Erick Colombo

Computer Scientist | Researcher | AI & Deep Learning

Computer Science student focused on research in Artificial Intelligence, Deep Learning, and computational methods applied to healthcare. This website serves as an open portfolio of my ongoing research, academic studies, and scientific experiments.

View Research Academic Projects

Research

Research in Progress Undergraduate Research

Influence of Chromatic Representations in Histopathological Breast Cancer Image Classification Using Convolutional Neural Networks

Research focused on evaluating how different chromatic representations of histopathological breast cancer images affect classification performance using Convolutional Neural Networks. The main objective is to perform binary classification between benign and malignant tissue.

Comparing Channels:
  • RGB (Baseline)
  • Grayscale
  • Red channel
  • Green channel
  • Blue channel
Evaluated Models:
  • ResNet50
  • EfficientNetV2
  • ConvNeXtV2
Evaluation Metrics:
  • Accuracy
  • Precision
  • Recall
  • F1-score
  • Confusion Matrix

Goal: Investigate the relationship between the reduction of chromatic information, input dimensionality, and computational requirements without compromising diagnostic accuracy.

Projects

Contributing to a private research project.

Project focused on medical image analysis using Vision Transformers and PyTorch for tumor classification, with a scalable FastAPI inference service and Docker-based infrastructure.

Faster R-CNN & ViT
Traffic Sign Recognition

Computer vision and deep learning study applied to the classification of traffic signs.

Tradicional CNN versus Resnet50
View Repository (GitHub)
(Currently private)

Research Interests

Artificial Intelligence Deep Learning Computer Vision Medical Image Analysis Digital Pathology Histopathological Image Analysis Convolutional Neural Networks Image Representation Machine Learning