Introduction to Biomedical Informatics


Published on

Lecture given to University of Iowa College of Medicine second year medical students in October 2015.

Published in: Health & Medicine, Business
  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Introduction to Biomedical Informatics

  1. 1. Introduction To Biomedical Informatics Michael P. D'Alessandro, M.D. University of Iowa College of Medicine / University of Iowa Children's Hospital @pedsimaging
  2. 2. Disclosures • Nothing to disclose
  3. 3. Goals + Objectives • Goal 1: Develop an understanding of health care system organization, policy and financing. • Objectives: • a. Analyze and evaluate the organization, financing, and performance of the US healthcare system, encompassing models of healthcare delivery as well as methods of financing and payment, including private and government health insurance programs. • b. Analyze and evaluate the impact of the US healthcare system on health outcomes at the population and individual patient level, and also its impact on health care providers. • c. Analyze and evaluate the effects of governmental and private-sector health policy on the medical system, biomedical research, and patient care, including the Affordable Care Act on delivery of and private/public funding of health care. • e. Characterize specific systems of care unique to particular patient populations and settings. • Goal 2: Understand how programs and methods of health care quality and safety contribute to the delivery of health care. • Objectives: • b. Assess the efficacy of programs and projects to maintain and improve patient safety in patient evaluations and treatments and in a variety of
  4. 4. The Good News • You will hear what is wrong with medicine many times • Problems • Informatics is what is going to be right about medicine • Solutions
  5. 5. Profession The Greatest Adventure in the World "I love what I do" Stay connected to world ~ Stay connected to family ~ Golden Rule ~ 80 percent of success is
  6. 6. About Me • Started programming in 1977 • B.S. in Computer Science in 1985 • Educational Informatics Lab's vision since 1989 • To improve patients' care, outcomes and lives; • By changing physician's knowledge, attitudes, and behaviors; • Through the creation and evaluation of tools, techniques, and procedures that shift learning from the classroom and lecture hall to the point-of-care and document + preserve this learning to create a personalized learning environment / knowledge management / e-memory system for every physician • Strong record of innovation in medical education • Resulting in grants, peer-reviewed publications, awards • Currently serve 2 million learners / year ~ highest impact factors on Internet
  7. 7. Disclaimer / Where I'm Coming From • I am a techno-optimist + techno- pessimist • Technology is neutral – we decide whether it does good or bad • I am happy for what informatics does for the top 5% • I am more interested + excited what it does for the other 95% in this country + in the world • I believe in the supremacy of the patient / doctor relationship • I think that will remain at center of medical
  8. 8. What is Biomedical Informatics? "Biomedical informatics…is the interdisciplinary, scientific field that studies and pursues the effective uses of biomedical data, information, and knowledge for scientific inquiry, problem-solving, and decision making motivated by efforts to improve human health." - American Medical Informatics Association •The application of computers in medicine, specifically in • Patient care (clinical informatics) • Education (educational informatics) • Basic research (bioinformatics)
  9. 9. Why is Biomedical Informatics Important? • In the 21st century, informatics is everyone's interface to health care and the health care system • Purchasing care ( • Accessing care • Reviewing your care (personal health record) • Understanding your care
  10. 10. Why is Biomedical Informatics Important? • As medicine becomes digital in the 21st century, a working knowledge of and basic literacy in informatics is a medical core competency, with its own place in physician's toolkits alongside anatomy + physiology, pathology, the physical exam and diagnostic reasoning
  11. 11. Overview • Biomedical Informatics • Past - Clinical Informatics • Electronic medical record • Present - Educational Informatics • Providers • Patients • Future - Bioinformatics • Quantified Self • Big Data • Personalized / Precision / Genomic Medicine • Tricorders • Artificial intelligence • Singularity • Assignments - Hands on Biomedical Informatics • To Learn More
  12. 12. What is Powering Informatics? Moore's Law • Gordon Moore - 1965 • Number of components that can be put on a computer chip doubles every 18 months while price remains same • Means computer power doubles every 18 months
  13. 13. What is Powering Informatics? Moore's Law • Technology in • 1963 - Born • Mainframe computers • 1981 – Start college / 1985 - Start medical school • Personal computers + networks + social networking • 2015 - Today • Ubiquitous computing • This is what happened computationally in my half century • What will happen in
  14. 14. The Past • Clinical Informatics
  15. 15. Electronic Medical Record • History • 1960's-1990's - Rare, institution-specific, handmade (HELP) • Medicine has been the last industry to go digital - Why? • 2000's-today - Universal, institution-independent, commercial • Components / Plumbing • Laboratory system • Documentation system • Computerized physician order entry • Communications system • User interface (requires training) • Patient portal • @UIHealthCare • Good news - you have the best – EPIC • Bad news - it is universally disliked • …still seems to be a lot of paper being carried around…
  16. 16. Electronic Medical Record The Medical-Industrial Complex • Monopolistic industry which hampers innovation + improvement • Moving glacially towards standards compliance + interoperability • EPIC Care Everywhere • User interface is abysmal "EMR vendors must realize that the human-computer interface in their systems is more than a marketing differentiator. It is instead, like cockpit controls, a critical component in a critical system that must be designed to be undemanding of attention and cognition. Anything less will create new cemetery plots, as surely as poor cockpit controls create smoking holes." - John Sotos MD, cardiologist + flight surgeon • Leads to rise of medical scribe industry • 20% of physicians use scribes, going from 15,000 today to 100,000 in 2020, little regulation + training
  17. 17. Old Computing Technology Doesn't Die… • …It doesn't even fade away... • MUMPS programming language created in 1966 • It is the core of EPIC today
  18. 18. Electronic Medical Record Enables • Billing • Disease correction - Consistency of care / improved patient outcomes • Disease prevention - health promotion • Answer to the question - what do you gain from going paperless?
