CN-120997831-B - Cell aging identification method and system based on image processing

Abstract

The application provides a cell aging identification method and system based on image processing, which are applicable to the technical field of image processing, wherein the method comprises the steps of sampling, enhancing and extracting features of current cell image information and historical cell image information to obtain current cell image feature information and historical cell image feature information; obtaining target cell image characteristic classifying parameter information and target cell aging identifying calculating parameter information according to historical cell image characteristic information, initial cell image characteristic classifying parameter information, initial cell aging identifying calculating parameter information, cell image characteristic classifying parameter adjusting step length and cell aging identifying calculating parameter adjusting step length, and obtaining current cell aging identifying image information according to current cell image characteristic information, target cell image characteristic classifying parameter information and target cell aging identifying calculating parameter information. The application can accurately capture the key characteristics of cell morphology, texture and the like, and improve the accuracy and reliability of cell aging identification.

Inventors

• ZHAO JIANQI
• WANG SHIHONG
• SUN JINGJING

Assignees

• 溯龄(北京)人体生物健康技术服务有限公司
• 威海格瑞安生物工程有限公司

Dates

Publication Date

20260505

Application Date

20250821

Claims (9)

1. 1. A method for identifying cellular senescence based on image processing, comprising: acquiring a plurality of pieces of current cell image information and a plurality of pieces of historical cell image information; based on a plurality of preset cell image sampling vectors, a preset cell image horizontal edge enhancement vector and a preset cell image vertical edge enhancement vector, sampling and enhancing the plurality of current cell image information and the plurality of historical cell image information to obtain a plurality of current cell image enhancement information and a plurality of historical cell image enhancement information; performing feature extraction processing on the current cell image enhancement information and the historical cell image enhancement information according to a plurality of preset cell image feature extraction vectors to obtain the current cell image feature information and the historical cell image feature information; Calculating to obtain target cell image characteristic classification parameter information and a plurality of target cell senescence recognition calculation parameter information according to the plurality of historical cell image characteristic information, the randomly generated initial cell image characteristic classification parameter information, the plurality of randomly generated initial cell senescence recognition calculation parameter information, the preset cell image characteristic classification parameter adjustment step length and the plurality of preset cell senescence recognition calculation parameter adjustment step length; Calculating to obtain current cell aging identification image information according to the current cell image characteristic information, the target cell image characteristic classification parameter information and the target cell aging identification calculation parameter information; The step of calculating to obtain the target cell image feature classification parameter information and the target cell aging identification calculation parameter information according to the plurality of historical cell image feature information, the randomly generated initial cell image feature classification parameter information, the plurality of randomly generated initial cell aging identification calculation parameter information, the preset cell image feature classification parameter adjustment step length and the plurality of preset cell aging identification calculation parameter adjustment step length specifically comprises the following steps: randomly extracting the plurality of historical cell image characteristic information according to the randomly generated initial cell image characteristic classification parameter information to obtain a plurality of extracted historical cell image characteristic information; obtaining a plurality of residual historical cell image characteristic information according to the historical cell image characteristic information and the extracted historical cell image characteristic information; calculating Euclidean distances of the extracted historical cell image characteristic information and the residual historical cell image characteristic information to obtain historical cell image characteristic distance information; Classifying the residual historical cell image characteristic information according to the historical cell image characteristic distance information and the extracted historical cell image characteristic information to obtain initial historical cell image characteristic category information; Calculating the median and the mean value of the characteristic category information of the plurality of initial historical cell images to obtain the number of the characteristic categories of the plurality of initial historical cell images and the mean value of the characteristic categories of the plurality of initial historical cell images; Determining central information of a plurality of initial historical cell image feature categories according to the number of the plurality of initial historical cell image feature categories, the average value of the plurality of initial historical cell image feature categories and a preset cell image feature category parameter difference threshold; Judging whether the central information of the characteristic categories of the plurality of initial historical cell images is the same as the characteristic information of the plurality of extracted historical cell images; If yes, the characteristic category information of the plurality of initial historical cell images is used as characteristic category information of a plurality of target historical cell images; if not, the step of taking the central information of the characteristic categories of the plurality of initial historical cell images as the characteristic information of a plurality of extracted historical cell images and returning to the step of obtaining a plurality of residual historical cell image characteristic information according to the characteristic information of the plurality of historical cell images and the characteristic information of the plurality of extracted historical cell images; obtaining a plurality of historical cell senescence recognition image information according to the characteristic category information of the plurality of target historical