More Related Content
Similar to Digital healthcare show - How will Artificial Intelligence in healthcare will impact patient outcomes in the future? (20)
Digital healthcare show - How will Artificial Intelligence in healthcare will impact patient outcomes in the future?
- 1. How will Artificial Intelligence in healthcare will impact patient outcomes in the future? Dominic Cushnan, Ameet Bakhai, Andy Wilkins Twitter: #AICommunity
- 2. The future is here: it is just not evenly distributed William Ford Gibson (born 17 March 1948) is an American-Canadian writer who has been called the "noir prophet" of the cyberpunk subgenre of science fiction. Gibson coined the term "cyberspace" in his short story "Burning Chrome" and later popularized the concept in his debut novel, Neuromancer (1984).
- 3. As healthcare professionals, what does this future hold for us and our patients? Write on the paper in front of you.
- 5. ADMIRALTY LIBRARY, NAVAL HISTORIC BRANCH Getting evidence in to practice
- 6. INSTITUTE OF NAVAL MEDICINE Getting evidence in to practice It took 200 years before the Royal Navy routinely used lemon juice to prevent scurvy. First study 1601
- 8. The internet has changed the way we do … everything. 1998 was also the year a little company called Google was born, although back then, it looked a little different than it does now.
- 9. Social media has taken over. One of the most influential applications of the internet is social media.
- 10. Source:OrganisationalNetworkAnalysisbyInnovisor Just 3% of people in the organisation or system typically influence 85% of the other people
- 13. What is AI?
- 15. The maned sloth, also known as the ai, is a three-toed sloth that lives only in Brazil.
- 16. What is AI AI describes a set of advanced technologies that enable machines to do highly complex tasks effectively – which would require intelligence if a person were to perform them. There is, however, “no standard definition of intelligence” and no single agreed definition of AI. In addition, the line between AI and other techniques, such as big data analytics, can be blurred.
- 17. AI Types • Weak AI (narrow AI) – non-sentient machine intelligence, typically focused on a narrow task (narrow AI). • Strong AI – (hypothetical) sentient machine (with consciousness and mind). • Artificial general intelligence (AGI) – (hypothetical) machine with the ability to apply intelligence to any problem, rather than just one specific problem, typically meaning "at least as smart as a typical human". • Superintelligence – (hypothetical) artificial intelligence far surpassing that of the brightest and most gifted human minds.
- 23. AI Organisations & Workstreams in Healthcare
- 26. “Despite central government’s enthusiasm for AI, current applications within the NHS are piecemeal.” Reform January 2018
- 27. Artificial intelligence / digital Health care Impact on professionals & future of NHS Healthcare Ameet Bakhai MBBS, MD, FRCP, FESC Consultant Cardiologist & Physician Cardiovascular Research & Development Director Amore Health Ltd / Royal Free London NHS Trust / Royal National Orthopeadic Hospital / Barnet CCG [email protected]
- 28. Patient Access Value Based Care Personalized Digital Health Preventative Medicine Drug Trials and Discovery data environment is rapidly changing Healthcare organizations are facing a deluge of rich data that is enabling them to become more efficient, operate with greater insight and effectiveness, and deliver better service Advances analytical and computing techniques coupled with the explosion of data in healthcare organizations can help uncover leading clinical practices, shrink research discovery time, streamline administration, and offer new personalized engagement paradigms at an industrial scale that align people’s decisions and actions in ways that improve outcomes and add value Sources of the data deluge Advances in computing power and techniques Smarter Algorithms Faster Processing Speed Improved Visualization Patient Centric Optimal Resource Structure Adaptive Organization * HP Autonomy, Transitioning to a new era of human information, 2013 ** Steve Hagan, Big data, cloud computing, spatial databases, 2012 Sensors / DevicesVideosImagesSocial MediaPaper / Text Documents EMRsMobile 40-50% Annual growth in digital data volume* ~9X of unstructured data vs. structured data by 2020** 62% Annual growth in unstructured data*
- 29. Technology – GAME CHANGERS / DISRUPTORS https://hbr.org/2011/01/reinvent-your-business-before-its-too-late http://marketingio.com/2016/11/07/martecs-law-the-greatest-management-challenge-of-the-21st-century-chief-marketing-technologist/
- 30. Technology Medicine • Gartner Hype Cycle for Telemedicine, 2014
- 31. Deep Learning Open Access Google Inc et al. May 2018 216K Adults 46K Million data 24 hours AUROC: (predict) 0.93 death 0.85 re-admit Beat clinical scores
