Dilvan de Abreu Moreira
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Dilvan de Abreu Moreira has a total of 53 co-authors. Here are the top co-authors by number of shared publications:
- Mark A. Musen
Stanford University, USA
3 shared publications - Vinicius Vanzin
University of São Paulo
3 shared publications - Caetano Mazzoni Ranieri
Institute of Geosciences and Exact Sciences, Sao Paulo State University (UNESP), Rio Claro, Brazil
2 shared publications - Jó Ueyama
University of São Paulo
2 shared publications - Alexandre Claudio Botazzo Delbem
Institute of Mathematics and Computer Science, University of São Paulo, Av. Trabalhador São-carlense, 400, São Carlos, 13566-590, SP, Brazil
1 shared publication - Antonio Mauro Saraiva
Universidade de São Paulo, Brazil
1 shared publication
Latest Publications
- 1 September 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Latin America Transactions
- Vol. 24 (9) , 893-904
A Decentralized Learning Architecture for Medical Prescription Anomaly Detection via Hybrid Federated-Swarm Learning
- 1 September 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Latin America Transactions
- Vol. 24 (9) , 893-904
A Decentralized Learning Architecture for Medical Prescription Anomaly Detection via Hybrid Federated-Swarm Learning
- Open Access
- 20 February 2026
- Published by Elsevier in Data in Brief
- Vol. 65, 112607
KGAP: An RDF knowledge graph of agricultural commodity prices
- 20 February 2026
- Published by Elsevier in Data in Brief
- Vol. 65, 112607
Open AccessKGAP: An RDF knowledge graph of agricultural commodity prices
- 8 February 2026
- Published by Springer Nature
- p. 470-484
Segmentation-Augmented Flood Risk Classification for Nighttime River Monitoring
- 8 February 2026
- Published by Springer Nature
- p. 470-484
Segmentation-Augmented Flood Risk Classification for Nighttime River Monitoring
- 2 January 2026
- Published by Springer Nature
- p. 409-420
Ontology Capture Based on Interpretive Structural Modeling
- 2 January 2026
- Published by Springer Nature
- p. 409-420
Ontology Capture Based on Interpretive Structural Modeling
- 18 July 2025
- Published by Elsevier in Artificial Intelligence in Medicine
- Vol. 168, 103204
LLM-based approaches for automated vocabulary mapping between SIGTAP and OMOP CDM concepts
- 18 July 2025
- Published by Elsevier in Artificial Intelligence in Medicine
- Vol. 168, 103204
LLM-based approaches for automated vocabulary mapping between SIGTAP and OMOP CDM concepts
- Open Access
- 17 March 2025
- Published by Universidade Federal de Santa Catarina (UFSC) in Encontros Bibli: Revista Eletrônica de Biblioteconomia e Ciência da Informação
- Vol. 30, 1-32
Gazetteer literário de Machado de Assis
Objetivo: Este estudo tem o objetivo de desenvolver uma aplicação web semânticaque mapeia localidades geográficas nas obras de Machado de Assis, armazenando-as em uma triplestore. A partir da integração dos dados disponibilizados pela enciclopédia MachadodeAssis.net com as coordenadas geográficas de Geonames.org e GoogleMaps, o projeto visa oferecer uma experiência de leitura através de mapas interativos que servirão de suporte para as menções aos espaços realizadas pelo escritor ao longo do Século XIX. Método: Utiliza a biblioteca Python BeautifulSoup para consultas e coleta dos dados da enciclopédia, estruturando-os de acordo com os parâmetros do schema.org. As citações coletadas são submetidas aos modelos gpt3.5-instruct e gpt4-turbo para obter os nomes atuais das localidades e a devida classificação destes espaços de acordo com a ontologia Geonames.org. Consultas SPARQL são realizadas ao portal dados.literaturabrasileira.ufsc.br para obter identificadores únicos para cada livro. Resultado: A aplicação oferece uma integração entre mapas, citações e textos completos, em consonância com os padrões Linked Data. Conclusões: A intersecção entre tecnologia, literatura e geolocalização pode oferecer experiências de leitura interessantes, proporcionando um terreno fértil para o desenvolvimento das chamadas humanidades digitais.
