Image Analysis

Advanced systems for comparative analysis and diagnosis. 
Analysis of both brightfield and fluorescence images, enabling the direct comparison of quantitative data between various groups. Our systems support cell segmentation, spatial analysis, pixel classification, and object measurement, facilitating in-depth insightful analytical data.


 

Cell Segmentation

In the context of cell segmentation, APS has the capability to identify the following components: positive cells, co-expression, and subcellular signals. Below, we provide a breakdown of the specifics of cell segmentation for the identification of each of these components. 

Positive cell detection

Quantify the signal intensity and define the positive cells in ROIs or whole slide. 

  • Quantify the positive cell number and the percentage among the whole cell population

  • Quantify the subclasses of positive cells with different expression level
CD-45 IHC Staining in Mouse Spleen
Quantification CD-45 (+) cells in Mouse Spleen
Cell detection

Co-expression detection

 Quantify the expression of multiple markers in a single cell level.

  • to classify different cell types
  • to quantify the level of co-expressed and co-localized
  • Exported data includes percentage of positive cells, percentage of co-expression in total cell population of a classified subgroup of cells in addition to the parameters mentioned in positive cell detection for each marker
Co-expression of CD-8 and Ki-67
Subcellular IHC signal of CD-8 and Ki-67

Subcellular signal detection

Quantify the signal intensity in different subcellar compartments, such as nucleus and cytoplasm

  • to classify different cell types

  • to determine the whether the markers are co-expressed and co-localized

  • common parameters of the exporting data includes percentage of positive cells, percentage of co-expression in total cell population of a classified subgroup of cells in addition to the parameters mentioned in positive cell detection for each marker

Spatial Analysis

 Spatial analysis is used to quantify the spatial associations within a tissue microenvironment, whether its between two categorized cell types of between a single cell type and a particular molecule. In the section below, we offer examples of spatial analysis within a single study. 
Original Multiplexed IHC Image
Nearest CD-68 (+) cells of each CD-20 (+) cells
Nearest PD-L1 (+) cells of each Ki-67 (+) cells

Pixel Classification

Pixel classification is the process of labeling individual pixels within a digital histological image. Each pixel is sorted into categories based on its attributes, which may include color, texture, or intensity. These categories correspond to specific elements, structures, or features within the studied tissue sample. This classification is commonly applied to segment and distinguish various elements in histological images, like cells, nuclei, connective tissue, or specific stains. Below, we offer an example where pixels within a tissue sample are classified as either negative or positive.

Area of sample to be selected
Classification of pixels as negative or postive

Object Measurement

Collaborating alongside pathologists, APS scientists have the ability to use various software tools to quantify the dimensions and spatial extent of damage or particular disease regions.
The length or depth of a wound
The size of germinal center of mouse spleen
The area of a wound
The size of mouse colon polyps
The angle of a wound
The size of necrosis in mouse liver