![]() From a user’s stand point, the parameters should be intuitive such as setting the minimum and maximum size of colonies expected in an assay. For the analysis of the whole data set, more flexible parameters must be chosen, since the individual images may vary greatly from one another in object size, object morphology and image intensity. For each individual step, algorithmic parameters are required to obtain the targeted colonies in each individual image. However, this leads to individual steps of a typical image analysis pipeline: Noise removal, image segmentation, feature extraction, feature selection, object counting/classification. Based on this segmentation, not only the counting of CFUs should be performed, but additional features of interest such as size, shape etc. ![]() In order to automate the analysis, one may need to develop an algorithm that automatically segments the colonies from the unwanted background in such a way that CFU boundaries are obtained. Therefore, an automatic, high accuracy analysis is needed. When the difference in change to be observed between the experiments is marginal, finding the size and number of CFUs/cells with great accuracy becomes imperative. For instance, in some experiments, it is desirable to see the change in cell/CFU morphology or size by the effects induced in the cells/CFUs under control and test conditions. However, to make a quantitative analysis based on other CFU features is a nontrivial and even more complex task than just enumerating all the colonies. ![]() One may be able to count the CFUs manually provided that adequate time is available. The colony forming units (CFUs) or cells found in images produced during microbiological assays are hard to analyze manually in an efficient way, especially when the images and/or the amount of CFUs/cells are large in number. However, this leads to a huge amount of image data containing many cells/colonies depending upon the type of experiment being performed and the effects under observation. Biologists perform numerous experiments in the laboratory based on cell culture under different conditions and acquire images for analysis. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |