Unobtrusive Fall Detection and Prevention: Extending From a Prototype Test to a Pilot Trial
Henriette Rau, Jacob Grieger, Christian Marzahl, Peter Penndorf, and Martin Staemmler
University of Applied Sciences, ETI, Stralsund, Germany
{henriette.rau, jacob.grieger, christian.marzahl, martin.staemmler}@fh-stralsund.de [email protected]
Abstract: Fall detection based on images is rated obtrusive and costly. This paper presents an unobtrusive fall detection system, which allows a nearly invisible positioning under furniture and cost-efficient integration and scalability in retirement homes via WLAN. The system classifies events using image analysis and notifies the caregiver if an alarming event occurred. It was successfully tested in a nursing and a retirement home and refined based on the experiences made.
Keywords: fall detection, fall prevention, context-aware, standards, AAL, complex event processing, pilot trial, evaluation
1 Introduction
Due to the demographic change in Germany the number of elderly people is continuously increasing. Statistics show that the risk to fall, the number of recorded falls and the seriousness of injuries increases exponentially starting at the age of 70 [Fun07]. Therefore the prevention and detection of falls is an important topic for ambient assisted living (AAL) research.
This paper presents lessons learned from a conducted prototype test of an unobtrusive fall detection system to design and implement a pilot trial in a retirement home. The initial system used for the protoype test is introduced in Chapter 2. The lessons learned presented in Chapter 3 detail on the re-engineering both of the system design and the integration for the pilot trial in a retirement home. Chapter 4 gives first trial results and a short discussion.
2 Initial Prototype Test
Based on previous work [Mar12] this chapter describes the architecture of the initial fall detection and prevention system and the experiences made from running the prototype test. The prototype test was performed from August to December 2011 in two inhabitant’s rooms of a nursing home. Its objective was to achieve first feedbacks regarding the prototype developed.
1456
2.1 System Design
Figure 1 shows the three main components of the system used for the prototype test. These components are described in the following.
Fall Detection Events, Status Central Processing AlarmSmart Phone / Work Station
Figure 1: Overview System Components
2.1.1 Fall Detection
The fall detection component involves the Microsoft Kinect [Kin12] camera and a computer for image analysis and event generation. These devices are connected via USB (Figure 2a).
KinectImage Analysis
Event Generation
USB
3D Images
Figure 2a: Fall Detection Components
Central Processing
Event Database
Event Storage / Event Analysis
Alarm Database
Alarm Storage / Alarm Configuration
Figure 2b: Central Processing Unit
The Kinect camera delivers 3D depth images to the image analysing unit. This unit processes the data image by image with distinct algorithms (e.g. segmentation, feature extraction) and classifies the current situation using a decision tree [Mar12]. The classification process generates an event and calculates the probability for the occurrence of this event. Implemented event types are:
Fall (person fell) Activity (person wandering in the room)
1457
Feet (feet in front of the bed) Bath (bath door opened) Exit (room door opened)
Generated events containing the event type and data such as timestamp, room ID and probability are sent cable-based or wireless to the central processing component. Furthermore, the fall detection component generates a heartbeat signal enabling the central processing component to detect unavailability of the fall detection system.
2.1.2 Central Processing
The central processing unit is implemented as a bundle of web services with distinct assignments (Figure 2b). A web service receives detected events and stores them in an event database. A time-triggered process analyses the stored events belonging to the same context (i.e. room or inhabitant) of the last ten minutes to detect conditions for raising an alarm (event-condition-action). These alarm condition rules have been manually configured in the alarm database for each context and consist of constraints (e.g. time or mutual exclusion of events) and if-then rules taking event probabilities into account. If the analysed data matches an alarm condition, an alarm actuator executes the appropriate alarm action. Those actions have also been configured manually in the alarm database and comprise of notifications sent to smart phones by using an Asterisk [Abo12] server or to work stations via a web interface.
