A method for white light-based quantitative analysis of cellular DNA

Abstract

The application provides a white light-based cell DNA quantitative analysis method, comprising the following steps: S1, acquiring an RGB color image of cells by using a color camera and a white light source; S2, performing gray scale processing on the color image by using a dark average elimination method; S3, generating a mask of a cell protoplast region by using a threshold segmentation method; S4, extracting all the cell nucleus candidate contours in the gray scale image by using a gray scale binary method and a gradient amplitude method respectively, and performing merging and de-duplication processing on the cell nucleus candidate contours extracted by the two methods to obtain a cell nucleus contour set; and S5, performing quantitative analysis on cell DNA in the cell nucleus contour region by using a light absorption entropy integral value algorithm. The application adopts a one-time imaging mode of a white light source combined with a color camera, replaces the complex process of multiple imaging by relying on multiple specific wavelength light sources in the traditional method, significantly reduces the equipment manufacturing cost and maintenance difficulty, and effectively improves the image acquisition efficiency and the stability of the system.

Description

Technical Field

[0001] This invention relates to the field of cell analysis technology, and in particular to a method for quantitative analysis of cellular DNA based on white light. Background Technology

[0002] Cervical cancer is a common malignant tumor that seriously threatens women's health. Early detection and diagnosis are crucial for improving cure and survival rates. Currently, the most widely used cervical cancer screening methods in clinical practice rely primarily on cell morphology observation, such as the TBS classification method. This method requires pathologists to manually examine slides under a microscope, identifying only a very small number of abnormal cells from tens of thousands of cells. This process is not only time-consuming and labor-intensive, but also highly dependent on the doctor's subjective experience and judgment, which can easily lead to visual fatigue and diagnostic discrepancies, resulting in missed or misdiagnosed cases.

[0003] Cellular DNA content is an objective biological indicator reflecting cell proliferation activity and malignancy. Quantitative cellular DNA analysis, by measuring abnormal changes in DNA within the cell nucleus, can detect precancerous cells earlier and more objectively than morphological observation, demonstrating a significant advantage in the early diagnosis of cervical cancer. This technology has become a routine auxiliary diagnostic tool in North America and Europe, and its application in cervical cancer screening is gradually being promoted in China. Numerous studies have shown that quantitative DNA analysis can effectively improve the sensitivity and specificity of screening, compensating for the shortcomings of traditional methods.

[0004] However, traditional methods for quantifying cellular DNA are mostly based on multispectral imaging technology. This method requires excitation with light of a specific wavelength (such as ultraviolet light), or relies on irradiating the sample with multiple spectra of light such as red, green, and blue light and performing multiple imaging operations before synthesizing the image for analysis. This technical approach has inherent drawbacks: First, multiple imaging operations result in slow detection speeds, making it difficult to meet the throughput requirements of large-scale screening; second, the system requires complex optical path switching devices and precise light source-camera synchronization control, resulting in complex equipment structures, high manufacturing costs, and difficult maintenance, limiting its widespread adoption in primary healthcare institutions. Summary of the Invention

[0005] This invention proposes a white light-based method for quantifying cell DNA, which solves the problems of low detection efficiency, complex equipment, and high cost in existing cell DNA quantification methods based on multispectral imaging technology.

[0006] The technical solution of this invention is implemented as follows:

[0007] This invention provides a method for quantitative analysis of cellular DNA based on white light, comprising the following steps:

[0008] S1, uses a color camera and a white light source to acquire RGB color images of cells;

[0009] S2, use the dark averaging method to convert the RGB color image to grayscale to obtain a grayscale image;

[0010] S3, a mask for the cell encapsulation region is generated using a threshold segmentation method;

[0011] S4. All candidate cell nuclei in the grayscale image are extracted using grayscale binarization and gradient amplitude methods, respectively. The candidate cell nuclei extracted by the two methods are then merged and deduplicated to obtain a set of cell nuclei contours.

[0012] S5, within the cell nucleus contour region, uses the light absorption entropy integral value algorithm to perform quantitative analysis of cell DNA.

[0013] Specifically, step S2 includes the following steps:

[0014] S201: For a specific pixel in an RGB color image, extract the values ​​of the corresponding R, G, and B channels respectively.

