(a)
(b)
(c)
Fig. 1. (a) A serial section. (b) The adjacent section after being stained. (c) The images after registration.
|
(a)
(b)
(c)
Fig. 2. (a), (b) Cells before and after staining. (c) Registration of the images.
|
To obtain a license for this software, place an order here =>
Description
Semi-automatic nonrigid multimodality image registration customized for use with the Arcturus Veritas Microdissection software
|
Image Registration and Fusion Systems
|
The Nonrigid multimodality image registration software has been customized for use with the Veritas(tm) Microdissection
Instrument software from Arcturus Bioscience, Inc. www.bucher.ch/en/products/Arcturus-Bioscience.html. The following two
capabilities have been added:
- In addition to a reference and a target image, an alternate target image can be loaded and registered with the
reference image . The target and the alternate target images are assumed to have the same geometry although they
may have different intensities. After finding the transformation function that registers the reference and target images,
the transformation is also applied to the alternate target image to register with the reference image. This enables the
capture of two images of the same sample under different lighting conditions while keeping the camera and the
sample stationary. In molecular imaging, this functionality makes it possible to register images of cells before and
after staining for various studies.
- In addition to providing the filenames interactively, this customized software can read filenames in a file generated by
the Arcturus Veritas Microdissection software.
Examples of image registration by this software are given in Figs. 1 and 2 below. These images show two consecutive
sections of a tissue block with a thickness of less than a cell. One section is left unstained and one section is stained. The
staining process makes the cells easily visible, but the process damages the cells. By registering the two images, it
becomes possible to analyze the cells while viewing their details. Some local misregistration may be observed in the
registration of such images because adjacent sections do not have exactly the same local geometry. The software will ignore
very small local geometric differences between images and nonrigidly align the images using larger local neighborhoods.







