July 24-26, 2018

Boston, USA

Day One
Wednesday, 25th July, 2018

Day Two
Thursday, 26th July, 2018

Breakfast & Networking

Chair’s Opening Remarks

Applying AI to Make Precision Medicine a Reality: The Intelligent Trial


Introduction and Purpose:

Slowly but surely, AI technologies are revolutionizing health care and advancing the clinical trial process. Cutting costs, improving trial quality, and reducing trial times by almost half are some of the promising potentials of an AI-driven clinical trial. Furthermore, as a disruptive technology, AI is an opportunity to precision medicine as it can deeply investigate  the roots of diseases and treatments and move the industry away form its current “one-size-fits-all” strategy.

Case Study: Machine Learning Based Patient Subgroup Identification for Precision Medicine

  • Jie Cheng Associate Director, Data & Statistical Sciences, Abbvie


  • What is patient subgroup identification?


  • What are the challenges and how machine learning can help?


  • How to incorporate patient subgroup identification in clinical development?


Reduction of High-dimensional Clinical RNA-seq Data by Deep Learning: Methods and Its Practical Application


  • Challenge of high-dimensional transcriptomic data in machine learning
  • Methods for reduction of high-dimensional transcriptomic data
  • Deep learning based approached for reduction of clinical RNA-seq data

Case Study: The Application of Machine Learning to the Development of Novel Imaging Biomarkers


  • AI can be used to create novel biomarkers from routine clinical scans


  • Valuable AI tools can be built using historical trial data


  • Data sharing and collaboration obstacles must be overcome


Case Study: The Application of Machine Learning Approaches for Enabling Translational Medicine

  • Raj Bandar Head of Data Sciences & Analytics , Bill & Melinda Gates Medical Research Institute


  • Application of machine learning approaches to understand patient journey, predict outcomes and elucidate biological mechanisms driving certain outcome and adverse events


  • The Use of patient electronic health records to explore advanced analytical methods


  • Developing the appropriate models for trial simulations requires access to a wide range of data from historical studies and patient health records


  • A showcase of our effort in Bayesian learning approaches to find and integrate historical clinical data in real time as well as our elastic computing platform development

Morning Refreshments & Networking

Current & Future Approaches to Real World Applications of AI


Introduction and Purpose:

The promise is that AI systems will deliver easily accessible, less expensive and higher quality care to patients while allowing for efficient data capturing with direct involvement of not just the clinicians but the patients themselves. Hence, further emphasising on the importance of not only optimization of candidate selection for clinical trials but also the pressing need for effective and practical connection between patients, clinicians and drug developers.

Case Study: Innovative Biomarkers in the Presurgical Evaluation of Children with Epilepsy

  • Christos Papadelis Head, Laboratory of Children’s Brain Dynamics, Boston Children’s Hospital; Assistant Professor of Pediatrics , Harvard Medical School


  • High frequency oscillations recorded invasively and noninvasively are promising interictal biomarker of epileptogenicity


  • AI can boost the development of epilepsy biomarker through training in large datasets from many epilepsy centres


  • Development of sensitive and specific epilepsy biomarker can dramatically improve the surgical outcome in patients with epilepsy

Case Study: Application of Deep Neural Networks for Developing Digital Biomarkers from Speech and Accelerometer Data


  • Deep learning enables development of biomarkers from raw sensor data without
    labour intensive signal analysis


  • The obtained biomarkers are competitive with hand-crafted ones


  • Data augmentation is critical

Case Study: Application of Medical Image Machine Learning in Clinical Development

  • Sehyo Yune Director of Research Translation, Laboratory of Medical Imaging & Computation, Massachusetts General Hospital


  • Medical imaging in clinical trials
  • Artificial intelligence and machine learning in medical imaging
  • Application of medical image AI in clinical development

Networking Lunch

Zooming the Spot Light on the Critical Issues: Voice, Exchange & Evaluate Ideas

Discuss: How AI & Machine Learning Could Present an Opportunity to Lower the Failure Rates of Clinical Trials Compared to the Contemporary Approaches?


  • What is the impact of AI on traditional drug development?


  • What is the regulatory and Commercialization outlooks of potential AI-developed drugs?


  • What is the impact of AI to clinically develop “value-based” therapies?

Discuss: The Application of AI & Machine Learning in Clinical Development Requires Large Amounts of Quality Data: How to Develop the Best Model of Practices in Terms of Data Access and Sharing Amongst the Industry?

  • Jie Cheng Associate Director, Data & Statistical Sciences, Abbvie
  • Leonardo Rodrigues Senior Director, AI & Machine Learning, Berg
  • Raj Bandar Head of Data Sciences & Analytics , Bill & Melinda Gates Medical Research Institute
  • Ed Addison CEO, Cloud Pharmaceuticals


  • What datasets are readily available for AI application in clinical development and what are the “low-hanging fruits”?


  • How to aggregate health data in a secure, trusted and automated style using AI & machine learning to benefit clinical research?


  • How practical is it to form comprehensive data partnership networks amongst the leading pharma and biotech companies and what is the way forward?

Chair’s Closing Remarks & Close of Inaugural AI Pharma Innovation: Clinical Development Summit 2018