Standardization of image file formats is one of the key challenges facing
neuroimaging software development. The NiFTI image format is mature and
versatile, offering a common platform for imaging data interchange. However,
making an imaging analysis application NiFTI-capable can involve changes that
extend far beyond image file IO: with its ability to represent spatial
transformations between image space and anatomical space and the capacity to
store multi-component imaging data, NiFTI forces developers to make extensive
changes to their application’s architecture. My talk will summarize our team’s
experience of enabling NiFTI support in ITK-SNAP, a leading 3D image
segmentation tool.