Erikson Aguiar

Erikson Aguiar

PhD Candidate

University of São Paulo -- ICMC/USP

About

Erikson Aguiar is a Ph.D. candidate at the University of São Paulo (USP), Brazil, in the Database and Image Group (GBDI). His research concerns security and privacy in Machine Learning for healthcare applications. He received his M.Sc. degree in Computer Science from the University of São Paulo (USP) in 2021. He completed his B.Sc. in Computer Science at the State University of Northern Paraná (UENP). His main research interest includes Security & Privacy, Machine Learning, Deep Learning, and Computer Vision. Also, he has technical skills in C, Golang, Node.js, Python, Git, TensorFlow, Pytorch, Scikit-learn, OpenCV, Docker, GCP, and AWS.

I relish coffee, sports, and music. When I leave the office, I enjoy watching football (supporting Flamengo🔴⚫ from Brazil), playing video games, and running.

Interests
  • Machine Learning & Deep Learning
  • Computer Vision
  • Security & Privacy
  • Distributed Systems
Education
  • PhD in Computer Science, ongoing

    University of São Paulo -- ICMC/USP

  • MSc in Computer Science, 2021

    University of São Paulo -- ICMC/USP

  • BSc in Computer Science, 2017

    State University of Nortern Parana -- UENP

Skills

Python
Golang
Pytorch/Tensorflow
Scikit-learn
OpenCV
Git

Recent Publications

Quickly discover relevant content by filtering publications.
(2023). Assessing Vulnerabilities of Deep Learning Explainability in Medical Image Analysis Under Adversarial Settings. CBMS'23.

(2023). A Deep Learning-based Radiomics Approach for COVID-19 Detection from CXR Images using Ensemble Learning Model. CBMS'23.

(2023). Security and Privacy in Machine Learning for Health Systems: Strategies and Challenges. IMIA'23.

(2022). Analysis of vertebrae without fracture on spine MRI to assess bone fragility: A Comparison of Traditional Machine Learning and Deep Learning. CBMS'22.

PDF Cite DOI

(2022). A blockchain-based protocol for tracking user access to shared medical imaging. FGCS.

PDF Cite DOI

Projects

Security and privacy in machine learning models to medical images against adversarial attacks
This project was financed by Sao Paulo Research Foundation (FAPESP) as Dotoracte Scholarship in Brazil (grant #21/08982-3) from Mach 01, 2022, to now.
Security and privacy in machine learning models to medical images against adversarial attacks
A trust protocol with multiple levels of access – case study with blockchain in medical applications
This project was financed by Sao Paulo Research Foundation (FAPESP) as Master Scholarship in Brazil (grant #18/18187-3) from April 01, 2019, to October 31, 2020.
A trust protocol with multiple levels of access -- case study with blockchain in medical applications

Teaching assistant

 
 
 
 
 
Image Processing
University of Sao Paulo - ICMC/USP
April 2023 – July 2023
 
 
 
 
 
Computer and Society
University of Sao Paulo - ICMC/USP
August 2022 – December 2022
 
 
 
 
 
Mining in large databases
University of Sao Paulo - ICMC/USP
August 2021 – December 2021
 
 
 
 
 
Introduction to programming (C language)
University of Sao Paulo - ICMC/USP
March 2020 – July 2020
 
 
 
 
 
Distributed Systems
University of Sao Paulo - ICMC/USP
August 2019 – December 2019

Contact

  • erjulioaguiar@usp.br
  • 400 Trabalhador São-carlense Avenue, Sao Carlos, SP 13566-590
  • Block 1, Office 1-116
  • Monday - Friday 08:00 to 17:00