  19. 19. Electronic Medical Record Supports • Provider-centric model of health care • Current paradigm of medical practice for providers and patients - disease correction
  20. 20. Electronic Medical Record Bottom Line • We are all sharing our information on patients • We are all wiser about patients • The patient is getting better care by some metrics • …but we have a long way to go • The road to hell is paved with good intentions • Questions now being asked • Are EMRs helping us save money / reduce errors / enhance research?
  21. 21. Shameless Radiology Plug • D'Alessandro's Theorem "The quality, usefulness and pertinence of the radiologist's report is directly proportional to quality of the history provided by the clinician" • D'Alessandro's Corollary: "The radiologist cannot answer a question the clinician has not asked!"
  22. 22. Electronic Medical Record State of the Art "Clinicians are overwhelmed by the current demands of Meaningful Use, hundreds of quality measures, population health, care management, and patient/family engagement. All of these are good ideas individually but the sum of their requirements overwhelms providers. In an era when we’re trying to control costs, adding more clinical FTEs to spread the work over a large team is not possible. The end result is that providers spend hours each night catching up on the day’s documentation and are demanding better tools /automation to reduce their strain. However, current EHRs are in an early stage of development and are data capture tools rather than customer relationship management systems." - John Halamka, M.D., 2015
  23. 23. Electronic Medical Record State of the Art
  24. 24. Electronic Medical Record State of the Art • Makes you susceptible to massive HIPAA violations • Makes you susceptible to cyber attacks "You have zero privacy anyway. Get over it." - Scott McNealy, Sun Microsystems
  25. 25. Electronic Medical Record The Dark Side - Look Up, Not Down! • Doctors often spend too much time looking at their computers and not enough time looking at and listening to their patients. • No one went into medicine to interact with an EMR "The i-patient is getting wonderful care, while the real patient is saying 'Where is everyone?'"
  26. 26. Electronic Medical Record The Dark Side • Residents spent 51% of their time interacting with computers + 9% of their time in direct contact with patients - Acad Med. 2016;91-827-832
  27. 27. Hands On • Clinical Informatics • Become familiar with patient-facing software in regards to you or your family's health • Login to your personal health record on EPIC MyChart • Walk a mile in the patient's shoes • Help you understand the benefits and challenges facing patients who do this
  28. 28. The Present • Educational Informatics for Providers • Achieving + Maintaining Mastery in Medicine • Tools, Techniques and Procedures to Create a Framework to Hang Your Learning On
  29. 29. Why We Need to Be Learning Every Day • The key ethical principle that drives the practice of medicine is "Primum non nocere" (first, do no harm). • Therefore in medicine we are always striving to develop or learn of new treatments for patients that are quicker, less painful, safer, easier to do consistently, etc…that ultimately lead to improved quality of care
  30. 30. How Do We Shift Learners From Being Extrinsically to Intrinsically Motivated? "The mind is not a vessel to be filled, but a fire to be lighted" - Plutarch, Greek historian "I think the big mistake in schools is trying to teach children anything, and by using fear as the basic motivation. Fear of getting failing grades, fear of not staying with your class, etc. Interest can produce learning on a scale compared to fear as a nuclear explosion to a firecracker."
  31. 31. Introduction How master clinicians achieve + maintain mastery• 1. Increase experience through simulation • 2. Stay current • With journals • Organize a personal medical library • With societies • 3. Maintain a learning portfolio • 4. Situate learning in practice • Perform point-of-care / just-in-time learning • Keep challenging yourself to find better ways to do things for patients, thus improving quality of care •
  32. 32. Step 1 • Increase experience through simulation • Learning through practice • Bootstrapping + then refining your knowledge
  33. 33. Do Cases! Apps iBooks Ebooks Web Radiology One Night In ED (iOS) (Kindle or EPUB) We know what we see, we see what we know How do you get to Carnegie Hall?
  34. 34. Microlearning Social Media Small units of learning over a short time ~ For when you have few free minutes anywhere #FOAMRad #PedsRad Radiology Tag Ontology – hashtags/ontology/radiology/
  35. 35. Hands On • Microlearning using social media • #FOAMed on Twitter
  36. 36. Step 2 • Stay current • With journals • Follow 3-5 journals most relevant to your specialty and read a few articles each month related to your practice • Organize a personal medical library • With societies • Be on lookout for new clinical practice guidelines
  37. 37. Staying Current With the Literature Really Simple Syndication (RSS) RSS Feedly (Android+iOS) • Is a newsfeed generated by a Web site when it is updated • Lets you subscribe to Web site + be notified when it is updated • Think of it as distributed version of Facebook + Twitter • Subscribe to RSS feeds of journal Web sites • Receive stream of journal articles • The Old Reader Web browser-based feed reader is an alternative to apps
  38. 38. Organizing a Library of Journal Articles Medical Literature Database Dedicated App Cloud-based file system Papers (iOS+MacOS+Win) Mendeley (Android+iOS+Win) Storage: Drive / Dropbox / iCloud / OneDrive (Android+iOS) "iTunes for PDFs" Viewer: ezPDF (Android) / GoodReader (iOS) • How to create a file system on a mobile device? • What is best way to handle PDFs organized in folders so you can read them when offline? • Nota Bene - Don't trust the cloud, always have backup of your data in
  39. 39. Staying Current With Societies Social Media Twitter (Android+iOS) Facebook (Android+iOS) • Subscribe to the accounts (Twitter) and pages (Facebook) of the ACR, ARRS, RSNA, and subspecialty societies • Receive stream of society news • Filter with lists
  40. 40. Listening to Podcasts + Vodcasts iTunes (iOS) Podcasts (iOS) Podcast Republic (Android) • Examples - CTisUS, LearningRadiology, Radiology, AJNR • Simple on iOS • All podcasts indexed in iTunes Store or Podcasts app • Difficult on Android + Windows • Find XML file containing RSS feed of podcast + type it into podcast aggregator (feed:// ) • Podcast Republic integrates with iTunes
  41. 41. Step 3 • Maintain a learning portfolio
  42. 42. How the Medical Apprentice Learns • Diagnosis • Clinical skills related to experience • When seeing a new case you encapsulate it in an "illness script" which you pattern match to previous cases "illness scripts" in order to diagnose • Case-based reasoning / Storytelling • Expert is the person who can best capture, organize, and retrieve their experience (cases) • Treatment • Work up driven by medical knowledge - Schmidt HG, Norman GR, Boshuizen HPA. A Cognitive Perspective on Medical Expertise: Theory and Implications. Acad Med 1990 Oct;65(10):611-21.
