MedImager is a powerful, user-friendly, and research-oriented open-source medical image viewer. It aims to provide smooth image interaction, multi-format support (DICOM, PNG, etc.), and advanced analysis features for academic and clinical workflows.

Create a powerful, user-friendly, and research-oriented open-source medical image viewer. MedImager aims to provide smooth image interaction, multi-format support (DICOM, PNG, etc.), and advanced analysis features for academic and clinical workflows, aspiring to be an open-source alternative to RadiAnt.

• File Handling:
  • Open and parse DICOM series from folders.
  • Open single image files (PNG, JPG, BMP).
  • DICOM tag viewer.
• Image Display:
  • Smooth pan and zoom 2D viewer.
  • Multi-viewport for image comparison with flexible layouts.
  • Display patient info and image overlays (scale, orientation marker).
• Image Interaction Tools:
  • Windowing: Interactive adjustment of HU window width/level (WW/WL).
  • Measurement Tools:
    • Ruler tool for distance measurement.
    • Ellipse/rectangle/circle ROI tools.
  • ROI Analysis: Calculate statistics within ROI (mean, std, area, max/min HU).
• Advanced Features:
  • Multi-Series Management: Load and manage multiple DICOM series simultaneously.
  • Series-View Binding: Flexible binding system with auto-assignment and manual control.
  • Synchronization: Cross-viewport sync for position, pan, zoom, and window/level.
  • Layout System: Grid layouts (1×1 to 3×4) and special layouts (vertical/horizontal split, triple column).
• User Interface:
  • Modern multilingual interface (Chinese/English).
  • Customizable theme system (light/dark themes) with real-time switching.
  • Complete settings system with tool appearance customization.
  • Unified toolbar with theme-adaptive icons.
  • Dockable panel layout.

• Multi-Planar Reconstruction (MPR): View axial, sagittal, and coronal planes from 3D volume data.
• 3D Volume Rendering: Basic 3D visualization of DICOM series.
• Image Fusion: Overlay two different series (e.g., PET/CT).
• Annotation & Export:
  • Save annotation info (ROIs, measurements).
  • Export annotated views as PNG/JPG images.
• Plugin System: Allow users to extend features via custom Python scripts for research.

• Language: Python 3.9+
• GUI Framework: PySide6 (LGPL)
• DICOM Parsing: pydicom
• Numerical/Image Processing: NumPy
• 2D/3D Visualization: Qt Graphics View Framework (2D), VTK 9+ (3D)
• Packaging: PyInstaller
• i18n: Qt Linguist (pylupdate6, lrelease)

The project follows an MVC-like pattern to separate data logic, UI, and user interaction.

medimager/
├── main.py                 # Application entry point
├── icons/                  # UI icons and SVG resources
├── translations/           # Translation files (.ts, .qm)
├── themes/                 # Theme configuration files
│   ├── ui/                 # UI themes (dark.toml, light.toml)
│   ├── roi/                # ROI appearance themes
│   └── measurement/        # Measurement tool themes
│
├── core/                   # Core logic, UI-independent (MVC Model)
│   ├── __init__.py
│   ├── dicom_parser.py     # DICOM loading/parsing via pydicom
│   ├── image_data_model.py # Data model for single image or DICOM series
│   ├── multi_series_manager.py # Multi-series management and layout control
│   ├── series_view_binding.py  # Series-view binding management
│   ├── sync_manager.py     # Cross-viewport synchronization
│   ├── roi.py              # ROI shapes and logic
│   └── analysis.py         # Statistical calculations (HU stats, etc.)
│
├── ui/                     # All UI components (MVC View & Controller)
│   ├── __init__.py
│   ├── main_window.py      # Main window with multi-series support
│   ├── main_toolbar.py     # Unified toolbar management (tools, layout, sync)
│   ├── image_viewer.py     # Core 2D image viewer (QGraphicsView)
│   ├── viewport.py         # Standalone viewport with image_viewer
│   ├── multi_viewer_grid.py# Multi-viewport grid layout manager
│   ├── panels/             # Dockable panels
│   │   ├── __init__.py
│   │   ├── series_panel.py     # Multi-series management panel
│   │   ├── dicom_tag_panel.py  # DICOM tag panel
│   │   └── analysis_panel.py   # ROI analysis panel
│   ├── tools/              # Interactive tool implementations
│   │   ├── __init__.py
│   │   ├── base_tool.py        # Abstract base class for tools
│   │   ├── default_tool.py     # Default pointer/pan/zoom/window tool
│   │   ├── roi_tool.py         # ROI tools (ellipse, rectangle, circle)
│   │   └── measurement_tool.py # Distance measurement tool
│   ├── dialogs/            # Dialog windows
│   │   ├── custom_wl_dialog.py # Custom window/level dialog
│   │   └── settings_dialog.py  # Application settings dialog
│   └── widgets/            # Custom UI widgets
│       ├── __init__.py
│       ├── magnifier.py        # Magnifier widget
│       ├── roi_stats_box.py    # ROI statistics display
│       └── layout_grid_selector.py # Layout selection widget
│
├── utils/                  # General utilities (MVC Model Support)
│   ├── __init__.py
│   ├── logger.py           # Global logging configuration
│   ├── settings.py         # User settings management
│   ├── theme_manager.py    # Theme system with icon management
│   └── i18n.py             # Internationalization utilities
│
├── tests/                  # Unit/integration tests
│   ├── __init__.py
│   ├── dcm/                # Test DICOM data
│   ├── scripts/            # Test data generation scripts
│   ├── test_dicom_parser.py
│   ├── test_roi.py
│   └── test_multi_series_components.py
│
├── pyproject.toml          # Project metadata and dependencies
└── README_zh.md            # Chinese documentation

First, ensure you have uv installed. It is an extremely fast Python package installer and resolver.

1. Clone the repo:

   git clone https://github.com/1985312383/MedImager.git
   cd MedImager

2. Setup Environment and Install Dependencies:

   # Create a virtual environment and sync dependencies from pyproject.toml
   uv venv
   uv sync

3. Run the app:

   # `uv run` executes the command within the project's virtual environment,
   # avoiding the need to activate it in your shell.
   uv run python medimager/main.py

   For developers who prefer an active environment:

   # To activate the environment in your current shell:
   # Windows
   .venv\\Scripts\\activate
   # macOS / Linux
   source .venv/bin/activate
   
   # Then you can run commands directly:
   python medimager/main.py

Contributions are welcome! Whether you're fixing a bug, adding a feature, or improving documentation, your help is appreciated. Please feel free to open an issue or submit a pull request.

This project is licensed under the GNU GENERAL PUBLIC LICENSE. See the LICENSE file for details.