Using the Python programming language for image processing in nuclear medicine

Learning Objectives After this presentation, attendees should 1) be able to list several advantages of the Python language, 2) be aware of how Python compares to other general purpose languages, 3) know how to obtain Python resources for their own use.

Python is a general purpose programming language with a number of potential advantages for image handling and data presentation in nuclear medicine. The language is open source, so its use incurs no cost or licensing. It runs under Microsoft Windows, MAC OS X, and linux, and has been ported to the Java and .NET virtual machines. Technical advantages, which will be demonstrated, include an English-like syntax, introspection, object orientation, dynamic typing, and a high degree of self-documentation. Most importantly for application to nuclear medicine, there are a large number of modules both included in the Python core and available as add-ons, for performing the kind of image manipulation and data presentation tasks useful in nuclear medicine contexts. These include numpy for the array manipulation needed in image processing, and matplotlib, for display of graphs and data plots. Examples of each of the basic tasks that can be performed will be demonstrated as part of a complete application, with example code and output, including: 1) interacting with a graphical user interface , 2) reading DICOM or Interfile images, 3) assigning regions of interest, 4) image processing functions, 5) time-activity curve construction and display, and 6) saving data and calculated results for later use. In this presentation, online sources for obtaining the language and various useful modules, and learning resources will be given.