Label-free cell two-dimensional scattering image inversion method based on gray level co-occurrence matrix

Abstract

The invention discloses a label-free cell two-dimensional scattering image inversion method based on a gray level co-occurrence matrix. The method comprises the following steps: acquiring a forward scattering image of a cell, calculating a gray level co-occurrence matrix of the image to solve an angular second moment, constructing a mathematical model between the cell volume and the angular secondmoment, and inverting the cell volume through the mathematical model and the angular second moment of the forward scattering image. Based on a forward two-dimensional scattering image of a cell, a quantitative law between the volume of the cell and forward scattering light of the cell is inversed; in order to find the quantitative law between the cell volume and forward scattering light thereof more simply and accurately, the scheme starts from global domain texture features of a cell two-dimensional forward scattering image, and proposes a gray-level co-occurrence matrix and a calculation method of an angular second moment of the gray-level co-occurrence matrix; and the angular second moment is used as an indirect representation cell volume, and a mathematical model between the cell volume and the angular second moment is constructed, so that the cell volume can be inversed more accurately and quickly.

Description

technical field [0001] The invention belongs to the research field of label-free cell scattering detection, and relates to a label-free cell two-dimensional scattering image inversion method based on a gray scale co-occurrence matrix. Background technique [0002] Cell light scattering is an important non-intrusive detection method. Light scattering is due to the interaction of electromagnetic waves with the medium, and the scattered waves carry a lot of information about the properties of the medium. The optical scattering information of cells not only contains the common characteristics of the same kind of cells, but also contains the individual characteristics of cells, and can accurately reflect the physical characteristics of biological cells in a non-intervention state, known as "cell fingerprint" information. [0003] The detection of human peripheral blood cells (circulating tumor cells, white blood cells, red blood cells, etc.) by means of fluorescent labeling and ...

AI Technical Summary

Problems solved by technology

This method has many processing steps, a large amount of calculation, and requires the establishment of a machine learning model, which cannot find the regular information corresponding to the cell volume

More importantly, although the machine learning method has good active intelligent discrimination ability, its discrimination criteria and conditions are a closed "black box", that is, the reflection between the two-dimensional scattering image information and the biological intrinsic physical characteristics cannot be known. law of play

Images