Skip to content

Sruthij93/BrainTumorClassification_model

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 

Repository files navigation

Brain Tumor Classification ML model

This project leverages advanced deep learning techniques to assist in the diagnosis of brain tumors using MRI scans. It aims to classify tumors into four categories: glioma, meningioma, pituitary tumors, or no tumor, providing an interactive interface for users, including patients and medical professionals.

Tech Stack

Backend

  • Developed using Python, TensorFlow, and Keras for building and training the deep learning models.

Frontend

  • Built with Streamlit, providing a user-friendly, interactive interface for uploading MRI scans and generating predictions.

Models

  • A pre-trained Xception model with transfer learning for high accuracy.
  • A custom-built CNN model for comparative insights.

Visualizations

  • Powered by Plotly for detailed bar charts of class probabilities.
  • Utilized OpenCV for generating saliency maps to highlight critical areas of the MRI scan.

LLM Integration

  • Incorporated Gemini, Llama, and Gemma2 for generating textual explanations and a comprehensive report.

Project Highlights

Model Predictions

  • The system classifies uploaded MRI scans and provides a detailed probability distribution for the predicted tumor type.

Saliency Maps

  • Generates visual overlays highlighting the critical regions the model focuses on during classification.
  • Aims to aid medical professionals in better understanding the results.

Interactive LLM-Powered Insights

  • Answers are dynamically generated by multimodal LLMs, combining text and image analysis.

Comprehensive Reporting

  • Provides a detailed report including:
    • Model predictions.
    • Additional insights.
    • Actionable next steps for patients and doctors.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published