- 32. 32 Example: Project Genesis • Scope of Computer Systems • Clinical • Power Chart - Orders and results • Clin Doc - Clinical documentation • PharmNet - Pharmacy • FirstNet: Emergency Dept. • RadNet: Radiology Dept. • SurgiNet: Operating Room • Inet: ICU • Profile - HIM application • EMPI • CPOE • Electronic Record - Clinical functions by pt. type - Current clinical documentation forms Implementation Readiness Process Requires 20-24 Months •People •Process •Technology Implementation Readiness Major Learning: Realizing clinical benefits of transformational change is a function of time
- 33. COGNITIVE LOAD THEORY: WHY USING YOUR ELECTRONIC HEALTH RECORD IS SO PAINFUL AND HOW TO FIX IT Michael Zimmerman, MD Temescal Creek Medicine
- 36. Amongst the Oldest Health Information Technologies • Original digital health device • Conceived and designed to solve a healthcare need, not fill a market gap • Strengths: • Fits into existing lifestyle • Passive (no patient action necessary) • It works really, really well • Weaknesses to overcome with new technologies • Security • One size does not fit all; QoL • Expense • Designed to be replaced – profit at the price of complications Oh, and by the way, making a great device is good business: 2014 WW sales > $6 billion https://globenewswire.com/news-release/2016/01/07/800112/0/en/Cardiac-Pacemaker-Market- to-Reach-US-12-85-bn-by-2023-Rising-Geriatric-Population-is-High-impact-Driver-Transparency- Market-Research.html
- 39. More Patients, Intelligently Navigated through surgery, with Better Outcomes, Shorter Stays, Better Staff Morale & Respectful Teamwork !
- 40. Care Algorithms / Decision Pathways: The Slide Towards AI in Healthcare Useful • Speed up decision making • Project sense of normality • Demonstrate predictability • Allow forecasting • Project an evidence base • Reduce variation in practice • Reduce ‘fud’ • Avoids senior input (GP + AI = Specialist?) Challenge • Speed up nature - impatience • Assume NO human error • Reduce checks • Debase impact of issue (MI) • Generalisability to individual • Divert individualised care • Reduce responsibility • Reduce experience base (GP – AI = Medical Novice?) MI = Myocardial Infarction – now only a 72 hr issue
- 41. When Big Data goes wrong: UK breast cancer screening IT error • All women aged 50 to 70 in the UK who are registered with a GP are automatically invited for breast cancer screening every three years • A “computer algorithm failure”, which dated back to 2009, meant an estimated 450,000 women aged between 68 and 71 were not invited to their final breast screening between 2009 and the start of 2018 • Initial estimates have suggested that between 135 and 270 women may have “had their lives shortened as a result” • If you don’t get the programming right, systems won’t work correctly Available from: https://www.digitalhealth.net/2018/05/14000-women-contact-breast-cancer-helpline/
- 42. AI: friend or foe? “What does your organization primarily plan to do in terms of employees that have been replaced with technology?” • Retrain them into a new role/area of the organization 34% • Redeploy within the same area of the organization 42% • Make them redundant 24% • Need to carefully consider how AI deployment could affect the workforce and ensure that the proper ethical checks for autonomous systems are in place. • AI will exist to support people in their jobs. For instance, AI will optimize clinical processes, such as recording patients’ vital signs or analysing scans and samples, but the doctor will decide the final line of treatment. • The purpose of AI will be to augment natural intelligence, and its role will always be subordinate to the human’s. Available from: https://www.infosys.com/smart-automation/Documents/ai-healthcare.pdf
- 43. AI in healthcare: peer-reviewed evidence PubMed search for clinical trials with the keywords “artificial intelligence” and “NHS” published in the last 5 years
- 44. Duty of candour: a level playing field? • Aim of the regulation: to ensure that providers are open and transparent with people who use services in relation to care and treatment. • Contains specific requirements that providers must follow when things go wrong with care and treatment, including: • informing people about the incident • providing reasonable support • providing truthful information and an apology. • Providers must promote a culture that encourages candour, openness and honesty at all levels. • What about companies who provide healthcare IT solutions? Available from: http://www.cqc.org.uk/sites/default/files/20150327_duty_of_candour_guidance_final.pdf
- 45. FDA report In 2010, 260 HIT reports, 44 injuries, 6 deaths in 2 years – Voluntary reporting system – likely underreported..