  43. 43. Analog Case Collections Slide and Film-Based Steve Fishman M.D. Correlapaedia
  44. 44. E-Memory System• A digital archive of your life • "Lifelogging" • Made possible / inevitable by • Most memories are digital • Near-infinite space to store them • Ever-improving technology to recall them • Captures, stores, organizes and makes retrievable your experience + reflected wisdom • You become the librarian, archivist, cartographer and curator of your professional life From Microsoft
  45. 45. Learning Portfolio A Framework to Hang Your Learning On Capturing + Organizing + Retrieving Your Experience / Cases Evernote (Android+iOS+Win) OneNote (Android+iOS+Win) Excel (Android+iOS+Win) •For each case • Image annotated with: Question ~ Story ~ Answer ~ Impact on Practice (Reflection) ~ Resources Used •Diary of your learning ~ Archive of your medical educational life + experience • Is a knowledge management system •Educational construct • Adult learning theory = Learning situated in practice • Schon's theory of clinical problem solving / model of reflective practice / learning cycle • Case-Based Learning • Constructionism ~ Learning by doing / Learning artifacts • Portfolio-based learning •Why? • To situate learning in practice - questions are starting point for learning ~ Point-of-care CME • For sharing - individuals in person, groups in conferences / lectures, globally on Internet
  46. 46. Evaluation of a Learning Portfolio• A pediatrician's clinical experiences coupled with reflection • 5 elements for each case • Evaluation shows unstructured curriculum unfolding in practice over 5 years (234 cases) • Covers 100% of age ranges (n=9) • 100% of specialties (n=42) • 98% of symptoms (n=127) • 55% of diseases (n=707 [50-60% are pediatric]) • 90% of topics in 3 national pediatric curricula • 20 hours of CME / year • "My reading is now focused on my patients" - D'Alessandro DM, D'Alessandro MP. Formative Evaluation of a Pediatric Digital Library's Educational Content and Comparison to National Curricular Standards. Medical Teacher. 2008;30(9- 10) 880-6. • Learner taking control of and assuming responsibility for their own learning by tying their learning to practice + receiving CME for it • Learning portfolio documents what you have learned •
  47. 47. Hands On • Educational Informatics / For Providers • Maintain a personal learning portfolio • For each case • Question ~ Story ~ Answer ~ Impact on Practice ~ Resources Used • Unstructured data • Evernote • OneNote • Structured data in Office Suite • Apple iCloud • Google Drive • Microsoft OneDrive • Be HIPAA compliant • Don't use protected health information • If you use protected health information make sure it is
  48. 48. Step 4 • Situate learning in practice
  49. 49. What Did You Read Last Night? "Read about your patients every night!" - Alan Gruskin, M.D.
  50. 50. Tie Your Learning to Your Practice Analog techniques in 1988 "Someday computers may help with this…"
  51. 51. Learning is an Apprenticeship "In what may be called the natural method of teaching, the student begins with the patient, continues with the patient, and ends his studies with the patient using books and lectures as tools, as means to an end." - Sir William Osler, 1903
  52. 52. Future of Medical Education Lifelong Apprenticeship Best way to change physicians' knowledge, attitude, and behaviors And thus positively influence patients' care, outcomes, and lives Is to connect education to practice + shift learning to the point- of-care - ACCME, CME as a Bridge to Quality, 2006 - Josiah Macy Foundation, Continuing Education in the Health Professions, 2008 - IOM, Redesigning Continuing Education in the Health Professions, 2009 - Molly Cooke, David M. Irby, Bridget C. O'Brien. Educating Physicians: A Call For
  53. 53. Apprenticeship and Active Learning "Activities and understandings do not exist in isolation; they are part of a broader system of relations in which they have meaning." "Learning is an improvised practice: A learning curriculum unfolds in opportunities for engagement in practice." - Lave and Wenger: Situated Learning - Legitimate Peripheral Participation
  54. 54. Personal Learning Environment Learning From Your Experience / Cases All the Apps / Ebooks / Journals / Podcasts / Web sites you use
  55. 55. Medical Student Initial Personal Learning Environment for Point-of-care / Just-in-time learning App – Medscape Textbook – Custom Search Engine - Web Site –
  56. 56. Radiologist Initial Personal Learning Environment for Point-of-care / Just-in-time learning App – Medscape Textbook – Search Engine - Web site –
  57. 57. Hands On • Educational Informatics / For Providers • Find a library of the best free medical textbooks on the Web • • Embrace the diversity of references on the Web • Web browser is the killer app • …Think back to the last patient you saw and a question you had about their care… • Am I making the right diagnosis? • Am I prescribing the right therapy? • Am I causing any untoward side effects?