- 17 March 2025
- Published by Universidade Federal de Santa Catarina (UFSC) in Encontros Bibli: Revista Eletrônica de Biblioteconomia e Ciência da Informação
- Vol. 30, 1-32
Open AccessGazetteer literário de Machado de Assis
Objetivo: Este estudo tem o objetivo de desenvolver uma aplicação web semânticaque mapeia localidades geográficas nas obras de Machado de Assis, armazenando-as em uma triplestore. A partir da integração dos dados disponibilizados pela enciclopédia MachadodeAssis.net com as coordenadas geográficas de Geonames.org e GoogleMaps, o projeto visa oferecer uma experiência de leitura através de mapas interativos que servirão de suporte para as menções aos espaços realizadas pelo escritor ao longo do Século XIX. Método: Utiliza a biblioteca Python BeautifulSoup para consultas e coleta dos dados da enciclopédia, estruturando-os de acordo com os parâmetros do schema.org. As citações coletadas são submetidas aos modelos gpt3.5-instruct e gpt4-turbo para obter os nomes atuais das localidades e a devida classificação destes espaços de acordo com a ontologia Geonames.org. Consultas SPARQL são realizadas ao portal dados.literaturabrasileira.ufsc.br para obter identificadores únicos para cada livro. Resultado: A aplicação oferece uma integração entre mapas, citações e textos completos, em consonância com os padrões Linked Data. Conclusões: A intersecção entre tecnologia, literatura e geolocalização pode oferecer experiências de leitura interessantes, proporcionando um terreno fértil para o desenvolvimento das chamadas humanidades digitais.
- Open Access
- 14 October 2024
- Published by Sociedade Brasileira de Computação
- p. 240-248
OntoDrug: Enhancing Brazilian Health System Interoperability with a National Medication Ontology
This paper presents OntoDrug, an ontology designed to enhance medicine management in Brazil by integrating regulatory frameworks and standardizing terminologies. OntoDrug improves patient safety and treatment efficacy by accurately identifying and classifying medications and supporting interoperability with health information systems. A proof-of-concept application integrated into the Hospital das Clínicas de Marília’s hospital EHR system demonstrated OntoDrug’s utility, achieving high precision and recall. An experimental study using large language models grounded on the ontology achieved, using GPT-4 turbo, 0.97 precision, 1.0 recall and an F1-score of 0.99. We also evaluated open-source models llama3-8b, llama3-70b, and gemma-7b-it. Their performance was close to GPT-4’s. The significant effectiveness is primarily due to the utilization of large language models (LLMs). While using these large language models enhanced performance, challenges related to cost, privacy, and service availability were identified. OntoDrug represents a significant advancement in Brazil’s medication information standardization and optimization.
- 14 October 2024
- Published by Sociedade Brasileira de Computação
- p. 240-248
Open AccessOntoDrug: Enhancing Brazilian Health System Interoperability with a National Medication Ontology
This paper presents OntoDrug, an ontology designed to enhance medicine management in Brazil by integrating regulatory frameworks and standardizing terminologies. OntoDrug improves patient safety and treatment efficacy by accurately identifying and classifying medications and supporting interoperability with health information systems. A proof-of-concept application integrated into the Hospital das Clínicas de Marília’s hospital EHR system demonstrated OntoDrug’s utility, achieving high precision and recall. An experimental study using large language models grounded on the ontology achieved, using GPT-4 turbo, 0.97 precision, 1.0 recall and an F1-score of 0.99. We also evaluated open-source models llama3-8b, llama3-70b, and gemma-7b-it. Their performance was close to GPT-4’s. The significant effectiveness is primarily due to the utilization of large language models (LLMs). While using these large language models enhanced performance, challenges related to cost, privacy, and service availability were identified. OntoDrug represents a significant advancement in Brazil’s medication information standardization and optimization.
- 8 August 2019
- Published by Springer Nature in Journal of Digital Imaging
- Vol. 33 (2) , 287-303
Automatic Staging of Cancer Tumors Using AIM Image Annotations and Ontologies
A second opinion about cancer stage is crucial when clinicians assess patient treatment progress. Staging is a process that takes into account description, location, characteristics, and possible metastasis of tumors in a patient. It should follow standards, such as the TNM Classification of Malignant Tumors. However, in clinical practice, the implementation of this process can be tedious and error prone. In order to alleviate these problems, we intend to assist radiologists by providing a second opinion in the evaluation of cancer stage. For doing this, we developed a TNM classifier based on semantic annotations, made by radiologists, using the ePAD tool. It transforms the annotations (stored using the AIM format), using axioms and rules, into AIM4-O ontology instances. From then, it automatically calculates the liver TNM cancer stage. The AIM4-O ontology was developed, as part of this work, to represent annotations in the Web Ontology Language (OWL). A dataset of 51 liver radiology reports with staging data, from NCI’s Genomic Data Commons (GDC), were used to evaluate our classifier. When compared with the stages attributed by physicians, the classifier stages had a precision of 85.7% and recall of 81.0%. In addition, 3 radiologists from 2 different institutions manually reviewed a random sample of 4 of the 51 records and agreed with the tool staging. AIM4-O was also evaluated with good results. Our classifier can be integrated into AIM aware imaging tools, such as ePAD, to offer a second opinion about staging as part of the cancer treatment workflow.
