CN119310075A

CN119310075A - A method for identifying slices using a hyperspectral microscope combined with HE staining

Info

Publication number: CN119310075A
Application number: CN202411329446.9A
Authority: CN
Inventors: 毛静涛; 张猛; 仲丹丹
Original assignee: Beijing Dragon Midas Science And Technology Development Co ltd
Current assignee: Beijing Dragon Midas Science And Technology Development Co ltd
Priority date: 2024-09-24
Filing date: 2024-09-24
Publication date: 2025-01-14

Abstract

The invention relates to the field of biomedical engineering technology, and discloses a method for identifying slices stained with a hyperspectral microscope combined with HE. The method comprises a data acquisition module, a data analysis module, an abnormal data processing module, a data management module and a control module, wherein the data acquisition module acquires data, the data analysis module calculates an automatic staining control index Ql, a microscope imaging stability coefficient Sl and an image quality assessment score Gl, the abnormal data processing module identifies abnormal staining data in an automatic staining process and abnormal data in a microscope imaging process according to the calculated numerical values, and judges whether the quality of the final imaging is up to standard, and the data management module performs data correction and fault diagnosis on the abnormal staining data in the automatic staining process and abnormal data in the microscope imaging process according to the abnormal data automatically identified by the abnormal data processing module, and sends the corrected parameters and fault diagnosis information to the control module, and the control module issues a repair instruction.

Description

Identification method for HE (high-intensity) stained sections by combining hyperspectral microscope with HE (high-intensity) staining

Technical Field

The invention relates to the technical field of biomedical engineering, in particular to a method for identifying HE (high-intensity) stained sections by combining a hyperspectral microscope.

Background

In traditional medical pathology, the morphological structure of tissue samples can be clearly observed by pathologists through H & E staining. Although the traditional method has important value, the traditional method depends on the effect of chemical dyeing, is easily influenced by various factors such as the type of fixing agent, the dyeing reagent, the performance of a slicing machine and the like, and leads to unstable imaging quality and complex operation. To address these issues, hyperspectral microscopy imaging techniques (HMI) have evolved and the application of multispectral analysis on HE stained sections provides new solutions. The HMI technology can capture more spectrum information so as to analyze and identify tissue types in more detail, control of automatic dyeing and an image processing technology are realized, and the identification efficiency and the consistency of results are remarkably improved. The introduction of the technology not only maintains the advantages of the traditional HE staining, but also strengthens the deep understanding of the characteristics of pathological tissues, and brings new visual angles and tools for pathological diagnosis and research.

Disclosure of Invention

(One) solving the technical problems

Aiming at the defects of the prior art, the invention provides a method for identifying the HE dyed section by combining a hyperspectral microscope, which has the advantages of automatic dyeing control and stable imaging quality, and solves the problems of dependence on chemical dyeing and unstable imaging quality in traditional medical pathology.

(II) technical scheme

In order to achieve the aim, the invention provides the following technical scheme that the identification method of the hyperspectral microscope combined with the HE staining section comprises the following steps:

Step one, a data acquisition module, a data analysis module, an abnormal data processing module, a data management module and a control module are established;

Step two, the data acquisition module is divided into an automatic staining data capturing unit, a microscope imaging data receiving unit and a staining slice data output unit;

Step three, the data analysis module is divided into an HE dyeing slice identification unit and a microscope imaging stabilization unit, and an automatic dyeing control index Ql, a microscope imaging stabilization coefficient Sl and an image quality evaluation score Gl are calculated;

and step four, respectively identifying abnormal dyeing data in the automatic dyeing process and abnormal data in the microscopic imaging process by the abnormal data processing module according to the automatic dyeing control index Ql and the microscopic imaging stability coefficient Sl, and judging whether the quality of final imaging meets the standard or not according to the image quality evaluation score Gl.

Preferably, the method comprises a data acquisition module, a data analysis module, an abnormal data processing module, a data management module and a control module;

The data acquisition module comprises an automatic dyeing data capturing unit, a microscope imaging data receiving unit and a dyeing slice data output unit, wherein the automatic dyeing data unit acquires automatic dyeing data through a hyperspectral imaging system, the microscope imaging data receiving unit acquires microscope imaging data through a CCD camera, the dyeing slice data output unit acquires dyeing slice data through an imaging spectrometer, and the spectral data capturing unit, the microscope imaging data receiving unit and the dyeing slice data output unit are connected with the data analysis module through a network;