  58. 58. Clinical Question Searches Using Google vs. Evidence-Based Summary Resources "The authors found no significant differences in speed or accuracy between searches initiated using Google versus summary resources." - Kim S Searching for Answers to Clinical Questions Using Google Versus Evidence- Based Summary Resources: A Randomized Controlled Crossover Study, Acad Med. 2014;89:940-943. •StatDx guides you immediately to the right answer vs. guides you to journal review articles for eventual answer in context • From which method do you better learn + retain the information?
  59. 59. Hands On • Educational Informatics / For Providers • Use medicine-specific search engine for point-of-care learning during case-based learning and in clinical rotations • • …Think back to the last patient you saw and a question you had about their care… • Am I making the right diagnosis? • Am I prescribing the right therapy? • Am I causing any untoward side effects?
  60. 60. Shift to the Mobile Internet • If > 30 years old, primary computing device is desktop or laptop • Internet is experienced though single app = Web browser • Tablet replaces desktop / laptop + still experience Internet through Web browser • If < 30 years old, primary computing device is mobile phone • Internet is experienced through numerous apps • Don't use Web browser, Web not important, leads to narrowing of information sources used - Owen Williams, Smartphones: The Silent Killer of the Web As You Know It, The Next Web, May 5, 2014
  61. 61. Hands On • Educational Informatics / For Providers • Use an app for point-of-care learning during case-based learning and in clinical rotations • Medscape • (Tic the box that downloads whole app into your mobile device) • …Think back to the last patient you saw and a question you had about their care… • Am I making the right diagnosis? • Am I prescribing the right therapy? • Am I causing any untoward side effects?
  62. 62. Check Your Institution First! • For subscriptions to • Apps • eBooks (including PDFs) • Web-based decision support tools • For discounts on all of the above • Your library should have the information
  63. 63. Step 5 • Participate in an educational social network
  64. 64. The 4 C's For the 21st Century • Critical thinking • Communication • Collaboration • Creativity
  65. 65. Entrepreneurial Learner "What does it mean to be an entrepreneurial learner? It means how do you constantly look around you all the time for new ways and new resources to learn new things. Entrepreneurial learners are basically fundamentally makers and tinkerers." - John Seely Brown, researcher
  66. 66. Dispositions of an Entrepreneurial Learner • Curiosity • Driven by awe • Questing • Seeking, uncovering, probing…but always doing (curiosity in action) • Connecting • Listening to others, engaging • Reflecting • On performance with help of cohorts. Reflective practitioner - John Seely Brown, researcher
  67. 67. Personal Knowledge Mastery • A lifelong learning strategy for individuals to control their professional development through continual process of • Seeking • Finding things out + keeping up to date • Use smart filters – in form of a network of trusted individuals - to sort out valuable information • Sensing • Personalizing information + using it • Includes reflection, based upon critical thinking • Can include blogging, tweeting, writing to contextualize + reinforce learning • Sharing • Exchanging resources, ideas, experiences with your networks + colleagues • Pass your knowledge forward, iterate + collectively learn • Build respect + trust by being relevant • Mastery in a digital age is only achieved if you know how to establish trust, respect + relevance in human networks - Kenneth Mikkelsen + Harold Jarche, Developing Mastery in a Digital
  68. 68. On Expertise • Hard work, practice, experience not enough to make an expert • Expert is recognized by ability to solve non-routine problems in given domain • Expert's secret is willingness to work at the edge of their competence + keep reconstructing their skills at higher levels • Ideal classroom culture is "knowledge-building community" which supports expert-like learning • Would assist creation of "expert society" where expertise is normal rather than exceptional • Expertise is an expression of uniquely human potential to go beyond competencies given us by nature - Carl Bereiter and Marlene Scardamalia, Surpassing Ourselves:
  69. 69. The Power of the Personal Learning Network "If individualized learning is chained to a social vision prompted by "prisoner dilemma" rationality, in which one cooperates only if it maximizes narrow self-interest, network learning is committed to a vision of the social – stressing cooperation, interactivity, mutuality, and social engagement for their own sakes and for the powerful productivity to which it more often than not leads. The power of ten working interactively almost invariably outstrip(s) the power of one looking to beat out the other nine." - Cathy N. Davidson and David Theo Goldberg, The Future of
  70. 70. Personal Learning Network "A personal learning network is at the same time my personally curated network of people I want to learn from and a network that learns together. It wasn't too far a leap from there to the notion of learning community." - Howard Rheingold, Net Smart, 2012
  71. 71. The Personal Learning Network Cultivation Process • Explore – multiple media • Search – after you have explored enough to get some sense of the field • Follow – candidates' activity streams • Tune – your network by dropping people who don't seem worth spending attention on regularly • Feed – the people who follow you by sharing value when you find or create it • Engage – the people you follow • Inquire – of the people you follow and those who follow you • Respond – to inquires made of you
  72. 72. Educational Social Network = Community of Practice "Communities of practice are groups of people who share a concern, a set of problems, or a passion about a topic, and who deepen their understanding of this area by interacting on an ongoing basis." - Wenger E. et. al. Cultivating Communities of Practice: A Guide to Managing Knowledge • Educational construct = Interdependent Learning Theory • Teaching each other = collaborative learning = sharing the wisdom • Can be easily done by sharing learning portfolios on educational social networks
  73. 73. Share With Your Community of Practice Sharing Your Experience / Cases Blog Twitter / Instagram Figure1 #FOAMRad #FOAMPed #PedsRad
  74. 74. Practice Social Media Hygiene • Be aware of your online persona + keep it professional + restricted • AMA, Opinion 9.124 - Professionalism in the Use of Social Media • Protect patient privacy • Use privacy settings to safeguard personal information to the extent possible • Maintain appropriate boundaries • Consider separating personal and professional information • Act on unprofessional posted content • Weigh how your actions online may negatively affect your reputation and undermine public trust in the medical profession • Never reveal protected health information / de- identify everything