Caregiver
Smart Phone
Work Station
Central Processing
Alarm Notification
LAN/WLAN
LAN/WLAN
Figure 3: Alarm Notification Process
1458
2.1.3 Smart Phone / Workstation
In most cases the receiver of an alarm notification is a caregiver assigned to the room, inhabitant or ward the alarm originates from. Figure 3 shows that the caregiver may either be notified by work station or smart phone in order to provide immediate assistance.
2.2 Test Accomplishments and Experiences of the Prototype Test Phase
The prototype test revealed some unwanted behaviour of the system, e.g. false-positive alarms, which resulted in an immediate correction of algorithms and rules. Thus, the false-positive alarm rate could be reduced. During the prototype test one real fall occurred, which was correctly detected. The images could also show what happened before, during and after the fall. Using only depth images ensures privacy through the anonymous detection of activities and falls. The prototype test also showed the need of protecting the system’s hardware from access through unauthorized persons or mechanical influences, e.g. shifting the camera or pulling the power plug. For reducing the computational effort the initial image analysis was enhanced to ignore non-moving objects and thereby to assign more effort to the event classification. Furthermore, following the need to synchronize between the client and server clocks, NTP has been established. Without proper synchronisation false-negative alarms existed due to a client’s time stamp virtually terminating earlier than the server’s event registration.
Calculating the probability of an event has been done for every image during the prototype test. To minimize fluctuating values for probabilities an analysis of two or more images captured in a row has been identified as a target for the pilot trial. For a better understanding of the context leading to the fall, the alarm notification of the pilot trial needs to be enhanced by saving captured 3D-images from a few minutes prior to the detected event using a data queue.
3 Designing the Pilot Trial
After improving the algorithms through the prototype test a pilot trial was performed to analyse the system’s suitability to operate permanently in a retirement home. The following descriptions are in general terms and do not focus on this particular home’s technical installation requirements, since these vary from home to home.
To eliminate possible obstacles in the new retirement home a two-week pre-trial with healthy test persons was performed in March 2012. After that the main field trial in two inhabitant’s rooms started in May and will end in August 2012.
1459
3.1 System Design
The key objective for the pilot trial was to further reduce the false-positive rate of 63% obtained in the pre-trial. Therefore both the certainty of the sensor data evaluation and the impact of context information had to be increased. The lessons learned from the prototype test lead to extensive changes of the system architecture on various levels. Table 1 shows the most significant changes.
Table 1: Changes to the System Architecture
Prototype Test Pilot Trial one sensor: Kinect + image processing various sensors: Kinect + image
processing and door contact sensors single data evaluation data fusion and evaluation sensor data is processed centrally with one rule set
sensor data is handled as an event, transported over event streams and processed in an event processing network (EPN) comprising of event processing agents (EPAs) with specialized rule sets
inference at context level inference at situation level alarm notification alarm notification and image display
The event-driven approach applied to the prototype test was extended by the complex event processing (CEP) architecture in order to achieve:
Actuality: Real-time data acquisition and analysis for immediate alarm notification
Efficiency: Ability to handle large data volumes from various sensors [Etz10] Scalability: Capability to scale-up the number of sensors and EPAs Agility : Possibility to add additional sensors and additional EPAs Maintainability: Separation of processing logic through EPAs
The initial design included one sensor, i.e. the Kinect and the image processing unit. To gain a better certainty, context information and additional sensors such as door contacts are included in the pilot trial. Sensor data with corresponding context is fused to generate relevant context information. A reasoning engine infers the current situation from different types of context information. According to the approach of CEP [Bru10] all sensor data is handled and represented as a simple event. Figure 4 shows the event streams associated with different levels of data processing.
1460
Sensor Data
Event Streams
Data Fusion Context Information
Simple Events Composed Events
Reasoning Situation
Complex Events
Level of Abstraction
Figure 4: Levels of Data Processing and According Event Streams
The context information is represented as a composition of simple events since this information consists of data from different sensors. An inferred situation is presented as a complex event representing the highest level of data abstraction.
