FIELD: medicine.
SUBSTANCE: invention relates to the field of medicine and can be used to recognise the structure of blood and bone marrow blast nuclei using light microscopy in combination with computer data processing. According to the invention, a colour image of blood and/or bone marrow smears is obtained using a computer analyser; leukocytes are isolated in the image, and a binary image thereof is obtained; a void-filling procedure is performed on the binary image of leukocytes by incremental scanning; the spatial brightness distribution of pixels of the image is measured, and the value of textural attributes characterising the nuclear structure is determined for the representations of colour models of leukocytes; based on the textural attributes, a matrix of numerical textural attributes of the nuclear structure is formed, wherein the columns contain the values of textural attributes and the rows contain the leukocyte images used to recognise the structure of blast nuclei. The values of textural attributes characterising the structure of leukocyte nucleus are determined for four directions of adjacency, wherein the textural attribute "local uniformity" for the red component of the colour image is determined using adjacency distances of eleven pixels, the textural attribute "moment of inertia" for the red component of the colour image is determined by calculating for the adjacency distance of two pixels, the textural attribute "moment of inertia" for the blue component of the colour image is determined using adjacency distances of seven pixels, the textural sign "entropy" for the blue component of the colour image is determined using adjacency distances of six pixels, and the textural attribute "energy" for the blue component of the colour image is determined using adjacency distances of six pixels.
EFFECT: increased reliability of diagnosing due to the use of attributes allowing for high informative value.
6 cl, 5 dwg
Title | Year | Author | Number |
---|---|---|---|
METHOD FOR RECOGNITION OF THE STRUCTURE OF BLOOD BLAST NUCLEI AND BONE MARROW USING LIGHT MICROSCOPY IN COMBINATION WITH COMPUTER DATA PROCESSING FOR THE DIAGNOSIS OF B- AND T-LINEAR ACUTE LYMPHOBLASTIC LEUKEMIA | 2017 |
|
RU2659217C1 |
METHOD OF DETAILING THE STRUCTURAL ELEMENTS OF BIOMEDICAL RESEARCH OBJECTS | 2023 |
|
RU2803277C1 |
METHOD FOR MICROSCOPIC EXAMINATION OF A SAMPLE WHICH CONTAINS MICRO-OBJECTS WITH HETEROGENEOUS ZONES | 2006 |
|
RU2308745C1 |
METHOD FOR RECOGNITION OF CELL TEXTURE IMAGE | 2008 |
|
RU2385494C1 |
METHOD OF DIAGNOSING ONCOLOGICAL DISEASES OF THE BLOOD | 2022 |
|
RU2803281C1 |
METHOD FOR AUTOMATED BLOOD CELLS ANALYSIS THROUGH LEUKOCYTES DESCRIPTION BASED ON OPTICAL FEATURES OF NUCLEI STRUCTURES | 2014 |
|
RU2612007C2 |
METHOD OF PIGMENT SKIN GROWTHS RECOGNITION | 2018 |
|
RU2712919C1 |
METHOD OF RECOGNITION AND MEASUREMENT OF DIAGNOSTIC PARAMETERS OF CYTOLOGICAL PREPARATIONS | 1993 |
|
RU2088922C1 |
METHOD FOR DIAGNOSING SKIN MELANOMA | 2021 |
|
RU2780367C1 |
METHOD OF ESTIMATING CLINICAL EFFECTIVENESS IN CHRONIC MYELOID LEUKEMIA | 2012 |
|
RU2495430C1 |
Authors
Dates
2022-12-09—Published
2021-11-19—Filed