  75. 75. So What? Does This Work? Educational Effectiveness of Educational Social Networks / Social Learning • Types of evaluation / assessment • Formative vs. Summative • Proposition: The house believes that social networking technologies will bring large [positive] changes to educational methods, in and out of the classroom. • Assessment for learning (formative assessment) • Learning in the classroom built around peer support, self-assessment + questioning, peer assessment -- all coupled with learning logs - with teacher as a guide = Learning in a social network - Ewan Mcintosh, The Economist Debate Series: Education - Social Networking, The Economist, January 2008 "Firm evidence shows that formative assessment is an essential component of classroom work and that its development can raise standards of achievement…" - Paul Black and Dylan Wiliam, Inside the Black Box: Raising Standards Through Classroom Assessment, Phi Delta Kappan, November 1998 "Teachers found that the motivation and attitudes of their students improved, and the students achieved higher scores on externally set tests and examinations" - Paul Black, Working Inside the Black Box: Assessment for Learning in the Classroom, Phi Delta Kappan Sept 2004 …So formative assessment improves summative assessment…
  76. 76. Educational Effectiveness of Educational Social Networks / Communities of Practice "Physicians interact with peers and mentors to frame issues, brainstorm, validate and share information, make decisions, and create management protocols, all of which contribute to learning in practice. It is likely that working together in this way creates the best environment for learning that enhances professional practice and professional judgment." - Parboosingh JT. Physician Communities of Practice: Where Learning and Practice are Inseparable. Journal of Continuing Education in the Health Professions. 2002. 22, pp. 230-230 • Sharing among health care providers is the best environment to move towards
  77. 77. The Learning Continuum Capturing + Organizing + Retrieving, Learning From, and Sharing Your Experience / Cases Data -> Information -> Knowledge - > Wisdom / Mastery [acquired] [organized] [reflected] [shared] Learning Portfolio (LP) -> Personal learning environment (PLE) -> Community of practice (COP) [focus on individual] [focus on group] Create cases in LP -> Learn about cases in PLE -> Share cases with COP
  78. 78. Conclusion How to Achieve + Maintain Mastery in Medicine • Components of ideal medical education • Increase experience through simulation • Stay current • With journals • Organize your own medical library • With societies • Maintain a learning portfolio • Situate learning in practice • Perform point-of-care / just-in-time learning • Participate in educational social networks • Goal is to create an environment which allows • Personal learning + reflection • Master-to apprentice and peer-to peer teaching • Conversations about art and science of medicine • Improvement in the quality of care you deliver • This is how to create and preserve a lifelong passion for medicine • This is how to create medical learning machines who strive for and achieve excellence in practice and pursue mastery of medicine over a lifetime
  79. 79. The Present • Educational Informatics for Patients
  80. 80. Health is Not Dependent on Physicians • 90% of variance in illness and premature death is related to factors other than access to medical treatment • 50% = Behavioral - Diet / Nutrition / Exercise • 25% = Environmental - Presence or absence of war, sanitation, toxins • 15% = Genetic • 10% = Access to medical treatment • Immunization mainly • Your physician is almost inconsequential • Unless you live in the western world and are susceptible to ills of affluence / lifestyle - MHS 2025 - Toward a New Enterprise, Section 4 page 2
  81. 81. Dose of Reality Regarding Compliance • Four good health rules • Don't smoke, maintain normal weight, eat fruit + vegetables, exercise • Only 3% of Americans follow all 4 rules - Archives of Internal Medicine 2005;165:854-7 [Two more rules - Don't drink to excess + Practice safe sex] • 40% of seniors don't take all the meds their physicians prescribe for them - Health Affairs 2005 April 19
  82. 82. Patient's Information Seeking Behaviors • There can be no more motivated learner than the patient • 81% of U.S. adults use the Internet • 59% of U.S. adults say they have looked online for health information in the past year • 35% of U.S. adults say they have gone online specifically to try to figure out what medical condition they or someone else might have - Pew Internet & American Life Project, Health Online 2013 • You should use Information Prescriptions to partner with them • Doctor means teacher - you must teach your patients •
  83. 83. Hands On • Educational Informatics / For Patients • Use Information Prescriptions to partner with them • • …Think back to the last time a family member asked you for health advice…write them an information prescription… • …Or the next time a family member asks you for health advice…write them an
  84. 84. The Future • Bioinformatics
  85. 85. Guild of Medicine is Under Siege "This is about the democratization of medicine. We are moving towards the individual taking charge." - Eric Topol MD, The Creative Destruction of
  86. 86. Guild of Medicine is Under Siege "Our generation is used to getting information they need when they want it, and the idea that we can't have access to our own health data is bewildering…If you can be proactive with your health, you can be more in control" - Ann Wojcicki, CEO of 23andme "If you're not sitting around the table you are on the menu"
  87. 87. The Future • Research Informatics • Paradigm of medical practice will change from disease correction to disease prevention • Informatics tools will allow super- empowered patients to live healthy lives • When patients become ill, informatics tools will help them take control of and guide their treatment and the role physicians will play
  88. 88. The Future "The least utilized resource in the healthcare system is the patient!" - Warner Slack, M.D. •Medical care was a doctor's effort where doctor was the whole team •Medical care is a multidisciplinary team's effort where doctor is quarterback of team •Medical care will be a team effort where the patient is the quarterback and the ball is the mobile device containing the
  89. 89. Quantified Self• Goals - Stay Healthy, Detect disease early, Personalize treatment • Step 1 - Sequence your genome "23andMe can help you manage risk and make informed decisions" "Change what you can, manage what you can't" Sequences portion of genome for $199 • 23andMe is creating vast database of genetic information of its customers + cross referencing it against the health history they provide • Is Big Data, CEO is former spouse of Larry Brin of Google • Goal was to offer info on any disease risks you may have…but FDA only allows them to give "carrier status" information on 35 genes that can cause disease...they have to prove they can explain test results • •Sequencing whole genome ~ $1,000 • Driven by Moore's Law • What computing was to the last 50 years, biology will be to the next 50 years •Will lead to personalized medicine • Treatment tailored to each person as unique individual suffering
