Sensor systems can be deployed in the homes of older adults living alone for functional health assessments. Their information is very useful for health care specialists. The problem lies in developing person independent models while facing a large variability in behavior. We address this problem by, first, proposing a new feature extraction method for data from ambient motion sensors. The method uses functional similarities between houses and daily structure to extract meaningful features. Second, we propose a change-based approach for analyzing data, taking difference scores of both the sensor features and health metrics. To evaluate our approach, experiments on longitudinal data were conducted, where the relationship between sensor data and health measurements was modeled with linear regression and (nonlinear) regression forests. These experiments show that the change-based approach yields better results and that the resulting models can be used as a reliable metric for (functional) health. In addition, feature analysis can help health care specialists understand relevant aspects of behavior. Prediction of health metrics is possible even with simple sensors. With such sensors, it is possible to detect problems and health decline in an early stage. This will have great impact on clinical practice.
Several studies have suggested that precision livestock farming (PLF) is a useful tool foranimal welfare management and assessment. Location, posture and movement of an individual are key elements in identifying the animal and recording its behaviour. Currently, multiple technologies are available for automated monitoring of the location of individual animals, ranging from Global Navigation Satellite Systems (GNSS) to ultra-wideband (UWB), RFID, wireless sensor networks (WSN) and even computer vision. These techniques and developments all yield potential to manage and assess animal welfare, but also have their constraints, such as range and accuracy. Combining sensors such as accelerometers with any location determining technique into a sensor fusion systemcan give more detailed information on the individual cow, achieving an even more reliable and accurate indication of animal welfare. We conclude that location systems are a promising approach to determining animal welfare, especially when applied in conjunction with additional sensors, but additional research focused on the use of technology in animal welfare monitoring is needed.
MULTIFILE
Sensor technology is increasingly applied for the purpose of monitoring elderly’s Activities of Daily Living (ADL), a set of activities used by physicians to benchmark physical and cognitive decline. Visualizing deviations in ADL can help medical specialists and nurses to recognize disease symptoms at an early stage. This paper presents possible visualizations for identifying such deviations. These visualizations have been iteratively explored and developed with three different medical specialists to better understand which deviations are relevant according to the different medical specialisms and explore how these deviations should be best presented. The study results suggest that the participants found a monthly bar graph in which activities are represented by colours as the most suitable from the ones presented. Although the visualizations of every ADL was found to be more or less relevant by the different medical specialists, the preference for focusing on specific ADL’s varied from specialist to specialist.
Management policy for protected species is currently often based on literature reviews and expert judgement, even though it requires tailor-made species knowledge on a local level. While wildlife management should preferably be evidence based, tailor-made field data is seldom used in current practices, because it is hardly available, difficult to collect and expensive. Recent development of digital technology is changing the field of wildlife management with “more, better, faster and cheaper” ways of data collection. Especially automated collection of field data with different types of sensors is promising, whereas miniaturization and low cost mass-production increase availability and use of these sensors. For collection of field data about predator-prey interactions, there is a need to develop wireless sensor networks that automatically identify different species in a community, while they record their spatially explicit data and their behaviour. Therefore, we will put together a consortium of partners that will develop a EU LIFE programme proposal, with the focus to develop a sensor network necessary to automatically monitor multiple species (i.e., species communities) for species conservation management. The consortium will consist of Van Hall Larenstein, Sovon Dutch Centre for Field Ornithology, the Dutch Mammal Society, Sensing Clues and DIKW intelligence. It will bring together a strong mix of expert knowledge on applied species conservation and wildlife management, ecological field research, wildlife intelligence, and handling and analysis of big data. This project matches the Top sector High-tech Systems & Materials, and revolves around 4 distinct phases: selection of potential consortium partners, exploration of the problem, working towards a common action perspective and writing a EU LIFE programme proposal. We will use knowledge co-creation techniques to explore the first three project phases.
Brandweermensen lopen het meeste gevaar als ze onder tijdsdruk een gebouw moeten verkennen, of een brand moeten blussen terwijl de situatie nog niet goed kan worden overzien. Omvallende muren, instortende plafonds of gewoon gestruikeld over door de rook onzichtbare brokstukken leiden tot vermijdbare letsels of zelfs slachtoffers. Met name de inzet bij branden in stedelijke parkeergarages onder woontorens vormen een enorm risico. Het inzetten van onbemande, op afstand bestuurbare voertuigen voor verkenning en bluswerk is een oplossing die binnen de brandweer breed wordt gedragen. De brandweer moet deze innovatieve technologie echter zien te omarmen. Zij werken nu vanuit hun intuïtie en weten direct hoe te acteren op basis van wat zij waarnemen. Praktijkgericht onderzoek heeft echter uitgewezen dat scepsis over de inzet van blusplatforms bij incidenten plaats heeft gemaakt voor zeker vertrouwen. Een blusplatform, voorzien van juiste sensoren kan de Officier van Dienst (OVD) ondersteunen bij het nemen van een beslissing om al dan niet tot een ‘aanval’ over te gaan. Praktijktesten hebben echter laten zien dat de huidige blusplatforms nog niet optimaal functioneren om als volwaardig ‘teamlid’ te kunnen worden ingezet. Dit heeft enerzijds met technologische ontwikkelingen (sensoren en communicatieverbindingen) te maken, maar anderzijds moet de informatievoorziening (human-machine interfacing) naar de brandweer beter worden afgestemd. In dit project gaan Saxion, het instituut fysieke veiligheid, de universiteit Twente, het bedrijfsleven en vijf veiligheidsregio’s onderzoeken hoe en wanneer innovatieve blusplatforms op een intuïtieve manier kunnen worden ingezet door training én (kleine) productaanpassing zodat deze een volwaardig onderdeel kunnen zijn van het brandweerkorps. Een blusplatform kan letselschade en slachtoffers voorkomen, mits goed ingezet en vertrouwd door de mensen die daarvan afhankelijk zijn. Het vak van brandweer, als beroeps of vrijwilliger, is een van de gevaarlijkste die er is. Laten we er samen voor zorgen dat het iets veiliger kan worden.
Wildlife crime is an important driver of biodiversity loss and disrupts the social and economic activities of local communities. During the last decade, poaching of charismatic megafauna, such as elephant and rhino, has increased strongly, driving these species to the brink of extinction. Early detection of poachers will strengthen the necessary law enforcement of park rangers in their battle against poaching. Internationally, innovative, high tech solutions are sought after to prevent poaching, such as wireless sensor networks where animals function as sensors. Movement of individuals of widely abundant, non-threatened wildlife species, for example, can be remotely monitored ‘real time’ using GPS-sensors. Deviations in movement of these species can be used to indicate the presence of poachers and prevent poaching. However, the discriminative power of the present movement sensor networks is limited. Recent advancements in biosensors led to the development of instruments that can remotely measure animal behaviour and physiology. These biosensors contribute to the sensitivity and specificity of such early warning system. Moreover, miniaturization and low cost production of sensors have increased the possibilities to measure multiple animals in a herd at the same time. Incorporating data about within-herd spatial position, group size and group composition will improve the successful detection of poachers. Our objective is to develop a wireless network of multiple sensors for sensing alarm responses of ungulate herds to prevent poaching of rhinos and elephants.
