Nick-Maximillien/medgate-compiler-data

MedGate Compiler Training Data

The Logic-Grounding Corpus for Computable Medical Law

This dataset is a hand-curated collection of 606 high-fidelity mappings designed to train Structural Compilers. It facilitates the translation of unstructured clinical guidelines and medical policy prose into machine-executable symbolic logic (JSON).

Dataset Summary

The MedGate Compiler Training Data provides the ground truth for Nexus Forensic – Layer 0 (Protocol Vault).

Each record contains:

• —A clinical instruction
• —A prose snippet (input)
• —A deterministic JSON object (output)

Together, these define the forensic logic gates, intent tags, and evidentiary requirements necessary for Computable Medical Law.

Dataset Structure

Data Fields

• —instruction The system prompt defining the persona Example: "You are a Forensic Logic Parser..."

• —input Raw clinical guideline text extracted from authoritative sources Sources include: MoH, NASCOP, KQMH

• —output A deterministic JSON string containing:
• —intent_tags: Objective of the rule (Safety, Quality, Compliance, Integrity)
• —logic_config: Triggers, required artifacts, thresholds
• —rule_type: Symbolic logic classification
• —scope_tags: Jurisdiction of the rule (Clinical, Facility, Legal)

Data Sample

{
  "instruction": "You are a Forensic Logic Parser...",
  "input": "Following an acute anterior MI, a contrast echocardiogram may be considered...",
  "output": "{\"intent_tags\": [\"safety\", \"quality\"], \"logic_config\": {\"required_artifact\": \"contrast echocardiogram\"}, \"rule_type\": \"conditional_existence\", \"scope_tags\": [\"clinical\"]}"
}

Dataset Statistics

Rule Type Distribution

The corpus is balanced across symbolic logic classifications to ensure robust handling of diverse legal and clinical constraints.

Intent Tag Distribution

Ethical & Safety Considerations

Non-Clinical Use

This dataset is intended for training legal and forensic compilers. It is not designed to provide clinical advice or patient diagnosis.

Grounding

All outputs are mapped to explicit policy standards to prevent hallucination in legal and regulatory contexts.

Privacy

Contains only clinical guidelines and public policy text. No Patient Identifiable Information (PII) is included.

Use Cases

Structural Compilation

Training models to convert PDF policies into executable Knowledge Graphs.

Neurosymbolic AI

Bridging LLM-generated prose with Python-based deterministic auditing gates.

Computable Medical Law

Automating medical insurance adjudication and regulatory compliance verification.