- 46. Example: Babylon Health to power NHS 111 with ‘AI triage’ bot https://www.digitalhealth.net/2017/01/babylon-health-to-power-nhs-111-with-ai-triage-bot/ • A chatbot to answer NHS non-emergency inquiries from more than a million Londoners as a new way to manage the growing health burden. • Driven by clinically based algorithms that triage patients without human intervention based on reported symptoms. • Based on the symptoms and its own algorithms, the app could refer the patient to hospital or recommend a GP appointment the next day. • Doctors have already expressed concerns about the reliance on algorithms and self-reported symptoms for determining the severity of a person’s illness. However, ? published evidence...
- 47. AI has huge potential Feasibility study of a randomised controlled trial to investigate the effectiveness of using a humanoid robot to improve the social skills of children with autism spectrum disorder (Kaspar RCT): a study protocol Mengoni SE, Irvine K, Thakur D, et al. BMJ Open 2017;7:e017376. doi:10.1136/bmjopen-2017-017376
- 48. Immediate example in cardiology management? • Research presented at the 2017 AHA Congress on pairing machine- learning algorithms with the Apple Watch’s heart-rate sensor and step counter to predict hypertension or sleep apnoea • Apple says it’s working on a study with Stanford that will test the gadget’s ability to detect atrial fibrillation • There are many theoretical possibilities for how AI could assist us with managing patients with or at risk from thromboembolic disease Available from: https://www.wired.com/story/ai-can-help-apple-watch-predict-high-blood-pressure-sleep-apnea/
- 49. Kintzugi – Japanese Art of Appreciating that which has Broken – for its wisdom & sacrifice. By Repairing it with Gold. The NHS may face disruption before it’s evolves to a new model of care. Schwartz Rounds - prior Barnet Clinical Lead - Compassion for the Caregivers
- 50. Disclosures Ameet Bakhai, Consultant Cardiologist & Physician / Cardiovascular R&D Director • I am employed by the NHS, CCG and work with AHSNs – UCL Partners, Imperial Health • Founder Amore Health Ltd • An ambassador for Digital Health London • I have advised pharma, device, advisory, strategy, health technology appraisal, government policy, commissioning groups and technology firms on innovations in healthcare from drugs, devices, diagnostics, decision pathways to digital technologies • Past & Present Committees: Research & Development, Governance and Risk, 18 week Pathway Champion, Medicine Management, D&TC, Thrombosis, Audit, Clinical Excellence Awards, Physician Associate Project, Work Experience, Education & Partnership for Cardiac & Stroke Network, Surrey Heath CCG Board Member, Task Forces in Cardiology for UCLp & Imperial Health, Horizon Scanning for NICE, End of Life Care for NHS England – London, Cardiovascular Research North Thames CRN, Education Standards ABPI • Studied Decision Analysis Modelling & Health Economics @ Harvard School of Public Health Advisor / Lecturer / Appraiser / Committee - Health technology appraisal groups - Economic modelling teams - Pharma, Device & Strategy companies - IT and Digital Health companies - ABPI, NICE, UCLp
- 51. Scope & purpose of report 51 A 10-15 year vision for AI powered person-centred public healthcare Andy Wilkins – Report Lead Author On behalf of the Royal Free Charity The Digital Healthcare Show 27th June 2018
- 52. 52 Royal Free North Central London New models of care • Integrated care • Whole person care • Population health New capabilities • New medical breakthroughs • New clinical & digital technologies Fog of Uncertainty The challenge – how to invest for the future when so much is changing?