The data analysis module comprises an HE (high-speed) dyed slice identification unit and a microscope imaging stabilization unit, wherein the HE dyed slice identification unit calculates an automatic dyed control index Ql according to automatic dyed data, the microscope imaging stabilization unit calculates a microscope imaging stabilization coefficient Sl according to microscope imaging data, the microscope imaging stabilization unit calculates an image quality evaluation score Gl according to dyed slice data, and the HE dyed slice identification unit and the microscope imaging stabilization unit are connected with the abnormal data processing module through a network;

The abnormal data processing module respectively identifies the dyeing abnormal data in the automatic dyeing process and the abnormal data in the microscopic imaging process according to the automatic dyeing control index Ql and the microscopic imaging stability coefficient Sl, judges whether the quality of the final imaging meets the standard according to the image quality evaluation score Gl, and is connected with the data management module through a network.

Preferably, the automatic dyeing data unit performs data numbering on the initially set dyeing time, the initial concentration of the dyeing liquid, the background dyeing value, the dyeing environment temperature influence value and the dyeing instrument correction factor according to the automatic dyeing data characteristics, the initially set dyeing time, the initial concentration of the dyeing liquid, the background dyeing value, the dyeing environment temperature influence value and the dyeing instrument correction factor are numbered T, C, B, mu and epsilon, and the HE dyeing slice identification unit calculates an automatic dyeing control index Ql according to the automatic dyeing data, wherein the calculation formula is as follows:

in the formula, ql represents an automatic dyeing control index, T, C, B, mu and epsilon respectively represent an initial set dyeing time, an initial concentration of a dyeing liquid, a background dyeing color value, a dyeing environment temperature influence value and a dyeing instrument correction factor, and (1-mu) represents a constant value excluding the temperature influence value.

Preferably, the microscope imaging data receiving unit respectively performs data numbering on gray values of all images according to the features of the microscope imaging data, and the gray value number of the images is f ₁、f₂、f₃、…f_n.

Preferably, the microscope imaging stabilization unit calculates a microscope imaging stabilization coefficient Sl according to microscope imaging data, and a calculation formula is as follows:

in the formula, sl represents a microscope imaging stability coefficient, f ₁、f₂、f₃、…f_n represents a gray value of an image, n represents the number of images, f _i represents a gray value of an i-th image, Representing the average gray value of all images.

Preferably, the stained section data output unit performs data numbering on the signal intensity of the stained area and the background noise level according to the characteristics of the stained section data, the signal intensity of the stained area is numbered as E ₁、E₂、E₃、…E_n, and the background noise level is numbered as R ₁、R₂、R₃、…R_m.

Preferably, the microscope imaging stabilization unit calculates an image quality evaluation score Gl according to the stained slice data, and the calculation formula is as follows:

Where Gl represents the image quality assessment score, E ₁、E₂、E₃、…E_n represents the signal intensity of the stained area, R ₁、R₂、R₃、…R_m represents the background noise level, Represents the average signal intensity of the stained areas, n represents the number of stained areas counted,Represents the average background noise level, and m represents the number of statistical background noise levels.

Preferably, the abnormal data processing module automatically identifies an initial concentration abnormal value or a background dyeing abnormal value of the dyeing liquid in the automatic dyeing process according to the automatic dyeing control index Ql, and automatically identifies a sharpening abnormal value or a contrast abnormal value of the imaging system according to the microscope imaging stability coefficient Sl.

Preferably, the abnormal data processing module judges whether the quality of final imaging meets the standard according to the image quality evaluation score Gl, when the image quality evaluation score Gl is more than or equal to 95, the imaging quality is excellent, when the image quality evaluation score Gl is less than or equal to 90 and less than or equal to 95, the imaging quality is good, and when the image quality evaluation score Gl is less than or equal to 95, the imaging quality meets the standard, and when the image quality evaluation score is 80.

Preferably, the data management module corrects the dyeing abnormal data in the automatic dyeing process and the abnormal data in the microscopic imaging process according to the abnormal data automatically identified by the abnormal data processing module, and performs fault diagnosis, and simultaneously sends the corrected parameters and fault diagnosis information to the control module, and the control module issues a repair instruction, and the control module is connected with the data acquisition module, the data analysis module, the abnormal data processing module and the data management module through a network.

