any internet-connected PC or appropriate smart device, independent of geographic location. The remote monitoring and data review applications of CARA are presented at as the main implementation examples. The results of experiments using the system are presented.

Research advances in smart home technologies are highly diverse and innovative[1,2,3]. Some of these novel ideas involve in creating new hardware to perform distinct functionality. In this paper, we present two effective methods for transforming these ideas and concepts into actual hardware: rapid prototyping and embedded systems. These two methods used in combination have the benefits of high fidelity, formal documentation (thus can be successively improved), quick turnaround time and low cost. To illustrate these benefits, we demonstrate two prototypes developed in our Smart Home Lab: a physical monitoring and logging device and a smart lamp. Since these two methods are applicable in creating wide varieties of prototypes, our discussion and demonstration should inspire other researchers in smart home areas to use similar methods to create new prototypes and products. To improve overall usability, some design principals and issues are also discussed.

This paper gives an overview of the recently granted AAL-JP project FEARLESS which stands for “Fear Elimination As Resolution for Loosing Elderly’s Substantial Sorrows”. The proposed project aims to reduce elderly’s fears within their homes. As elderly potentially refuse or forget to wear any additional sensors to activate alarm calls, FEARLESS will visually and acoustically detect and handle risks by contacting the relatives or care taker organization automatically - without the need of any user intervention. This is done by using only one single type of sensor making the system affordable for everyone. It increases the feeling of safety, reduces fears, enhances the self-efficacy and thus enables elderly to be more active, independent and mobile in today’s self-serve society.

Currently smart home are created and deployed on each user’s specifications, which guarantees their complete satisfaction, but incurs a very high deployment cost and non-usable resources. Techniques from product line engineering provide a framework to promote resource reusability by identifying reusable common features. They help to reduce the cost but can compromise users’ satisfaction, especially when they need special accommodations or have unique preferences. In this paper, we propose wrenching, a transient relaxation of common features into variability, so uncommon, ad hoc request can be satisfied. Based on wrenching, we further devise Smart Variability Model, that can accommodate situations where existing models are not applicable.

Smart homes of the future will have a number of different types of sensors. What types of sensors and how they will be used depends on the behaviour needed from the smart home. Using the sensors to automatically determine if a home is occupied can lead to a wide range of benefits. For example, it could trigger a change in the thermostat setting to save money, or even a change in security monitoring systems. Our prototype

Home Occupancy Agent

(HOA), which we present in this paper, uses a rule based system that monitors power consumption from meters and ambient light sensor readings in order to determine occupancy.

With the advance of IT technologies, a new type of computing device will be introduced for smart environments in the near future. In this paper, we outline our on-going development of the robotic computer that naturally interacts with users, understands the current situation of users and environments, and proactively provides users with services. The proposed robotic computer consists of a control unit and an agent unit. The control unit, which is a remote processing server connected to the host server of the smart environment, provides memory (high-level information) and processing (computing resources) capabilities to the agent unit. The agent unit is a portable device which is wirelessly connected to the control unit, and interacts with users and physical objects. We describe the system architecture and the implementation of a proof-of-concept prototype of the proposed robotic computer.

A new architecture is suggested in this paper for an Assistive Platform in an Ambient Computing for elderly or disabled people. In this situation, interaction between the user, the environment and the platform is guided by means of a generated process. Depending on the context, the elements of the process are dynamically assembled. The aspect of assembly is based on rules that are set through an online learning operation. This new approach of designing an assistive platform is validated through devices integrated in a powered wheelchair with those implemented within the environment. Rules of operation evolve depending on a data base which is continuously updated depending on user’s daily activities.

The vision of smart living promises innovative services from providers in energy, healthcare, entertainment and surveillance sectors. While smart homes used to equal home automation, evolving ICT technologies now enable truly adaptive and intelligent services that are integrated in several industries. Sector-specific service platforms are emerging that provide basic intelligence for services. However, smart living services should not be constrained to a specific industry sector, and hence the service platforms should be easily accessible for service providers regardless of their industry. In addition, smart living services should not be constrained to residents within the home, but should take advantage of outdoors position information to open up a range of novel service concepts. Establishing such a vision of smart living involves various technological and organizational challenges. This paper gives an overview of the current service platforms for smart homes and positioning information. Based on this overview, we propose a research agenda for enabling smart living services. The two major issues are how to achieve collective action between players from multiple sectors in order to set up a common service platform, and how to design business models that allow adding GNSS-based positioning information to the service platform.

