Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Assessment Test Framework for Collecting and Evaluating Fall - Related Data using Mobile Devices

2,034 views

Published on

  • Be the first to comment

  • Be the first to like this

Assessment Test Framework for Collecting and Evaluating Fall - Related Data using Mobile Devices

  1. 1. Assessment Test Framework for Collecting and Evaluating Fall-Related Data Using Mobile Devices DI Stefan Almer July 11th, 2012Graz University of Technology Central European Institute of TechnologyInstitute for Information Systems and Computer Media Institute for Rehabilitation and Ambient Assisted Living TechnologiesUniv.-Doz. Dipl.-Ing. Dr.techn. Martin Ebner Dipl.-Ing. Dr.techn. Johannes OberzaucherDipl.-Ing. Dr.techn. Josef Kolbitsch
  2. 2. Assessment Test Framework for Collecting and Evaluating Fall-Related Data Using Mobile Devices 2 Agenda• Introduction• Mobile Devices for Fall Detection• Assessment Test Framework• Mobile Device Client• Evaluation• Summary Stefan Almer July 11th, 2012
  3. 3. Assessment Test Framework for Collecting and Evaluating Fall-Related Data Using Mobile Devices 3 Introduction• Motivated by the demographic trend [van den Broek et al., 2009] • average age will increase • impact on healthcare systems, retirement plans • more people will need assistance or support• Falls and fall-related injuries• Mobile Devices • device of the future: “the steady companion” Stefan Almer July 11th, 2012
  4. 4. Assessment Test Framework for Collecting and Evaluating Fall-Related Data Using Mobile Devices 4 Fall Prevention [Todd and Skelton, 2004; WHO, 2007; Tremblay Jr. and Barber, 2005; LeMier et al., 2002; BRAID, 2010]• Common methods • assessment tests • adjustment of environment and walking aids • gait analysis • education • exercise/training• Differ in usage based on context Stefan Almer July 11th, 2012
  5. 5. Assessment Test Framework for Collecting and Evaluating Fall-Related Data Using Mobile Devices 5 Fall Detection• Five phases of a fall 1) Activity 2) Hard- 3) Free-fall 4) Impact 5) Optional of daily living predictable event recovery Fig. 1: Fall Phases [Abbate et al., 2010]• Classification of fall detection methods [Yu, 2008] • wearable device / camera-based / ambience device• Important to differentiate between a fall and activities of daily living Stefan Almer July 11th, 2012
  6. 6. Assessment Test Framework for Collecting and Evaluating Fall-Related Data Using Mobile Devices 6Mobile Devices for Fall Detection [Columbus, 2011; Noury et al., 2007; Kangas et al., 2007]• Classic approach • “Individual” devices and sensors• New approach: Mobile Devices • equipped with required hardware: accelerometer • software capabilities to read acceleration data• Method: measure body acceleration • fall has higher acceleration • acceleration threshold to determine fall • problem: position of sensor Stefan Almer July 11th, 2012
  7. 7. Assessment Test Framework for Collecting and Evaluating Fall-Related Data Using Mobile Devices 7 Test Framework• Fall detection is complex • many parameters • no general fall detection algorithm• Aim of the framework • collecting fall-related data • easily set up of tests settings • integration with different systems and devices Stefan Almer July 11th, 2012
  8. 8. Assessment Test Framework for Collecting and Evaluating Fall-Related Data Using Mobile Devices 8 Test Framework (cont.)• Assessment test-based approach Motion Data Sensors User Test Device Position Test Type Sample Rate Fig. 2: Test Properties and Device Relation• Provides Interface (API) • well defined interface • integrate various devices with different sensors • stored data can be accessed later Stefan Almer July 11th, 2012
  9. 9. Assessment Test Framework for Collecting and Evaluating Fall-Related Data Using Mobile Devices 9 Framework Architecture [Helic, 2008] • Based on 3-tier architecture Data Tier Application Tier Client Tier Interface HTTP Database JDBC Java (Web service) Client (Browser/ iOS) Static Content HTTP (Backend)Fig. 3: Framework Architecture Stefan Almer July 11th, 2012
  10. 10. Assessment Test Framework for Collecting and Evaluating Fall-Related Data Using Mobile Devices 10 Proof-of-Concept• Integration of different devices• Mobile Device Client • demonstrates functionality of the framework • shows capabilities and sensor accuracy• Developed on iOS Platform • uses the possibility to receive high- rate continuous motion data Fig. 4: Apple iOS Client Stefan Almer July 11th, 2012
  11. 11. Assessment Test Framework for Collecting and Evaluating Fall-Related Data Using Mobile Devices 11 Evaluation• 3 clinical mobility tests were performed [Podsiadlo and Richardson, 1991; Whitney et al., 2005; Lewis and Shaw, 2005] • “Sit-to-Stand 5”, “Timed Up and Go”, “2-Minute-Walk”• iPhone 4, worn on hip height, 50Hz sample rate• Tests analyzed afterwards Fig. 5: User wears Device while performing test Stefan Almer July 11th, 2012
