João Vitor Andrioli de Souza

João Vitor Andrioli de Souza

Auto-generated profile, based on publicly available information

João Vitor Andrioli de Souza

Pontificia Universidade Católica do Parana

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 June 2023
Verified
Conference Paper
Closed access
ESElisa Terumi Rubel Schneider
Elisa Terumi Rubel Schneider
YGYohan Bonescki Gumiel
Yohan Bonescki Gumiel
JSJoão Vitor Andrioli de Souza
João Vitor Andrioli de Souza
LMLilian Mie Mukai
Lilian Mie Mukai
LOLucas Emanuel Silva e Oliveira
Lucas Emanuel Silva e Oliveira
EPEmerson Cabrera Paraiso
Emerson Cabrera Paraiso
Contextual word embeddings and the Transformers architecture have reached state-of-the-art results in many natural language processing (NLP) tasks and improved the adaptation of models for multiple domains. Despite the improvement in the reuse and construction of models, few resources are still developed for the Portuguese language, especially in the health domain. Furthermore, the clinical models available for the language are not representative enough for all medical specialties. This work explores deep contextual embedding models for the Portuguese language to support clinical NLP tasks. We transferred learned information from electronic health records of a Brazilian tertiary hospital specialized in cardiology diseases and pre-trained multiple clinical BERT-based models. We evaluated the performance of these models in named entity recognition experiments, fine-tuning them in two annotated corpora containing clinical narratives. Our pre-trained models outperformed previous multilingual and Portuguese BERT-based models for cardiology and multi-specialty environments, reaching the state-of-the-art for analyzed corpora, with 5.5% F1 score improvement in TempClinBr (all entities) and 1.7% in SemClinBr (Disorder entity) corpora. Hence, we demonstrate that data representativeness and a high volume of training data can improve the results for clinical tasks, aligned with results for other languages.

24 February 2020
Verified
Book Chapter
Closed access
PDF file
JSJoão Vitor Andrioli de Souza
João Vitor Andrioli de Souza
LOLucas Emanuel Silva E Oliveira
Lucas Emanuel Silva E Oliveira
YGYohan Bonescki Gumiel
Yohan Bonescki Gumiel
DCDeborah Ribeiro Carvalho
Deborah Ribeiro Carvalho
CMClaudia Maria Cabral Moro
Claudia Maria Cabral Moro

11 June 2019
Verified
Conference Paper
Open access
JSJoão Vitor Andrioli De Souza
João Vitor Andrioli De Souza
YGYohan Bonescki Gumiel
Yohan Bonescki Gumiel
LOLucas Emanuel Silva e Oliveira
Lucas Emanuel Silva e Oliveira
CMClaudia Maria Cabral Moro
Claudia Maria Cabral Moro
Considering the difficulties of extracting entities from Electronic Health Records (EHR) texts in Portuguese, we explore the Conditional Random Fields (CRF) algorithm to build a Named Entity Recognition (NER) system based on a corpus of clinical Portuguese data annotated by experts. We acquaint the challenges and methods to classify Abbreviations, Disorders, Procedures and Chemicals within the texts. By selecting a meaningful set of features, and parameters with the best performance the results demonstrate that the method is promising and may support other biomedical tasks, nonetheless, further experiments with more features, different architectures and sophisticated preprocessing steps are needed.

4 January 2024
Verified
Book Chapter
Closed access
PDF file
CPCarolina Braun Prado
Carolina Braun Prado
YGYohan Bonescki Gumiel
Yohan Bonescki Gumiel
ESElisa Terumi Rubel Schneider
Elisa Terumi Rubel Schneider
LCLilian Mie Mukai Cintho
Lilian Mie Mukai Cintho
JSJoão Vitor Andrioli de Souza
João Vitor Andrioli de Souza
CMClaudia Moro
Claudia Moro