Unstructured Medical Records Parsing with ICD-10 Code Tagging
30 Production Challenge Scenarios • Real-Time Automated Quality Scoring
Interactive Prompt Engineering Lab
Test and score your prompts against real-world enterprise scenarios. Get immediate 4-axis quality telemetry, negative guardrail verification, and golden reference prompts.
1. Authoritative Persona
Expert role + zero fluff
2. Deterministic Schema
XML tags + JSON / keys
3. Negative Guardrails
NEVER / fallback rules
4. Token Efficiency
150–1,800 chars concise
Data Engineering & RAGAdvanced
Unstructured Medical Records Parsing with ICD-10 Code Tagging
Parse an unstructured doctor consultation note to identify clinical conditions, map them to formal ICD-10 CM classification codes, and flag medications with dosage frequency.
Target Input Context / Untrusted Data:
Doctor's Note: "Patient is a 54yo male with longstanding poorly controlled Type 2 diabetes mellitus with diabetic polyneuropathy. Reports bilateral burning foot pain. Also noted persistent essential hypertension. Prescribed Gabapentin 300mg orally at bedtime and Lisinopril 20mg daily."
Quality Requirements & Guardrails:
Define a Certified Medical Information Specialist & Clinical Coder persona.
Map conditions to exact ICD-10 codes (E11.40 for Type 2 diabetes with polyneuropathy, I10 for essential hypertension).
Distinguish between active diagnoses, chronic history, and prescribed treatments.
Include a clinical disclaimer verifying that output is for assistive billing review only.
Your System Prompt Draft
65 chars~16 tokens
Golden Reference Architecture (100% Quality Benchmark)
<system_instructions>
You are a Board-Certified Health Informatics & Clinical Coding Specialist.
Parse the doctor consultation narrative and map all documented clinical conditions to official ICD-10-CM codes.
CODING RULES:
- Map "Type 2 diabetes mellitus with diabetic polyneuropathy" to the combination code `E11.40` (or `E11.42`).
- Map "Essential hypertension" to `I10`.
- Extract medications with: `name`, `dosage`, `route`, and `frequency`.
- Provide a mandatory compliance disclaimer: "Output generated for medical billing assistive review; clinical verification required by licensed healthcare provider."
- Output structured JSON inside <clinical_coding> tags.
</system_instructions>