Projects

This is an ongoing project of self-study/improvement where I create notebooks/markdowns of the techniques I’m learning. The initial idea is to have a cheat-sheet with notes for my own reference.

Code created for the papers are listed at the end of this page.

Study/Portifolio

• Survival Analysis: [Jupyter Notebook] [GitHub]
  • Language: Python
  • Library: LifeLines
  • Methods: Kaplan-Meier, Weibull, Exponential, LogNormal
  • Evaluation of covariates: Weibull and Cox Proportional Hazards models.

• Bee Images Classification: [Jupyter Notebook] [GitHub]
  • Language: Python
  • Features: HOG and Color pixel intensities (vector size of 31296 elements)
  • Feature transform: PCA (the best number of components was analyzed)
  • Classifier: SVM (linear kernel, binary classification)
  • Evaluation: Accuracy, ROC curve and AUC ROC

Papers

For a complete list of publications check my Google Scholar.

Sensitivity Analysis of Stroke Predictors Using Structural Equation Modeling and Bayesian Networks

• Conference: 2022 IEEE Conf. on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB)
• Code: Jupyter Notebook
• DOI: CIBCB55180.2022.9863028

Decision Support for Infection Outbreak Analysis: the case of the Diamond Princess cruise ship

• Conference: 2021 IEEE Symp. Series on Computational Intelligence (SSCI)
• Code: Jupyter Notebook
• DOI: SSCI50451.2021.9660140

A cross-cutting approach for tracking architectural distortion locii on digital breast tomosynthesis slices

• Journal: Biomedical Signal Processing and Control
• DOI: 10.1016/j.bspc.2019.01.001

Exploratory learning with convolutional autoencoder for discrimination of architectural distortion in digital mammography

• Conference: SPIE Medical Imaging 2019: Computer-Aided Diagnosis
• DOI: 10.1117/12.2513021

Reduction of false-positives in a CAD scheme for automated detection of architectural distortion in digital mammography

• Conference: SPIE Medical Imaging 2018: Computer-Aided Diagnosis
• DOI: 10.1117/12.2293388

Validation of no-reference image quality index for the assessment of digital mammographic images

• Conference: SPIE Medical Imaging 2016: Image Perception, Observer Performance, and Technology Assessment
• DOI: 10.1117/12.2217229

Use of Wavelet Multiresolution Analysis to Reduce Radiation Dose in Digital Mammography

• Conference: 2015 IEEE 28th Intl Symp. on Computer-Based Medical Systems
• Code: Matlab
• DOI: 10.1109/CBMS.2015.35

Feasibility study of dose reduction in digital breast tomosynthesis using non-local denoising algorithms

• Conference: SPIE Medical Imaging 2015: Physics of Medical Imaging
• DOI: 10.1117/12.2082398

© Helder C. R. Oliveira | Website made with Pandoc and R Markdown