AUTOMATIC CELL SEGMENTATION TO ANALYSE SINGLE CELL HETEROGENEITY BASED ON MORPHOLOGICAL FEATURES

The cytoskeleton is a structure that helps cells to maintain their shape and internal organization, and it also provides mechanical support that enables cells to carry out essential functions like division and movement. All cytoskeletons consist of two major classes of elements that differ in size and in protein composition. They are microtubules and microfilaments. Microtubules are the largest type of filament, with a diameter of about 25 nanometres (nm), and they are composed of a protein called tubulin. Microfilaments are the actin filaments, the smallest type, with a diameter of only about 6 nm, and they are made of a protein called actin. The ensemble behaviors of the cell population may not represent the behaviors of an individual cell. The behavior of each and every cell in a population is important to understand the accurate behavior of the cells. Automatic segmentation and extraction of multiple morphological features from cells using MATlab was performed to understand the behavior of cells.

Our dataset comprised of the images from two different stress levels i.e. 0.03 and 0.3 which corresponds to the Normal and diseased condition consecutively. For each stress level, the cells were imaged at three-time points i.e. 3 hours, 9 hours and 24 hours for actin, tubulin, and nucleus. We calculated all the morphological features and stored the values to excel to perform the normalization for each feature selected. The analysis of the features showed that single and overlapped cells have no difference.

Work flow of the process