[0015] S202, remove the minimum value among the R, G, and B channel values, and take the average value of the remaining two channel values ​​as the gray value of the pixel.

[0016] S203, using the methods of steps S201 and S202, perform grayscale processing on each pixel in the RGB color image to obtain a grayscale image.

[0017] Specifically, in step S3, a set threshold is used to perform binary segmentation on the grayscale image. The grayscale values ​​of pixels with grayscale values ​​greater than the set threshold are set to 255, and the grayscale values ​​of pixels with grayscale values ​​less than or equal to the set threshold are set to 0. The grayscale values ​​of the binary segmented image are then inverted. The grayscale values ​​of the cell capsule region with grayscale values ​​less than or equal to the set threshold are inverted to 255, and the grayscale values ​​of the background region with grayscale values ​​greater than the set threshold are inverted to 0, thus obtaining a mask for the cell capsule region.

[0018] Specifically, in step S4, the method for extracting all candidate cell nucleus contours in the grayscale image using grayscale binarization includes the following steps:

[0019] The grayscale image of S2 is processed using the mask of S3. The grayscale values ​​outside the mask in step S2 are set to 0, while the grayscale values ​​inside the mask remain unchanged.

[0020] The processed grayscale image is then subjected to Gaussian smoothing.

[0021] The grayscale image after Gaussian smoothing was binarized by using multiple different grayscale thresholds in sequence, and candidate contours of cell nuclei were extracted from the binarized image using OpenCV.

[0022] Each candidate contour is subjected to feature filtering to remove abnormal or duplicate contours, resulting in a set of candidate contours for cell nuclei.

[0023] Specifically, in step S4, the method for extracting all candidate cell nucleus contours in the grayscale image using the gradient magnitude method includes the following steps:

[0024] The three channels of the RGB color image of S1 are preprocessed using the mask of S3. The gray values ​​outside the mask of the three channel images are set to 0, while the gray values ​​inside the mask remain unchanged.

[0025] The gradient magnitude of each pixel in the three channels is calculated using the Sobel operator;

[0026] Take the maximum gradient magnitude of the three channels as the comprehensive gradient magnitude of the pixel, and obtain a gradient magnitude map containing the comprehensive gradient magnitude of each pixel.

[0027] Multiple different gradient magnitude thresholds were used to binarize the gradient magnitude map, and OpenCV was used to extract candidate contours of cell nuclei from the binarized image.

[0028] Each candidate contour is subjected to feature filtering to remove abnormal or duplicate contours, resulting in a set of candidate contours for cell nuclei.

[0029] Specifically, the method for merging and deduplicating candidate cell nucleus contours extracted by the two methods includes the following steps:

[0030] Traverse the candidate cell nuclei contours extracted by the two methods, calculate the center point coordinates and area of ​​each candidate contour in the image, and save the center point coordinates and area of ​​the first candidate contour.

[0031] Starting from the second candidate contour, calculate the distance between the center point of the new contour and the center point of the saved contour, and determine whether the distance is less than the set threshold.

[0032] If the distance is less than the set threshold, the new contour is considered to be a duplicate contour of the saved contour, and only the contour with the larger area is retained.

[0033] If the distance is greater than the set threshold, the new contour is considered not to be a repeating contour, and the center point coordinates and area of ​​the new contour are directly saved.

[0034] Repeat the above steps until all candidate contours have been traversed, resulting in a merged and deduplicated set of cell nucleus contours.

[0035] Specifically, step S5 includes the following steps:

[0036] S51 merges and fills all the cell nucleus outlines to generate a mask for the cell nucleus region;

[0037] S52, calculate the integral value of the light absorption entropy in the cell nucleus region;

[0038] S53, calculate the difference between the integral value of the light absorption entropy in the cell nucleus region and the standard reference value to obtain the cell DNA content.

[0039] Further, in step S51, the method for generating the mask of the cell nucleus region is as follows: merge all intersecting cell nucleus contours into a new contour, set the gray value of the pixels within the contour to 255, and set the gray value of the pixels outside the contour to 0, thereby obtaining the mask of the cell nucleus region.

