Automated SNOMED CT Concept Annotation in Clinical Text Using Bi-GRU Neural Networks

2025-08-04

AI Summary

This study addresses the high cost and poor scalability of manual SNOMED CT concept annotation in clinical text. We propose a lightweight, efficient sequence labeling method that replaces computationally intensive Transformers with a bidirectional GRU architecture, significantly reducing inference overhead while preserving performance. To enhance input representation, we integrate domain-adapted tokenization—combining SpaCy and SciBERT—and incorporate contextual, syntactic, and morphological features, thereby improving robustness against lexical ambiguity and orthographic errors. Evaluated on a MIMIC-IV subset, our model achieves an F1-score of 90%, outperforming conventional rule-based systems and matching state-of-the-art neural models. The approach thus delivers both high accuracy and strong deployability, offering a practical solution for large-scale clinical concept extraction.

Technology Category

• Natural Language Processing: Sentence-level Semantics, Textual Inference, etc.
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning
• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning

Application Category

• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling
• Web Mining and Content Analysis: Normalization, clustering, classification, and summarization of Web text

Abstract

Automated annotation of clinical text with standardized medical concepts is critical for enabling structured data extraction and decision support. SNOMED CT provides a rich ontology for labeling clinical entities, but manual annotation is labor-intensive and impractical at scale. This study introduces a neural sequence labeling approach for SNOMED CT concept recognition using a Bidirectional GRU model. Leveraging a subset of MIMIC-IV, we preprocess text with domain-adapted SpaCy and SciBERT-based tokenization, segmenting sentences into overlapping 19-token chunks enriched with contextual, syntactic, and morphological features. The Bi-GRU model assigns IOB tags to identify concept spans and achieves strong performance with a 90 percent F1-score on the validation set. These results surpass traditional rule-based systems and match or exceed existing neural models. Qualitative analysis shows effective handling of ambiguous terms and misspellings. Our findings highlight that lightweight RNN-based architectures can deliver high-quality clinical concept annotation with significantly lower computational cost than transformer-based models, making them well-suited for real-world deployment.

Problem

• Automated SNOMED CT annotation in clinical text for structured data extraction
• Overcoming labor-intensive manual SNOMED CT labeling at scale
• Lightweight Bi-GRU model outperforms rule-based systems in concept recognition

Innovation

• Bidirectional GRU model for SNOMED CT annotation
• Domain-adapted SpaCy and SciBERT tokenization
• Lightweight RNN outperforms transformer-based models