Melina Frenken

Modeling individual healthy behavior using home automation sensor data: Results from a field trial

A technical system for unobtrusive presence measurement and two novel models for describing user behavior in domestic environments are presented. Within the developed models user behavior is either described as the probability of being present at a certain location within an environment at a certain time on a day of the week or being present at a location for a certain number of times with a certain duration. The models are called timeslot-based and duration-based. Both models have been applied to presence information gathered by a technical system using home automation sensors. The system was installed into two flats of older people during a field trial for eight months. Results of the experiment show that the two models can be applied to describe individual user behavior. The influence of data structure and model quality on the detection of anomalies and the generation of alarms is discussed. On the long-term, the approach aims at detecting cutbacks in self-care ability and changes in health state by autonomously learning typical user behavior from presence information in a spatial model and by detecting untypical behavior, called anomalies, and generating alarms for caretakers. Such automatic assessment of self-care ability and health state is required in order to meet the increased challenges imposed to the decreasing number of care personal during the progress of the demographic change.

Performing gait analysis within the timed up & go assessment test: comparison of aTUG to a marker-based tracking system

Results from a technical validation of the aTUG (ambient Timed Up & Go) system are presented. The approach’s gait analysis capabilities were compared to a gold standard: SIMI Motion, a marker-based motion tracking system. Seven people participated and computation of step length and step duration happened with a median error of 3 cm (IQR 3 cm) respectively 0.08 s (IQR 0.07 s). These results show that aTUG has a measurement precision which is sufficient for use in clinical gait analysis and enables the use of the device without a gold standard, i.e. in hospitals outside laboratories or in the homes of patients. aTUG is an approach and system that utilizes only ambient sensor technologies to support the execution of geriatric mobility assessment tests and to perform a gait analysis simultaneously. Such capabilities are strongly demanded in order to support physicians in executing the geriatric assessment tests frequently and objectively in professional and domestic environments. The latter may enable more early prevention and more sustainable rehabilitation.

Towards Pervasive Mobility Assessments in Clinical and Domestic Environments

This paper provides an overview of current research and open problems in sensor-based mobility analysis. It is focused on geriatric assessment tests and the idea to provide easier and more objective results by using sensor technologies. A lot of research has been done in the field of measuring personal movement/mobility by technical approaches but there are few developments to measure a complete geriatric assessment test. Such automated tests can very likely offer more accurate, reliable and objective results than currently used methods. Additionally, those tests may reduce costs in public health systems as well as set standards for comparability of the tests. New sensor technologies and initiatives for data standardization in health processes offer increased possibilities in system development. This paper will highlight some open problems that still exist to bring automated mobility assessment tests into pervasive clinical and domestic use.

Functional Assessment in Elderlies’ Homes: Early Results from a Field Trial

Early results from a field trial regarding the assessment of functional status relevant to self-care ability in domestic environments are presented. A previously developed technical system for unobtrusively recording location information using home automation data was installed in the homes of five participants aged 64–84 years over a period of partially more than nine month. The recordings are manually evaluated to check whether items of geriatric assessment tests relevant to self-care ability can be assessed using the sensor recordings. The evaluation is a preliminary step to develop an automatic assessment algorithm and to develop a model for mapping domestic assessment results to result scales of clinical assessment tests. The mapping is required since most clinical assessment tests are not suitable for execution in domestic environments and thus new approaches are required which do also account for the difference between performance and capacity in functional abilities as proposed within WHO’s ICF.

A workflow for design and evaluation of embedded control systems in medical devices

The recent development of more and more autonomous medical devices facilitates unsupervised treatment and thereby saving of costs for healthcare. While flawless control software is needed for such life-critical systems, no specific development processes are established in the medical domain. The here presented approach founds on the functional simulation of system components and their virtual integration as well as modelling of the decisive parts of the physiology influenced by the device. Additionally, a virtual execution platform is integrated enabling to develop the embedded control software and evaluate it online to the functional simulation. Finally, the coupling to an assertion engine for semi-formal verification allows validation of the system’s behaviour and its compliance with requirements, specified as assertions. The utilisation of this proposed methodology is demonstrated on the control development for an autonomous dialysis system. The applicability is evaluated and insights gained through its utilisation exemplary demonstrate its benefits.

Designing and Introducing an Activity Motivating Mobile Web App Platform for Seniors

In this paper, we present results from the VERA project regarding design, usability, and motivation for long-term usage of a mobile web app platform that encourages physical and mental activity. The system is designed for elderly inhabitants of senior residences to motivate them to increase their physical and mental activity in everyday life. Activities can be logged in five different categories: everyday activities, endurance, strength, mental, and special activities. The data of a residential home and an assisted living facility and a total of 29 users were analyzed. The results show insight view about the usage and suggest that accompanying help can improve the long-term sustainability of the motivating effect.

LsW: Networked Home Automation in Living Environments

Due to demographic changes a lot of research in the field of Ambient Assisted Living (AAL) has been done within the last years. Many of the project systems never left the status of a research prototype nor reached real home environments. The “Langer selbstbestimmt Wohnen” (LsW) project is one step beyond the targets of typical research projects and is integrating preexisting (home automation) technologies in home environments to support the elderly residents. Requirements, selection of possible scenarios and evaluation were done in close cooperation with the residents. The integrated AAL system will remain within the flats of the residents beyond the end of the LsW project. Main points of the LsW project were the analysis of current mobile devices like tablets as human machine interfaces for AAL systems, the integration of existing home automation technologies in existing buildings, the connection between the tablet and the home automation components, and the interoperation between several flats.

Ambient water usage sensor for the identification of daily activities

Dementia patients, like most older adults, prefer to live in their own home as long as possible. This requires, however, that they are able to perform activities of daily living (ADL). Therefore, many research projects install different sensor setups to identify ADLs. Though the water usage correlates with many ADLs (i.e.: bathing, cooking) only few of these systems use water usage sensors. The reason is that there is no water usage sensor available that is unobtrusive, ambient and precise. In this article, we propose a water usage sensor that is based on a piezoelectric element that fulfills these requirements. We describe the implementation of the sensor system in a living lab. Additionally, we discuss different features that were extracted from the sensor signal and different machine learning algorithms that were used to classify the data. Finally, we present the results to several tests we performed to determine the accuracy of our sensor system under different environmental conditions.

Perception based method for indoor air quality control

This work describes the design of a building automation controller for indoor air quality, which is different from common control strategies. Normative strategies are focused solely on the management of ventilation rates and the reduction of pollutant concentrations. The positive effect of air movement and the negative effect of high temperatures on the perceived air quality are not considered. In this paper, we present a novel approach to design a controller based on fuzzy logic, which adapts the air movement velocity depending on the temperature. Data about the perceived indoor air quality during varying thermal conditions are used to model the fuzzy controller.

Self-configuring home automation networks based on psychophysical principles

Home automation networks are designed to meet standards of comfort and energy efficiency. Hence, the interaction between home automation and the inhabitant is based on rudimentary knowledge about the condition and the perception of the human. In order to improve the human-to-automated-home interaction, psychophysical measurements can be helpful to design a human-oriented home automation network. In particular elderly residents and people with impairments may have specific requirements and altered expectations towards comfort as defined in the standard guidelines. In this paper, a the concept of a systematic process model, leading to an improved interaction between human and building environment, is presented. This process model has the purpose to create a selfconfiguring home automation network by using already existing psychophysical knowledge about the inhabitant.