Machine Learning Healthcare




Machine Learning Healthcare

10 December 2018

New York

Added 01-Jan-1970

Healthcare is a field of incredible challenges and life changing opportunities and machine learning can help many facets.

We have THREE excellent speakers from Flatiron Health, Pfizer, and Talkspace covering:

  • Machine learning techniques to reduce bias in Electron Medical Records (EMR) analysis
  • ML improving ROI for drug R&D and even design of new drugs
  • Psychotherapy techniques leveraging a HIPAA-BAA compliant chatbot in the cloud

Please note the location at Betaworks Studios.

Event Schedule:

6:00 PM Networking & Food, Beer
6:30 PM Kickoff & Sponsors (IBM, Betaworks Studios)
6:40 PM Lucy He, Flatiron Health
7:15 PM Michael Frank, Pfizer
7:40 PM Nick Lamm, Talkspace
8:00 PM Extended Q&A and Open Member Discussion

Presentation Summaries:

1) When machine learning (ML) techniques are applied to electronic medical records (EMRs) for research, studies can be larger or done faster than non-automated studies on EMR data. However, using ML may add clinical bias into the data extracted from EMRs. In this session, we will present how to limit and measure such bias, which enables ML to be used in high-stakes research areas like cancer.

2) Across the value chain of drug discovery, AI/ML is being tested as a way to improve the return on R&D investment and bring drugs to patients faster. In this talk we will explore our experience and the challenges faced with applying ML across the phases of drug development. We will look at how it can be used as a decision support tool, and lastly we will explore the disruptive impact that generative methods are having in creative fields and art, including the art of designing a drug.

3) Psychotherapy, as a primarily conversation-based method to treat mental illness, is a tempting arena for conversational AI. In fact, one of the earliest chatbots, ELIZA, was believed to simulate (or perhaps parody) the interactions between a psychotherapist and client. At Talkspace, the leading online therapy platform, we are interested in how traditional psychotherapy techniques can be translated to a digital medium. We will discuss how we designed a HIPAA-BAA compliant chatbot that takes cues from psychotherapy, and how these principles might be applied to other contexts.

Speaker Bios:

1) Currently a software engineer at Flatiron Health, Lucy He is on a team applying machine learning in oncology to ultimately improve patient care and accelerate cancer research. The team helps build machine learning infrastructure for the company, and data products for research partners. Prior to Flatiron, Lucy was at Jigsaw, Alphabet’s’ geopolitical think tank and technology incubator, and at Google for Nonprofits. While at Jigsaw, she mostly focused on projects concerning digital censorship.

Lucy received a B.A. in computer science from Columbia University, though spent her first year in college studying biology.

2) Michael is a Director of Strategy in Pfizer’s World Wide R&D organization. His group’s mission is to advance strategy and innovation in pharmaceutical development across all therapeutic areas. His current focus is expanding the role of Artificial Intelligence in drug discovery to increase productivity and identify new therapies to help patients.

Prior to Pfizer, Michael was a Strategy Consultant at Strategy& (a PWC company) and PA Consulting. He was also VP of technology at Phosplatin Therapeutics, a NYC biotech startup, and Director of BD and Commercial Operations at Forma Therapeutics (formerly CombinatoRx Singapore).

Michael lives in Brooklyn, and in his spare time enjoys helping entrepreneurs get their ideas to market; with several successful launches, from KickStarter to biotech IPOs.

His own inventions include a kitchen gadget for SkyMall, a dual language eReader platform, and several medical diagnostics and products. He has an MBA from Columbia Business School, and MS in Biomedical Engineering from Boston University, and a BA in Biology from University of Michigan.

3) Nick Lamm is a Data Scientist at Talkspace, the leading online therapy platform. As a former marketer, Nick is interested in the intersection between machine learning and user-facing products. He is a strong supporter of Talkspace’s mission to increase access to mental healthcare.