cell images and a plurality of initial cell senescence recognition calculation parameter information which are randomly generated; Calculating a plurality of historical cell senescence recognition accuracy information according to the plurality of historical cell senescence recognition image information and a preset cell senescence recognition information base; Calculating the average value of the historical cell aging identification accuracy information to obtain the average value of the historical cell aging identification accuracy; Judging whether the average value of the historical cell aging recognition accuracy is larger than or equal to a preset historical cell aging recognition accuracy threshold; If yes, taking the randomly generated initial cell image feature classification parameter information as target cell image feature classification parameter information, and taking the plurality of randomly generated initial cell aging identification calculation parameter information as a plurality of target cell aging identification calculation parameter information; If not, adjusting the initial cell image characteristic classification parameter information and the initial cell aging identification calculation parameter information according to a preset cell image characteristic classification parameter adjustment step length and a plurality of preset cell aging identification calculation parameter adjustment step lengths to obtain intermediate cell image characteristic classification parameter information and intermediate cell aging identification calculation parameter information; and taking the intermediate cell image characteristic classification parameter information as initial cell image characteristic classification parameter information which is randomly generated, taking a plurality of intermediate cell aging identification calculation parameter information as a plurality of initial cell aging identification calculation parameter information which is randomly generated, returning to the step of randomly extracting the plurality of historical cell image characteristic information according to the initial cell image characteristic classification parameter information which is randomly generated, and obtaining a plurality of extracted historical cell image characteristic information.
2. 2. The method for recognizing cell aging based on image processing according to claim 1, wherein, The plurality of preset cell image sampling vectors comprise preset first-scale cell image sampling vectors and preset second-scale cell image sampling vectors; The step of sampling and enhancing the current cell image information and the historical cell image information based on the preset cell image sampling vectors, the preset cell image horizontal edge enhancing vectors and the preset cell image vertical edge enhancing vectors to obtain the current cell image enhancing information and the historical cell image enhancing information specifically comprises the following steps: Sampling the current cell image information and the historical cell image information based on a preset first-scale cell image sampling vector to obtain the current cell image sampling information and the historical cell image sampling information; Sampling the current cell image information and the historical cell image information based on a preset second-scale cell image sampling vector to obtain the second-scale current cell image sampling information and the second-scale historical cell image sampling information; Splicing the plurality of first-scale current cell image sampling information and the plurality of second-scale current cell image sampling information to generate a plurality of current cell image sampling information; Splicing the plurality of first-scale historical cell image sampling information and the plurality of second-scale historical cell image sampling information to generate a plurality of historical cell image sampling information; Performing enhancement processing on the plurality of current cell image sampling information and the plurality of historical cell image sampling information according to a preset cell image horizontal edge enhancement vector and a preset cell image vertical edge enhancement vector to obtain a plurality of current cell image edge enhancement information and a plurality of historical cell image edge enhancement information; And obtaining a plurality of current cell image enhancement information and a plurality of historical cell image enhancement information according to the plurality of current cell image edge enhancement information, the plurality of historical cell image edge enhancement information, the plurality of current cell image information and the plurality of historical cell image information.
3. 3. The method for recognizing cell aging based on image processing as claimed in claim 1, wherein the step of extracting the feature of the plurality of current cell image enhancement information and the plurality of history cell image enhancement information according to a plurality of preset cell image feature extraction vectors to obtain a plurality of current cell image feature information and a plurality of history cell image feature information comprises: pixel extraction is carried out on the plurality of current cell image enhancement information and the plurality of historical cell image enhancement information, so that a plurality of current cell image pixel information and a plurality of historical cell image pixel information are obtained; Extracting color information from the pixel information of the current cell images and the pixel information of the historical cell images to obtain the pixel color information of the current cell images and the pixel color information of the historical cell images; according to the pixel color information of the current cell images and the pixel color information of the historical cell images, carrying out matching processing on the pixel information of the current cell images and the pixel information of the historical cell images to obtain a plurality of pixel information matched with the current cell colors and a plurality of pixel information matched with the historical cell colors; Generating a plurality of current cell color matching pixel vectors and a plurality of historical cell color matching pixel vectors according to the plurality of current cell color matching pixel information and the plurality of historical cell color matching pixel information; And carrying out feature extraction processing on the current cell color matching pixel vectors and the historical cell color matching pixel vectors according to a plurality of preset cell image feature extraction vectors to obtain a plurality of current cell image feature information and a plurality of historical cell image feature information.
4. 4. The method for recognizing cell aging based on image processing according to claim 3, The plurality of preset cell image feature extraction vectors comprise preset cell image locking feature extraction vectors, preset cell image associated feature extraction vectors and preset cell image convergence feature extraction vectors; The step of performing feature extraction processing on the plurality of current cell color matching pixel vectors and the plurality of historical cell color matching pixel vectors according to a plurality of preset cell image feature extraction vectors to obtain a plurality of current cell image feature information and a plurality of historical cell image feature information specifically comprises the following steps: obtaining a plurality of current cell image locking feature information and a plurality of historical cell image locking feature information according to the plurality of current cell color matching pixel vectors, the plurality of historical cell color matching pixel vectors and a preset cell image locking feature extraction vector; Obtaining a plurality of current cell image associated feature information and a plurality of historical cell image associated feature information according to the plurality of current cell color matching pixel vectors, the plurality of historical cell color matching pixel vectors and a preset cell image associated feature extraction vector; obtaining a plurality of current cell image convergence feature information and a plurality of historical cell image convergence feature information according to the plurality of current cell color matching pixel vectors, the plurality of historical cell color matching pixel vectors and a preset cell image convergence feature extraction vector; Obtaining a plurality of current cell image locking related characteristic information according to the plurality of current cell image locking characteristic information and the plurality of current cell image related characteristic information; obtaining a plurality of historical cell image locking related characteristic information according to the plurality of historical cell image locking characteristic information and the plurality of historical cell image related characteristic information; Locking the associated characteristic information and the convergent characteristic information of the current cell images according to the current cell images to obtain the characteristic information of the current cell images; And locking the associated characteristic information and the convergent characteristic information of the plurality of historical cell images according to the plurality of historical cell images to obtain the characteristic information of the plurality of historical cell images.
5. 5. The method for identifying cellular aging based on image processing as claimed in claim 1, wherein the step of classifying the plurality of remaining historical cellular image feature information according to the plurality of historical cellular image feature distance information and the plurality of extracted historical cellular image feature information to obtain a plurality of initial historical cellular image feature class information comprises: The historical cell image characteristic distance information corresponding to the residual historical cell image characteristic information is arranged in an ascending order to obtain a plurality of historical cell image characteristic distance sequences; Taking the first element values of the plurality of historical cell image feature distance sequences as a plurality of historical cell image feature classification discrimination distances; And classifying the plurality of residual historical cell image characteristic information according to the plurality of historical cell image characteristic classifying and distinguishing distances and the plurality of extracted historical cell image characteristic information to obtain a plurality of initial historical cell image characteristic category information.
6. 6. The method for identifying cellular aging based on image processing according to claim 1, wherein the step of determining the central information of the plurality of initial historical cellular image feature categories according to the number of the plurality of initial historical cellular image feature categories, the average value of the plurality of initial historical cellular image feature categories and the preset cellular image feature category parameter difference threshold value specifically comprises: Calculating the number of the plurality of initial historical cell image characteristic categories and the difference value of the average value of the plurality of initial historical cell image characteristic categories to obtain a plurality of cell image characteristic category parameter difference values; Judging whether the difference value of the cell image characteristic category parameters is larger than a preset cell image characteristic category parameter difference value threshold value or not; If yes, determining the number of the initial historical cell image characteristic categories as initial historical cell image characteristic category central information; If not, determining the average value of the characteristic categories of the initial historical cell images as central information of the characteristic categories of the initial historical cell images.
7. 7. A cell senescence recognition system based on image processing, comprising: The cell image information acquisition module is used for acquiring a plurality of pieces of current cell image information and a plurality of pieces of historical cell image information; The cell image enhancement information generation module is used for carrying out sampling processing and enhancement processing on the current cell image information and the historical cell image information based on the preset cell image sampling vectors, the preset cell image horizontal edge enhancement vectors and the preset cell image vertical edge enhancement vectors to obtain the current cell image enhancement information and the historical cell image enhancement information; the cell image feature information generation module is used for carrying out feature extraction processing on the current cell image enhancement information and the historical cell image enhancement information according to a plurality of preset cell image feature extraction vectors to obtain the current cell image feature information and the historical cell image feature information; The target cell image characteristic classification parameter information and target cell aging identification calculation parameter information determining module is used for calculating to obtain target cell image characteristic classification parameter information and target cell aging identification calculation parameter information according to the plurality of historical cell image characteristic information, the randomly generated initial cell image characteristic classification parameter information, the plurality of randomly generated initial cell aging identification calculation parameter information, a preset cell image characteristic classification parameter adjustment step length and a plurality of preset cell aging identification calculation parameter adjustment step length; the current cell aging identification image information generation module is used for calculating current cell aging identification image information according to the plurality of current cell image characteristic information, the target cell image characteristic classification parameter information and the plurality of target cell aging identification calculation parameter information; The step of calculating to obtain the target cell image feature classification parameter information and the target cell aging identification calculation parameter information according to the plurality of historical cell image feature information, the randomly generated initial cell image feature classification parameter information, the plurality of randomly generated initial cell aging identification calculation parameter information, the preset cell image feature classification parameter adjustment step length and the plurality of preset cell aging identification calculation parameter adjustment step length specifically comprises the following steps: randomly extracting the plurality of historical cell image characteristic information according to the randomly generated initial cell image characteristic classification parameter information to obtain a plurality of extracted historical cell image characteristic information; obtaining a plurality of residual historical cell image characteristic information according to the historical cell image characteristic information and the extracted historical cell image characteristic information; calculating Euclidean distances of the extracted historical cell image characteristic information and the residual historical cell image characteristic information to obtain historical cell image characteristic distance information; Classifying the residual historical cell image characteristic information according to the historical cell image characteristic distance information and the extracted historical cell image characteristic information to obtain initial historical cell image characteristic category information; Calculating the median and the mean value of the characteristic category information of the plurality of initial historical cell images to obtain the number of the characteristic categories of the plurality of initial historical cell images and the mean value of the characteristic categories of the plurality of initial historical cell images; Determining central information of a plurality of initial historical cell image feature categories according to the number of the plurality of initial historical cell image feature categories, the average value of the plurality of initial historical cell image feature categories and a preset cell image feature category parameter difference threshold; Judging whether the central information of the characteristic categories of the plurality of initial historical cell images is the same as the characteristic information of the plurality of extracted historical cell images; If yes, the characteristic category information of the plurality of initial historical cell images is used as characteristic category information of a plurality of target historical cell images; if not, the step of taking the central information of the characteristic categories of the plurality of initial historical cell images as the characteristic information of a plurality of extracted historical cell images and returning to the step of obtaining a plurality of residual historical cell image characteristic information according to the characteristic information of the plurality of historical cell images and the characteristic information of the plurality of extracted historical cell images; obtaining a plurality of historical cell senescence recognition image information according to the characteristic category information of the plurality of target historical cell images and a plurality of initial cell senescence recognition calculation parameter information which are randomly generated; Calculating a plurality of historical cell senescence recognition accuracy information according to the plurality of historical cell senescence recognition image information and a preset cell senescence recognition information base; Calculating the average value of the historical cell aging identification accuracy information to obtain the average value of the historical cell aging identification accuracy; Judging whether the average value of the historical cell aging recognition accuracy is larger than or equal to a preset historical cell aging recognition accuracy threshold; If yes, taking the randomly generated initial cell image feature classification parameter information as target cell image feature classification parameter information, and taking the plurality of randomly generated initial cell aging identification calculation parameter information as a plurality of target cell aging identification calculation parameter information; If not, adjusting the initial cell image characteristic classification parameter information and the initial cell aging identification calculation parameter information according to a preset cell image characteristic classification parameter adjustment step length and a plurality of preset cell aging identification calculation parameter adjustment step lengths to obtain intermediate cell image characteristic classification parameter information and intermediate cell aging identification calculation parameter information; and taking the intermediate cell image characteristic classification parameter information as initial cell image characteristic classification parameter information which is randomly generated, taking a plurality of intermediate cell aging identification calculation parameter information as a plurality of initial cell aging identification calculation parameter information which is randomly generated, returning to the step of randomly extracting the plurality of historical cell image characteristic information according to the initial cell image characteristic classification parameter information which is randomly generated, and obtaining a plurality of extracted historical cell image characteristic information.
8. 8. A terminal device, characterized in that it comprises a memory, a processor, on which a computer program is stored which is executable on the processor, the processor executing the computer program to carry out the steps of the method according to any one of claims 1 to 6.
9. 9. A computer readable storage medium storing a computer program, characterized in that the computer program when executed by a processor implements the steps of the method according to any one of claims 1 to 6.