Figure 5 depicts an exemplary evaluation of the situation “Person-Left-Room”, i.e. the inhabitant left the room. The Kinect in correspondence to the image processing detects a door-opened event (i.e. exit door) with a specific certainty. Furthermore, a door contact assigned to the exit door signals the same type of event. Even though the information of the events is similar, the events contain different types of data. In order to enable the CEP to handle event data from different sensors, adapter components translate the sensor data in heterogeneous formats into event data in a homogeneous format. The system design for the pilot trial is comprised of an EPN with two types of EPAs. The low level type aggregates simple events to represent context information; the high level type infers the current situation taking diverse context information into account.
1461
KinectImage Analysis
USB
Sensor 1 Sensor 2
Door Contact Sensor
Complex Event ProcessingAdapter Adapter
EPA: Exit Door Opened
Rule Repository
Event Processing Engine
Event Model
EPA: Activity
EPA: Room Left
Exit Door Opened
No Activity
Event Repository
......
Adapter
Web Service
Event Repository Alarm Configurations
Smart PhoneWork Station
AsteriskGUI
Processor
WS IF
Event: Exit Door Opened; Value: 76%
Event: Exit Door Opened; Value: true
Room Left
...
Figure 5: Exemplary Situation Evaluation
1462
The CEP component was implemented using the Esper [Eve12] framework. Figure 5 shows the EPA for generating the “Exit-Door-Opened”-event. To increase the certainty of this context information this EPA takes data from different sources into consideration. Each EPA is comprised of an event processing engine (EPE), a rule repository and event model specifications. The EPE processes the event stream applying rules and constraints from the rule repository and generates events according to the event model specifications. The EPE queries the event stream for event patterns specified in the event processing language (EPL) EQL which is included in the Esper framework.
The composed event “Exit-Door-Opened” is streamed to the EPA dedicated to detect “Person-Left-Room”-events. This agent accesses different context information to infer a situation. In addition, high level EPAs store arriving events to detect temporal and causal relation among events. The “Person-Left-Room”-EPA in Figure 5 evaluates the “Exit-Door-Opened”-event and the following “No-Activity”-event detected by another EPA. Furthermore, this EPA takes persisted events from the event repository into account to infer the “Person-Left-Room”-situation. In case of the situation, in which a person occupies the bathroom (inferred from “Activity”-event and “Bath”-event), the time the person spends in the bathroom can indicate a fall which might be undetected by the Kinect camera. The rules for the evaluation of the time spent in the bathroom are configured individually for every inhabitant since an unusual timespan is a personalized parameter.
Recognized high level events respectively situations are sent to a web service that stores high level events and raises alarms according to alarm configurations. Similar to the prototype test the system for the pilot trial includes caregiver notification using a web interface or smart phones. In addition to the notification the web service allows to transfer images from the sensor containing the Kinect to a workstation for display (Figure 6). The image repository on the image analysis unit provides this image retrieval functionality.
KinectImage Analysis
USB
Sensor 1Web Service
Event Repository
Work Station
GUIImage Request
Image Response
Image Repository
Figure 6: Image Retrieval via Web Service
1463
3.2 System Integration
This chapter details the integration into a retirement home and more importantly the necessary steps to implement the unobtrusive fall detection in an individually furnished room and with regard to different physiques of the inhabitant.
3.2.1 Technical Integration
The technical integration, in particular the establishment of network connections was one of the most important prerequisites of the pilot trial since network failures were one of the main sources of error during the prototype test. To implement a network architecture which could be used in every room and or retirement home a repeater-based network is used. As described in Figure 7 one repeater was connected to the physical network cable and generates thereby the needed network for the fall detection sensor (Kinect and image analysis device) to communicate with the event processing unit. Events are sent via WLAN from the fall detection devices to the physical network-cable and then via LAN from the inhabitant’s room to the event processing unit located in the server room. A remote access to the server was implemented for maintenance purposes.
