PinkCC - Medical Image Viewer

A comprehensive medical imaging application for viewing and analyzing CT scans and segmentation data, with support for 2D, Multi-Planar Reconstruction (MPR), and 3D visualization.

Features

• Data Management: Load and browse patient CT data from different institutions (MSKCC, TCGA)
• Visualization Modes:
  • 2D Slice Viewer: View axial, coronal, and sagittal slices with adjustable window level/width
  • MPR View: Synchronized multi-planar reconstruction with crosshairs
  • 3D Visualization: Interactive 3D rendering with adjustable threshold and opacity
• Segmentation Overlay: Display tumor (red) and metastasis (green) segmentation overlays
• Performance Optimization: Image caching for improved responsiveness
• Customization: Adjustable window presets, quality settings, and display options

Project Structure

• local_view.py: Main application with the medical image viewer interface
• explore_dataset.py: Helper functions for dataset exploration and patient ID management
• assets/: Contains CSS files for styling
• DatasetChallenge/: Dataset directory containing CT scans and segmentation files
  • CT/: CT scan files in NIfTI (.nii.gz) format
    • MSKCC/: Memorial Sloan Kettering Cancer Center dataset
    • TCGA/: The Cancer Genome Atlas dataset
  • Segmentation/: Segmentation mask files
    • MSKCC/: Segmentation files for MSKCC dataset
    • TCGA/: Segmentation files for TCGA dataset

Requirements

The application requires the following Python packages:

• numpy
• nibabel
• matplotlib
• tkinter
• pillow
• plotly
• scikit-image
• pandas

Install the requirements using:

pip install -r requirements.txt

Usage

1. Run the local viewer application:

   python local_view.py
   

2. The application will open with the following interface:

   • Select an institution and patient ID from the dropdown menus
   • Use the tabs to switch between 2D, MPR, and 3D visualization modes
   • Adjust window level/width using sliders or presets
   • Toggle segmentation overlay and other display options

3. Explore the dataset separately:

   python explore_dataset.py
   

Implementation Details

local_view.py

The main application file implementing the medical image viewer with a tkinter GUI. Key components:

• MedicalImageViewer class: Main application class
• Helper functions for loading and processing medical images
• 2D, MPR, and 3D visualization implementations
• Image windowing and segmentation overlay functionality

explore_dataset.py

Utility module for exploring and managing dataset information:

• scan_dataset(): Scans dataset directory structure
• get_patient_ids(): Extracts unique patient IDs from available files
• generate_dataset_report(): Creates a summary of available images and patients

License

This project is for educational and research purposes.