Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Medical AI & Healthcare Support

Medical AI & Healthcare Support

Artificial Intelligence, once a mere curiosity, is gaining a great deal of interest in the field of medicine. In this talk we will explore some of the practical applications of medical AI, the research surrounding this exciting field and the potential for AI to be utilised as a support tool in healthcare and medicine. The talk will take high level view of the technology and it’s application as apposed to a low level technical analysis, making it accessible to everyone.

Nader Al-Shamma

February 02, 2017
Tweet

More Decks by Nader Al-Shamma

Other Decks in Technology

Transcript

  1. Copyright © 2017 Nader Al-Shamma 1 Medical AI & Healthcare

    Support Nottingham AI February 28, 2017
  2. Copyright © 2017 Nader Al-Shamma 2 Why Care About Chronic

    Illness? Long term chronic disease is one of the biggest challenges health systems, globally, will face during the twenty-first century.
  3. Copyright © 2017 Nader Al-Shamma 3 What we are dealing

    with 89% of Deaths Each Year ~15 Million in England 12% of Premature Deaths “Tobacco use, physical inactivity, the harmful use of alcohol and unhealthy diets all increase the risk of dying from an NCD.” *UK Figures
  4. Copyright © 2017 Nader Al-Shamma 4 Three Areas of Focus

    Diet and Nutrition Lifestyle and Self-management Patient Monitoring and Insight
  5. Copyright © 2017 Nader Al-Shamma 5 Artificial Intelligence and Medicine

    • Researchers have been talking about this since the 1950s • Successfully implimented in 1976 • Research is fosucing on a number of areas including diagnostics, prognostics and clinical support. • Research is looking into specific diseases.
  6. Copyright © 2017 Nader Al-Shamma 6 AI in clinical care

    Johnson et al. (2016) conducted a thorough review of Machine Learning in critical care (ICU). The challenge: • Data comes from a variety of different departments • Data is often missing or corrupted • It’s an environment that lends itself to human error Conclusion: • Machine Learning is effective in gathering, analysing, interpreting and presenting a wide range of clinical data in a relevant and useable format.
  7. Copyright © 2017 Nader Al-Shamma 7 AI in Diagnostic Artificial

    Neural Network (ANN) and a Probabilistic Neural Network (PNN) used to diagnose IBD. 98.31% Rate of Accuracy.
  8. Copyright © 2017 Nader Al-Shamma 8 AI in Prognostics The

    Fight Aginst Drug Resistant Bacteria. Hidden Markov and Fuzzy logic can help determine proper antibiotic dosage.
  9. Copyright © 2017 Nader Al-Shamma 10 AI in Healthcare Support

    Case Based Reasoniong is being used in the fight against diabetes.
  10. Copyright © 2017 Nader Al-Shamma 11 AI in Healthcare Support

    AI and the Xbox Kinect helping to treat Parkinson's
  11. Copyright © 2017 Nader Al-Shamma 12 Where are we going?

    “We are moving towards an era of Cognitive Systems” Dr Laura Haas, IBM Future Technology Conference San Francisco 2016
  12. Copyright © 2017 Nader Al-Shamma 13 References: • SAIConference, Cognitive

    Computing - Laura Haas, IBM. 2016. • ‘WHO | Noncommunicable diseases’, WHO. [Online]. Available: http://www.who.int/nmh/en/. [Accessed: 27-Feb-2017]. • ‘2010 to 2015 government policy: long term health conditions - GOV.UK’. [Online]. Available: https://www.gov.uk/government/publications/2010-to-2015-government-policy-long-term-health-conditions/2010-to-2015-government- policy-long-term-health-conditions. [Accessed: 27-Feb-2017]. • ‘NHS Outcomes Framework 2016 to 2017 - GOV.UK’. [Online]. Available: https://www.gov.uk/government/publications/nhs-outcomes- framework-2016-to-2017. [Accessed: 27-Feb-2017]. • ‘The Evolution of Bacteria on a “Mega-Plate” Petri Dish (Kishony Lab) - YouTube’. [Online]. Available: https://www.youtube.com/. [Accessed: 28-Feb-2017]. • A. P. Eremeev and V. N. Vagin, ‘Common Sense Reasoning in Diagnostic Systems’, 2011.