Search academic texts

Information × Registration Number 0222U001677, 0121U110569 , R & D reports Title Verification of key pathomorphological characteristics of tumor growth as innovative elements of artificial intelligence in the optimiziation of breast cancer diagnostic technology popup.stage_title Head Chekhun Vasyl F., Доктор медичних наук Registration Date 31-01-2022 Organization R.E. Kavetsky Institute of Experimental Pathology, Oncology and Radiobiology National Academy of Sciences of Ukraine popup.description2 For the diagnosis of breast cancer, adequate therapy and clinical prognosis, the size of the tumor, the status of regional lymph nodes and the degree of histological gradation, which is assessed by histological examination of slides (microphotographs). This process is a time-consuming complex task, and therefore encourages the development systems such as a computer program to automate the processing and analysis of histological sections of tumors. One of the stages of solving the problem is to create a database (DB) to support the researcher's decisions during pathomorphological verification and to "teach" the artificial neural network. An array of data on clinical and pathological characteristics of the tumor process, as well as indicators of overall and relapse-free survival of 310 patients with breast cancer (BC) I-II stages was analyzed and the signs associated with the degree of malignancy of this form of cancer were identified. It was selected a collection of tumor tissue samples from patients with breast cancer with different clinical status and degree of malignancy (n=31) (age 42-80 years, 36% of whom had regional metastases, 74% had invasive ductal, and 26% - invasive lobular breast cancer). The database, which was formed during the project implementation for the development of approaches to computer diagnostics and forecasting of breast cancer, had a complex hierarchical branched structure, the components of which were general information and information about the tumor accompanied by at least 5 slides of different breast cancer fields for each patient. Thus, the latest data were obtained, which allowed determining the system of informative attributes for identifying signs associated with aggressive breast cancer, to select a collection of tumor tissue samples from patients with different clinical statuses and degree of aggressiveness of the tumor process, and to create a database for the development of machine technology. Product Description popup.authors Zadvornyi Taras V. Kunska Lyubov M. Lukianova Nataliya Yu. Piatchanina Tetiana V. popup.nrat_date 2022-03-09 Close