Elamur Research⌕
☰
home / papers / 2026 / From Annotated Clinical Narratives to Ontology: Structuring Brazilian Portuguese Clinical Data

From Annotated Clinical Narratives to Ontology: Structuring Brazilian Portuguese Clinical Data

Paper · 2026 · Fernando Henrique Moura de Oliveira and Cleyton Mário de Oliveira Rodrigues

Summary

Clinical NLP for Brazilian Portuguese has few semantically structured resources. Existing annotated corpora use flat tag schemes that are hard to connect to standard terminologies. The authors address this by converting the SemClinBr corpus (1,000 annotated clinical notes) into a lightweight OWL domain ontology. They follow a corpus-driven approach: classes and relations come from the corpus's own annotation schema rather than from an external design.

A Python pipeline built on Owlready2 reads the XML files and extracts texts, entity annotations and relations. It then creates one individual per annotated span and one ClinicalEpisode per document, and links them with object properties. Concept-to-concept links such as associated_with and negation_of are also asserted. The HermiT reasoner in Protégé found the resulting ontology logically consistent, and SPARQL queries confirmed it is densely connected.

The authors also report weaknesses. Linguistic and biomedical categories are mixed in one hierarchy, formal SNOMED CT alignment is missing, and surface variants create duplicate instances. The work is an early-stage bridge between corpus annotations and formal ontologies. It supports tasks like entity normalization, semantic search and ontology-guided annotation.

Key points

  • The resulting ontology (ClinicalKnowledge.owl) has 57 classes (built on 14 top-level abstract classes), 17,663 individuals, 23 object properties, 7 data properties and 87,634 axioms, with a maximum hierarchy depth of 6.
  • It encodes 1,000 clinical episodes, each with about 16 connections on average.
  • associated_with is the most frequent relation (9,852 occurrences), against 1,606 for negation_of. Negation is represented by linking a finding to a negation individual such as Concept_sem ("without").
  • Tags without a direct class mapping fall back to the Finding superclass. Variation in case, accents and word order (e.g., "estável hemodinamicamente") inflates instance counts.
  • Planned work includes lexical normalization, disjointness and domain/range axioms, explicit disease–symptom relations, and alignment with ICD and SNOMED CT. The underlying corpus, obtained from HAILab-PUCPR, cannot be shared publicly.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Fields, topics and concepts

📚 Shelves 2The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 works

🧭 Topics 1Research subjects and languages the work is about.Knowledge graphs & ontologies · 192 works

Cite

Cite as Fernando Henrique Moura de Oliveira and Cleyton Mário de Oliveira Rodrigues. 2026. From Annotated Clinical Narratives to Ontology: Structuring Brazilian Portuguese Clinical Data. Proceedings of the 17th International Conference on Computational Processing of Portuguese (PROPOR 2026) - Vol. 2.

The paper itself, not this page. BibTeX from the library's record of the work.

@inproceedings{oliveira2026annotated,
  title = {{From Annotated Clinical Narratives to Ontology: Structuring Brazilian Portuguese Clinical Data}},
  author = {Fernando Henrique Moura {de Oliveira} and Cleyton Mário {de Oliveira Rodrigues}},
  booktitle = {Proceedings of the 17th International Conference on Computational Processing of Portuguese (PROPOR 2026) - Vol. 2},
  year = {2026},
  url = {https://aclanthology.org/2026.propor-2.20/}
}