Figure 7: Architecture of the installed Hardware
This approach enables data transmissions over larger distances in a retirement home. Two or more rooms may use the same WLAN repeater. As a result this architecture is easily expandable in a cost-efficient way.
To minimize physical influences on the camera system (e.g. shifting the camera while mopping) and visual impact on the inhabitant’s environment (e.g. loose cables) a special housing for the Kinect and PC was developed. The housing base used aluminium combined with an acrylic glass window for the Kinect. The network connection via WLAN was ensured.
Thanks to the system’s design only a power plug and a connection from the WLAN access point to the server room are required, minimizing necessary cabling and potential tripping hazard.
1464
3.2.2 Positioning
Positioning the camera turned out to be a key issue with regard to the quality of alarm detection. The initially planned camera position under the inhabitant’s bed couldn’t be used due to the extra-low electric beds and or unfavourable room topology. To see all important areas (bed, bathroom door and the room’s exit) at least one camera had to be placed under the inhabitant’s furniture as depicted in Figure 8.
The best camera position and the areas to detect alarms have to be individually planned and configured. It’s of great significance to talk to care givers or the inhabitants to learn of habitual places for e.g. wheelchairs or rollators which may obstruct the camera.
3.2.3 Configuration
Due to the divergent furnishing found in the inhabitant’s rooms every detection area had to be manually drawn. For this purpose a specialized tool was developed to support the technician in specification using a step by step approach. The technician uses room points to indicate where the area, e.g. bed, should be.
Figure 8: Inhabitant Room Topology and Kinect’s Field of View
The prototype test showed that the configuration has to be customized to the inhabitant’s body height and width to improve the alarm rate. That way even falls of smaller persons will be accurately detected.
Exit
1465
Following this individual configuration process, which takes some minutes, the system is ready for permanent operation without the need of further configuration steps.
3.3 Preliminary Trial Accomplishment
The pre-test showed that the accuracy to detect real falls of persons with different height was about 96%. But the initial system also raised alarms when no fall occurred with a false positive rate of 63%. As previously described the probability calculation of an event has been done for every image during the prototype test. This approach was modified and the system designed for the pilot trial uses more than two images captured in a row to calculate an event probability and reduce false-positive alarms. Due to additional sensors and therefore more context information the false-positive alarm rate should be further reduced. The pilot trial is still ongoing. Therefore no data analysis concerning the false-positive alarms can be presented yet.
Additional analyses, e.g. daily activities or the lack thereof, are also possible and allow the recognition of behavioral changes. Those changes often indicate a worsened health state and are, therefore, important for caregivers. The pilot trial indicated that a second Kinect should be used in every inhabitant’s room to avoid blind spots.
The scalable integration of the system in the retirement home was thanks to the repeater-based network easily implemented and can be repeated in other retirement homes without the need of constructional measures.
The pilot trial was showed so far a lot potential for additional analyses, enhancement with sensors and implementing in more rooms throughout the retirement home.
4 Discussion
The lessons learned from the prototype test lead to a new pilot trial design concerning both the system architecture and the system integration. The architecture based on CEP provides scalability and higher certainty of context information compared to the initial design. In future work the HL7 Personal Healthcare Monitoring Report (PHMR) standard [Hea12] will be used for communication between the client and the server. Though this CEP approach does not use a semantic event model or sensor model based on ontologies [Sit12], the system is able to extract relevant context information using data fusion and event aggregation. In future work, the recognition of acoustic events will be implemented, though the combination of Kinect data and additional sensor data provides a reliable situation evaluation. In [Pfl12] audio and video sensors are combined already, but neither a Kinect camera is used nor are pilot trial results present yet. In addition, future work will involve the analysis of behaviour patterns to generate an inhabitant profile [Che08]. This profile could provide statistics such as movement tracking and tracing to support dementia research. Future trials will involve at least two Kinect cameras per inhabitant room to compensate for the limited field of view and to cover all important room areas.
