Whole-slide digital imaging is the process by which digital images of entire histologic slides are created by highresolution scanning of the slide. Digital imaging vendors may use slightly different technologies to produce these digital images, but most vendors use a tiling system, in which multiple individual high-resolution images are acquired as tiles, or a line-scanning system that creates linear scans of tissue areas. Both methods require the tiles and line scans to be stitched together and smoothened to create a single digital image of the histologic section.
Enables users to reconstruct big images/stacks from an arbitrary number of tiled input images/stacks. Stitching is based on the Fourier Shift Theorem, and computes all possible translations between pairs of 3D images. It avoids the propagation of errors by consecutive registration steps. Moreover, this tool uses phase correlation to find the translation between all image pairs and registers multi-tile acquisitions globally minimizing all pairwise registration errors.
Evaluates the precision of serial registration algorithms for 3D histology reconstructions based on serial histological sections. RegBenchmark is a framework that provides an estimation based on a panel of metrics including landmark (fiducial) correspondences and pixel-wise similarity metrics. Results provide a basis for development and application of 3D histology algorithms.
Serves for stitching of large 2D time-lapse mosaics. MIST can estimate the mechanical stage model parameters from the computed translations to limit the optimization search space. It also minimizes stitching errors by constraining and optimizing the translations within a square area. It can be used on any grid-based dataset like those acquired by a microscope. Moreover, this tool utilizes normalized cross correlations to compute image registrations.
Allows to run an ImageJ macro, possibly developed and tested on a single image, on a full digital slide. SlideJ exploits Bio-Formats basic input capabilities. It splits the slide in square tiles and iterates a user-configurable macro on each of them. The tool allows to construct rapid prototyping of image processing and analysis of digital slides using ImageJ. It is not able to consider digital slides with more than one focal plane.
Reconstructs and makes complex datasets easily accessible. BigStitcher is an ImageJ plugin that enables interactive visualization, precise alignment, real-time fusion, deconvolution, support for alignment of multi-tile acquisition taken from different physical orientations, or so-called multi-tile views. It can also be used for doubling the size of specimens that can be imaged, while further orthogonal views can render the resolution isotropic.
You can access more results by creating a free plan account or unlimited content via a premium account.