Hosted by NumFOCUS

SciPy brings together attendees from industry, academia and government to showcase their latest projects, learn from skilled users and developers, and collaborate on code development. We are excited to have a poster accepted this year co-authored by Matt McCormick, Ph.D. “Obtain quantitative insights through image registration in python.”

To understand changes that occur in 2D and 3D scientific imaging datasets, registration is required. Registration is the task of finding a spatial transformation that aligns one image to another by optimizing relevant image similarity metrics. Whether the images datasets come from biological microscopy, neuroscience medical imaging volumes, remote sensing images in geoscience, or microscopy images encountered in materials science or chemistry, image registration enables quantification of changes that occur under different experimental conditions or over time. For example, alignment of mouse liver volumes across subjects given an experimental treatment for cirrhosis can provide insights into the drug’s effectiveness. Or, registration of lung volumes over time can characterize diseases like asthma.

In this talk, Matt will present itk-elastix, an easy-to-use and mature python package for registration of 2D and 3D scientific images. This open source tool for rigid and nonrigid registration is generally effective on a variety of imaging datasets for rigid, affine, or b-spline deformable registration. And, itk-elastix’s modular design allows users to quickly configure, test, and compare different registration methods for a specific application. Moreover, the elastix Model Zoo is available to find parameters that work well with specific datasets.

This talk will also introduce relevant concepts and practices for image registration in python. At the end of the talk, attendees unfamiliar with registration will have working knowledge of the concepts involved and how to perform registration in python with itk-elastix.

Physical Event

AT&T Hotel and Conference Center
Austin, Texas
Virtual Location