Формування концепції автоматизованої обробки та аналізу цифрових зображень субстратзалежних клітинних систем

Abstract

Розроблено загальну концепцію автоматичної обробки та аналізу цифрових зображень в мікроскопії з подальшим створенням алгоритмів кількісного аналізу візуальних даних щодо живих субстратзалежних клітинних систем на прикладі перещеплювальної культури клітин аденокарциноми гортані людини HEP-2.

Разработана общая концепция автоматической обработки и анализа цифровых изображений в микроскопии с последующим созданием алгоритмов количественного анализа визуальных данных о живых субстратзависимых клеточных систем на примере перепрививаемой культуры клеток аденокарциномы гортани человека HEP-2.

Background. Cell systems are widely used in various fields of biology and medicine, physiology, pharmacology, classical virology etc. Microscope remains the main instrument of the researcher. But today automated analysis of digital images for evaluating the results of microscopic studies is used more and more often. Because of this, obtaining the accurate images, providing transfer of images from the optical system of microscope to the computer and further automated analysis with the extensive use of computer technologies are necessary. Statement of the problem. Our previous studies described a method of standardizing numerical methods for evaluation of microscopic images of anchorage-dependent cell systems in vitro. The proposed approach of obtaining and processing images of cell monolayers gives the promising results. However, the general concept of automatic processing and analysis of digital images remains undeveloped. Solving this problem first of all will significantly facilitate the work and reduce the research time due to the fact that most of mechanical steps on edge detection and calculating main geometric shape parameters will become automated providing significant increase of the array of data being processed. In addition this approach will improve the accuracy of edge detection thus providing representativeness of the sample. So the aim of the work is to develop a common concept of automated processing and analyzing digital images of anchorage-dependent cell systems. Research methodology. Research is conducted using inoculated cell cultures of adenocarcinoma of human larynx HEP-2 obtained from the collection of cell cultures of the RE Kavetsky Institute of Experimental Pathology, Oncology and Radiobiology, National Academy of Sciences of Ukraine, Kyiv. The cell monolayers are grown in culture polystyrene mattress with growth surface area of 425 cm2 and crop concentration of 106 cells/cm2. The cells were cultured for 72 hours at 37° C followed by microscopic examination after 48 and 72 hours from the time of inoculation. Microscopic study of native cell monolayers were carried out under an inverted microscope PrimoVert in transmitted bright field illumination. Image of the cell monolayer visualized using digital color cameras Digital Microscopy Camera AxioCam ERc5s with using Carl Zeiss software. Preprocessing of digital images was performed using Otsu’s method. Conclusions. The overall concept of automatic processing and analyzing of digital images in microscopy followed by the creation of algorithms for quantitative analysis of visual data on living anchorage-dependent cell systems is presented. The proposed solutions will improve the accuracy of automatic edge detection in the cells images with the possibility of correction in manual mode. They also will automate statistical analysis of the cells parameters in large samples.