![]() ![]() ![]() Trainable WEKA segmentation comes preinstalled with the FIJI build of ImageJ and is convenient to use, since the training dataset is created by annotating the objects and background directly in ImageJ. Through labeling objects in the image and then making it a training set for the classifier, once the result has been obtained and improved by providing feedback, the classifier model can be used to classify similar images. Advances in machine learning potentially provide a way to overcome this issue by transforming the segmentation problem into a pixel classification problem. The built-in ImageJ-based tool by itself, however, encounters limitations for correctly segmenting overlapping objects. ![]() Furthermore, the setting and operating steps in ImageJ can be recorded and composed into a batch of commands, or macros, for easy execution in the subsequent, repeat analysis. The Watershed tool, also preinstalled in ImageJ, applies the watershed algorithm to separate overlapping objects based on the edges. Other preinstalled tools, Threshold and Analyze Particles, can be applied to highlight and count the objects according to the size and circularity threshold. For example, the Find Maxima tool of ImageJ determines the local intensity of the image, and users can adjust the prominence level for the object of interest (OOI) as a threshold for counting. Available from the public domain, ImageJ can be adopted to manual counting or semi-automated counting using various built-in tools and community-supplemented plugins. Software such as ImageJ provides an array of tools to assist the counting process, ranging from object counter, image editing and object calculation. Furthermore, they are not always suitable for day-to-day operations, since tweak and adjustment are often required or a larger volume of cells is needed.ĭigital image analysis can facilitate manual counting to improve the efficiency and consistency. These expensive and specialized devices, however, may not be accessible for laboratories with limited resources or under an educational setting. Meanwhile, advanced machineries, such as the Coulter counter, flow cytometer, image cytometer and microfluidic cytometer, have been applied to quantify cell numbers. This method only requires basic equipment and a dozen microliters of cell culture but has a disadvantage of being low throughput, prone to human errors and subject to high variation. The most common method to count Tetrahymena (or cultured cells in general) is using the hemocytometer, which is a popular laboratory technique for counting cells manually. By counting the cell number under different treatments and time courses, it is easy to monitor growth inhibition and assay the effect of a potential toxicant. As a eukaryotic microorganism, Tetrahymena grows rapidly in the laboratory and divides every 2–3 h in the optimal condition, making it a superb experimental system for toxicological analysis. These properties make the ciliated protozoan an appropriate organism to determine the health of an aquatic environment. Additionally, they are able to consume free organic material from the environment if necessary. They play an integral part in the community by connecting the food chain between bacteria and small phytoplankton to larger metazoa and zooplankton. ![]() The optimized methods reported in this paper provide scientists with a convenient tool to perform cell counting for Tetrahymena ecotoxicity assessment.Ĭiliated protozoa are unicellular eukaryotes commonly found in aquatic environments. Among the five methods tested, all of them could yield decent results, but the deep-learning-based SDM displayed the best performance for Tetrahymena cell counting. We conducted Tetrahymena number measurement using five different methods: particle analyzer method (PAM), find maxima method (FMM), trainable WEKA segmentation method (TWS), watershed segmentation method (WSM) and StarDist method (SDM), and compared their results with the data obtained from the manual counting. In this study, we established the ImageJ-based workflow to quantify ciliate numbers in a high-throughput manner. Although advanced counting devices are available, the specialized and usually expensive machinery precludes their prevalent utilization in the regular laboratory routine. Previous methods to measure protozoan numbers mostly rely on manual counting, which suffers from high variation and poor efficiency. ![]()
0 Comments
Leave a Reply. |