In this paper, we present the UbiSMART (Ubiquitous Service MAnagement and Reasoning archiTecture) framework that aims to provide a highly flexible context-aware service interaction mechanism. Pervasive computing enables assistive environments with extensive possibilities, especially when seizing opportunities to enhance quality of life for ageing people with dementia. However, we have to design novel interfaces to aid the engagement of targeted users towards the provided services. A natural, seamless interactive experience must be created. Our focus lies mainly on the reasoning mechanisms performing user interface plasticity, i.e. context-aware adaptation of the interaction to create a seamless user experience.

This paper presents a design of an architecture that facilitates the work of a fusion engine. The logical combination or merging of input streams invoked by the fusion engine is based upon the definition of a set of patterns and its similarity with previously collected data from various modalities. Previous fusion engine designs had many weaknesses, among them their being specialized on a specific domain of application. The proposed architecture addresses such weakness and provides additional features, namely its ability to handle large number of modalities and due to its using knowledge base and standardization characteristics; it becomes suitable to various types of multimodal systems. The techniques used to achieve these features are discussed in this paper.

BlindShopping is a mobile low-cost easily-deployable system devised to allow visually impaired people to do shopping autonomously within a supermarket. Its main contributions are: a) a user navigation component combining an RFID reader on the tip of a white cane and mobile technology, and b) a product recognition component that uses embossed QR codes placed on product shelves and an Android phone camera for their identification. Furthermore, it provides a web-based management component to easily configure the system, generating and binding barcode tags for product shelves and RFID tag markers attached to the supermarket floor.

As part of a PhD thesisat the Loria research center, National Polytechnic Institute of Lorraine (INPL), “Maisons Vill’Âge

®

” a new concept of building smart home by integrating telehealthcare and home automation systems, is developed in France. The segment of population, we are targeting is the senior citizens. The healthcare system uses home automation sensors and other environmental sensors like bed and chair sensors to monitor the activity level of the elderly. Activity patterns are analyzed by an intelligent application which is based on Fuzzy Logic to find any unusual behavior. The system is also equipped with wireless medical sensors to monitor the health situation of the elderly; it uses also a wireless sensor network to detect falls. The system detects health abnormalities at an early stage through the frequent monitoring of physiological data. This paper presents a brief description of the system.

With growing aging population, elderly with physical or cognitive impairments may end up staying alone at home. Kitchen is generally the most vulnerable place at home as mishandling of devices or improper kitchen activities could lead to hazardous and life threatening situations. In this paper, we have proposed pervasive intelligence system that augment existing kitchen environment with sensors, actuators and processing intelligence. We have first identified possible safety issues and, then deploy sensors to recognize what happened and actuators to control kitchen settings. We have developed 3-stage processing with hierarchical inference approach to determine abnormalities in the kitchen. According to the severity of abnormalities, the proper appliances control or reminder escalation to appropriate personal will be issued to alleviate or minimize the undesirable consequences.

The multi-sensor fusion can provide more accurate and reliable information compared to information from each sensor separately taken. Moreover, the data from multiple heterogeneous sensors present in the medical surveillance systems have different degrees of uncertainty. Among multi-sensor data fusion techniques, Bayesian methods and evidence theories such as Dempster-Shafer Theory (DST), are commonly used to handle the degree of uncertainty in the fusion processes. Based on a graphic representation of the DST called evidential networks, we propose a structure of heterogeneous multi-sensor fusion for falls detection. The proposed Evidential Network (EN) can handle the uncertainty present in a mobile and a fixed sensor-based remote monitoring systems (fall detection) by fusing them and therefore increasing the fall detection sensitivity compared to the a separated alone system.