Description

Cell aging identification method and system based on image processing Technical Field The application belongs to the technical field of image processing, and particularly relates to a cell aging identification method and system based on image processing. Background In the field of cell image analysis, along with the continuous development of microscopic imaging technology and related research, the demands for accurate processing and analysis of cell images are becoming urgent. From cell image acquisition to subsequent analysis procedures, innovations continue to be pursued for more efficient, accurate results. In the prior art, a classical machine learning classifier, such as a support vector machine, is generally used to classify cell images and determine cell states so as to realize cell aging identification. However, in the prior art, the support vector machine is difficult to adapt to complex and changeable cell image data, so that the reliability and accuracy of cell aging identification are greatly reduced. Disclosure of Invention In view of the above, the embodiment of the application provides a cell aging identification method and system based on image processing, which aim to solve the problem of inaccurate cell aging identification in the prior art. A first aspect of an embodiment of the present application provides a cell senescence recognition method based on image processing, including: acquiring a plurality of pieces of current cell image information and a plurality of pieces of historical cell image information; based on a plurality of preset cell image sampling vectors, a preset cell image horizontal edge enhancement vector and a preset cell image vertical edge enhancement vector, sampling and enhancing the plurality of current cell image information and the plurality of historical cell image information to obtain a plurality of current cell image enhancement information and a plurality of historical cell image enhancement information; performing feature extraction processing on the current cell image enhancement information and the historical cell image enhancement information according to a plurality of preset cell image feature extraction vectors to obtain the current cell image feature information and the historical cell image feature information; Calculating to obtain target cell image characteristic classification parameter information and a plurality of target cell senescence recognition calculation parameter information according to the plurality of historical cell image characteristic information, the randomly generated initial cell image characteristic classification parameter information, the plurality of randomly generated initial cell senescence recognition calculation parameter information, the preset cell image characteristic classification parameter adjustment step length and the plurality of preset cell senescence recognition calculation parameter adjustment step length; And calculating the current cell aging identification image information according to the plurality of current cell image characteristic information, the target cell image characteristic classification parameter information and the plurality of target cell aging identification calculation parameter information. A second aspect of an embodiment of the present application provides a cell senescence recognition system based on image processing, including: The cell image information acquisition module is used for acquiring a plurality of pieces of current cell image information and a plurality of pieces of historical cell image information; The cell image enhancement information generation module is used for carrying out sampling processing and enhancement processing on the current cell image information and the historical cell image information based on the preset cell image sampling vectors, the preset cell image horizontal edge enhancement vectors and the preset cell image vertical edge enhancement vectors to obtain the current cell image enhancement information and the historical cell image enhancement information; the cell image feature information generation module is used for carrying out feature extraction processing on the current cell image enhancement information and the historical cell image enhancement information according to a plurality of preset cell image feature extraction vectors to obtain the current cell image feature information and the historical cell image feature information; The target cell image characteristic classification parameter information and target cell aging identification calculation parameter information determining module is used for calculating to obtain target cell image characteristic classification parameter information and target cell aging identification calculation parameter information according to the plurality of historical cell image characteristic information, the randomly generated initial cell image characteristic classification parameter information, the plur