A Tumor Detection Model/Application

• 

  An overview

Inspiration

Machine learning and deep learning technologies are increasing at a fast pace with respect to the domain of healthcare and medical sciences. These technologies sometimes even out perform medical doctors by producing results that might not be easily notable to a human eye. Tumor (polyp) recognition and segmentation is one great technology which helps doctors identify tumors from colonoscopic images.

What it does

The core of the application is a deep learning model trained to detect tumors. The model was implemented using PyTorch, a powerful open-source machine learning library for Python.

How we built it

This application uses Deep Learning and Computer Vision techniques to detect cancerous & non-cancerous tumors in medical images.

Tech Stack; Language used : Python Deep learning library used : PyTorch Computer vision library used : OpenCV Other python libraries : Scikit-learn, pandas, numpy, albumentations etc. Deployment: Streamlit

Challenges we ran into

One interesting challenge faced during the project was Training the Model (A GPU was required since model training takes a really long time in CPUs). Also, a lot of issues occurred when pushing the codes to GitHub, which slowed down deployment. Apparently I wasn't supposed to push environment (.env) files, which was later understood.

Accomplishments that we're proud of

Accomplishing this project and a successful deployment.

What we learned

Learned a lot about Git/Github Usage and Image Processing using Opencv Libraries

What's next for A Tumor Detection Model/Application

Project Improvement and Advancement in a large-scale setting.

Built With

• albumentations
• numpy
• opencv
• pandas
• python
• pytorch
• scikit-learn
• streamlit