Intelligent Systems and Inflammatory Bowel Disease: Exploring the Potential for Outpatient Support

This talk outlines the findings from research which seeks to understand how AI and ML have been applied to the realm of patient care within the context of chronic disease, with a specific focus on Inflammatory Bowel Disease (IBD).

Transcript

1. Nader Al-shamma Dr Ali Jwaid Intelligent Systems and Inflammatory Bowel Disease: Exploring the Potential for Outpatient Support Computing Conference 2017 18-20 July 2017 | London UK
2. What is the problem? “[…]a diagnosis of IBD is not a death sentence, but it can often be a life sentence. [..] For patients and their families, the impact of a chronic disease will bring changes in their personal, social and emotional lives.” Thomas, G (2008) Counselling and Reflexive Research in Healthcare : Working Therapeutically with Clients with Inflammatory Bowel Disease, Jessica Kingsley Publishers, London. The solution? Can advances in machine learning and artificial intelligence help tackle some of the challenges faced by those affected by IBD?
3. How big is the problem? Global rates of IBD are increasing, as a result the illness is fast becoming an international problem. 3 Million Across Europe €4.6 to €5.6 Billion Per year 300,00 0 In UK
4. Areas of focus: Diet and Nutrition Lifestyle and Self-management Patient Monitoring and Insight
5. Applied AI and Machine Learning: Application Domain Prognostics Diagnostics Outpatient Support (?)
6. AI and Machine Learning in Action Diagnosis: • Matalka et al. (2013) Artificial Neural Network (ANN) and a Probabilistic Neural Network (PNN) used to diagnose IBD. • 98.31% Rate of Accuracy. Prognosis: • Waljee et al. 2010 used Random Forest to analysis laboratory data to determine the success of thioprine treatment. Results: Area under Receiver Operating Characteristic (AUROC) curve of 0.856 vs AUROC of 0.594 from traditional testing techniques. • Hardalaç et al 2015 used ANN to determine the success of treating patients with the immunosuppressant drug Azathioprine. Taking into account patient health and lifestyle metrics. Results: predict the efficacy of Azathioprine use with a precision of 79%. Outpatient Support: ?????
7. Next Steps:
8. Contact Nader Al-Shamma Nader.alshamma@ieee.org twitter.com/nadershamma Nadershamma.com uk.linkedin.com/in/nadershamma
9. Credits http://fontfabric.com/aleo-free-font/ Aleo Free Font Lato Font http://www.fontsquirrel.com/fonts/lato FONTS Images: • E-healthcare concept with hand holding smart phone #115660297, Author: iconimage , https://en.fotolia.com/id/115660297 Icons: • Run, apple, google pulse https://materialdesignicons.com/ • Search , timeline, groups https://material.io/icons/