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.

Jain Socal HIMSS keynote 1805018

375 views

Published on

What if an app could guide you to better health, similar to how GPS navigation directs you to your desired destination? What if the app could use real-time information to redirect you around a disease, just as you’re rerouted to avoid traffic? What if the app could provide step-by-step directions to get you to your optimal health state, whether you’re a professional athlete or retired school teacher? We discuss how this navigational approach to healthcare could become a reality by combining emerging technology with well-established cybernetic principles.

Published in: Healthcare
  • Be the first to comment

Jain Socal HIMSS keynote 1805018

  1. 1. Ramesh Jain Institute for Future Health University of California, Irvine Jain@ics.uci.edu
  2. 2. We believe that… •  A holistic focus on lifestyle, environment, social, and medical aspects is now feasible and essential for human health. •  The translation of research should start right from the beginning. •  Working with companies and communities to make this a new model for health research is essential.
  3. 3. The fundamental drivers of health are intertwined. •  Quality of Life •  Cost •  Productive Lifespan Health Productive Lifespan Cost Quality of Life
  4. 4. Healthcare was supposed to be about people. Employers Health Providers Insurance Companies
  5. 5. But healthcare became a financial enterprise. Employers Health Providers Insurance Companies
  6. 6. Important Turning Point in Health In the mid-20th century, the primary causes of death worldwide shifted from infections to chronic conditions.
  7. 7. Evolution Continues The Economist Dec. 13, 2003
  8. 8. Changing Times. CDC’s Division of Diabetes Translation. National Diabetes Surveillance System available at http:// www.cdc.gov/diabetes/statistics Obesity (BMI ≥30 kg/m2) 1994 2000 No Data <14.0% 14.0-17.9% 18.0-21.9% 22.0-25.9% >26.0% 2008 Diabetes No Data <4.5% 4.5-5.9% 6.0-7.4% 7.5-8.9% >9.0% 1994 2000 2008
  9. 9. INFECTIOUS DISEASES ILLNESS OCCURS TEST & DIAGNOSIS CARE $3.15 Trillion •  High fixed costs (hospitals, equipment) •  Hotel model of occupancy for revenue •  Relies on intuitive medicine for diagnosis – REACTIVE MEET •  Outpatient model is better suited •  Extremely expensive to care for a patient in an infectious disease care model – THAT IS WHAT WE DO TODAY CHRONIC DISEASES Heart Disease Cancer Type 2 Diabetes Brain Diseases Lung Diseases Bone Diseases Treating Chronic Disease in an Infectious Disease Model
  10. 10. Medicine-Genetics-Environment-Lifestyle
  11. 11. The problem: What is right Vs. What is Easy.
  12. 12. Data, information, and knowledge Continue to Transform Industries. Artificial Intelligence: Converting Available Data to Useful Information.
  13. 13. Popular Episodic Approach was OK in 20th Century! v You go to a doctor because you feel sick. v The doctor asks you questions about your health status and history while observing your reactions. v The doctor examines your body and may order some tests (such as x-rays or blood work). v Based on your exam and test results, the doctor estimates your health state (disease and its severity). v A prescription and regimen are stipulated and explained. v You try to follow those instructions. v The doctor sees you periodically as needed (possibly recommending continued visits, hospital admission, or surgery followed by a stay in ICU).
  14. 14. Episodic Health Cycle: MEGI Estimate (Disease) Guidance Influence and Compliance Measure Collect all Relevant data Health: Collect Data; Derive Actionable Information; Provide Guidance at Right Time, in Right Situation, in a right way.
  15. 15. 2017: New Cars(17.85M) and New Humans(3.98M) q  Most cars send more than 25 gigabytes of data/hour (130TB per year) to the cloud. q  The data is used for longitudinal studies as well as for population and social studies. q  Most people collect almost zero data. q  The data collected is not usable. Who do we value more: cars or humans?
  16. 16. A New Perspective for Health Health Personal Models and Cybernetics. Focus on:
  17. 17. How did you get to your destination?
  18. 18. Turn-by-turn directions felt so advanced! Maps Points of Interest Starting Point Destination Mapping System Just 10 years ago!
  19. 19. Cybernetics: Feedback revolutionizes system design Modern control theory and AI came from Cybernetics.
  20. 20. Basic Systems Theory Where x, u, and y are state, input, and output vectors; A, B, C, and D are appropriate matrices for transformation. Observability: Determine state using measurements. Controllability: Get to desirable state.
  21. 21. Magic Happened: Can you drive without a navigation system? Maps Points of Interest Current State : GPS Current Traffic: Waze Destination Cybernetic loop Navigation system
  22. 22. Health Navigator • What if an app could guide you to better health, similar to how GPS navigation directs you to your desired destination? • What if the app could use real-time information to redirect you around a disease, just as you’re rerouted to avoid traffic? • What if the app could provide step-by-step directions to get you to your optimal health state, whether you’re a professional athlete or retired school teacher?
