Introduction
Artificial intelligence promises to revolutionize healthcare delivery, yet most organizations struggle to move beyond isolated pilot projects. The gap between potential and practice stems not from technological limitations but from fragmented implementations that create silos rather than solutions. Industry leaders at Becker’s Fall 2025 Payer Issues Roundtable revealed how strategic infrastructure investments and deeper clinician-engineer collaboration can finally unlock AI’s transformative capabilities at scale.
Representatives from Raport and Precision Medicine Diagnostic Partners shared critical insights about building reliable, repeatable AI systems that address real clinical challenges. Their discussion highlighted why understanding fundamental AI concepts, eliminating redundant development, and prioritizing human expertise remain essential for sustainable healthcare automation.
Defining AI Agents Versus Automated Workflows
What True AI Agents Actually Do
Healthcare organizations frequently mischaracterize their technology investments as “AI agents” when they’ve actually deployed rigid automated workflows. Nathan Feldt, chief technology officer and co-founder of Raport, explained that authentic AI agents interpret contextual information, develop strategic plans, utilize appropriate tools, and act autonomously toward defined objectives.
“What you’re expecting is that the agent can come up with its own solution to the problem,” Mr. Feldt explained. “In contrast, a workflow is more predefined — you know exactly what the criteria are and the steps to meet them.”
Why This Distinction Matters Clinically
The difference carries significant implications for complex healthcare applications. Workflows operate within predetermined parameters and execute scripted sequences. While valuable for standardized processes, they lack the adaptability required for nuanced clinical decision-making where patient presentations vary significantly.
Many current market solutions claim agent capabilities while delivering workflow automation built in low-code or no-code environments. These systems lack the computational rigor necessary for clinical scenarios demanding reliability and repeatability across diverse patient populations and care settings.
The Fragmentation Challenge in Healthcare AI
Vendor Silos Create Operational Inefficiencies
Specialized AI vendors typically develop narrowly focused tools that operate independently, creating technical debt and operational complexity. “Managing these systems adds complexity, and honestly takes away a little bit of the benefits that you’re getting from these systems when you’re having to switch off to all these different integrations,” Mr. Feldt noted.
Redundant Development Wastes Resources
Mauricio Garcia Jacques, MD, CEO and co-founder at Raport, identified context retrieval as a particularly problematic area. This essential function that powers clinical AI workflows gets repeatedly rebuilt across different vendor platforms, consuming resources while producing incompatible data models that prevent interoperability.
“Ultimately, what we hope to avoid is rebuilding infrastructure only to achieve similar capabilities between these systems,” Dr. Garcia Jacques explained. This redundancy increases costs, extends implementation timelines, and limits the scalability potential that makes AI investment worthwhile.
Building Unified AI Infrastructure for Scale
Creating Shared Ecosystems for Intelligence
Raport’s platform addresses fragmentation by establishing common ecosystems where tools, subagents, and workflows share contextual information and outputs. This architectural approach enables intelligent automation that learns and improves across interconnected processes.
For example, intake summarization workflow outputs can directly feed prior authorization processes, eliminating redundant documentation while maintaining clinical context throughout the care continuum. This integration reduces administrative burden and accelerates care delivery.
Making Custom Solutions Economically Viable
Unified infrastructure makes customization affordable and scalable. Rather than building separate systems for each use case, organizations can leverage shared components to develop tailored solutions that address specific clinical workflows while maintaining interoperability with existing processes.
“We believe AI has the capability to transform healthcare,” Mr. Feldt stated. “But the current landscape of these systems makes it incredibly difficult to scale. Our goal at Raport is to be able to build a system that allows for performance and scalable AI systems that also support interoperability and avoid this utility fragmentation.”
Cancer Care Demands Smarter Data Integration
Rising Complexity in Oncology Workflows
Precision Medicine Diagnostic Partners highlighted modern oncology’s growing challenges: increasing cancer incidence rates, expanding biomarker testing requirements, and care journeys involving multiple specialists, diagnostic tests, and treatment planning reports. This complexity demands sophisticated data management and workflow coordination.
Co-founders Judy Largen, BSN, RN, and Dot Guccione, MSN, RN, emphasized that inconsistent data capture and poor care coordination frequently extend diagnosis and treatment timelines from optimal two-week periods to nearly two months—delays that significantly impact patient outcomes.
Analytics That Surface Actionable Insights
Their team developed specialized analytics algorithms, disease-specific data models, and visualization tools that identify testing gaps, process bottlenecks, and guideline compliance issues. By translating complex real-time clinical data into clear, actionable insights, these systems help organizations pinpoint delay sources and implement targeted improvements.
Why Clinical Expertise Must Drive AI Development
Pairing Domain Knowledge With Technical Capability
The collaboration between Precision Medicine Diagnostic Partners and Raport demonstrates how combining deep clinical expertise with scalable AI infrastructure produces superior outcomes. Precision Medicine Diagnostic Partners contributes oncology domain knowledge and clinical data methodologies, while Raport provides workflow engines capable of operationalizing those insights across care teams and health systems.
Building Extensible Solutions for Multiple Diseases
Their partnership aims to create interoperable, extensible workflows reflecting real-world cancer care patterns that can expand to additional diseases. “Our collective intent is to collaborate on our methodologies, technologies and solutions because we really do want to identify those patterns in the data that help us point to specific gaps in cancer care,” Ms. Largen explained.
These streamlined workflows and clinical support tools show promise for replication across rare diseases, chronic conditions, and other complex care scenarios requiring sophisticated coordination and data integration.
Balancing Automation With Human Judgment
Ms. Largen emphasized that technological advancement cannot replace clinical expertise: “AI is so valuable, but human insight is critical. You can’t forget that.” This reminder underscores the importance of positioning AI as an augmentation tool that enhances rather than replaces human decision-making in healthcare delivery.
Conclusion
Successfully scaling healthcare AI requires strategic infrastructure investments that eliminate fragmentation, enable interoperability, and maintain clinical rigor. Organizations must distinguish between true AI agents and automated workflows, build unified platforms that share context across processes, and ensure deep clinician involvement throughout development cycles. As cancer care complexity demonstrates, the most impactful solutions combine sophisticated technical capabilities with irreplaceable human expertise—a partnership model that promises better outcomes for patients and more sustainable operations for healthcare systems.
Discover the latest payers’ news updates with a single click. Follow DistilINFO HealthPlan and stay ahead with updates. Join our community today!