This study aimed to examine the feasibility of using and evaluating SMART (Supported Mental Assessment, Rehabilitation and Treatment) electronic technology as part of providing cognitive remediation for tertiary mental health care in-patients diagnosed with schizophrenia-related cognitive impairments that are considered a barrier to their independent living. This was an uncontrolled intervention feasibility and exploratory study. The study involved eight participants in total. Both qualitative and quantitative research strategies were used. Results to date are promising. Participating patients expressed satisfaction with the simulated apartment and the smart technology and patients were able to learn and use skills relevant to independent living.

Diabetes is one of the chronic diseases that have an effect on the life of a large number of people around the world. This chronic disease causes approximately 5% of all deaths globally each year[1], and is most likely to increase exponentially within the coming few years. Reduction to the effect of such a disease, a close monitoring to patients is essential in order to control its level via increasing patients’ awareness. Thus a smart system that is capable of collecting medical readings from patients, analyzing the data, and suggesting the appropriate diagnosis has become a necessity. This work advises a system that is capable of meeting all of the above along with reducing the physical access load on healthcare centers via instructing the system to notify both healthcare provider and patients with the acquired results.

In previous works overall activity and inactivity levels of users living in Ambient Assisted Living (AAL) enabled flats were determined using standard home automation sensors. The flats are regular dwellings for long-term use by approximately 30 tenants located in Kaiserslautern, Germany. In this real-world AAL project it was shown that basic inactivity alarms based on linear thresholds can be triggered within 30 to 180 minutes after the occurrence of a potential emergency. However, inactivity alarms are somewhat coarse and do not make full use of additional information inherent in the raw sensor data: spatial and temporal information regarding the location of a tenant in their flat and the time spent in a room. Using that information, it can be determined in which room a tenant has resided for how long at a given time. Hence, in this paper a method for location tracking is proposed, forming a novel alarming criterion.

A growing number of the elderly with dementia are accommodated to group homes and live with caregivers in Japan. We are concerned with Activities of Daily Living(ADL) of these residents and installed into a group home a sensor network as a pilot study to record and analyze their whereabouts. Twenty one antennas were placed under mats and five residents wore slippers into which Radio Frequency IDentification(RFID) tags were inserted. We collected location data for thirty months and looked for long term tendencies in their activities. We found that the information concerning their whereabouts may indicate their waning health conditions and may reflect caregivers’ treatment. We also found a seasonal effect on residents’ mobilities, a decline in summer. The result shows that location data may be exploited as a valuable source to recognize residents’ slowly changing conditions.

Smart home has become a very active topic of research in the past few years. The problem of recognizing activity inside a smart home is one of the biggest challenge researchers have to face in this discipline. Many of them have presented approaches exploiting temporal constraints in order to maximize the efficiency and the precision of their recognition model. However, only few works investigate the spatial aspects characterizing the habitat context. In this paper, we present a new algorithm and a complete experiment showing the importance of taking into account spatial constraints in the recognition process. The goal of this paper is to demonstrate that recognition algorithms will benefit from exploiting spatial constraints.

Multiple people activity recognition system is an essential step in Ambient Assisted Living system development. A possible approach for multiple people is to take an existing system for single person activity recognition and extend it to the case of multiple people. One approach is Multiple Hypothesis Tracking (MHT) which provides capabilities of multiple people tracking and activity recognition based on the Dynamic Bayesian Network Model. The advantage of such systems is that the number of people can vary, while the disadvantage is that the activity recognition configuration cannot be done if only multiple people data is available for training.

Single-person households have appeared recently as one of new distinct family types in Korea and lived mainly small scales of studio apartments and flats. However, we expect that normal large scale of houses or apartments will be inhibited by the residents of them in Korea. Accordingly the demands for their home would become dynamic like those of the other family types because their housings will not be temporary choices. The research proposes a direction for the development of smart homes in Korea, especially, for single households with a focus on their home workspace. Based on the results of a case study, a home workspace was identified as the highest need of the single households. Thus, this research emphasizes the effective functioning of home workspace and suggests intelligent components for it by introducing enabling technologies and notions to the development of the smart homes.