AI in Healthcare Day One


AI in Healthcare 2018 Day One

08:00 Registration opens

Opening keynotes – AI in healthcare


Opening remarks from Terrapinn

Georgia Papathomas

How will AI disrupt the healthcare industry?

Rangaprasad Sarangarajan

Combining patient biology and AI based analytics for therapeutic discovery

Philipp M Diesinger

Artificial Intelligence transforming healthcare and the pharmaceutical industry

  • A real-world application of artificial intelligence that helps Boehringer Ingelheim save hundreds of thousands of lives across Europe and globally
  • Emphasize the interface between AI and human roles and discuss the transformation the pharmaceutical industry is facing
Baiju Parikh

Cancer genomics and cognitive computing: bending the analysis curve to accelerate personalised medicine

  • 10 minutes vs. 160 expert human hours: interpreting whole genome sequencing data for a glioblastoma patientWatson for Genomics was tested on 1,018 retrospective patient cases. In more than 300 cases
  • Watson found clinically actionable therapeutic options that the physicians had not identified
  • Expediting patient screening for clinical trial eligibility down from 110 minutes to 24 minutes

10:40 Networking Break

round tables

AI round table discussion session

Table 1: Transforming health care delivery through artificial intelligence
William Paiva

William Paiva, Executive Director, Center for Health Systems Innovation, Oklahoma State University

Table 2: Is AI safety akin to clinical safety? Current issues in healthcare governance
Sam Smith

Sam Smith, Co-ordinator, MedConfidential

Table 4: What will the cognitive hospital of the future look like?
David Cole

David Cole, Head Innovation Lead, Europe, IBM Watson Health


Speed Networking

13:00 Networking lunch

AI pharma and research - drug discovery and biomarkers

Healthcare delivery - using AI in hospitals


Integrative big data for drug discovery

  • Innovation in drug discovery is becoming increasingly challenging requiring comprehensive objective decision making
  • Cancer therapy is becoming increasingly complex with drug resistance and tumor heterogeneity and tumor evolution
  • We have developed integrative data and machine learning / AI approaches to address some key bottle necks in drug discovery and adaptive cancer therapy
Using AI in hospitals

Our partnership with Deepmind

  •     Current use of the Streams application
  •     Future development of the Streams application

Strategy for AI-driven drug discovery

  • Outline of GSK’s in-silico drug discovery strategy
  • How can AI be integrated into the drug discovery workflow?
Using AI in hospitals

Data analytics and AI for healthcare transformation

  • The healthcare crisis can only be addressed if healthcare model is transformed
  • Value-based healthcare is the new paradigm
  • Data are essential to implement the new model
  • Business intelligence and AI are two basic tools for value-based healthcare
  • A specific architecture for simulation, emulation and big data set generation will be presented

Improving translation rates into the clinic through AI modeling of ADME/Tox

  • Discuss the current state-of-the-art modeling of ADME and toxicity
  • Examine some of the challenges/wins in applying AI modeling to ADME and toxicity
  • Explore case studies involving the use of these models to solve key challenges

15:55 Networking Break


Reimagining drug discovery through AI

  • AI has vast potential, but we’re still at least a decade away fully in silico drug discovery efforts
  • Massively expanding and accelerating traditional approaches like phenotypic screening provides a feasible near-term solution to bringing substantial improvements to the efficiency of discovery and development efforts
  • In this talk, I will detail how Recursion sees the use of AI in discovery and development, as well as describe in some detail our technical strategies to accelerate discovery using AI, including our image-based phenotypic screening platform
  • Overviews of internal successes in rare disease, immunology, and immuno-oncology will be discussed. Partnerships with large pharmaceutical companies including Sanofi and Takeda may also be detailed
Using AI in hospitals

The Wearable Clinic: Digital phenotyping using mobile health technologies

  • People with long-term conditions interact with the NHS through rigid pathways that poorly match the dynamic nature of their condition
  • Our vision is to enable new forms of collaborative care through dynamic personal care plans and mobile/wearable health technologies
  • Focusing on patients with servere mental illness, we present machine learning algorithms to assess social functioning and risk of psychotic relapse using community-acquired geolocation data

Digitised biology and image analysis augment drug discovery

  • What does it mean to ‘see’ for a machine?
  • Impact of AI in image and video analysis for pharmaceutical industry
  • Our work in image analysis is currently focusing on a range of enabling topics in computer vision, image processing and pattern recognition with the goal of automation
Using AI in hospitals

Artificial Intelligence at the point of care

  • GOFAI: Good Old Fashioned AI has had many successes but has struggled to get professional adoption in healthcare
  • NEWFAI: New Fangled AI (or machine intelligence) promises a revolution but will face its own challenges
  • Human centred cognitive systems combine the strengths of both – time to invest?

Drug discovery to improve patient outcomes

Lee Chapman, Global Head of Discovery & IP, Celixir
Using AI in hospitals

Building explainable models of disease trajectories

  • Data can be collected in a multitude of ways: one-off population studies, in-clinic patient monitoring, personalised health apps ...
  • There is an urgent need to focus on models that predict early stages of disease for more effective intervention
  • We are developing novel AI methods that allow us to integrate data to build personalised trajectories of patients with a focus on early stages
  • Building explanation into complex models will be necessary to navigate new data protection laws
  • We suggest a trade-off: Techniques that exploit the power of "deep" methods but incorporate explanation

Modeling pharmaceutical drug discovery datasets with machine learning

  • Application of neural networks to pharmaceutical datasets
  • Comparison of deep-learning based approaches to baseline methods
  • Discussion of relevant issues when applying machine learning in an industrial setting
Using AI in hospitals

Extended Q&A


18:00 Offsite drinks reception

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