EDEN: A Large-Scale Corpus of Clinical Notes for Italian

2026-06-10 · Source: Computation and Language · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics, Cybersecurity & Data Privacy · Depth: Advanced, quick

Summary

EDEN (Emergency Department Electronic Notes) is a new, large-scale corpus comprising approximately 4 million anonymized clinical notes from Italian Emergency Departments. This dataset covers diverse patient care phases and includes a subset of about six thousand notes manually annotated by clinical experts using a structured Case Report Form (CRF) with 132 items, focusing on dyspnea and loss of consciousness situations. The annotation process involved multiple clinicians and iterative revision. EDEN aims to address a significant data gap, supporting the development and application of Large Language Models in concrete medical contexts. The authors detail the data collection, anonymization pipeline, corpus statistics, and annotation scheme, proposing CRF-filling as a novel structured information extraction benchmark. Baseline results are provided using Gemma-27B and MedGemma-27B. EDEN is currently the largest freely available corpus of Italian clinical notes.

Key takeaway

For NLP engineers and research scientists developing medical AI solutions for Italian healthcare, EDEN offers an unprecedented resource. You should integrate this large-scale, anonymized corpus to train and validate Large Language Models, particularly for structured information extraction tasks like CRF-filling. This dataset fills a critical gap, enabling more accurate and contextually relevant AI applications in emergency department settings, potentially improving patient care workflows.

Key insights

EDEN provides the largest freely available Italian clinical note corpus, crucial for medical LLM development.

Method

The method involves a data collection protocol, an on-site anonymization pipeline, and a multi-clinician iterative annotation scheme for a subset of notes.

In practice

• Develop Italian medical LLMs
• Benchmark structured information extraction
• Analyze emergency department patient care

Topics

• Clinical Notes
• Italian Language
• Large Language Models
• Medical NLP
• Dataset
• Anonymization
• Emergency Department

Best for: AI Scientist, NLP Engineer, Research Scientist

Related on AIssential

• Large AI Models in Dental Healthcare: From General-Purpose Systems to Domain-Specific Foundation Models — Effective dental AI systems combine general-purpose and dental-specific models within structured pipelines.
• NormaTex-MapSNOMED: Bridging the Gap Between Brazilian Portuguese Clinical Narratives and SNOMED CT — LLMs effectively map Brazilian Portuguese clinical terms to SNOMED CT categories, outperforming lexical and contextual baselines.
• Ten Headache Specialists versus Artificial Intelligence for Clinical Literature Summarization: A Critical Evaluation and Comparison — Expert-written clinical summaries remain preferred over RAG-LLM outputs, despite AI's growing proficiency and occasional indistinguishability.
• ChatHealthAI: Aligning Electronic Health Record Representations with Large Language Models for Grounded Clinical Reasoning — ChatHealthAI integrates EHR representations with LLMs for grounded, interpretable clinical reasoning and accurate patient prediction.
• Using natural language processing to identify emergency department patients with incidental lung nodules requiring follow-up — An NLP pipeline using RoBERTa and spaCy can automate incidental lung nodule follow-up identification from ED CT reports.
• OpenMed: Running Medical NLP Locally Without Sending Patient Data to the Cloud — OpenMed enables secure, local medical NLP by processing sensitive patient data entirely within hospital systems.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Computation and Language.