1466
The unobtrusive fall detection system presented in this paper has benefits compared to various approaches described in literature. Whereas in other projects sensors have to be carried by the inhabitant, this system will not interact with the inhabitant’s daily life being positioned under the inhabitant’s furniture. Also integrating the unobtrusive fall detection system does not need any constructional measures (e.g. false floors for cable lying, specific floors acting as sensors) and works in a plug-and-play-way assuming that a network infrastructure is given. To increase the portability of the system prospective studies will investigate the integration both of the universAAL [uni12] platform and the openAAL [ope12] middleware.
Additional sensors such as bed sensors, smart meters or door contacts can be integrated in the system parallel to the Kinect. Nonetheless sensors such as motion detectors, which use differences in heat radiation between a person and its environment for detection, as described in [Bau12] may be replaceable through the Kinect’s ability to register motions, which can then be further analyzed. Another advantage of the Kinect is that it does not use the heat radiation principle, which was a source of error according to the test results in [Bau12]. The infrared capability of the Kinect is a plus when compared to other systems [Pha08] which require light.
Notifying a caregiver on a smart phone or workstation in case of an alarm has already been implemented. The system architecture is not limited to these two means of notification and therefore allows the notification through the retirement home’s nurse call system which, however, might lead to additional costs for interfacing.
Whereas most projects are developed without a test in real environments or in so called living labs, this system for unobtrusive fall detection was integrated in one nursing home and one retirement home with different technical equipment. These tests enabled a better understanding of the functionalities needed and the restrictions imposed by those homes.
During the successful pilot trial new ideas to expand the unobtrusive fall detection system such as integrating sensors for vital parameters arose, like work presented by the authors Busch et al. [Bus11]. The realization of these ideas and further long-term tests will be performed in the near future.
Acknowledgements. The authors are deeply indebted to the AWO nursing home "Haus am Bodden" for the prototype test and to the KerVita retirement home “Am Tierpark“ for the pilot trial. Both trials had been approved by the corresponding organisations.
1467
References
[Abo12] „About The Asterisk Project,“ [Online]. Available: http://www.asterisk.org/asterisk. [last access 1 5 2012].
[Bau12] P. Bauer, T. Rodner and L. Litz, „Fitness-analysis of home automation sensors for AAL purposes - Requirements and reality,“ 5. Deutscher AAL-Kongress, 2012.
[Bru10] R. Bruns and J. Dunkel, Event-Driven Architecture - Softwarearchitekturen für ereignisgesteuerte Geschäftsprozesse, Berlin Heidelberg: Springer-Verlag, 2010.
[Bus11] B.-H. Busch, A. Kujath, H. Witthöft and R. Welge, „Preventive Emergency Detection Based on the Probabilistic Evaluation of Distributed, Embedded Sensor Networks,“ Ambient Assisted Living, p. 163, 2011.
[Che08] L. Chen, C. Nugent, M. Mulvenna, D. Finlay, X. Hong and M. Poland, „Using Event Calculus for Behaviour Reasoning and Assistance in a Smart Home,“ Smart Homes and Health Telematics, pp. 81-89, 2008.
[Etz10] O. Etzion and P. Niblett, Event Processing in Action, Greenwich: Manning, 2010.
[Eve12] „Event Stream Intelligence: Esper & NEsper,“ [Online]. Available: http://esper.codehaus.org/. [last access 1 5 2012].
[Fun07] M. Funk and A. Pierobon, Sturzprävention bei älteren Menschen: Risiken - Folgen - Maßnahmen, Stuttgart: Georg Thieme Verlag, 2007.
[Fut12] „Future Shape,“ [Online]. Available: http://www.future-shape.com/de/technologies/11. [last access 1 5 2012].
[Hea12] „Health Level Seven International,“ [Online]. Available: http://www.hl7.org/implement/standards/product_brief.cfm?product_id=33. [last access 4 5 2012].
[Kin12] „Kinect für Xbox 360 - xbox.com,“ [Online]. Available: http://www.xbox.com/de-DE/Kinect/. [last access 1 5 2012].
