Skip to content

Patient Diagnosis

Bayesian Pattern Recognition for Real World Applications

📚 View all posts in the Graph-based Healthcare Series

Graph-based Healthcare Series — 5

This is the fifth post in an ongoing series on graph-based healthcare tools. Stay tuned for upcoming entries on clinical modeling, decision support systems, and graph-powered AI assistants.

In our previous post, we explored how large language models (LLMs) can simulate realistic pediatric patient encounters based on the IMNCI guidelines. These synthetic notes were grounded in real clinical logic, labeled with structured IMNCI classifications, and validated using a multi-agent verification strategy inspired by the Bayesian Truth Serum (BTS). The result: a high-fidelity dataset of richly annotated, clinically plausible pediatric cases.

In this post, we put that dataset to work—prototyping a Bayesian diagnostic engine that quantifies clinical evidence, scores conditions, and updates probabilities in a way that mirrors how clinicians think.

Managing Agentic Flows with Pydantic Graph

📚 View all posts in the Graph-based Healthcare Series

Graph-based Healthcare Series — 3

This is the third post in an ongoing series on graph-based healthcare tools. Stay tuned for upcoming entries on clinical modeling, decision support systems, and graph-powered AI assistants.

In our previous post, we demonstrated how the IMNCI graph model could power a graph-based retrieval-augmented generation (graph RAG) pipeline. By combining structured clinical knowledge with large language models (LLMs), we laid the foundation for a system that supports real-world diagnostic workflows.

In this installment, we take that idea further by introducing agentic flows—a new phase in our clinical decision support pipeline. Here, an intelligent, dialogue-capable assistant doesn’t just answer queries; it actively guides the diagnostic process. This assistant leverages the structured IMNCI graph as its reasoning backbone and uses pydantic graph to statefully orchestrate a set of modular, task-specific assistants (tools).

Enhancing Patient Diagnosis with Graph-based Retrieval-Augmented Generation

📚 View all posts in the Graph-based Healthcare Series

Graph-based Healthcare Series — 2

This is the second post in an ongoing series on graph-based healthcare tools. Stay tuned for upcoming entries on clinical modeling, decision support systems, and graph-powered AI assistants.

In our previous post, we explored how the Integrated Management of Neonatal and Childhood Illness (IMNCI) guidelines were transformed from static, text-heavy documents into an interactive graph model. This structure enabled more intuitive navigation of clinical logic, laying the groundwork for advanced applications in AI-assisted patient diagnosis.

In this follow-up, we demonstrate how that graph model serves as the foundation for a graph-based retrieval-augmented generation (graph RAG) system. By combining the structured clinical knowledge encoded in Neo4j with the generative capabilities of large language models (LLMs), we create a framework that supports transparent, context-aware patient diagnosis at the point of care.