- 53. 53 The answer – look beyond the “fog” to shine a light on a vision of the long term future
- 54. A 10-15 year vision for person-centred public healthcare Scope & purpose of report 54 New major report coming soon!! Report sets out a future that delivers: 1. Transformational improvements in the nation’s levels of health and wellbeing 2. A pathway to building a sustainable, person-centred 21st Century public health and care system 3. An engine for economic growth and social renewal The report describes a vision based on the transformational new capabilities arriving in the next 10- 15 years
- 55. 55 We rolled forward 6 key trends… to imagine a world where they had all “landed” at scale We asked ourselves: 1. What would health look like for the individual 2. What would this mean for the future delivery of health & care?
- 56. 56 The 21st Century health challenges call for a wider framing of the healthcare landscape
- 57. 57 A new generation of sensors will enable revolutionary new sources of data and possibilities to improve care
- 58. 58 Real time data will make it possible to dynamically simulate health
- 59. 59 A Digital health coach enables always-on personalised care support My health context Holistic care support Decision Support Integrated Care Teams
- 60. 60 The three core elements of the Vision 1. Personalised Health and Wellbeing2. AI Mediated Health Coaching 3. Future Health and Care System
- 61. 61 As PoC technologies miniaturise and are combined with AI powered Decision support systems then: 1. Care becomes more integrated 2. LTC care becomes more health/ life coaching based and moves into community settings 3. Digital health coaches manage day to day and moment to moment support What impact will this have on the management of chronic disease?
- 62. 62 Size of the prize = better quality of life + sustainable healthcare system
- 63. Building the vision will require a phased transformation journey
- 64. System Framing Making prevention and population health a priority Integrating prevention, healthcare and social care into a unified care system Taking a longer term investment perspective Public engagement Making the case for self-management of health and wellbeing Acceptance of sensors Acceptance of sharing of data Acceptance of a AI powered digital health coach New healthcare ecosystem New data skills and capabilities New medical and wellbeing skills Transformational funding New clinical care models New funding and governance models New clinical organisational structures Person-centred data strategy Radical new innovation 15 Building Blocks – challenges to overcome not reasons to hold back
- 65. 65 The next stage in our journey – creating a movement for change! 2. Working with Junior Doctors at Barnet Hospital on a Story Platform – CPD opportunity 1. Created a cross sector steering group to plan the provocation 3. Exploring joint initiatives with the RSM 4. Engaging with Politicians, Healthcare leaders & Industry on the need for a new vision for 21st Century healthcare
- 66. 66 We’d love you to be involved… go to www.beyondthefog.org to stay in touch
- 67. Snowstorm Write down one key thing you have learnt or will think about when you go back to your organisation?
- 68. Snowstorm • Write down one key thing you have learnt or will think about when you go back to your organisation? • Screw the paper up • On the signal, throw your snowball in the air.
- 69. Snowstorm • Write down one key thing you have learnt or will think about when you go back to your organisation? • Screw the paper up
- 70. Snowstorm • Write down one key thing you have learnt or will think about when you go back to your organisation? • Screw the paper up • On the signal, throw your snowball in the air. • Pick up a snowball and read it the person next to you
- 71. Snowstorm • Write down one key thing you have learnt or will think about when you go back to your organisation? • Screw the paper up • On the signal, throw your snowball in the air. • Pick up a snowball and read it the person next to you • On your way out please hand the snowballs to us.
- 72. Thank you Dominic Cushnan, Ameet Bakhai, Andy Wilkins