Sensors and Big Data for Health and Well-being
Download as PPTX, PDF1 like1,410 views
This document discusses the potential for sensors and big data in health and well-being. It highlights Google Flu Trends, which was found to be wrong in its predictions 100 times out of 108 weeks. It also discusses the challenges of big data in health, including issues with data variety, veracity (accuracy of sensors, manual entries, and missing data), and privacy concerns. The document promotes tracking health metrics like exercise, weight, and calorie intake but notes the ecosystem is still facing challenges to realize its full potential.
Sensors and Big Data for Health and Well-being
- 1. Sensors and Big Data for Health and Well-being Alasdair J G Gray http://www.alasdairjggray.co.uk @gray_alasdair A.J.G.Gray@hw.ac.uk
- 2. Google Flu Predictions 20 June 2014SICSA THAW Workshop 1
- 3. Google Flu Predictions 20 June 2014SICSA THAW Workshop 2 http://www.google.org/flutrends/intl/en_us/about/how.html http://www.bbc.co.uk/news/business-27683581 Google Flu Prediction wrong 100 times in a 108 week study!
- 4. Big Data 20 June 2014SICSA THAW Workshop 3 http://www.datasciencecentral.com/profiles/blogs/data-veracity
- 5. Data Science: Extracting Value 20 June 2014SICSA THAW Workshop 4http://senderocorp.com/images/uploads/bigdata_v9.png Volume Velocity Variety Veracity Value VisualizationAnalytics Big Data Technology
- 6. Warning! 20 June 2014SICSA THAW Workshop http://businessoverbroadway.com/wp-content/uploads/2013/05/veracity.png 5
- 7. Healthy Living 20 June 2014SICSA THAW Workshop 6 http://weightlosswithexercise.com/images/WeightLoss-45119_11.jpg http://m.cdc.gov/en/HealthSafetyTopics/Healthy Living/HealthyWeight/BalancingCalories
- 8. Exercise Tracking 20 June 2014SICSA THAW Workshop 7 http://www.fitbit.com/ http://www.nike.com/gb/en_gb/c/n ikeplus-fuelband
- 9. Intelligent Scales 20 June 2014SICSA THAW Workshop 8
- 10. Calorie Counting 20 June 2014SICSA THAW Workshop 9
- 11. Dashboard Ecosystem 10 20 June 2014SICSA THAW Workshop
- 12. Challenges: Variety 11 20 June 2014SICSA THAW Workshop
- 13. Challenges: Data Veracity Messy Data Best guess at sleep activity Accuracy of sensors Manual entry Missing Data Forgot to put my fitbit on Didn’t enter my lunch Skipped weighing myself Dead battery 20 June 2014SICSA THAW Workshop 12
- 14. Challenges: Privacy 20 June 2014SICSA THAW Workshop 13
- 15. Challenges: Privacy 14 20 June 2014SICSA THAW Workshop
- 16. Thank you http://www.alasdairjggray.co.uk A.J.G.Gray@hw.ac.uk @gray_alasdair Acknowledgement David McGookin 20 June 2014SICSA THAW Workshop 15
Editor's Notes
- #3: Animation of predicted flu occurrences by State Predication based on 45 Google search terms
- #4: Google results published in Nature in 2008 Study by Northeastern and Havard Universities (published in Science) Prediction too high, sometimes double Previous week’s occurrences more accurate Google tweak their algorithms – increases uncertainty Accuracy far greater before autocomplete introduced in 2009
- #6: Deriving value from the data
- #8: We are what we eat…and do Eating Exercise Weight
- #9: What about other exercise? Weights, pilates, circuits …
- #12: Incentive: Gameification/competition
- #13: All different!
- #14: Manual entry: Fine for foods Provenance of data, what sensor generated the data?
- #15: Who can see my data? How is my data used?
- #16: As much about educating the public Public relations
- #17: Digital Health in a Connected Hospital https://www.innovateuk.org/competition-display-page/-/asset_publisher/RqEt2AKmEBhi/content/digital-health-in-a-connected-hospital?p_p_auth=rNL7u7Xw