Diese Präsentation wurde erfolgreich gemeldet.
Wir verwenden Ihre LinkedIn Profilangaben und Informationen zu Ihren Aktivitäten, um Anzeigen zu personalisieren und Ihnen relevantere Inhalte anzuzeigen. Sie können Ihre Anzeigeneinstellungen jederzeit ändern.

Using The Internet of Things for Population Health Management - StampedeCon 2016

745 Aufrufe

Veröffentlicht am

The Internet of (Human) Things is just beginning to take shape. The human body is an inexhaustible source of data about personal health, and the healthcare industry is just beginning to scratch the surface of the potential insights and value that will come from that data.  While much of healthcare traditionally focuses on the episodic delivery of services, the Affordable Care Act is pushing healthcare providers, payers, and self-funded employer groups to look at ways to proactively encourage healthy behaviors. Providing personal health devices as a way to promote individual health is one way that healthcare is beginning to take advantage of IoT technologies.  This session provides insight into how IoT is being leveraged in population health management through a solution jointly delivered by Amitech Solutions and Big Cloud Analytics.  Attendees will learn how Hadoop is being used to gather personal device from various vendors, integrate and analyze that information, differentiate trends across regional and cultural diversity, and provide personal recommendations and insights into health risks. This session presents one important way the healthcare industry is leveraging IoT.