[Mar12] C. Marzahl, P. Penndorf, I. Bruder and M. Staemmler, Unobtrusive Fall Detection Using 3D Images of a Gaming Console: Concept and First Results, R. Wichert und B. Eberhardt, Hrsg., Berlin Heidelberg: Springer-Verlag, 2012,
1468
pp. 135-146.
[ope12] „openAAL,“ [Online]. Available: http://wiki.openaal.de/index.php/Main_Page. [last access 1 5 2012].
[Pfl12] M. Pflüger, J. Kroll and B. Steiner, „Automatic Recognition of emergencies with the Help of Optical and Acoustic Sensors,“ Ambient Assisted Living, pp. 30-41, 2012.
[Pha08] Q.-C. Pham, Y. Dhome, L. Gond and P. Sayd, „Video Monitoring of Vulnerable People in Home Environment,“ Smart Homes and Health Telematics, pp. 90-98, 2008.
[Sch11] J. Schroeder, S. Wabnik, P. W. van Hengel and S. Goetze, „Detection and Classification of Acoustic Events for In-Home Care,“ Ambient Assisted Living, p. 181, 2011.
[Sit12] G. F. Sit, C. Shen, H. Storf and C. Hofmann, „Application-Oriented Fusion and Aggregation of Sensor Data,“ Ambient Assisted Living, p. 5, 2012.
[uni12] „universAAL,“ [Online]. Available: http://universaal.org/about/about. [last access 1 5 2012].
1469
Datenmanagement und Interoperabilität im Gesundheitswesen
1470
GI Workshop Sicherheit, Datenschutz, Management und
Interoperabilität medizinischer Daten
Medizinische Daten spielen nicht nur bei der ärztlichen Untersuchung und Behandlung von Patienten, bei der Abrechnung der medizinischen Leistungen, in Krebsregistern, Biobanken und anderen Bereichen des Gesundheitswesens eine Rolle, sondern werden auch immer wichtiger in anderen Anwendungszusammenhängen: Gesunde Ernährung und Fitness werden mit Lifestyle verbunden, Systeme des Ambient Assisted Living sollen Menschen in der alternden Gesellschaft ein selbstbestimmtes Leben ermöglichen, über soziale Netzwerke findet nicht nur medizinische Beratung statt, sondern darüber organisieren sich auch Gruppen zum gemeinsamen Abnehmen der Selbsthilfegruppen für Kranke.
Die Datennutzung im Gesundheitssektor für die Forschung und eine moderne umweltbezogene Gesundheitsberichterstattung umfasst die Sammlung, Aufbereitung, Darstellung und Bewertung von Informationen über den Gesundheitszustand der Bevölkerung sowie mögliche Risikofaktoren und deren Abhängigkeiten untereinander. So liefern Gesundheitsberichte detaillierte Informationen über Krankheiten, mögliche Ursachen, Lebenslage der Betroffenen, Versorgung in medizinischen Einrichtungen und über Kosten und Finanzierung. Weitere Daten fallen in der Entwicklung neuer Medikamente und Verfahren wie bspw. in Independent Image Reviews an.
Mit der fortschreitenden Entwicklung komplexer IT-Systeme im Gesundheitswesen und der Zunahme von komplexen Datenmengen in der Gesundheitsbranche haben sich neue Herausforderungen gebildet, wenn es um die Zusammenarbeit verschiedener Teilnehmer geht und um die stufenweise Erweiterung der IT-Landschaften. Gerade die Interoperabilität von Softwaresystemen stellt im Gesundheitswesen eine große Herausforderung dar. Bestehende Standards wie DICOM oder HL7 können hierbei hilfreich sein, reichen jedoch in der Praxis oftmals nicht aus, um eine flexible Grundlage der Systemintegration zu bieten. Durch die Zunehmende Verknüpfung von Daten spielen Datenschutz- und Sicherheitsüberlegungen eine immer wichtigere Rolle.
