Automatic GUI Based Counting Software for Immunostaining Analysis Using Matlab GUIDE

Abstract

Typically, researchers need to the manually count the number of cells in many histological sections. However, this manual counting method is labor intensive and time consuming. Manual counting is also subject to bias, where two individuals (who are blinded to each other’s results) normally count each section. Considering all these reasons, a user friendly automated Graphical User Interface (GUI) based counting software has been developed using the MATLAB® environment that greatly accelerates the process and improves the preciseness of data determination. Manual counting does not consider color intensity and requires approximately thirty minutes per image to calculate the cell counts. Whereas the user-friendly automated GUI based counting software can implement an identical analysis with enhanced quantitative capabilities in less than five minutes. The enhancements of this automated GUI software are greatly escalated speed of counting, rerun the process at a specific point if anything is selected incorrectly, minimization of counter bias and the ability to get accurate quantitative estimates of staining intensity. After uploading an image in the automated GUI software, the user has to first select either a color channel (i.e. red channel, green channel and blue channel) or the original image itself. Then the image go through manual segmentation process, where the user has the ability to remove iv unwanted regions and select any area of the stained image. Later, the image is thresholded to spot prominent stains where the user can remove unwanted background or noise. Finally, the software counts the cells using various cell counting methods such as entire area cell counts, drawing a free hand ROI, drawing a box, using the watershed algorithm, and manual counting. Several correlation studies have been performed to compare manual versus automated GUI cell counting. The coefficients of determination for DAPI and β–galactosidase positive cells are good, but moderate in case of Sclerostin. This automated GUI software method is a major advancement in terms of osteocyte counting from histological sections, and the procedure is explicitly compatible to diverse staining methods and color combinations. A standalone application is created for the automated GUI with the help of MATLAB® Compiler that can run in any computer.