Compared with the prior art, the invention provides a method for identifying the HE stained sections by combining a hyperspectral microscope, which has the following beneficial effects:

1. According to the invention, the automatic dyeing control index Ql is calculated, the abnormal data processing module compares the automatic dyeing control index Ql with a historical preset standard range, when the automatic dyeing control index Ql exceeds the preset range, the initial concentration of the dyeing liquid is indicated to be too high, the dyeing is too deep, abnormal data is transmitted to the data management module, the data management module can automatically reduce the concentration of the dyeing liquid to adjust the dyeing depth to a proper dyeing depth, when the automatic dyeing control index Ql is lower than the preset range, the background dyeing value is indicated to be too low, the dyeing is insufficient, the data management module can correspondingly adjust the background dyeing value to ensure full dyeing, the adjustment process is favorable for realizing quality control of the dyeing process, so that each dyed slice can achieve the optimal effect, and the accuracy of pathological diagnosis is improved.

2. According to the invention, by calculating the microscope imaging stability coefficient Sl, when the microscope imaging stability coefficient Sl is smaller than 1, the condition that the imaging system has fluctuation or blurring is indicated, so that the image quality is reduced, when an abnormal condition is identified by the abnormal data processing module, the stability of the imaging system (comprising a light source, a microscope optical system and a sample position) can be automatically checked, so that all components are in a stable state, when the microscope imaging stability coefficient Sl is larger than 1, the condition that the imaging system is over-sharpened or the contrast is over-high is indicated, at the moment, image details can be lost, the data management module can automatically adjust the focusing or optical setting of the microscope, so that a more balanced image effect is obtained, and the problems of dependence on chemical dyeing and unstable imaging quality in traditional medical pathology are solved.

Drawings

FIG. 1 is a schematic diagram of the structure of the present invention.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

Referring to fig. 1, a method for identifying a slice by combining a hyperspectral microscope with HE staining includes the following steps:

The method comprises a data acquisition module, a data analysis module, an abnormal data processing module, a data management module and a control module;

The abnormal data processing module respectively identifies dyeing abnormal data in the automatic dyeing process and abnormal data in the microscopic imaging process according to the automatic dyeing control index Ql and the microscopic imaging stability coefficient Sl, judges whether the quality of final imaging meets the standard according to the image quality evaluation score Gl, and is connected with the data management module through a network;

The automatic dyeing data unit carries out data numbering on the initially set dyeing time, the initial concentration of the dyeing liquid, the background dyeing value, the dyeing environment temperature influence value and the dyeing instrument correction factor according to the characteristic of the automatic dyeing data, wherein the initially set dyeing time, the initial concentration of the dyeing liquid, the background dyeing value, the dyeing environment temperature influence value and the dyeing instrument correction factor are numbered T, C, B, mu and epsilon, and the HE dyeing slice identification unit calculates an automatic dyeing control index Ql according to the automatic dyeing data, and the calculation formula is as follows:

in the formula, ql represents an automatic dyeing control index, T, C, B, mu and epsilon respectively represent the initial set dyeing time, the initial concentration of the dyeing liquid, the background dyeing color value, the dyeing environment temperature influence value and the dyeing instrument correction factor, and (1-mu) represents a constant value excluding the temperature influence value;

The automatic dyeing control index Ql is compared with a historical preset standard range by the abnormal data processing module, when the automatic dyeing control index Ql exceeds the preset range, the initial concentration of the dyeing liquid is indicated to be too high, the dyeing is too deep, abnormal data are transmitted to the data management module, the data management module can automatically reduce the concentration of the dyeing liquid to adjust the proper dyeing depth, when the automatic dyeing control index Ql is lower than the preset range, the background dyeing value is indicated to be too low, the dyeing is insufficient, the data management module can correspondingly adjust the background dyeing value to ensure the full dyeing, the adjustment process is favorable for realizing the quality control of the dyeing process, so that each dyed slice can achieve the best effect, and the accuracy of pathological diagnosis is improved.

And the microscope imaging data receiving unit respectively carries out data numbering on the gray values of all the images according to the characteristics of the microscope imaging data, and the gray value number of the images is f ₁、f₂、f₃、…f_n.

The microscope imaging stabilization unit calculates a microscope imaging stabilization coefficient Sl according to microscope imaging data, and a calculation formula is as follows:

in the formula, sl represents a microscope imaging stability coefficient, f ₁、f₂、f₃、…f_n represents a gray value of an image, n represents the number of images, f _i represents a gray value of an i-th image, Representing the average gray value of all images;

The method has the advantages that by calculating the microscope imaging stability coefficient Sl, when the microscope imaging stability coefficient Sl is smaller than 1, fluctuation or blurring exists in the imaging system, so that the image quality is reduced, when an abnormal condition is identified by the abnormal data processing module, the stability (comprising a light source, a microscope optical system and a sample position) of the imaging system can be automatically checked, so that all components are in a stable state, when the microscope imaging stability coefficient Sl is larger than 1, the image detail can be lost at the moment, and the focusing or optical setting of the microscope can be automatically adjusted by the data management module, so that a more balanced image effect can be obtained.

