AI in Healthcare Day Two


Ai in Healthcare 2018 Day Two

08:00 Registration opens

Day 2 Keynotes - Science disrupters

Abraham Heifets

Using artificial intelligence to discover new medicines

Patrick Keohane

Using pharma data for machine learning

Daniel Ray

AI – a game changer in healthcare?

  • Effective use case examples
  • Implications for health data and technology
  • Opportunity for AI and value for the NHS
  • What is NHS digital doing in this space?
Theo Blackwell

The Smart London Plan – meeting London’s healthcare needs

10:40 Networking Break

Saman Farid

Break out panel: Investment into AI for Healthcare

  • Panel session lead by 6 senior investors, actively investing within the AI for healthcare space
  • What do investors look out for in start-ups?
  • What are the current trends in AI for healthcare?
  • Where do we see the industry moving to in the next 5-10 years?
  • How can you gain investment?

Machine learning and data

Patient monitoring

Chaired by Wei-Yi Cheng, Senior Data Scientist, Roche Innovation Centre New York

Chaired by Jim Fackler, Associate Professor, John Hopkins School of Medicine


Machine learning for big pharma big data

  • How machine learning facilitated the development of digital biomarkers for more objective, continuous assessment of Parkinson's disease progression
  • Opportunities of bringing value from medical images, genomics, EMR for drug development
  • The key challenges of internalizing state-of-art machine learning capabilities on pharma big data

Building a Cognitive Hospital - How Alder Hey are using Artificial Intelligence to develop a hospital capable of thinking, sensing and feeling in order to care for the people within it

  • The magical world of a "Living Hospital” and improving patient experience
  • Adapting emerging technologies to enable a building to care for those within it both medically and holistically
  • Building AskOli - a conversational interface for patients and families
  • Practical implications of using AI and overcoming barriers to adoption in the NHS

Making good decisions with Bayesian AI

  • Bayesian/probabilistic machine learning models can incorporate all sources of uncertainty, including noisy data and misclassified outcomes
  • This uncertainty is propagated to give a realistic assessment of the confidence in a prediction, leading to better decisions
  • Examples from safety pharmacology will be used to demonstrate the advantages of these models

Artificial intelligence for diabetes control

  • Type 1 diabetes sufferers rely on multiple daily insulin injections to control their blood glucose levels
  • There is a clear need for more automated and intelligent systems to support people with type 1 diabetes to manage their glucose levels
  • The latest advances in intelligent decision support systems for type 1 diabetes management developed at Imperial College London will be introduced
  • Results of our latest clinical trial evaluating an AI-enhanced artificial pancreas system will be presented

Speaker TBC


Objective assessment of biological and chemical feature spaces towards drug combination synergy predictions using large-scale Oncology screening data

  • Assess biological feature spaces from a large-scale internal combination screening panel towards identifying novel synergistic combinations
  • Discuss validity and robustness of the Machine Learning algorithm by assessing translatability to other large-scale combination screens, as well as limitations of the approach
  • Discuss case studies across different modes-of-action mechanisms

AI at the bedside: from early recognition to value delivery

  • It is necessary but not sufficient to use “big data” and machine learning techniques to predict conditions best avoided (e.g. sepsis, hypotension, cardiac arrest)
  • It is further necessary but not sufficient to follow the above early detection with Augmented Intelligent Decision Support that more reliably delivers the correct therapies in a timelier manner
  • Decision support solutions must improve the value we deliver for the patient

13:00 Networking Lunch

Janette Rawlinson

Lunchtime break out panel: Patient involvement in precision medicine

Break out panel: Patient involvement in precision medicine
  • What are the benefits of involving the public in precision medicine
  • The patient experience with precision medicine and patient involvement
  • How clinicians can improve the patient experience
  • What are the barriers to involvement and how can they be overcome

Regulation and data for AI


Transforming healthcare delivery through artificial intelligence

  • History of Artificial Intelligence and Machine Learning and the applications within Health Care
  • An overview of Oklahoma State University’s (OSU) Center for Health Systems Innovation’s (CHSI) efforts to apply these approaches to the transformation of healthcare delivery with an emphasis on rural health care
  • Examples utilising the largest clinical database representing clinical information from 63 million patients collected over 16 years through electronic medical record systems

“Trust me, I’m a machine”: Public confidence and AI using patient data

  • Challenges for public, patient and clinicians’ confidence in the use of AI tools in healthcare and research
  • Research into public attitudes to ML and other data-driven technologies in health
  • What helps create a trustworthy system for using patient data – and what doesn’t

Machine learning in genomics and proteomics

  • See how is deep learning being applied
  • See how relationships in proteomics/genomics are being learned via network structure
  • See applications with potential clinical implications

Disrupting the invasion of AI in healthcare – examples from cardiology

  • Machine Learning and Digital Technologies have been common place in the management of patients with cardiovascular disease.
  • Innovations in cardiology bring convenience, improved outcomes and are soundly researched for safety and efficacy – examples of good practice
  • Security risks, employment impact of automation, alarm fatgue regarding events captured by monitoring technologies all pose issues
  • Does ML solve the human need condition or provide a lower cost, lower quality substitute to the vulnerable?
  • ML, digital technologies and healthcare professionals symbiosis – proposed meeting of minds

Multi-modal deep learned biomarkers of aging and disease

  • Using age as a feature to aggregate static and dynamic biological data
  • Assessing the biological relevance of the chronological age predictors
  • Using the biological age predictors for target identification and for clinical trials enrollment 
  • Using networks trained on age to generate the virtual synthetic patient data​

15:50 Networking Break

Closing keynotes – AI, pharma and healthcare working together

17:35 End of Conference

last published: 15/May/18 12:55 GMT




Sign Up for Event Updates

World Precision Medicine Congress


Erica Baeta
+44 (0)207 092 1152


Joan Shutt
t/ +44 (0)207 092 1134


Issa Mauthoor
+44 (0)207 092 1257