- 8 August 2019
- Published by Springer Nature in Journal of Digital Imaging
- Vol. 33 (2) , 287-303
Automatic Staging of Cancer Tumors Using AIM Image Annotations and Ontologies
A second opinion about cancer stage is crucial when clinicians assess patient treatment progress. Staging is a process that takes into account description, location, characteristics, and possible metastasis of tumors in a patient. It should follow standards, such as the TNM Classification of Malignant Tumors. However, in clinical practice, the implementation of this process can be tedious and error prone. In order to alleviate these problems, we intend to assist radiologists by providing a second opinion in the evaluation of cancer stage. For doing this, we developed a TNM classifier based on semantic annotations, made by radiologists, using the ePAD tool. It transforms the annotations (stored using the AIM format), using axioms and rules, into AIM4-O ontology instances. From then, it automatically calculates the liver TNM cancer stage. The AIM4-O ontology was developed, as part of this work, to represent annotations in the Web Ontology Language (OWL). A dataset of 51 liver radiology reports with staging data, from NCI’s Genomic Data Commons (GDC), were used to evaluate our classifier. When compared with the stages attributed by physicians, the classifier stages had a precision of 85.7% and recall of 81.0%. In addition, 3 radiologists from 2 different institutions manually reviewed a random sample of 4 of the 51 records and agreed with the tool staging. AIM4-O was also evaluated with good results. Our classifier can be integrated into AIM aware imaging tools, such as ePAD, to offer a second opinion about staging as part of the cancer treatment workflow.
- Open Access
- 4 July 2016
- Published by Sociedade Brasileira de Computação
BioDSL: A Domain-Specific Language for mapping and dissemination of Biodiversity Data in the LOD
Currently, Linked Open Data (LOD) have enabled integrated data sharing across disciplines over the Web. However, for LOD users, in areas such as biodiversity (which massively use the Web to disseminate data), the task of transforming data file contents in CSV (Comma Separated Value) to RDF (Resource Description Framework) is not trivial. We have developed a new approach to map data files in CSV to RDF format based on a domain-specific language (DSL) called BioDSL. Using it, biodiversity data users can write compact programs to map their data to RDF and link them to the LOD. Biodiversity vocabularies and ontologies, such as Darwin Core and OntoBio, can be used with BioDSL to enrich user data. Existing tools are exclusively focused on mapping (CSV to RDF), offering little or no support for linking data to the LOD (interconnecting user entities to LOD entities). They also are more complex to use than BioDSL.
- 4 July 2016
- Published by Sociedade Brasileira de Computação
Open AccessBioDSL: A Domain-Specific Language for mapping and dissemination of Biodiversity Data in the LOD
Currently, Linked Open Data (LOD) have enabled integrated data sharing across disciplines over the Web. However, for LOD users, in areas such as biodiversity (which massively use the Web to disseminate data), the task of transforming data file contents in CSV (Comma Separated Value) to RDF (Resource Description Framework) is not trivial. We have developed a new approach to map data files in CSV to RDF format based on a domain-specific language (DSL) called BioDSL. Using it, biodiversity data users can write compact programs to map their data to RDF and link them to the LOD. Biodiversity vocabularies and ontologies, such as Darwin Core and OntoBio, can be used with BioDSL to enrich user data. Existing tools are exclusively focused on mapping (CSV to RDF), offering little or no support for linking data to the LOD (interconnecting user entities to LOD entities). They also are more complex to use than BioDSL.
- 12 July 2015
- Published by Springer Nature
- p. 112-126
User Extensible System to Identify Problems in OWL Ontologies and SWRL Rules
- 12 July 2015
- Published by Springer Nature
- p. 112-126
User Extensible System to Identify Problems in OWL Ontologies and SWRL Rules