The stained section data output unit numbers the signal intensity of the stained area and the background noise level according to the characteristics of the stained section data, the signal intensity of the stained area is numbered as E ₁、E₂、E₃、…E_n, and the background noise level is numbered as R ₁、R₂、R₃、…R_m.

The microscope imaging stabilization unit calculates an image quality evaluation score Gl according to the stained slice data, and the calculation formula is as follows:

Where Gl represents the image quality assessment score, E ₁、E₂、E₃、…E_n represents the signal intensity of the stained area, R ₁、R₂、R₃、…R_m represents the background noise level, Represents the average signal intensity of the stained areas, n represents the number of stained areas counted,Representing the average background noise level, m representing the number of statistical background noise levels;

The method has the advantages that the quality grade of an image is evaluated by calculating the image quality evaluation score Gl, so that the method is applied to subsequent required pathological analysis according to the grade required by the pathology, the risks of misdiagnosis and missed diagnosis can be reduced, the parameters in the formula are helpful for positively understanding the relation between the signal intensity and background noise in the imaging process, the experimental conditions set by a sample preparation, dyeing process and a microscope can be adjusted according to actual needs, so that a higher-quality image is obtained, and an abnormal data processing module can automatically evaluate the image quality according to the image quality evaluation score Gl and identify abnormal data at the same time, so that manual intervention is reduced, and the data processing speed is improved.

The abnormal data processing module automatically identifies an initial concentration abnormal value or a background dyeing abnormal value of the dyeing liquid in the automatic dyeing process according to the automatic dyeing control index Ql, and automatically identifies a sharpening abnormal value or a contrast abnormal value of the imaging system according to the microscope imaging stability coefficient Sl.

The abnormal data processing module judges whether the quality of final imaging meets the standard according to the image quality evaluation score Gl, when the image quality evaluation score Gl is more than or equal to 95, the imaging quality is excellent, when the image quality evaluation score Gl is less than or equal to 90 and less than or equal to 95, the imaging quality is good, and when the image quality evaluation score Gl is less than or equal to 80, the imaging quality meets the standard.

The data management module carries out data correction and fault diagnosis on the dyeing abnormal data in the automatic dyeing process and the abnormal data in the microscopic imaging process according to the abnormal data automatically identified by the abnormal data processing module, and simultaneously sends corrected parameters and fault diagnosis information to the control module, the control module gives a repair instruction, and the control module is connected with the data acquisition module, the data analysis module, the abnormal data processing module and the data management module through a network.

Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.

Claims

1. A method for identifying slices using a hyperspectral microscope combined with HE staining, characterized in that it comprises the following steps:

Step 1: Establish a data acquisition module, a data analysis module, an abnormal data processing module, a data management module and a control module;

Step 2, the data acquisition module is divided into an automatic staining data capture unit, a microscope imaging data receiving unit and a staining slice data output unit;

Step 3: The data analysis module is divided into an HE staining slice recognition unit and a microscope imaging stabilization unit, and calculates the automatic staining control index Ql, the microscope imaging stability coefficient Sl and the image quality assessment score Gl;

Step 4: The abnormal data processing module identifies abnormal staining data during the automatic staining process and abnormal data during the microscope imaging process according to the automatic staining control index Ql and the microscope imaging stability coefficient Sl, and determines whether the quality of the final imaging meets the standard according to the image quality evaluation score Gl.

2. A method for identifying slices by combining a hyperspectral microscope with HE staining according to claim 1, characterized in that: the method comprises a data acquisition module, a data analysis module, an abnormal data processing module, a data management module and a control module;

The data acquisition module includes an automatic staining data capturing unit, a microscope imaging data receiving unit and a staining slice data output unit, wherein the automatic staining data unit collects automatic staining data through a hyperspectral imaging system, the microscope imaging data receiving unit collects microscope imaging data through a CCD camera, and the staining slice data output unit collects staining slice data through an imaging spectrometer, and the spectrum data capturing unit, the microscope imaging data receiving unit and the staining slice data output unit are connected to the data analysis module through a network;

The data analysis module includes a HE-stained slice identification unit and a microscope imaging stabilization unit, wherein the HE-stained slice identification unit calculates an automatic staining control index Ql according to the automatic staining data, the microscope imaging stabilization unit calculates a microscope imaging stability coefficient Sl according to the microscope imaging data, and the microscope imaging stabilization unit calculates an image quality assessment score Gl according to the stained slice data, and the HE-stained slice identification unit and the microscope imaging stabilization unit are connected to the abnormal data processing module through a network;

The abnormal data processing module identifies abnormal staining data in the automatic staining process and abnormal data in the microscope imaging process according to the automatic staining control index Ql and the microscope imaging stability coefficient Sl, and determines whether the quality of the final imaging meets the standard according to the image quality evaluation score Gl. The abnormal data processing module is connected to the data management module through a network.

