Segmentation Based on Determinate Chaos of Histological Specimen Images
Abstract: Neuroblastoma is one of the most abundant tumors in infancy and ranks the fourth place among all malignant tumors of children, after acute leukemia, tumors of central nervous system and malignant lymphomas. Histological examination of atypical cell structures is mandatory for diagnosing and forecasting of neoplasm growth .. At present, the image processing of histological specimens is made by a professional histologist manually and is based on visual perception of the histologist. Naturally, visual approach has some shortcomings. Therefore, for objective estimation of medical images, increasing diagnosis accuracy and information processing speed, the computer analysis of histological images is a current need. In this paper, the features of the digital images of histological specimens and methods for digital analysis of medical images are presented. An innovative information system for instant diagnosis of the stage of neuroblastoma based on digital images analysis of histological specimens is developed. In presented studies, a new algorithm, based on deterministic chaos of digital images of histological specimens which uses original formalized indicators (irregularity of contoured zone of diagnostic interest, its structures heterogeneity including cells, nuclei, itochondrion
etc.) have been implemented.
Keywords: image analysis, segmentation, histological specimen, neuroblastoma
Area: Biomedical Engineering
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.