Breno Nunes de Sena Keller

Mr. Breno Nunes de Sena Keller

Departamento de Computação, Universidade Federal de Ouro Preto, Ouro Preto, Brazil
orcid

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

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

Representative Publications

View All Publications
30 November 2022 in AppliedMath
Verified
Journal Article
Open access
PDF file
DDDébora N. Diniz
Débora N. Diniz
BKBreno Nunes de Sena Keller
Breno Nunes de Sena Keller
MRMariana T. Rezende
Mariana T. Rezende
ABAndrea Gomes Campos Bianchi
Andrea Gomes Campos Bianchi
CCclaudia carneiro
claudia carneiro
Marcone Jamilson Freitas SouzaMarcone Jamilson Freitas Souza
Marcone Jamilson Freitas Souza
Screening of Pap smear images continues to depend upon cytopathologists’ manual scrutiny, and the results are highly influenced by professional experience, leading to varying degrees of cell classification inaccuracies. In order to improve the quality of the Pap smear results, several efforts have been made to create software to automate and standardize the processing of medical images. In this work, we developed the CEA (Cytopathologist Eye Assistant), an easy-to-use tool to aid cytopathologists in performing their daily activities. In addition, the tool was tested by a group of cytopathologists, whose feedback indicates that CEA could be a valuable tool to be integrated into Pap smear image analysis routines. For the construction of the tool, we evaluate different YOLO configurations and classification approaches. The best combination of algorithms uses YOLOv5s as a detection algorithm and an ensemble of EfficientNets as a classification algorithm. This configuration achieved 0.726 precision, 0.906 recall, and 0.805 F1-score when considering individual cells. We also made an analysis to classify the image as a whole, in which case, the best configuration was the YOLOv5s to perform the detection and classification tasks, and it achieved 0.975 precision, 0.992 recall, 0.970 accuracy, and 0.983 F1-score.

8 July 2022
Verified
Book Chapter
Closed access
PDF file
RVRafael F. Vitor
Rafael F. Vitor
BKBreno Nunes de Sena Keller
Breno Nunes de Sena Keller
DBDébora L. M. Barbosa
Débora L. M. Barbosa
DDDébora N. Diniz
Débora N. Diniz
MSMateus C. Silva
Mateus C. Silva
SDSaul Delabrida
Saul Delabrida

22 October 2018
Verified
Conference Paper
Open access
BKBreno Nunes de Sena Keller
Breno Nunes de Sena Keller
JLJoão Pedro Lonczynski
João Pedro Lonczynski
TDThiago D'ANgelo
Thiago D'ANgelo
SDSaul Delabrida
Saul Delabrida

3 April 2024
Verified
Conference Paper
Open access
PDF file
DSDanilo César S. Soares
Danilo César S. Soares
GFGabriel M. S. Ferreira
Gabriel M. S. Ferreira
AEAlan Erse
Alan Erse
BKBreno Nunes de Sena Keller
Breno Nunes de Sena Keller
MRMariana T. Rezende
Mariana T. Rezende
ABAndrea G. C. Bianchi
Andrea G. C. Bianchi
O aplicativo CitoFocus foi desenvolvido visando conectar profissionais em um ambiente propício a discussões que contribuem para a construção de diagnósticos citopatológicos. Logo, a aplicação exige níveis de qualidade elevados para não interferir na atividade principal, o diagnóstico. Este artigo propõe a remodelação do back-end, utilizando NestJS, com o objetivo de alcançar padrões de qualidade de software rigorosos. Adicionalmente, foram implementados testes de experiência do usuário para assegurar que a interação do usuário seja otimizada e atenda às suas necessidades de forma eficaz.

10 September 2026 in JMIR Formative Research
Verified
Journal Article
Open access
BKBreno Nunes de Sena Keller
Breno Nunes de Sena Keller
MRMariana T Rezende
Mariana T Rezende
RORenata Rr Oliveira
Renata Rr Oliveira
CCCristina Rp Costa
Cristina Rp Costa
CCclaudia carneiro
claudia carneiro
ABAndrea Gomes Campos Bianchi
Andrea Gomes Campos Bianchi
Background: Advances in hardware and software have transformed how users perform activities across various domains by integrating technology in ways that enhance task execution. In education, these processes facilitate pedagogical innovation and learning through diverse interaction models, such as web systems, augmented or virtual reality, and mobile devices. An example of these approaches occurs in the context of cytopathology. Digitization and slide visualization technologies allow the use of real images in a computational environment, enabling an alternative model of interaction between the professional and the samples. Objective: This study aims to propose a computational framework to support the teaching-learning process of visual perception skills, using cervical cytology as a case study. Methods: The framework described in the study provides interactive, adaptive exercises that simulate practical laboratory experiences, enabling students to engage with specific diagnoses and rare scenarios. A proof-of-concept system was implemented and evaluated with undergraduate students and professionals in cervical cytology. Results: Test participants answered an average of 88.36% (SD 10.30%) of the activities and correctly responded to an average of 69.86% (SD 14.78%) of the questions. Compared with previous approaches, these results indicated no significant difference in student performance compared with this work approach, suggesting that the proposed framework offers similar, if not improved, performance, while providing better usability and overall user experience. Additionally, the results demonstrate the system’s effectiveness in supporting cervical cytology learning while minimizing its impact on the user’s routine. Furthermore, the test participants reported a positive impression of the system and its approach to supporting the learning process in cervical cytology. Conclusions: The study highlights the framework’s potential for scalability and adaptation across other disciplines that require visual analysis, offering a promising avenue for enhancing education through technology.

11 November 2019
Verified
Conference Paper
Open access
PDF file
BDBruno Dias
Bruno Dias
BKBreno Nunes de Sena Keller
Breno Nunes de Sena Keller
SDSaul Delabrida
Saul Delabrida
A evolução dos conceitos pedagógicos deve ser fortemente suportada por tecnologias que se tornaram populares. Uma destas tecnologias é a Realidade Aumentada que se trata da mistura de elementos virtuais em ambientes reais. Este trabalho tem como principal motivação a oportunidade de usar Re- alidade Aumentada dentro de sala de aula como suporte ao ensino presencial. O acesso à tecnologia ainda precisa de novas interfaces para uso de profissio- nais dedicados ao ensino. Este trabalho apresenta uma ferramenta móvel para avaliação quantitativa e qualitativa dos ambientes de ensino presencial. Além da ferramenta, apresentamos a metodologia proposta que é implementada e um caso de estudo.