Breast Cancer Diagnosis

• Based on Mammograms, I and my team developed a system that can diagnose the probability of cancer.

• We've reached to 85% sensitivity as the result matrix.

• Taking the unusual approach by using Visual Transformer as the base model, the results got improved.

A-Eye

• A wearable device for visually impaired people, that guides them to recognize, count, and read objects and text in front of them.

• This project contained two deep learning models, that work simultaneously, one for object detection, counting, and locating, and the second for OCR. The project uses an NVIDIA Jetson Nano to deploy deep learning models.

• The project was ranked 40th globally in the OpenCV AI Competition 2020.

Text to Image Generator

• Trained a GAN model to generate images based on text inputs.

• The project contains two models, 1) NLP model for text analysis, and 2) GAN architecture for image generation.

• It was my final year academic project I did this under the director of the L. J. Institute of Engineering and Technology Dr. Anil Suthar.

Monet Art Generator

• Developed a CycleGAN model that converts a picture into monet art.

• Trained the data on 300 the Monet paintings and generated more than 7000 images.