3. A method for identifying a hyperspectral microscope combined with HE-stained slices according to claim 2, characterized in that: the automatic staining data unit performs data numbering on the initially set staining time, the initial concentration of the staining solution, the background staining value, the staining environment temperature influence value and the staining instrument correction factor according to the automatic staining data characteristics, and the initially set staining time, the initial concentration of the staining solution, the background staining value, the staining environment temperature influence value and the staining instrument correction factor are numbered as T, C, B, μ, ε, and the HE-stained slice identification unit calculates the automatic staining control index Ql according to the automatic staining data, and the calculation formula is:

In the formula, Ql represents the automatic dyeing control index, T, C, B, μ, and ε represent the initial set dyeing time, the initial concentration of the dyeing solution, the background dyeing value, the dyeing environment temperature influence value, and the dyeing instrument correction factor, respectively, and (1-μ) represents the constant value that excludes the temperature influence value.

4. A method for identifying slices with hyperspectral microscopy combined with HE staining according to claim 1, characterized in that: the microscope imaging data receiving unit performs data numbering on the grayscale values of all images according to the characteristics of the microscope imaging data, and the grayscale values of the images are numbered as _f1 , _f2 , _f3 , ... _fn .

5. A method for identifying a hyperspectral microscope combined with HE-stained sections according to claim 4, characterized in that: the microscope imaging stabilization unit calculates the microscope imaging stability coefficient S1 according to the microscope imaging data, and the calculation formula is:

In the formula, Sl represents the microscope imaging stability coefficient, _f1 , _f2 , _f3 , ... _fn represent the grayscale value of the image, n represents the number of images, _fi represents the grayscale value of the i-th image, Represents the average gray value of all images.

6. A method for identifying HE-stained sections using a hyperspectral microscope according to claim 1, characterized in that: the stained section data output unit performs data numbering on the signal intensity and background noise level of the stained area according to the stained section data characteristics, the signal intensity of the stained area is numbered as E ₁ , E ₂ , E ₃ , ...E _n , and the background noise level is numbered as R ₁ , R ₂ , R ₃ , ...R _m .

7. A method for identifying a hyperspectral microscope combined with HE-stained slices according to claim 6, characterized in that: the microscope imaging stabilization unit calculates the image quality evaluation score Gl according to the stained slice data, and the calculation formula is:

In the formula, Gl represents the image quality assessment score, _E1 , _E2 , _E3 , ... _En represents the signal intensity of the stained area, _R1 , _R2 , _R3 , ... _Rm represents the background noise level, represents the average signal intensity of the stained area, n represents the number of statistically stained areas, represents the average level of background noise, and m represents the number of statistical background noise levels.

8. A method for identifying HE-stained sections in combination with a hyperspectral microscope according to claim 7, characterized in that: the abnormal data processing module automatically identifies the initial concentration abnormal value or the background staining abnormal value of the staining solution during the automatic staining process according to the automatic staining control index Ql, and automatically identifies the imaging system sharpening abnormal value or the contrast abnormal value according to the microscope imaging stability coefficient Sl.

9. According to the method for identifying a hyperspectral microscope combined with HE-stained slices in claim 7, it is characterized in that: the abnormal data processing module determines whether the quality of the final imaging meets the standard according to the image quality evaluation score Gl. When the image quality evaluation score Gl ≥ 95 points, the imaging quality is excellent; when the image quality evaluation score Gl < 95 points, the imaging quality is good; when the image quality evaluation score Gl < 80 points, the imaging quality meets the standard.

10. A method for identifying slices using a hyperspectral microscope combined with HE staining according to claim 9, characterized in that: the data management module performs data correction and fault diagnosis on the abnormal data in the automatic staining process and the abnormal data in the microscope imaging process according to the abnormal data automatically identified by the abnormal data processing module, and sends the corrected parameters and fault diagnosis information to the control module, which issues a repair instruction, and the control module is connected to the data acquisition module, the data analysis module, the abnormal data processing module, and the data management module through a network.