Welcome to Pixel Diagnose! 🧠Our mission is to empower radiologists with state-of-the-art artificial intelligence, enhancing the accuracy of differential diagnoses for cancer patients.
This repository is dedicated to our Jupyter notebooks, pivotal in visualizing, segmenting, and modeling for Pixel Diagnose. Utilizing Google Cloud, including Google Colab and Colab Enterprise, these notebooks detail our methodologies in preprocessing image data, developing models, and generating embeddings, all within a cloud-based environment.
- Notebooks Folder: Our Jupyter notebooks here are central to our workflows, covering image preprocessing (applying segmentation masks, extracting slices), model development, and embedding generation.
- Helper Folder: Contains bash shell scripts for effective file management on Google Cloud Storage, ensuring efficient data handling.
For more details on the Pixel Diagnose project, including our overarching goals, a demo video, and information on the frontend and backend development, please see our organization profile.
Empowering Radiologists, One Pixel at a Time. 🧠💡