  90. 90. Quantified Self • Step 2 - Engineering approach to life • Quantify it with biosensors, analyze it, act on it • • "Self-Knowledge Through Numbers" • Self-tracking is a practice of self- observation leading to self-awareness + hopefully some sort of behavioral change or self-improvement • Apps / devices aspire to be motivational • Gamification for individuals • Competitions for groups
  91. 91. Quantified Self • Definition • Ordinary people recording + analyzing numerous aspects of their lives to understand + improve themselves • Types of data • Exercise, activity, idleness • Medications • Nutrition • Sleep • Mood • Cognition speed + ability • Blood pressure + blood glucose • Challenge • Data mining analytic capabilities that search for patterns over gathered data are currently primitive
  92. 92. Quantified Self • Result - Empowered patients who will present with data driven chief complaint • Gordon Bell - • Larry Smarr - hacking his own body "Basically, we will have personalized doctors with us at all times, instead of two 15-minute visits a year." - Larry Smarr Ph.D. "You now have, for the first time in history, a scientific basis for medicine." - Larry Smarr Ph.D.
  93. 93. Quantified Life• Replace guesswork presently guiding individual health decisions with specific guidance tailored to details of each person's body • A genome is an individual's basic program • Once you know your genome and begin monitoring your bodily systems with a computer, the computer will know you better than any doctor could + will spot disease long before you feel sick • From a low-res to a high-res image of your health • Doctor goes from oracle to consultant • Essence of life is variability • Constant monitoring is recipe for all of us to be judged "sick," that will lead to a lot of interventions in people who are basically healthy + creating epidemic of anxiety "I can conceive of this happening. But is this the model we want for good health? What does it mean to be healthy? Is it something we learn from a machine? Is it the absence of abnormality? Health is a state of mind. I don’t think constantly monitoring yourself is the right path to that state of mind. Data alone is not the answer." - Dr. H. Gilbert Welch, author of Overdiagnosed: Making People Sick in the Pursuit of Health.
  94. 94. Quantified Self Downsides • How do we make the data understandable to patients? • How do we make the data manageable by physicians? • Does this lead to a new kind of surveillance state? • Who will own / will be able to access this data? • Creators of data? Their employer? Their insurer? Researchers? • Who will preserve the data? • Insurance based upon community pooling of risk because we don't know what our individual risk is…but now we will know what
  95. 95. Personal Health Record
  96. 96. (Alphabet) Productizing Quantified Self• Verily • "The mission of Google X Life Sciences is to change healthcare from reactive to proactive. Ultimately it’s to prevent disease and extend the average lifespan through the prevention of disease, make people live longer, healthier lives." • Nanoparticles built to attach to particular types of cells (i.e. cancer cells) circulate in the blood • Wearable device detects nanoparticles + provides info to physicians • Let's you go from reactive to proactive medicine – to prevent disease instead of finding ways to treat it • Starts with Baseline Study to identify biomarkers of disease • Calico • "Calico’s mission is to improve the maximum lifespan, to make people live longer through developing new ways to prevent aging." - Steven Levy, We're Hoping to Build the Medical Tricorder, Medium, Oct. 28,
  97. 97. Hands On • Bioinformatics • Let's make *you* a patient! • Consider undergoing genetic testing to learn what diseases you are at risk of developing and then develop strategies to manage your disease risk over your life time as a super-empowered patient would do •
  98. 98. Hands On • Quantify Yourself • Get an Apple Watch or Fitbit or Microsoft Band or Android equivalent • Open a Microsoft HealthVault account for your own personal health record • Help you understand the benefits and challenges facing patients who do this •
  99. 99. Patients Taking Control of Their Care • Internet was designed to route around obstructions • Now patients are using the Internet to route around obstructions like you • They are running their own clinical trials • - "Share your experiences with treatments. Find patients just like you. Learn from others who know." • Social network for sick people
  100. 100. Future - For Patients Tricorder X Prize • Goal - create a mobile platform that will enable people to diagnose a set of 15 conditions (pneumonia to diabetes to sleep apnea) without having to rely on
  101. 101. The Next Step: Big Data + Data Mining • Definitions • Big Data – complex datasets characterized by four V's: Volume (exabytes = 1 billion gigabytes), Velocity (data generated at high speed), Variety (data from multiple sources), Veracity (uncertainty in some data elements) • Demands cost-effective, innovative forms of information processing for enhanced insight + decision making • Data Mining - Algorithms extract and synthesize knowledge from analyzing large numbers of individual cases • Applications • Non-medical • Targeted advertising, personalized consumer recommendations, real-time traffic maps • Medical – The Dream: Big Data meets Personalized Medicine -> Precision Medicine • Disease surveillance • Speed learning benefits + harms of current treatments • Speed development of new treatments • Decision support • Support efforts to improve quality of care • Advantages • Non-medical companies (Google) may bypass HIPAA • Disadvantages • How do you protect patient privacy? • Barriers • Interconnecting disparate databases • Overcoming HIPAA
  102. 102. Examples of Big Data Google auto-predict Amazon recommendations Google Flu Trends Google Translate
  103. 103. Personalized / Precision / Genomic Medicine The Coming Revolution in Health Care • How health care will be reformed • Your genome sequenced to tell you what diseases you are at risk for • Your health will be continuously monitored by wearable computing devices + sensors • Your information will be stored in a personal health record • Decision support tools will help you live a healthier life and decrease your risk of disease • The refrigerator will be your health hub • The physician will interface with patient, rather than that patient with the physician • When the patient is ill they will run their own clinical trials • Bottom line - Taking responsibility for your own health ~ influencing your destiny
  104. 104. Future - For Providers Artificial Intelligence / Machine Learning / Watson "I for one welcome our new computer overlords." – Ken Jennings •Will Watson serve as an assistant to, or replacement for physicians?