  12. 12. Assessment Test Framework for Collecting and Evaluating Fall-Related Data Using Mobile Devices 12 Evaluation (cont.)• Movement analysis of recorded gait data • sensor accurate enough to perform fall detection• Data analysis and future extraction (peak detection, knowledge methods, statistical analysis) 1.4 3 m straight walk turn phase 1.2 3 m straight walk sit phase sit down phase sit phase 1 0.8SVM [g] 0.6 0.4 0.2 0 0 100 200 300 400 500 600 700 800 900 1000 datapoints [n] - movement data recorded with 50 Hz Fig.6: Timed Up and Go Test (Individual Phases) Stefan Almer July 11th, 2012
  13. 13. Assessment Test Framework for Collecting and Evaluating Fall-Related Data Using Mobile Devices 13 Summary• One framework for various devices • support for different sensors • data collected or later analysis • Web-Service API• Backend for administrative tasks• Proof-of-Concept • Mobile Device Client (iOS platform) • hardware well-suited for fall detection Stefan Almer July 11th, 2012
  14. 14. Assessment Test Framework for Collecting and Evaluating Fall-Related Data Using Mobile Devices Stefan Almer stefan@almer.cc @stefalmer Slides available at: http://elearningblog.tugraz.atGraz University of Technology Central European Institute of TechnologyInstitute for Information Systems and Computer Media Institute for Rehabilitation and Ambient Assisted Living Technologieshttp://www.iicm.tugraz.at http://www.ceit.at
  15. 15. [Abbate et al., 2010] Abbate, S., Avvenuti, M., Corsini, P., Vecchio, A., and Light, J., 2010. Monitoring of Human Movements for Fall Detection and Activities Recognition inElderly Care using Wireless Sensor Network: A Survey. InTech. ISBN 9789533073217.[BRAID, 2010] BRAID, 2010. Falls Prevention. http://capsil.braidproject.eu/index.php?title=Falls_Prevention&oldid=6250. Last accessed November 2, 2011.[Columbus, 2011] Columbus, L., 2011. Gartner Releases Hype Cycle for Networking and Communications, 2011. http://softwarestrategiesblog.com/2011/08/27/gartner-releases-hype-cycle-for-networking-and-communications-2011/. Last accessed October 12, 2011.[Helic, 2008] Helic, D., 2008. Software Architecture VO/KU. http://coronet.iicm.tugraz.at/sa/s5/sa_styles1.html. Last accessed October 6, 2011.[Kangas et al., 2007] Kangas, M., Konttila, A., Winblad, I., Jämsa, T., 2007. Determination of Simple Thresholds for Accelerometry-Based Parameters for Fall Detection. In Proc.of the 29th Annual International Conference of the Engineering in Medicine and Biology Society, volume 2007, pages 1367–1370. doi:10.1109/IEMBS.2007.4352552.[LeMier et al., 2002] LeMier, M., Silver, I., Bowe, C., 2002. Falls Among Older Adults: Strategies for Prevention. Technical Report, Washington State Department of Health.http://www.doh.wa.gov/hsqa/emstrauma/injury/pubs/FallsAmongOlderAdults.pdf. Last accessed November 2, 2011.[Lewis and Shaw, 2005] Lewis, C., Shaw, K., 2005. Benefits of the 2-Minute Walk Test. Physical Therapy & Rehab Medicine, 16(16). http://physical-therapy.advanceweb.com/Article/Benefits-of-the-2-Minute-Walk-Test.aspx. Last accessed October 1, 2011.[Noury et al., 2007] Noury, N., Fleury, A., Rumeau, P., Bourke, A.K., Ó Laighin, G., Rialle V., Lundy, J.E., 2007. Fall Detection - Principles and Methods. In Proc. of the 29thAnnual International Conference of the Engineering in Medicine and Biology Society, pages 1663–1666. IEEE. doi:10.1109/IEMBS.2007.4352627.[Podsiadlo and Richardson, 1991] Podsiadlo, D., Richardson, S., 1991. The Timed ’Up & Go’: A Test of Basic Functional Mobility for Frail Elderly Persons. American GeriatricsSociety, 39(2), pages 142–148. ISSN 00028614. http://www.ncbi.nlm.nih.gov/pubmed/1991946. Last accessed October 1, 2011.[Todd and Skelton, 2004] Todd, C., Skelton, D., 2004. What are the main risk factors for falls among older people and what are the most effective interventions to preventthese falls? Technical Report, WHO Regional Office for Europe. http://www.euro.who.int/document/E82552.pdf. Last accessed November 2, 2011.[Tremblay Jr. and Barber, 2005] Tremblay Jr., K.R., Barber, C.E., 2005. Preventing Falls in the Elderly. http://www.ext.colostate.edu/pubs/consumer/10242.pdf. Last accessedNovember 2, 2011.[van den Broek et al., 2009] van den Broek, G., Cavallo, F., Odetti, L., Wehrmann, C., 2009. Ambient Assisted Living Roadmap. http://www.aaliance.eu/public/documents/aaliance-roadmap/aaliance-aal-roadmap.pdf. Last accessed October 20, 2011.[WHO, 2007] WHO, 2007. WHO Global Report on Falls Prevention in Older Age. http://www.who.int/ageing/publications/Falls_prevention7March.pdf. Last accessedNovember 2, 2011.[Whitney et al., 2005] Whitney, S., Wrisley, D.M., Marchetti, G.F., Gee, M.A., Redfern, S.M., Furman J.M., 2005. Clinical Measurement of Sit-To-Stand Performance in PeopleWith Balance Disorders: Validity of Data for the Five-Times Sit-To-Stand Test. Physical Therapy, 85(10), pages 1034–1045. ISSN 00319023. http://www.ncbi.nlm.nih.gov/pubmed/16180952. Last accessed October 1, 2011.[Yu, 2008] Yu, X., 2008. Approaches and Principles of Fall Detection for Elderly and Patient. In Proc. of the 10th International Conference on e-Health Networking,Applications and Services, pages 42–47. IEEE. doi:10.1109/HEALTH.2008.4600107.

×