Projects

Deep Neural Network in C with 2 layers

A deep neural network implementation in C, designed with a focus on parallelizing the training process. This project demonstrates the use of parallel computing techniques to enhance the efficiency and speed of neural network training using OpenMP and CUDA.

COPD Severity Staging using U-net

This project utilizes the U-Net architecture to stage the severity of Chronic Obstructive Pulmonary Disease (COPD) based on medical imaging data. The U-Net model is used for precise image segmentation, aiding in the automated classification of COPD severity, which can be crucial for clinical decision-making.

Perona Malik with Mimetic Method

An implementation of the Perona-Malik anisotropic diffusion model using the Mimetic Method. This project primarily focuses on denoising in image processing, enhancing image quality while maintaining important structural features.