Skip to content

README.md

Latest commit

 

History

History
87 lines (53 loc) · 3.59 KB

README.md

File metadata and controls

87 lines (53 loc) · 3.59 KB
Raw

Python 3.8+ VTK License: MIT

3D Volumetric Medical Image Visualization

Overview

3D Volumetric Medical Image Visualization is a powerful Python-based pipeline for processing and rendering medical imaging data (Computed Tomography). By leveraging GPU-accelerated Volume Rendering via Ray Casting, it transforms raw 2D DICOM series into a fully interactive 3D environment, allowing for dynamic segmentation and exploration of anatomical structures.

Features

  • Reliable Data Unification: Seamlessly process and decompress raw DICOM slices (including JPEG Lossless scans) into a clean, single .mha volumetric file using SimpleITK.
  • High-Performance Rendering: Offloads heavy spatial calculations and ray-casting directly to the GPU via vtkGPUVolumeRayCastMapper.
  • Dynamic Tissue Segmentation: Isolate specific anatomical structures instantly using transfer functions based on the Hounsfield (HU) scale.
  • Zero-Reload Interactivity: Switch between visualization modes on-the-fly using keyboard observers—no memory reloading or application restarts required.
  • Trackball Camera: Smooth, native 3D interaction for rotating, zooming, and panning around the medical model.

Installation

Prerequisites

  • Python 3.8 or higher
  • A dedicated or modern integrated GPU to handle VTK Ray Casting

Setup

Clone the repository and install the required dependencies:

# Clone the repository
git clone [https://github.com/your-username/3d-medical-visualization.git](https://github.com/your-username/3d-medical-visualization.git)
cd 3d-medical-visualization

# Install the required libraries
pip install vtk SimpleITK

Usage

The application flow is divided into data preparation and visual inspection.

1. Data Preprocessing

Before rendering, compile your individual DICOM slices into a continuous 3D volume. Edit conversor.py to point to your local DICOM directory, then run:

Bash

python conversor.py

This will output a volume.mha file in your specified directory.

2. 3D Visualization

Once the .mha volume is generated, launch the interactive viewer:

Bash

python main.py

Controls & View Modes

Interact with the 3D scene using your mouse (Left-click to rotate, Scroll to zoom). Use the keyboard numbers to switch between specialized medical filters:

  • [ 1 ] Bone Mode: Filters out air and soft tissues, rendering only the skeletal structure with maximum opacity.

    Bone Mode Preview

  • [ 2 ] Soft Tissue Mode: Hides the internal skeletal structure, highlighting the patient's epidermis and facial muscles.

    Bone Mode Preview

  • [ 3 ] X-Ray (Translucent) Mode: Overlays views by assigning 15% opacity to the skin, revealing the solid bones inside.

    Bone Mode Preview

Contributing & Maintenance

This project is actively maintained for educational purposes in computer graphics, biometrics, and healthcare technology. We welcome contributions from software engineers and medical professionals!

If you'd like to improve the shading algorithms (ShadeOn), optimize performance, or add UI elements, please read our Contribution Guidelines before submitting a Pull Request.

License

This project is open-source and licensed under the MIT License. See the LICENSE file for details.