Project DetailsThe main goal of this work is to develop an algorithm which classifies depth of indentation for each shown cell in a given surface image. In the first step, we developed an algorithm that receives surface images and identifies which of the cells penetrated the gel substrate. In second step, we tried to calibrate the pushing depth of cells by a blur map.
A close relationship can be found between the penetration depth of cancer cells and their metastatic nature. The deeper the cells penetrate, the more metastatic they are in nature, meaning more dangerous for the patient. Until now, analyzing such images in laboratories was partly manual and partly automatic and took a long time. Therefore, we aimed to create an algorithm that would shorten the analysis while maintaining accuracy and correct identification. Unlike the current analysis which included microscope images from several different depths in the gel, we only had a single surface image at our disposal, which made the task very challenging. We found that the most successful way to implement the first step is to combine the algorithm created in a previous work (using a blur map) together with an algorithm based on the variance that can identify smooth areas in the image. In the second part of the project, we tried to understand if it is indeed possible to identify a relationship between the depth of the push and the degree of blurring. Since the difference between the pushing depths is very small (1-6 microns), our conclusion is that it is not possible to achieve such a calibration using only the blur map. However, computer vision and deep learning algorithms can identify feature vectors that we have difficulty identifying in the image and thus perhaps characterize different push depths better.
