2D Automated Analysis

Fast. Accurate. Repeatable.

Change the way you segment, classify and analyze images

Introducing the most efficient and repeatable process for analyzing complex measurements across multiple images and channels.

NEW

Analysis Protocols

Powered by Machine Learning

Use Smart Segmentation Machine Learning to find and extract relevant regions per channel, properly classify, and accurately measure using an intelligent pixel classification recipe.

Multi-Channel by Design

Analyze relationships between and across multiple channels, wells/sites, and regions seeding the results of one channel with those of another.

Customizable Across Similar Routines

Adjust settings and then save new versions of any Protocol to create your own customized Protocol. Custom name your channels, segmentations, and results.

Easy, Repeatable Performance

Batch Process a folder full of single images or a single image set with many wells/sites contained within — delivering consistent and repeatable results every time.

Fast Analysis in Parallel

Multi-threaded processing runs analysis or multiple images in parallel across all available CPU Cores.

Relevant Data at your Fingertips

Quick access to powerful relationship data in Data Collector tables, Graphs, Charts, and Custom reports auto-generated for each Protocol.

See Protocols in Action

So Many Protocols to Choose From

Image-Pro's library of Protocols is rapidly expanding and making image analysis easier than ever. Each Protocol is packed with powerful analysis tools designed to deliver flexible and intuitive options.

INCLUDED

Essentials Collection

Counts and classifies objects in the region of interest.

Measures objects (children) within objects (parents).

Measure percentage area covered by segmented objects.

OPTIONAL

Cell Biology Collection

Measures apoptosis based on caspase activation of fluorescently labeled cells.

Measure the number of fluorescently labeled autophagosomes in cells containing fluorescently labeled nuclei.

Count fluorescently labeled nuclei.

Measure the morphology of fluorescently labeled cells containing fluorescently labeled nuclei.

Measure the fluorescently labeled total cell population and the proportion of fluorescently labeled proliferating cells.

Measure the area and estimated cell count of cells in brightfield images.

Measure the proportion of fluorescently labeled cells that contain three or more fluorescently labeled lipid droplets (adiposomes).

Measure fluorescently labeled populations of cells. Display as Total, Live and Dead (all cells), Live and Dead, Total and Live, and finally Total and Dead.

Measure a fluorescently labeled target within a ring mask outside, inside or overlapping fluorescently labeled nuclei, cells, or other structures.
Use with 1 Ring and 1 to 2 Targets, or with a cell surface or plasma membrane ring mask and a cytoplasm ring mask.

Measure the proportion of cells in transmitted light images expressing a fluorescently labeled recombinant protein or 2 fluorescent labels.

OPTIONAL

Cell Biology Plus Collection

Measures the growth and diameter of fluorescently labeled blood vessels.

Measure the correlation and overlap of two fluorescently labeled molecules, both as object-based and image-based.

Measure fluorescently labeled neurons and cell bodies containing fluorescently labeled nuclei.

Measure either the total cellular area of transport of a fluorescently labeled target protein between fluorescently labeled nuclear and cytoplasmic compartments or just the area above threshold.

Measure cell migration rates as the gap left by a “scratch” closes.

OPTIONAL

Materials Collection

Measure the area of pores and constituents and calculate the area fraction of a composite material.

Measure the thickness of concentric layer(s).

Measure the thickness of curved layer(s).

Measure the thickness of horizontal, vertical or curved layers.

Separate crossing and overlapping fibers, and measure the fiber thickness, length and orientation in optical or SEM micrographs.

Measure both the thickness and orientation of fibers in optical or SEM micrographs.

Measure and classify particles.

Measure the phases and calculate the area fraction with individual particles.

Measure phases and calculate the area fraction.

Measure pores in a range of materials such as composites, porous silicon and manufactured holes in substrates.

Smart Segmentation for Free Form Analysis

Whether using segmentation within a Protocol or as a free form segmentation tool, Image-Pro’s Smart Segmentation technology uses a proven method of pixel classification to identify hard-to-segment objects and regions. This Machine Learning-based method helps identify faintly colored objects, textured objects, and accounts for uneven backgrounds.

This is the foremost solution for gathering data from images and our simple stepwise approach is designed for ultimate flexibility to analyze nearly any image type while remaining simple to learn and train.

How it Works

A simple four-step process that uses a pixel classification algorithm to identify hard-to-segment objects and regions:

Paint target classes and a background
Name your classes, easily identify representative pixels of your objects, and set a background. Paint more pixels to compensate for uneven backgrounds and colors across images.

Define recipe parameters
Build a unique “recipe” including Intensity, Color, Background, and Morphological Filters. For more complex samples, pre-apply custom filters for even better results.

Train over multiple images
It’s rare when one image fully represents a large set acquired over time, so Smart Segmentation uses a Global Recipe option that can be trained from multiple images in the set.

Filter by Measurement
Filter out objects to be counting or sized using specific range restrictions (graphically or numerically) by any number and combination of parameters.

Create object outlines
Once objects are identified and outlined you can count and measure areas, percent area, regions, intensity values, and more.

Split & Merge
Splitting objects that are touching is necessary in many images so use watershed or boundary shape-based splitting techniques when possible, or simply draw a split line to get the job done.

Learning Classification
Choose your reference objects, set the parameters to be used for building a classification recipe, and then automatically apply classes to all objects.

Auto Classification
Define groups to be created, choose the measurement parameters for the classification, and then apply hierarchical clustering to the objects.

Single Variable Classification
Define classification bins, set measurement ranges, and assign classes to objects according to the bin ranges they fall in. Use any measurement to classify objects by that “single” variable.