  105. 105. Artificial Intelligence Based on Rules Isabel DXplain
  106. 106. From Search Engine to Question Answering System Through Artificial Intelligence• Artificial Intelligence • Arises not through rules but through statistics • Example question answering systems • Watson - IBM • Inspiration - computer on Star Trek • Interface - natural language processing • Approach - statistical computation - using hundreds of statistical algorithms simultaneously, then chooses best answer from amongst them • Hardware - very fast, lots of memory - Clive Thompson, What is IBM's Watson? New York Times Magazine, June 20, 2010
  107. 107. How Watson Works • Expert systems definition - computer systems that can reason over knowledge • Expert systems failed in 1980's because it was difficult to capture and maintain a mathematization of a domain • The promise of Watson is that it can reason over unstructured information, in the form that humans are creating it in textbooks and articles in natural language • If we have a computer that can reason over natural language it can scale and do useful things without a lot of work on our part • Watson will never be as precise as an expert system - it won't have a chain of reasoning - it will deliver explanations in your words • Why medical expert systems have not been widely adopted so far - primary reason is that they are very narrow and hard to keep up to date and don't show you the context for the rules - they don't let you judge the evidence, which is changing all the time • We don't want to take judgment away from humans, we want to enable better judgment and provide the content that persuades them • Prior expert systems are brittle and narrow and don't give you the kind of information you are comfortable with • Watson is not rules based - A Computer Called Watson - Presentation at the Computer History Museum on Nov 11, 2011 by David Ferrucci - starting at 42:30 minutes
  108. 108. "People ask me if this is HAL. HAL's not the focus; the focus is on the computer on 'Star Trek,' where you have this intelligent information seeking dialogue, where you can ask follow-up questions and the computer can look at all the evidence and tries to ask follow-up questions. That’s very cool." "The computer on ‘Star Trek’ is a question-answering machine. It understands what you’re asking and provides just the right chunk of response that you needed. When is the computer going to get to a point where the computer knows how to talk to you? That’s my question." "Wouldn't it be great to be able to communicate with the computer like Captain Picard or Captain Kirk does on "Star Trek," where you can fluently dialogue with an information-seeking computer that can understand what you're asking, ask follow up questions, and get exactly at the information that you need? That would be incredible. That's kind of this motivating vision, and whether Watson loses or not in this big game is really not the point. The point is we were able to take a step forward in that direction, and I think that's what we're most excited about." - David Ferrucci, PhD, IBM, principal investigator DeepQA /
  109. 109. Cognitive Computing • By 2029, Kurzweil believes computers will be able to do all the things that humans do, only better • He is now Google's Director of Engineering and his job description is to help bring natural language understanding to Google "And what's not generally appreciated is that Watson's knowledge was not hand-coded by engineers. Watson got it by reading. Wikipedia – all of it. Computers are on the threshold of reading and understanding the semantic content of a language, but not quite at human levels. But since they can read a million times more material than humans they can make up for that with quantity. So IBM's Watson is a pretty weak reader on each page, but it read the 200m pages of Wikipedia. And basically what I'm doing at Google is to try to go beyond what Watson could do. To do it at Google scale. Which is to say to have the computer read tens of billions of pages. Watson doesn't understand the implications of what it's reading. It's doing a sort of pattern matching. It doesn't understand that if John sold his red Volvo to Mary that involves a transaction or possession and ownership being transferred. It doesn't understand that kind of information and so we are going to actually encode that, really try to teach it to understand the meaning of what these documents are saying." - Ray Kurzweil - Carole Cadwalladr, Are the Robots About to Rise? Googles' New Director of
  110. 110. Artificial Intelligence Based on Machine Learning / Deep Learning • Machine learning - teaching computers to crunch vast amounts of data, recognize patterns, get better at what they do • Deep learning, a step beyond machine learning, uses unsupervised learning to automatically convert unstructured information into useful actionable knowledge, will lead to a system that knows the world • Ultimate goal is create a system that inhales world's information, structures it into a form it understands, then takes action • 2012 – Google trains computer to identify cats by watching YouTube videos • 2014 – Deep Mind (now part of Google) trains computer to learn to play classic video games by watching videos of games being played • 2015 – California Health Care Foundation using Kaggle trains computer to learn how to diagnose diabetic retinopathy by reading thousands of images of retinas • 2015 – IBM buys Merge, a PACS vendor with billions of radiology images, to feed to Watson…
  111. 111. Deep Learning • Deep learning systems - a modern refinement of machine learning systems, which mimic layers of neurons in brain, by crunching vast amounts of data, can learn to perform some tasks as well as humans such as pattern recognition, translation, playing videogames • These learned tasks are narrow + specific • Powers Google's search engine, Facebook's automatic photo tagging, Apple's Siri voice assistant, Amazon's shopping recommendations, Tesla's self-driving cars "Instead of people writing software, we have data writing software" - Jen- Hsun Huang, CEO of NVIDIA • Long term goal is to create an artificial general intelligence which can solve a wide variety of tasks • Most optimistic view is that this will take a decade - Dawn of Artificial Intelligence, The Economist, May 9, 2015 - Rise of the Machines, The Economist, May 9, 2015 - March of the Machines, The Economist, Jun. 25, 2016 - The Return of the Machine Question, The Economist, Jun 25, 2016 - From Not Working to Neural Networking, The Economist, Jun 25, 2016
