Image Analysis Software for Cell Biology Experiments

Using image analysis software in cell biology experiments

Image-based experiments underpin much of the work conducted to deeply study and understand cell biology. Image analysis software allows scientists to delve deeper into the cells on microscope samples, enabling the evaluation of important, quantitative information about the target cell subjects. This information depends on an investigator’s research question and could include data on a cell’s and tissues’ physical features, cell-to-cell interactions, intracellular functioning, whole organism physiology and even a cell’s dynamics with viral or bacterial agents, such as in disease research.

Automated Image analysis software technology can support analysis based on either single cells or a cell population, analyzing numerous images in a single go.  Coupling such software with automated microscopes for real-time analysis simultaneous to image acquisition streamlines experimental process in the laboratory.

Our previous post about High Content Imaging (HCI) introduced High Content Analysis (HCA) as it involves the use of automated image analysis software.  Automated image analysis is completely controlled by computers algorithms that receive one or more images as an input and extract information from objects within those images.  These algorithms begin by segmenting images into objects of interest, which are selected given certain pixel value properties.  Fluorescence images, having specific labelling and high contrast, are ideal for such analysis.  Indeed, the analysis of fluorescence images imparts an important aspect to the power of fluorescence labelling and the continued need to develop new fluorescent labels.  In fluorescence images, bright objects of interest are segmented away from the dark background that is ignored.  Transmitted light images are can also be analyzed to identify structures by identifying changes in pixel intensity values, which often identify the edges of objects.

Once segmented, each individual object is measured to extract metrics such as statistics regarding pixel intensity values, and object shape and morphology (e.g. area, shape, elongation, etc.).  Advanced analysis can also study the interaction of objects identified in different color channels, such as colocalization of multiple signals in overlapping regions.  Data obtained using multiple imaging channels can also allow for interpretation of the signals, such as presence within a cellular location such as the nucleus vs. cytoplasm.  Automated image analysis is powerful because it is unbiased.  Once segmentation parameters are defined and set, researchers can regularly and reliably apply them to multiple datasets.  So long as the experimental preparation and imaging conditions are consistent, the data extracted can be directly compared to one another.  While some misidentification can occur with automated approaches, with appropriate controls and sample size the resulting systematic error can be reduced to provide statistically significantly results.  Biological differences or changes identified in this fashion are more reliable than those obtained through manual image analysis, which always involve the subjective decisions of a researchers performing the analysis. 

Our WiScan® Athena image analysis software can capture, quantify, and extract morphological and intensity-based data from cells using advanced analysis algorithms. Furthermore, visualization tools offered by the Athena software allows for analysis results to be displayed in an easily presentable and shareable format. With this software’s ability to be multi-functional and malleable to best serve the needs of its users, life sciences experiments save on time and resources. This in turn, makes space for more innovation, creativity, and critical thought around the results being provided by cutting edge image analysis software.