The Pulmonary Toolkit is a software suite for the analysis of 3D medical lung images for academic research use.

It consists of:

• a GUI application for visualising and analysing clinical lung images (CT & MRI);
• a rapid prototyping framework for running existing algorithms and developing new algorithms;
• an API which allows you to use the framework from within your own code, independent of the GUI application;
• a library of lung analysis algorithms which can be used independently.

This is experimental research software and is primarily intended to support our own work. However, we are happy for you to make use of the software, and we have therefore made the source code available for free under the open-source licence (GNU-GPL3).

This software requires:

• Matlab (version R2011a or later)
• A C++ compiler is needed for a number of the segmentation routines
• Matlab Image Processing Toolbox
• Matlab Statistics and Machine Learning toolbox is currently needed to support some of the analytics (specifically prctile)

This software is intended for research purposes only. It is not intended for clinical use.

Online manuals

PDF tutorials can be found in the Downloads folder after checking out the project, or you can download them directly here:

Installing the Pulmonary Toolkit

Tutorial 1 - Loading and visualising data

Tutorial 2 - Exporting data

Tutorial 3 - Programming with the Pulmonary Toolkit

Tutorial 4 - Lobar analysis of CT data

Please note that the tutorials only cover a few of the features of the Toolkit.

What can I do with the Pulmonary Toolkit?

There are many ways of using the Toolkit, for example:

• Use the GUI to load lung images from Dicom or mhd/raw files, perform automated analysis such as lobe segmentation or emphysema detection, and then save the results out;
• Write your own plugins to perform image analysis tasks, such as regional detection of lung disease;
• Write a Matlab script to perform automated analysis on hundreds of datasets using the Toolkit's API, for example gathering airway measurements;
• Using the PTKViewer tool to quickly view 3D datasets from the Matlab command window;
• Build your own medical application, by adding the Toolkit's image viewing panel (PTKViewerPanel) to your application;
• Use the Toolkit's suite of library functions to help in loading/saving, image processing (e.g. 3D watershed transforms) and image analysis

Requirements

To run the current alpha version you will need the following:

• Matlab version R2010b or later
• The Matlab Image Processing Toolbox
• A C++ compiler
• (recommended) a Git client

Releases

You can download and run the software but please be aware the Toolkit is currently in alpha. There is currently no stable release. We recommend you check out the latest version using Git and pull regularly from the master branch to obtain new features and bug fixes.

See the GitHub website for more information on how to obtain the source code. While you can download a zip file, I recommend you use Git as it is easier to obtain updates. Git clients are available for all operating systems

Please pull changes regularly from the GitHub repository to receive new features and fixes.

Support

Support is provided via the Tutorials and the wiki.

If you are experiencing problems, please make sure you have the required version of Matlab and the Imaging Processing Toolbox. Please also ensure you have a suitable C++ compiler installed and set up. The Toolkit will not work correctly without these.

Please update your git checkout to obtain the latest bug fixes.

The toolkit works primarily with medical Dicom images, but there is also limited support for mhd/mha files

License

You may download and use the Toolkit subject to the conditions of the GNU GPL v3 license. Note that under this license you can use the Toolkit in your own software, but if you do, and if you distribute your software to anyone else, then you must also make your software source code freely available. See the GNU GPL v3 license for details.

Note: Some parts of the software in the External folder are covered by different licences - see the licence files in the External folder for details.