Back to Projects
Deep Learningintermediate

END 2 END Tumor Detection with XAI and GCP

OncoScan AI is a full-stack medical imaging application for detecting tumors in MRI (Brain) and Breast Scan images. It leverages Explainable AI (XAI) with GradCAM to provide heatmaps highlighting regions influencing AI's predictions, crucial for medical professionals, and features a production-grade MLOps pipeline.

24 lectures

What You Will Learn

Mastering the use of TensorFlow for implementing CNNs for medical image analysis.
Implementing GradCAM for generating Explainable AI (XAI) heatmaps to understand model decisions.
Building a REST API using FastAPI for handling prediction requests.
Orchestrating Docker containers using Kubernetes for scalable deployment.
Automating the CI/CD pipeline using GitLab CI/CD for seamless updates.
Implementing image processing techniques using OpenCV and Pillow for medical images
Understanding the complete MLOps pipeline from model development to deployment.

System Architecture

END 2 END Tumor Detection with XAI and GCP Architecture Diagram

High-level architecture overview of the END 2 END Tumor Detection with XAI and GCP .

What You'll Build

  • A tumor detection system for MRI and Breast Scan images.
  • A REST API endpoint for receiving image data and returning predictions.
  • A GradCAM heatmap visualization tool integrated into the frontend.
  • A Dockerized application for consistent execution across environments.
  • A Kubernetes deployment for scalable and highly available service.

Project Instructor

Divesh Jadhwani

Divesh Jadhwani

3+ years exp
LinkedIn
END 2 END Tumor Detection with XAI and GCP
Premium
One Subscription. 40+ Projects. Unlimited Access.
AccessMobile & Web