Projects developed during the “Computational Intelligence Applied to Healthcare” course in my master’s program.
Tabular dataset#
In the first project, we worked with a tabular health-indicator dataset to classify individuals with diabetes, the CDC Diabetes Health Indicators dataset. We evaluated Random Forest, Support Vector Machines, Logistic Regression, and Extreme Gradient Boosting. Additionally, due to the class imbalance problem (diabetic vs. non-diabetic individuals), we explored a strategy to develop a machine learning model capable of handling this issue effectively.
- Source code: intel-comp-saude-ufes/2024-1-P1-classificador-diabetes
- Tools: Python, Matplotlib, Scikit-learn, Numpy, Pandas, Seaborn, Jupyter Notebook, Git, GitHub
Image dataset#
In the second project, we explored convolutional neural networks (CNNs), including ResNet and VGG19, using the LC25000 histopathological lung cancer image dataset. We applied transfer learning techniques and evaluated models results through cross-validation.
- Source code: intel-comp-saude-ufes/2024-1-P2-classificador-cancer-de-pulmao
- Tools: Python, Pytorch, Matplotlib, Numpy, Pandas, Seaborn, Jupyter Notebook, Git, GitHub