Veröffentlicht in: Technologie
  • Als Erste(r) kommentieren

Using The Internet of Things for Population Health Management - StampedeCon 2016

  1. 1. IoT for Population Health Paul Boal, Amitech Solutions, @paulboal StampedeCon 2016 1
  2. 2. Topics •IoT Across Healthcare •IoT Technologies •An IoT and Population Health Example •Wearable Fitness Devices in Apache NiFi 2
  3. 3. IoT in Healthcare 3 http://mynutratek.com/welcome/health-providers-healthcare-plans/
  4. 4. IoT in Healthcare • Smart beds • Smart pumps • Robots • Smart Monitors • Smart Soap Dispensers 4 Clinical Monitoring R/T Location Systems At Home Monitoring Personal Fitness Ingestible Devices
  5. 5. IoT in Healthcare 5 Clinical Monitoring R/T Location Systems At Home Monitoring Personal Fitness Ingestible Devices • Assets • Inventory • Patients • Visitors • Clinicians
  6. 6. IoT in Healthcare 6 Clinical Monitoring R/T Location Systems At Home Monitoring Personal Fitness Ingestible Devices
  7. 7. IoT in Healthcare • Clinical monitoring • Real-time location systems • EHR transactions • At home monitoring 7
  8. 8. IoT in Healthcare • Clinical monitoring • Real-time location systems • EHR transactions • At home monitoring 8 https://www.researchgate.net/figure/272386643_fig2_Figure-3-Left-demo- set-up-with-belt-prototype-worn-by-a-12-week-old-baby-Right
  9. 9. IoT in Healthcare • Heart rate • Sleep • Perspiration • Temperature • Activity 9 Clinical Monitoring R/T Location Systems At Home Monitoring Personal Fitness Ingestible Devices
  10. 10. IoT in Healthcare • Chemistry Sensors • Medicine Dispenser • Cameras 10 Clinical Monitoring R/T Location Systems At Home Monitoring Personal Fitness Ingestible Devices
  11. 11. IoT Data Processing •Transactional vs Micro-batch •Development Environment •Connectors and Processors •Durability •Out of Order Processing •Scalability 11
  12. 12. IoT Data Processing • UC Berkley AMPLab • Databricks • Airbnb • Autodesk • Concur • eBay • MyFitnessPal • NASA JPL • Opentable • University of MO 12
  13. 13. IoT Data Processing • Twitter • Groupon • The Weather Channel • Yahoo! • WebMD • Spotify • Klout • NaviSite • PARC • Wayfair • Cerner • Yelp 13
  14. 14. IoT Data Processing • NSA • Dar Group • MD Anderson • Xavient Information System • Lowes • Schlumberger 14
  15. 15. IoT Data Processing • LinkedIn • Intuit • MobileAware • Project Florida • Happy Pancake • TiVo • Uber • Netflix 15
  16. 16. IoT Data Processing • dataArtisans • Capital One • Ericsson • king.com (CandyCrush) • Portugal Telecom • ResearchGate • Okkam SRL • Google Gloud Dataflow 16
  17. 17. Amitech Solutions and Big Cloud Analytics • Collects millions of data points from thousands of deployed wearable devices that capture 50+ biometric data points • Computes advanced population health management analytics, scores and coefficients • Manages population’s wellness • Groups cohorts by sleep, activity level and resting heart rate • Alerts and triggers for conditions such as device abandonment, elevated resting heart rate and others • Guides users to better health with event-triggered messaging 17
  18. 18. From Accelerometers to Cash 18
  19. 19. Future Data Ingest Architecture 19
  20. 20. Introduction to NiFi 20 Flow File Processor Connections Flow Controller
  21. 21. Introduction to NiFi 21 Data Transformation Routing and Mediation Database Access Attribute Extraction System Interaction Data Ingestion Data Egress / Sending Splitting Aggregation HTTP Flow File Processor Connections Flow Controller
  22. 22. Introduction to NiFi 22 Flow File Processor Connections Flow Controller
  23. 23. Introduction to NiFi 23 Flow File Processor Connections Flow Controller
  24. 24. From POJO to NiFi Processor 1. Extend AbstractProcessor 2. Configure pom.xml for NiFi 3. Build and Deploy 24
  25. 25. Code Walk Through 25 Properties and Relationships Initialize Read Config Process Request Return Results POM and Metadata
  26. 26. Code Walk Through 26 Properties and Relationships Initialize Read Config Process Request Return Results POM and Metadata
  27. 27. Code Walk Through 27 Properties and Relationships Initialize Read Config Process Request Return Results POM and Metadata
  28. 28. Code Walk Through 28 Properties and Relationships Initialize Read Config Process Request Return Results POM and Metadata
  29. 29. Code Walk Through 29 Properties and Relationships Initialize Read Config Process Request Return Results POM and Metadata
  30. 30. Code Walk Through • pom.xml configuration • Processor file for nar metadata 30 com.bca.etl.nifi.processors.WearableDeviceProcessor Properties and Relationships Initialize Read Config Process Request Return Results POM and Metadata
  31. 31. NiFi Configuration 31 Flow Controller Extract Properties Processor Config
  32. 32. NiFi Configuration 32 bca.device=Garmin bca.username=me@me.com bca.password=XXX bca.startdate=2016-04-01 bca.enddate=2016-04-02 Flow Controller Extract Properties Processor Config
  33. 33. NiFi Configuration 33 Properties match the properties in the WearableDeviceProcessor class Flow Controller Extract Properties Processor Config
  34. 34. NiFi Output Data from vendor API output • Write the JSON to a file • Write to NoSQL DB • Write to Hbase • Make a REST call with this payload • Send to Kafka queue • Extract with JSON Path • Process with Spark or Storm
  35. 35. NiFi Results • Easy to setup and run locally for development. • From existing code to NiFi processor took less than a day (including making several dump mistakes along the way). • Framework will enable scale. • Lots of flexibility in where the data goes next.
  36. 36. Summary •IoT will save the healthcare industry •It doesn’t have be like Y2K •Go try something other than Twitter! 36
  37. 37. References • https://www.cbinsights.com/blog/iot-healthcare-market-map-company-list/ • http://www.cakesolutions.net/teamblogs/comparison-of-apache-stream-processing- frameworks-part-1 • http://www.kdnuggets.com/2016/03/top-big-data-processing-frameworks.html • http://events.linuxfoundation.org/sites/events/files/slides/JoeWitt_apr2015_apachecon_be tteranalytics-betterdataflow_v1.pdf • http://www.slideshare.net/JenAman/airstream-spark-streaming-at-airbnb • http://www.slideshare.net/edvorkin/learning-stream-processing-with-apache-storm • http://www.slideshare.net/HadoopSummit/from-zero-to-data-flow-in-hours-with-apache- nifi-64032731 • https://qconsf.com/system/files/presentation-slides/qconsf-2015- stream_processing_in_uber.pdf • https://techblog.king.com/rbea-scalable-real-time-analytics-king/ • http://www.zdnet.com/article/nsa-partners-with-apache-to-release-open-source-data- traffic-program/ • https://samza.apache.org/learn/documentation/0.10/comparisons/storm.html 37
  38. 38. Paul Boal paul.boal@amitechsolutions.com @paulboal Paul has been architecting healthcare analytics solutions for 15 years, implementing a range of technologies from traditional data warehouses to Hadoop-based solutions, advanced analytics, and real-time clinical data integration. Paul is now a practice lead with Amitech Solutions focused on delivering big data solutions for healthcare, including a healthcare IoT platform that leverages data from personal wearable devices for population health management. 38

×