Kitware is the lead developer of 3D Slicer, an open-source application for medical image visualization, segmentation, and registration. Built using the Insight Segmentation and Registration Toolkit (ITK), the Visualization Toolkit (VTK) and CMake, in collaboration with the National Alliance for Medical Image Computing and Isomics, 3D Slicer has become a foundation for cutting-edge research and numerous pre-clinical and clinical products.
Last night, 3D Slicer was featured in a Podcast that demonstrated yet another use of 3D Slicer. Michael Balzer, host and founder of the ‘All Things 3D’ (Podcast) created a podcast entitled “Learn to turn your CAT (CT) or MRI scan into a 3D model”.
As highlighted in our previous post, printing three-dimensional (3D) anatomic models for medical scans is quickly becoming a popular and effective tool for surgical planning. Such patient-specific 3D models are increasing surgeons’ understanding of the critical anatomy involved and the surgical approaches that should be taken for each case, potentially enabling less invasive procedures to be determined and conducted with confidence.
During the two-hour live podcast, Michael demonstrated how to load a medical scan (Figure 1), visualize it and generate a model suitable for 3D printing using 3D Slicer.
The first half of the podcast focused on the creation and visualization (Figure 2) of a model of a brain tumor, using MRI data that is distributed with 3D Slicer. The output of their work was then saved in the STL format, which is commonly used by 3D printers (Figure 3).
In the second half of the podcast, Michael processed high-resolution computed tomography computed tomography (CT) data. Michael visualized it using 3D Slicer’s Volume Rendering module (Figures 4 and 5) and then applied a series of filters to generate another 3D model suitable for 3D printing.
As mentioned by Michael, 3D Slicer was particularly effective for generating these models of human anatomy not only because it provides tools for segmenting that anatomy, but also because it provided effective tools for “cleaning up” the noisy data and complex structures that are common to CT and magnetic resonance imaging (MRI) medical data. In particular, there is no “single click” solution to accurately extract, visualize and quantify anatomic structures in medical images for diagnosis, treatment planning, or 3D printing. It is often necessary to tune the parameters of existing medical image segmentation and registration algorithms and to combine those algorithms into data processing pipelines to achieve desired results. 3D Slicer makes that work easier by providing advanced algorithms such as gradient anisotropic diffusion, smoothing recursive gaussian, and opening by reconstruction, built using ITK or the python-wrapped version of ITK known as SimpleITK.