  23. 23. Measure Collect all data Perpetual Health Guidance Guidance Estimate (Health State) Influence and Compliance How do we close the loop? Inspiration: Navigation Systems.
  24. 24. Key Components • Data Source Selection and Ingestion • Building Personal Models: • Short term (Maybe for specific health situation) • Long term • Health State estimation • Recommendations • Influence and Adherence
  25. 25. Key Technology Components •  Measure •  Multimodal semantics, Synchronization •  Semantic compression •  Event Recognition in multimodal signals •  Event mining: learning and causality, Associative rule mining •  Estimate •  Learning for diagnostics •  Causality from event characteristics •  Guide •  Recommendations in context •  Situational knowledge in recommendation engines •  Optimization of health and quality of life •  Influence •  Gamification •  Communicating right information at right time in a right way
  26. 26. Personal Health Navigator MEASURE: COLLECT DATA STREAMS EVENTSHOP ENVIRONMENT KNOWLEDGE ESTIMATION (HEALTH STATE) MICRO-REPORTS LOCAL MEDICAL DEMOGRAPHICS ENVIRONMENT REGULATIONS DATA STREAM PROCESSING EVENT DETECTION EVENT MINING PERSONICLE ‘OMICS’ INFLUENCE AND ADHERENCE PERSONAL MODEL GUIDANCE HEALTH CARE PROVIDER supervisory system
  27. 27.   –Daniel Kahneman, Awarded 2002 Nobel Memorial Prize in Economic Sciences   “Quantifying information about time usage and its frequency, as well as stress level, pleasure, and other affective states of each individual, is potentially useful for health research” Our system automatically builds Personicle on Android. Recording Daily Activities: Personicle
  28. 28. Automatic Creation of Personicle
  29. 29. Unobtrusive collection of Personicle
  30. 30. Pulling data from ALL sources in Personicle
  31. 31. Food is an important MISSING LINK No good technique for logging food diary Food Recognition Nutrition in Food People’s preferences: what when, where, weather, company, … Grand Challenge: Create a Detailed Personal Food Log.
  32. 32. Personal Model Builder ‘Omics’ Personal Model Environment Knowledge Personicle AI + Learning •  Propensity to diseases •  Rule Based Health Behavior EventShop for Situation Recognition and geo-spatial event analysis. Could be for Whole Life Recent Past After a particular health Condition (such as heart attack)
  33. 33. Pattern Mining High-level Pattern Formulation Pattern Query … Data Streams Event Streams Semi-interval Event Sequences Data-Driven Analysis Hypothesis-Driven Analysis While air pressure is high, pollution starts increasing gradually, within T time units asthma outbreak happens. ((pollution_inc_steadily ;ωT asthma_outbreak ) || airpressure_stayshigh) Example: Interactive Visualization Interactive Event Mining Bottom Up Top Down
  34. 34. Event Stream Mining for Rule-Based Actionable Insights Spicy food and 2 glasses of wine result in sleepless nights. Personicle Food Stream t1 t2 t3 t4 t5 Warn him when he is at an Indian restaurant. Sleep
  35. 35. Personal Health Navigator: Diabetes MEASURE: COLLECT DATA STREAMS EVENTSHOP ENVIRONMENT KNOWLEDGE ESTIMATION (HEALTH STATE) MICRO-REPORTS LOCAL MEDICAL DEMOGRAPHICS ENVIRONMENT REGULATIONS DATA STREAM PROCESSING EVENT DETECTION EVENT MINING PERSONICLE ‘OMICS’ INFLUENCE AND ADHERENCE PERSONAL MODEL GUIDANCE HEALTH CARE PROVIDER supervisory system Activity Food Medication Glucose Level Activity Food Medication Glucose Level in last 3 hours Sensitivity to different foods. Sensitivity to medication. Activity Checks. Glucose Level Checks.
  36. 36. PHN and •  Guide to better health, better quality of life. •  Uses real-time information to redirect you around a disease, as if a doctor is with you 24/7. •  Provides step-by-step directions to get you to your optimal health state.
  37. 37. PHN and Employers • Employee satisfaction and wellbeing. • Potential reduction in cost of healthcare. • Productive and predictive relations with employees.
  38. 38. PHN and Insurance Companies •  Personalized Protection and Incentives •  Data and Insights for Value-based reimbursements •  Preventive, Predictive, Personalized •  Use of SDOH in Insurance
  39. 39. PHN and Health Providers •  Patient Engagement for Reducing Readmissions •  Patient Relationship Management •  Precise information about patient conditions •  Perpetual care rather than episodic subjective information
  40. 40. Health Navigator Your food and activity scores for the last 3 days are on low side. Health Navigator in Action My recommendations for you are.
  41. 41. Our Goal • A Personal Health Navigator as an early embodiment of a novel approach for Future Health. • Build a community around Future Health that will bring medical science and all relevant technologies to advance Future Health. • Identifying realistic approaches to work with like-minded groups and individuals to realize Future Health.
  42. 42. Thanks. jain@ics.uci.edu

×