[0040] Furthermore, in step S52, the formula for calculating the integral value of the light absorption entropy of the cell nucleus region is as follows:

[0041] ;

[0042] Wherein, OAEI is the integral value of the light absorption entropy of a certain nuclear region. for The light absorption entropy at the location, For the mask at position The value of , within the cell nucleus region, Outside the cell nucleus region, ; For position pixel grayscale values, This represents the background grayscale value outside the cell nucleus region.

[0043] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0044] (1) The present invention uses a white light source combined with a color camera to perform a single imaging method, which replaces the complex process of multiple imaging based on multiple specific wavelength light sources in the traditional method. This greatly simplifies the system structure and reduces the need for precision optical components and complex synchronous control systems. It not only significantly reduces the equipment manufacturing cost and maintenance difficulty, but also effectively improves the image acquisition efficiency and system stability, laying the foundation for lowering the threshold of clinical testing and realizing the popularization of equipment.

[0045] (2) The grayscale processing method of "de-darkening averaging" proposed in this invention can adaptively enhance the characteristics of cell images. This method effectively suppresses the influence of overexposed areas and effectively improves the contrast of cell boundaries by removing the minimum value in the three channels of each pixel and calculating the average value of the remaining two channels, thus providing a high-quality image basis for subsequent accurate segmentation of cell nuclei and ensuring the accuracy of subsequent quantitative analysis;

[0046] (3) The present invention adopts a strategy combining grayscale binarization and gradient amplitude method for cell nucleus segmentation, and significantly improves the comprehensiveness and accuracy of cell nucleus recognition through merging and deduplication processing mechanism; the two methods complement each other, can effectively deal with cell nuclei with different contrast and boundary clarity, overcome the limitations of single method, reduce missed detection and false detection, and provide a reliable guarantee for realizing high-throughput automated analysis.

[0047] (4) This invention uses the integral value of light absorption entropy for DNA quantitative analysis. This method converts gray information into DNA content based on the principle of light absorption, which has clear physical meaning and good linear relationship. The algorithm is scientific and reliable, and can output stable and objective quantitative results, significantly reducing the error caused by human subjective judgment, and providing repeatable quantitative basis for early cancer diagnosis and efficacy evaluation. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This is a schematic flowchart of a white light-based method for quantitative analysis of cellular DNA according to the present invention.

[0050] Figure 2 The image shows an RGB color image of a cell acquired in an embodiment of the present invention.

[0051] Figure 3 This is a grayscale image obtained by processing an RGB color image using the darkening averaging method in an embodiment of the present invention.

[0052] Figure 4 This is a mask image of the cell capsule region generated by the threshold segmentation method in an embodiment of the present invention;

[0053] Figure 5 This is a grayscale image of the area outside the mask after setting the grayscale value to 0 in an embodiment of the present invention;

[0054] Figure 6In this embodiment of the invention, the gradient magnitude method is used to extract all candidate cell nucleus contours from the grayscale image;

[0055] Figure 7 This is a mask image of the cell nucleus region generated after merging and filling all cell nucleus outlines in an embodiment of the present invention;

[0056] Figure 8 This is a statistical chart of cellular DNA content on the entire glass slide in an embodiment of the present invention. Detailed Implementation

[0057] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0058] Reference Figure 1 This invention provides a method for quantitative analysis of cellular DNA based on white light, comprising the following steps:

[0059] S1, using a color camera and a white light source, acquires RGB color images of cells on a glass slide, such as... Figure 2 As shown;

[0060] S2, use the dark averaging method to convert the RGB color image to grayscale to obtain a grayscale image;

[0061] S3, a mask for the cell encapsulation region is generated using a threshold segmentation method;

[0062] S4. All candidate cell nuclei in the grayscale image are extracted using grayscale binarization and gradient amplitude methods, respectively. The candidate cell nuclei extracted by the two methods are then merged and deduplicated to obtain a set of cell nuclei contours.

[0063] S5, within the cell nucleus contour region, uses the light absorption entropy integral value algorithm to perform quantitative analysis of cell DNA.

[0064] This invention uses a white light source combined with a color camera for single-image imaging, replacing the complex process of multiple imaging based on multiple specific wavelength light sources in traditional methods. This greatly simplifies the system structure, reduces the need for precision optical components and complex synchronous control systems, significantly reduces equipment manufacturing costs and maintenance difficulty, and effectively improves image acquisition efficiency and system stability, laying the foundation for lowering the threshold for clinical testing and achieving widespread equipment adoption.

