SVCell seamlessly executes automated, high performance analysis, on par with custom solutions, for a broad range of microscopy applications. Customized algorithms and protocols for an entire application are stored in recipes, and the entire analysis can be executed with a single click. In addition, SVCell provides tools for image and data analysis, and visualization, within an innovative RecognitionFrame interface that is ideal for imaging experiments.

Recipes automate image import, enhancement, segmentation, object qualification, measurement and object classification. Individual sub-recipes are organized into a single recipe list for one click execution.

Import
Automate the import of individual image frames on file into multi-channel and time-lapse FOVs.

Enhance
Automate the application of SVCell’s sophisticated grayscale and binary processing functions, as well as common tasks like image bit depth conversion

Segment
Automate the detection and identification of biological objects of interest in the images through the application of SVCell’s machine learning enabled confidence mapping functions. The soft matching based confidence mapping is a revolution in the way image analysis is done in the Life Sciences.

Object Qualification
Automatically exclude objects from further analysis using basic morphological and grayscale intensity criteria.

Measure
Apply just the measurements you need every time. Each measurement recipe encodes a subset of a wide range of Field and object measurements. 16 bit and raitometric measurements are supported. Support is also provided for multi-channel, temporal and multi-mask objects.

Classify
Automatically classify complex phenotypes using SVCell’s unique decision teaching technology, the regulated decision tree. The decision recipe provides a way to accurately and robustly classify complex and subtle phenotypes that cannot be easily characterized with simple rules.

Image and Data Analysis
SVCell is designed for image recognition, and all the application elements are visualized through a single interface called the RecognitionFrame (or RFrame). The RFrame makes it easy to apply a recipe to a set of experimental images and quickly understand the experimental result. The data is presented hierarchically allowing you to drill down from group statistics, to field measurements to individual object measurements. The object region of interest overlays, charts and spreadsheets are all linked to facilitate data driven image investigation — a major time saver!