NormaTex-MapSNOMED: Bridging the Gap Between Brazilian Portuguese Clinical Narratives and SNOMED CT

2026-04-12 · Source: Paper Index on ACL Anthology · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Health & Medical Research · Depth: Expert, quick

Summary

NormaTex-MapSNOMED is a new component of the NormaTex framework designed to map clinical terms from Brazilian Portuguese free-text narratives to predefined categories aligned with SNOMED CT. This method addresses the challenge of processing unstructured clinical data and linguistic variability in non-English languages. It utilizes large language models (LLMs) guided by a structured prompt to assign extracted clinical terms to target categories. Experiments were conducted on Portuguese clinical narratives and evaluated using three strategies: lexical similarity (Levenshtein distance), contextual similarity (BERT-based model), and semantic validation (LLMs). The LLM-based evaluation consistently outperformed both lexical and contextual baselines, demonstrating higher precision for disease-related terms than for symptom-related expressions. These results suggest LLMs are a promising approach for semantic mapping and can enhance clinical term normalization and interoperability with standardized terminologies in Brazilian Portuguese.

Key takeaway

For NLP Engineers working with clinical data in Brazilian Portuguese, NormaTex-MapSNOMED offers a robust LLM-based approach to standardize clinical terms. Your team should consider integrating LLMs with structured prompts for mapping free-text narratives to SNOMED CT, especially for disease-related expressions where higher precision is observed, to improve data interoperability.

Key insights

LLMs effectively map Brazilian Portuguese clinical terms to SNOMED CT categories, outperforming lexical and contextual baselines.

Principles

• LLMs excel at semantic mapping.
• Structured prompts guide LLM term assignment.

Method

The method uses LLMs with structured prompts to assign extracted clinical terms to SNOMED CT-aligned categories, evaluated via lexical, contextual, and LLM-based semantic similarity.

In practice

• Use LLMs for clinical term normalization.
• Prioritize disease terms for higher precision.

Topics

• NormaTex-MapSNOMED
• SNOMED CT
• Brazilian Portuguese
• Clinical Narratives
• Large Language Models

Best for: AI Scientist, NLP Engineer, Research Scientist

Related on AIssential

• EuropeMedQA Study Protocol: A Multilingual, Multimodal Medical Examination Dataset for Language Model Evaluation — EuropeMedQA is a new multilingual, multimodal medical dataset for evaluating LLMs on European regulatory exams.
• The Slop Paradox: How Synthetic Standardization Erodes Clinical Uncertainty and Cross-Modal Alignment in AI-Rewritten Radiology Reports — AI-rewriting of clinical text for standardization can paradoxically degrade cross-modal alignment despite appearing cleaner.
• MedArena: Comparing LLMs for Medicine-in-the-Wild Clinician Preferences — Real-world clinician preferences reveal that medical LLM utility extends beyond factual recall to include depth, clarity, and clinical nuance.
• Language Models as Semantic Teachers: Post-Training Alignment for Medical Audio Understanding — Aligning audio encoders with medical LLMs via a "semantic teacher" paradigm significantly enhances clinical diagnostic performance.
• Language Models as Interfaces, Not Oracles: A Hybrid LLM-ML System for Pediatric Appendicitis — Hybrid AI systems improve clinical decision support by using LLMs for feature extraction and validated ML models for reliable, auditable prediction.
• From Annotated Clinical Narratives to Ontology: Structuring Brazilian Portuguese Clinical Data — A new ontology structures Brazilian Portuguese clinical narratives, enhancing NLP interoperability and semantic expressiveness.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Paper Index on ACL Anthology.