  112. 112. Deep Learning "Perhaps the best way to think about AI is to see it as simply the latest in a long line of cognitive enhancements that humans have invented to augment the abilities of their brains. It is a high-tech relative of technologies like paper, which provides a portable, reliable memory, or the abacus, which aids mental arithmetic. Just as the printing press put scribes out of business, high-quality AI will cost jobs. But it will enhance the abilities of those whose jobs it does not replace, giving everyone access to mental skills possessed at present by only a few. These days, anyone with a smartphone has the equivalent of a city-full of old-style human "computers" in his pocket, all of them working for nothing more than the cost of charging the battery. In the future, they might have translators or diagnosticians at their beck and call as well." - Dawn of Artificial Intelligence, The Economist, May 9, 2015 - Rise of the Machines, The Economist, May 9, 2015 - March of the Machines, The Economist, Jun. 25, 2016 - The Return of the Machine Question, The Economist, Jun 25, 2016 - From Not Working to Neural Networking, The Economist, Jun 25, 2016
  113. 113. The Coming Singularity? "Kurzweil believes that we're approaching a moment when computers will become intelligent, and not just intelligent but more intelligent than humans. When that happens, humanity - our bodies, our minds, our civilization - will be completely and irreversibly transformed. He believes that this moment is not only inevitable but imminent. According to his calculations, the end of human civilization as we know it is about 35 years away." "Here's what the exponential curves told him. We will successfully reverse-engineer the human brain by the mid- 2020s. By the end of that decade, computers will be capable of human-level intelligence. Kurzweil puts the date of the Singularity - never say he's not conservative - at 2045. In that year, he estimates, given the vast increases in computing power and the vast reductions in the cost of same, the quantity of artificial intelligence created will be about a billion times the sum of all the human intelligence that exists today." "One of the goals of the Singularity Institute is to make sure not just that artificial intelligence develops but also that the AI is friendly. You don't have to be a super-intelligent cyborg to understand that introducing a superior life-form into your own biosphere is a basic Darwinian error." - Lev Grossman, 2045: The Year Man Becomes Immortal, Time, Feb 10, 2011
  114. 114. ASIMO
  115. 115. The Future of Medicine • Quantified Self – everything will be measured + recorded • Big Data – this will be used for population health • Personalized medicine – will turn patients into extremely unique individuals • Expert systems will be tool that will help physicians work together with patients to manage their health + disease – but we must not let them become our overlords
  116. 116. Where Will We Be in 30 Years? • Ray Kurzweil says we will be pets • Machine learning / Deep learning through descendants of Watson may have taken over diagnosis + treatment • ASIMO may have taken over some procedures • What will that leave doctors to do? • Spend more time with + listen to our patients • Become a guide on the side rather than a sage on the stage
  117. 117. Conclusion Why is Biomedical Informatics Important? • In the 21st century, informatics is everyone's interface to health care and the health care system
  118. 118. Conclusion Who Will Biomedical Informatics Impact the Most? • The greatest impact will be not on the top 5% but instead on everyone else • How did mobile technology change Bill Gate's life? / How did mobile technology change the lives of those in the developing world? • How has solar energy changed your life? / How is solar energy changing the lives of those in the developing world? • Informatics can help provide medical care where there is none…
  119. 119. Conclusion Goals For The Future • Computers should be used to augment rather than replace physicians • Stanford vs. MIT schools of artificial intelligence • Computers should be used to increase humanity in medicine rather than reduce it • Informatics is one way to reduce our bias towards high tech high cost medicine
  120. 120. Conclusion • Medicine – It's not just a job, it's a profession • They can stop you from coming to work but they can't stop you from learning at home - so read about your patients! • Spend less time looking down and more time looking up • Let's model good informatics behavior for our patients • Join me in being an advocate for excellent informatics for our patients + ourselves • 30 years ago it was just me, now we have a critical mass • You can make a difference. You can make it better. Let's change the world!
  121. 121. One More Thing… "Here's to the crazy ones. The rebels. The troublemakers. The ones who see things differently. While some may see them as the crazy ones, we see genius. Because the people who are crazy enough to think they can change the world, are the ones who do."
  122. 122. To Learn More • Twitter • • Life as a Healthcare CIO – John Halamka M.D. • • TED Talks • – look at health, health+care, medicine, medical+research • Textbook • Biomedical Informatics by Shortliffe + Cimino (4th edition) • American Medical Informatics Association • • University of Iowa Health Care Chief Medical Information Officer • Maia Hightower MD MBA MPH • University of Iowa Interdisciplinary Graduate Program in Informatics • • National Library of Medicine informatics rotations for medical students • • Fellowships in biomedical informatics – this is now a boarded specialty • • How to live your life from Steve Jobs • • If you are interested in becoming a biomedical informatician • Join AMIA, do informatics fellowship as medical student, read Shortliffe's book • Complete your M.D. degree, then complete a medical residency, then do an informatics fellowship • Don't have to learn computer science - do have to learn computer literacy (see D is for Digital by Brian Kernighan) – learn a little computer programming (in a MOOC)
  123. 123. Ask Me Anything • Email • • These slides • ro/presentations