Creation of an Annotated Corpus of Spanish Radiology Reports

TL;DR

This paper introduces a new annotated corpus of 513 Spanish radiology reports, providing valuable resources for developing and evaluating NLP algorithms in the biomedical domain, especially for Spanish language data.

Contribution

It creates and shares a manually annotated Spanish radiology report corpus with entities, negation, and relations, addressing the scarcity of biomedical annotated resources.

Findings

Corpus enables evaluation of NER and relation extraction algorithms

Provides guidelines for creating similar biomedical annotated datasets

Facilitates supervised learning approaches in Spanish biomedical NLP

Abstract

This paper presents a new annotated corpus of 513 anonymized radiology reports written in Spanish. Reports were manually annotated with entities, negation and uncertainty terms and relations. The corpus was conceived as an evaluation resource for named entity recognition and relation extraction algorithms, and as input for the use of supervised methods. Biomedical annotated resources are scarce due to confidentiality issues and associated costs. This work provides some guidelines that could help other researchers to undertake similar tasks.