2019 Data Science Summit

Data access and liquidity in health information systems continue to create challenges for AI development. To better understand the technical and ethical issues confronting both institutions and AI developers in the US and internationally, join us to learn up to date information and hear about potential solutions.

2019 Data Science Summit: Data Access in Healthcare – Implications for the AI Ecosystem

• The current state of data access and liquidity in the US and abroad
• The ethical issues regarding data ownership, privacy, sharing, bias and stewardship
• The democratization of AI using tools built for practicing radiologists to facilitate distributed AI development, data sharing and model validation while keeping patient data on premises. Any institution will be able to
• Contribute to AI development by annotating local cases for AI development
• Participate in independent testing and validation of AI algorithms
• Institute safeguards to protect developers' intellectual property
• Create AI solutions to meet local institution needs
• More

The Summit brings together thought leaders and attendees for a robust discussion and evaluation of where we are today and what you can expect in the future. Whether you are a developer, looking for insights from radiology leaders, or a radiologist with an informatics background seeking best practices, the Summit can help you understand how data access and liquidity will impact you in the coming year, and provide strategies for data sharing to overcome potential barriers to AI development.

In Depth Discussion Includes:

• Full-day focused content
  • Brief keynote presentations
  • In depth panel discussions
• Conversations with faculty and experts
• Opportunity to forge long-term relationships with peers

Those who should attend the Summit include:

• Industry partners and developers interested in pursuing data sharing arrangements
• Clinical informaticists and fellows, residents, enterprise IT, clinical applications professionals, technologists, and all those who want to learn more about AI data sharing, the ethics of AI and the democratization of AI in health care

Upon completion of the Summit on Data Access in Healthcare, the participant will be able to:

• Outline the current state of data liquidity in health information systems define the challenges for access to data for AI development
• Understand the ethical issues regarding data ownership, privacy, sharing, bias and stewardship
• Describe how to manage issues related to conflicts of interest
• Outline the varying international standards for data access and sharing
• Define the opportunities and challenges for AI democratization using cross-site learning tools for distributed computing and on-premises data for algorithm training

Data Liquidity and Data Access: Setting the Stage
10:00 AM – 10:35 AM

• Welcome (Bibb Allen)
• Introduction to the Challenges of Data Privacy, Access and Liquidity (Amy Kotsenas)
• Data Sharing for Research and Clinical Trials: Lessons Learned (Etta Pisano)
• Data Liquidity, Data Standards and CDEs – Normalizing the Data (Chris Treml)

The AI Ecosystem Meets HIPAA (Moderator – Kotsenas)
10:35 AM – 12:00 PM

• Data Ownership and Regulatory Implications for Data Sharing in the US (Ben Harvey)
• Data Ownership and Regulatory Implications for Data Sharing Internationally (Erik Ranschaert)
• Data Privacy and De-identification: How Far Do We Go? (Steve Langar)
• Patient Perspectives on Privacy and Data Sharing (Andrea Borondy Kitts)
• Data Sharing and Data Use Strategies in AI (Tessa Cook)

Panel Discussion (35 minutes):

ACR Data Science Institute Overview and Updates: Democratizing AI (Moderator Allen)
12:00 PM – 12:30 PM

• What’s New at the ACR Data Science Institute (Bibb Allen)
• Democratization of AI (Keith Dreyer)

Lunch and Networking
12:30 PM – 1:30 PM

A Look to the Future: Distributed Learning for AI Model Development: Training, Testing and Validation (Moderator – Allen)
1:30 PM – 3:15 PM 

• Distributed Learning Architecture – Crowd Sourcing AI to Radiological Practices (Mike Tilkin)
• Data Needs in Algorithm Training (Tessa Cook)
• Distributed Learning in AI Model Development (Jayashree Kalpathy-Cramer)
• Independent Validation of AI Algorithms: Centralized and Distributed Solutions (Laura Coombs)
• Protecting Intellectual Property in Distributed Learning (Ben Harvey)
• A Look to the Future (Amy Kostenas)

Panel Discussion: (40 minutes)

Break and Networking
3:15 PM – 3:30 PM

Ethics of AI in Healthcare: Aspirational Goals for Data Sharing, Algorithm Development and Radiological Practice (Moderator  – Geis)
3:30 PM – 5:00 PM

• Overview of Ethical Considerations in AI Development (Raym Geis)
• Managing Conflict of Interest in Algorithm Development (Amy Kotsenas)
• Mitigating Bias In Algorithm Development 
• Ethics in Algorithm Training (Jayashree Kalpathy-Cramer)
• Ethics of AI Use in Radiological Practice (Nina Kottler)

Panel Discussion: (40 minutes)

Bibb Allen
Keith Dreyer
Amy Kotsenas
Raym Geis
Tessa Cook
Ben Harvey
Etta Pissano
Nina Kottler
Steve Langar
Erik Ranschaert
Andrea Borondy Kitts
Jayashree Kalpathy-Cramer
Mike Tilkin
Laura Coombs
Chris Treml