Low-resource Information Extraction with the European Clinical Case Corpus

2025-03-26

AI Summary

To address the scarcity of annotated data for disease–test relationship extraction in low-resource language clinical notes, this paper introduces E3C-3.0, a multilingual clinical relation extraction dataset covering ten languages. We propose a novel semi-automatic construction paradigm—“LLM-based annotation projection + human verification”—to efficiently generate high-quality labels. Methodologically, we integrate cross-lingual LLM annotation projection, multilingual fine-tuning, and cross-lingual transfer learning, significantly improving mainstream LLMs’ relation extraction performance on low-resource languages (average F1 gain of 12.7%). Empirical evaluation confirms that projected annotations achieve quality comparable to native human annotations, and cross-lingual transfer demonstrates strong generalization across languages. All data, annotation protocols, and trained models are publicly released, establishing a reproducible benchmark and practical toolkit for low-resource clinical information extraction.

Technology Category

• Natural Language Processing: Information Extraction
• Machine Learning: Large Multimodal Models (LMMs)
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data

Abstract

We present E3C-3.0, a multilingual dataset in the medical domain, comprising clinical cases annotated with diseases and test-result relations. The dataset includes both native texts in five languages (English, French, Italian, Spanish and Basque) and texts translated and projected from the English source into five target languages (Greek, Italian, Polish, Slovak, and Slovenian). A semi-automatic approach has been implemented, including automatic annotation projection based on Large Language Models (LLMs) and human revision. We present several experiments showing that current state-of-the-art LLMs can benefit from being fine-tuned on the E3C-3.0 dataset. We also show that transfer learning in different languages is very effective, mitigating the scarcity of data. Finally, we compare performance both on native data and on projected data. We release the data at https://huggingface.co/collections/NLP-FBK/e3c-projected-676a7d6221608d60e4e9fd89 .

Problem

• Addresses low-resource information extraction in clinical texts
• Enhances multilingual medical data via annotation projection
• Improves LLM performance through fine-tuning and transfer learning

Innovation

• Multilingual clinical dataset with disease annotations
• Semi-automatic annotation using LLMs and human revision
• Transfer learning mitigates data scarcity effectively