Der Workshop richtet sich sowohl an Wissenschaftler als auch an Entwickler und Anwender, die sich mit dem Themengebiet Sicherheit, Datenschutz, Management und Interoperabilität medizinischer Daten beschäftigen. Besonders beachtet werden dabei Lösungen und Ansätze und Best-Practices, die zu erhöhter Flexibilität im Gesundheitswesen führen.
Eingereichte Beiträge wurden vom Programmkomitee sorgfältig geprüft, nach dem Peer-Review-Prinzip ausgewählt und ggf. überarbeitet, um eine hohe Qualität sicherzustellen.
1471
Organisation
Die Organisatoren des Workshops sind (in alphabetischer Reihenfolge):
Dr. Sven Abels Ascora GmbH (www.ascora.de)
Prof. Dr. Uwe Nestmann TU Berlin (www.tu-berlin.de)
Felicitas Vlachakis, Joannis Vlachakis sofd GmbH (www.sofd.de)
GI-Fachgruppe PET: Marit Hansen, Sebastian Pape, Dr. Stefan Weiss (http://fg-pet.gi.de/)
GI-Fachgruppe SECMGT: Bernhard C. Witt, Claus Stark (http://fg-secmgt.gi.de/)
GMDS-AG DGI: Prof. Dr. Bernd Blobel, Prof. Dr. Klaus Pommerening (http://www.gmds.de/fachbereiche/informatik/AGsundPGs/AG_Datenschutz_in_Gesundh/AG_Datenschutz_in_Gesundh.php)
Programmkomitee
Das Programmkomitee des Workshops besteht aus (in alphabetischer Reihenfolge):
Dr. Sven Abels, Ascora GmbH
Dr. Oliver Blankenstein, Charité Berlin
Ass. Prof. Ph.D. Bernd Blobel, Competence Center University Hospital Regensburg
Prof. Dr. Michael Clasen, Abteilung Wirtschaftsinformatik, Hochschule Hannover
Marit Hansen, Unabhängiges Landeszentrum für Datenschutz Schleswig-Holstein
Prof. Dr. Wilhelm Hasselbring, Lehrstuhl für Software Engineering, Universität Kiel
Kristijan Jakic, Normize GmbH
Dr. Thomas Lux, Competence Center eHealth Ruhr, Ruhr-Universität Bochum
Sascha Koch, InfoAnalytics AG
Dr. Jochen Kokemüller, Fraunhofer Institut für Arbeitswirtschaft und Organisation, Stuttgart
Dr. Christian Mathis, SAP AG, Walldorf
1472
Isabel Münch, Bundesamt für Sicherheit in der Informationstechnik (BSI)
Prof. Dr. Uwe Nestmann, Institut für Softwaretechnik und Theoretische Informatik, Technische Universität Berlin
Prof. Dr. Boris Otto, Kompetenzzentrum Corporate Data Quality am Institut für Wirtschaftsinformatik, Universität St. Gallen
Sebastian Pape, TU Dortmund
Prof. Dr. Andreas Polze, Betriebssysteme und Middleware, Hasso-Plattner-Institut, Universität Potsdam
Prof. Dr. Klaus Pommerening, Johannes-Gutenberg-Universität Mainz
Prof. Dr. Kai Rannenberg, Goethe-Universität Frankfurt am Main
Dr. Stefan Schulte, Institut für Informationssysteme, Technische Universität Wien
Dieter Schuller, Technische Universität Darmstadt
Hans v. Sommerfeld, DIN NIA-01-27 IT-Sicherheitsverfahren
Claus Stark, Citigroup AG
Prof. Dr. Ali Sunyaev, Wirtschaftsinformatik und Information Systems Quality, Universität zu Köln
Joannis Vlachakis, sofd GmbH
Dr. Stefan Weiss, Swiss Re, Zürich
Bernhard C. Witt, it.sec GmbH & Co. KG
Die Organisatoren danken dem Programmkomitee für die vorbildliche Unterstützung und den Autoren für die eingereichten Beiträge.
1473