References

17 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Léo Bouscarrat, Antoine Bonnefoy, Cécile Capponi, and Carlos Ramisch. 2020. Multilingual enrichment https://icd.who.int/en/ https://github.com/HAILab-PUCPR of disease biomedical ontologies. In Proceedings of the lrec 2020 workshop on multilingual biomedical text processing (multilingualbio 2020), pa… ↗
  2. Naila Camila Da Rocha, Abner Macola Pacheco Barbosa, Yaron Oliveira Schnr, Juliana Machado-Rugolo, Luis Gustavo Modelli de Andrade, José Eduardo Corrente, and Liciana Vaz de Arruda Silveira. 2022. Natural language processing to extract information from portuguese-language medical records. Data, 8(1)… find
  3. Rildo Pinto da Silva and Antonio Pazin-Filho. 2025. Dataset of anonymized discharge summaries of sepsis patients from a brazilian tertiary hospital for nlp applications. Data in Brief, 61:111804. find
  4. Clément Dalloux, Vincent Claveau, Natalia Grabar, Lucas Emanuel Silva Oliveira, Claudia Maria Cabral Moro, Yohan Bonescki Gumiel, and Deborah Ribeiro Carvalho. 2021. Supervised learning for the detection of negation and of its scope in french and brazilian portuguese biomedical corpora. Natural Lang… find
  5. Joao Vitor Andrioli de Souza, Yohan Bonescki Gumiel, Lucas Emanuel Silva, Claudia Maria Cabral Moro, and 1 others. 2019. Named entity recognition for clinical portuguese corpus with conditional random fields and semantic groups. In Simpósio Brasileiro de Computação Aplicada à Saúde (SBCAS), pages 31… find
  6. Lívia Dutra, Arthur Lorenzi, Lorena Larré, Frederico Belcavello, Ely Edison da Silva Matos, Amanda Pestana, Kenneth Brown, Mariana Gonalves, Victor Herbst, Sofia Reinach, and 1 others. 2023. Building a frame-semantic model of the healthcare domain: Towards the identification of gender-based violence… find
  7. Ian Horrocks, Boris Motik, and Zhe Wang. 2012. The hermit owl reasoner. Journal of Automated Reasoning, 858(3). find
  8. Amila Kugic, Stefan Schulz, and Markus Kreuzthaler. 2024. Disambiguation of acronyms in clinical narratives with large language models. Journal of the American Medical Informatics Association, 31(9):2040–2046. find
  9. Prashanti Manda. 2025. Large language models in bio-ontology research: A review. Bioengineering, 12(11):1260. find
  10. Mark A Musen. 2015. The protégé project: a look back and a look forward. AI matters, 1(4):4–12. find
  11. Aurélie Névéol, Hercules Dalianis, Sumithra Velupillai, Guergana Savova, and Pierre Zweigenbaum. 2018. Clinical natural language processing in languages other than english: opportunities and challenges. Journal of biomedical semantics, 9(1):12. find
  12. Lucas Emanuel Silva e Oliveira, Ana Carolina Peters, Adalniza Moura Pucca Da Silva, Caroline Pi-latti Gebeluca, Yohan Bonescki Gumiel, Lilian Mie Mukai Cintho, Deborah Ribeiro Carvalho, Sadid Al Hasan, and Claudia Maria Cabral Moro. 2022. Semclinbr-a multi-institutional and multi-specialty semantica… find
  13. Felipe Carvalho Pellison, Rui Pedro Charters Lopes Rijo, Vinicius Costa Lima, Nathalia Yukie Crepaldi, Filipe Andrade Bernardi, Rafael Mello Galliez, Afrânio Kritski, Kumar Abhishek, and Domingos Alves. 2020. Data integration in the brazilian public health system for tuberculosis: use of the semanti… find
  14. Filipe Santana, Fred Freitas, Roberta Fernandes, Zulma Medeiros, and Daniel Schober. 2012. Towards an ontological representation of morbidity and mortality in description logics. Journal of biomedical semantics, 3(Suppl 2):S7. find
  15. Filipe Santana, Daniel Schober, Zulma Medeiros, Fred Freitas, and Stefan Schulz. 2011. Ontology patterns for tabular representations of biomedical knowledge on neglected tropical diseases. Bioinformatics, 27(13):i349–i356. find
  16. Hugo Sousa, Alipio Mario Jorge, Arian Pasquali, Catarina Santos, and Mario Lopes. 2023. A biomedical entity extraction pipeline for oncology health records in portuguese. In Proceedings of the 38th ACM/SI-GAPP Symposium on Applied Computing, pages 950– 956. find
  17. Diego Bettiol Yamada, Filipe Andrade Bernardi, Newton Shydeo Brandão Miyoshi, Inácia Bezerra de Lima, André Luiz Teixeira Vinci, Vinicius Tohoru Yoshiura, and Domingos Alves. 2020. Ontology-based inference for supporting clinical decisions in mental health. In International Conference on Computation… find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/oliveira-2026-annotated-clinical-narratives-ontology-structuring/