Key Insights from the St. Louis Healthcare Data Series

On October 8th, healthcare leaders from across the region gathered for the latest edition of the St. Louis Healthcare Data Series, hosted at Husch Blackwell’s stunning office in Clayton. The event focused on Balancing Executive Pressure for AI & Data Wins with the Need for Sustainable Strategies, a topic that resonates with healthcare leaders everywhere as they navigate the fast-paced world of AI and data.

While many of the attendees represented organizations from St. Louis, our blog readers span across the nation, and we’re excited to share the insights from this local event more widely. This article is for healthcare leaders and professionals anywhere who are interested in leveraging AI and data for impactful, sustainable change.

Panelists and Key Perspectives

The panel brought together experts with deep experience in healthcare and data:

• Tricia Copeland, Chief Information and Risk Officer at McCormack Baron Salazar, where she oversees information management for mixed-income housing developments.
• Dr. Josh Wymer, Chief Health Information and Data Strategy Officer at the Missouri Department of Health and Senior Services, who focuses on how data, people, and processes support the health of Missourians.
• Jacob Gower, Director of Data and AI at Technology Partners, moderated the event. With a background in healthcare data engineering and AI consulting, Jacob guided the discussion to focus on practical, sustainable strategies.

Although Dr. Leslie Hinyard from Saint Louis University was unable to attend, the conversation remained dynamic, with panelists sharing valuable insights on how to balance the drive for immediate AI wins with the need for lasting, long-term data strategies.

1. Building the Right Data Strategy: The Foundation of AI

The panel quickly agreed that no AI initiative can succeed without a strong data strategy. While many healthcare organizations still rely on traditional data warehouses, the conversation centered on the importance of moving toward centralized, cloud-based platforms. These platforms enable organizations to better manage data at scale, increase security, and allow AI models to thrive in environments that support new varieties of data.

The takeaway: AI success starts with a strong, scalable data foundation, and healthcare organizations that haven’t yet made the leap to the cloud will find it increasingly difficult to compete in the AI-driven future.

2. Scaling AI Care Models: The Customization Challenge

As AI continues to evolve, its application in care models is a promising area of development. However, panelists highlighted that scalability is a challenge. Many healthcare systems find that AI care models, though valuable, fail when applied across multiple hospital systems due to variations in demographics and operational practices.

The solution? Training AI models on hospital-specific data. By customizing models to meet the unique needs of each hospital, healthcare organizations can ensure that AI delivers the best possible outcomes for both patient care and operational efficiency.

3. Securing AI Model Training: Overcoming Data Sensitivity

Security concerns are at the forefront of any AI initiative, particularly in healthcare, where sensitive patient data is involved. The panel discussed several approaches to training AI models securely without compromising data privacy:

• Containerized AI Models: One approach involves deploying AI models within the healthcare organization’s own cloud environment. This keeps sensitive data secure but may require vendors to maintain multiple versions of their models across different organizations.
• Anonymized or Synthetic Data: Another option is to use anonymized or synthetic data for model training. Tools like MDClone can generate high-quality synthetic data that maintains statistical relevance, making this a promising option for organizations that want to protect patient privacy.
• Federated Learning: While still in its early stages for many organizations, federated learning allows models to be trained across multiple institutions without sharing sensitive data. This method holds potential for the future as healthcare organizations explore ways to collaborate without compromising security.

4. Workflow Integration Specialists: The Human Side of AI

The panel emphasized that AI is not just about technology—it’s about processes and people. Rather than rushing to implement AI products, healthcare organizations should first invest in workflow integration specialists who can analyze current processes and identify areas where AI can make the biggest impact.

These specialists, who often have backgrounds in human-centered design, can help organizations document opportunities for improvement and identify AI use cases that not only justify the technical investment but also bring measurable ROI. By focusing on early success stories, healthcare leaders can build trust around AI and ensure that their investments are both strategic and sustainable.

Expanding the Conversation Nationwide

Although this discussion took place in St. Louis, the insights shared by our panelists are relevant to healthcare organizations across the country. As more organizations look to balance the pressure for immediate AI and data wins with the need for sustainable, long-term strategies, these takeaways offer a practical roadmap.

Looking Ahead: What’s Next for Healthcare AI?

The conversation ended with optimism about the future of healthcare AI. Although challenges remain—such as scaling AI care models, securing data, and aligning AI initiatives with broader business goals—there’s a growing recognition that the key to success lies in building a strong data foundation and focusing on workflow integration. As healthcare organizations continue to explore AI and data solutions, the future promises even more innovative and impactful applications.