Detection of MCF-7 breast cancer cells and monitoring of area, radius, and perimeter developments of them in different days

Breast cancer is the most common cancer among women. In the world, these cells need to be analyzed well in order to develop effective treatment methods against cancer cells. In this study, image processing techniques were applied to breast cancer cells called MCF-7 cultured in the Lab-on-a-chip environment, and their development on 3 different days was examined. 3-dimensional images of breast cancer cells within spheroid used in the study were obtained with an optical-based imaging tool called mini-Opto tomography platform. Images containing three-dimensional spheroid structures were separated into two-dimensional layers with the ImageJ program before processing, and the images belonging to these layers were subjected to three main processes. In the first step, the pre-processing was performed on the images and with this operation, the disorders in the images were eliminated with the help of various filters and the images were adapted to the segmentation step. The main purpose of the segmentation step is to separate the main tumor masses in the spheroid structure from the image background. For this, the main tumor masses was brought to the fore by applying the thresholding, various morphological, and contour operations. The next step after the segmentation step is the "feature extraction" step that numerical data were obtained from some geometric properties of main tumor masses such as area, perimeter, and radius. As a result of the main operations applied to the images, the boundaries of the main tumor masses were drawn automatically and successful results were obtained compared to the manual drawings. When analyzed the volumetric development of 3-dimensional cancerous structures for three different days, the main tumor mass has grown approximately 121.72 percent from the second day to the fourth day and approximately 85.80 percent from the fourth day to the sixth day. These growth rates indicate that the main tumor masses used in the study tend to grow continuously. The data obtained from the MCF-7 breast cancer cells used in the study are very important for the volumetric development of cancer cells and for the classification processes that will be used in determining the type of cancer in the following stages. In the thesis study, using Python programming language and open source OpenCV library, early detection of cancer cells and a good analysis of their properties are very important in terms of providing clear and detailed information to experts. This information provided to the experts is vital in increasing the success of the treatment with early diagnosis, avoiding unnecessary biopsies, and reducing the time that the specialists interpret cancerous images.