Accelerating Value, Reducing Risk: SLUHN AI Innovation Lab

Through the Innovation Lab, we explored early-stage AI concepts related to documentation, abstraction, and operational workflows in a controlled, governance-focused environment. While the work was exploratory in nature, it provided valuable insight into potential opportunities and considerations as we continue to thoughtfully evaluate this emerging technology.

CHERYL DAVIDSON,MSN, RN, Sr. Network Director, Data Abstraction & Registry Services

Background and Challenge 

Like many large health systems, St. Luke’s University Health Network (SLUHN) faced growing operational and financial pressures alongside a surge of AI vendors promising automation across revenue cycle, patient access, registries, and contract management. While leadership recognized the potential of AI and advanced analytics, they also faced significant risks: 

• High uncertainty and risk in committing to multi-year AI platform investments without clear evidence of value, data readiness, or workflow fit 

• Vendor overload, with limited internal capacity to objectively compare “build vs. buy” options or evaluate Epic-native capabilities versus niche AI products 

• Governance and trust concerns, including the need for strong data stewardship, compliance, and human oversight in AI-enabled workflows 

SLUHN needed a structured way to move from AI curiosity to action—without disrupting care, compromising governance, or over-investing in unproven solutions. 

Solution: The AI Innovation Lab

In partnership with Tegria, SLUHN launched an AI Innovation Lab—a sprint-based, governed framework designed to safely test and validate AI use cases before scaling. 

The Innovation Lab provided SLUHN with a low-risk environment to: 

• Validate AI opportunities through focused proofs-of-concept (POCs) 

• Demonstrate measurable clinical, operational, or financial value 

• Maintain Epic-first principles and data governance standards 

• Create clear go/no-go decisions and scalable roadmaps 

Each Innovation Lab cycle moved from intake to insight through executive governance, structured prioritization, and close collaboration between SLUHN leaders, subject matter experts, and Tegria consultants. 

Innovation Lab Cycle

1. Identify: Where can AI make a measurable difference?
2. Integrate: Bring together the right data, tools, and expertise to explore the problem.
3. Validate: Test the idea in a controlled, real-world setting—small scale, quick results.
4. Reflect: What worked, what didn’t, and what should change?
5. Operationalize: Successful POCs become roadmaps for enterprise adoption.

Approach

Sprint-Based POC Model

• Each POC was scoped to a defined business problem and executed over a four- to five-week sprint, with clear success criteria focused on feasibility, impact, and risk. 

Cross-Functional Teams

• POCs brought together operational leaders, IT and analytics teams, Epic and data platform experts, and Tegria specialists across revenue cycle, access, care operations, and data and analytics. 

Embedded Governance 

• Existing SLUHN committees provided oversight and prioritization, ensuring alignment with enterprise strategy, Epic-first standards, and compliance requirements. 

Impact and Outcomes

The AI Innovation Lab delivered value beyond individual POCs, creating a repeatable enterprise model for AI evaluation and adoption. 

Risk Reduction 

• Scoped data access, short POC timelines, and human-in-the-loop review minimized operational and compliance risk. 

• SLUHN gained clarity on feasibility and limitations before making investment decisions. 

Epic-First and Platform-First Insights 

• Multiple POCs demonstrated that Epic-native workflows, combined with analytics and AI within existing platforms (e.g., Databricks, Power BI), can meet or exceed the value promised by external AI vendors. 

Actionable Roadmaps 

• Each POC concluded with clear recommendations, value estimates, and next-step options—whether to scale, refine, or stop. 

Improved Data Governance Awareness 

• The lab surfaced gaps in documentation standards, metadata, and data stewardship, enabling SLUHN to address foundational issues critical for safe AI adoption. 

Looking Ahead 

Through the AI Innovation Lab, SLUHN moved from fragmented AI experimentation and vendor noise to a governed, enterprise-ready innovation pipeline. The Medical Registry POC, alongside other lab initiatives, helped leaders understand where AI can deliver real value today, and what groundwork is required to scale tomorrow.