[0065] Specifically, step S2 includes the following steps:

[0066] S201: For a specific pixel in an RGB color image, extract the values ​​of the corresponding R, G, and B channels respectively.

[0067] S202, remove the minimum value among the R, G, and B channel values, and take the average value of the remaining two channel values ​​as the gray value of the pixel.

[0068] S203, using the methods of steps S201 and S202, performs grayscale processing on each pixel in the RGB color image to obtain a grayscale image, such as... Figure 3 As shown.

[0069] The proposed "de-darkening averaging method" grayscale processing method can adaptively enhance cell images based on their characteristics. This method effectively suppresses the influence of overexposed areas and significantly improves the contrast of cell boundaries by removing the minimum value from each pixel's three channels and calculating the average of the remaining two channels. This provides a high-quality image foundation for subsequent precise cell nucleus segmentation, thus ensuring the accuracy of subsequent quantitative analysis.

[0070] Specifically, in step S3, a set threshold (205 in this embodiment, but the specific value can be adjusted flexibly according to the actual situation) is used to perform binary segmentation on the grayscale image. Pixels with grayscale values ​​greater than the set threshold of 205 are set to 255 (white), and pixels with grayscale values ​​less than or equal to the set threshold of 205 are set to 0 (black). The grayscale values ​​of the binary segmented image are then inverted. The grayscale values ​​of the cell capsule region with grayscale values ​​less than or equal to the set threshold of 205 are inverted to 255 (white), and the grayscale values ​​of the background region with grayscale values ​​greater than the set threshold of 205 are inverted to 0 (black), thus obtaining a mask for the cell capsule region, as shown below. Figure 4 As shown, Figure 4 The white area in the middle is the cell capsule area.

[0071] Specifically, in step S4, the method for extracting all candidate cell nucleus contours in the grayscale image using grayscale binarization includes the following steps:

[0072] The grayscale image from step S2 is processed using the mask from step S3. The grayscale values ​​outside the mask corresponding to the grayscale image in step S2 are set to 0 (black), while the grayscale values ​​inside the mask remain unchanged. The processed grayscale image is shown below. Figure 5 As shown, this greatly enhances the contrast between the patina area and the background, providing a clearer boundary for subsequent cell nucleus segmentation.

[0073] The processed grayscale image is then subjected to Gaussian smoothing.

[0074] The Gaussian-smoothed grayscale image was binarized by using multiple different grayscale thresholds (such as 100, 125, and 150), and candidate contours of cell nuclei were extracted from the binarized image using OpenCV.

[0075] For each candidate contour, feature filtering is performed (based on features such as contour area and shape) to remove abnormal or duplicate contours, resulting in a set of candidate cell nucleus contours based on grayscale binarization segmentation.

[0076] Specifically, in step S4, the method for extracting all candidate cell nucleus contours in the grayscale image using the gradient magnitude method includes the following steps:

[0077] The three channels of the RGB color image of S1 are preprocessed using the mask of S3. The gray values ​​outside the mask of the three channel images are set to 0, while the gray values ​​inside the mask remain unchanged.

[0078] The gradient magnitude of each pixel in the three channels is calculated using the Sobel operator and normalized to 0~255.

[0079] Take the maximum gradient magnitude of the three channels as the comprehensive gradient magnitude of the pixel (it can be appropriately enhanced, such as by multiplying by 4), and obtain a gradient magnitude map containing the comprehensive gradient magnitude of each pixel;

[0080] Multiple gradient magnitude thresholds (e.g., 70~90) were used to binarize the gradient magnitude map to accommodate cell nuclei with different contrasts. OpenCV was used to extract candidate contours of cell nuclei from the binarized image and extract their convex hulls (opencv ConvexHull).

[0081] For each candidate contour, feature filtering is performed to remove abnormal or duplicate contours, resulting in a set of candidate contours for cell nuclei, such as... Figure 6 As shown.

[0082] The method for calculating the gradient magnitude of each pixel in the three channels using the Sobel operator is as follows:

[0083] The color image is split into three independent single-channel grayscale images: R, G, and B.

[0084] For the R-channel image, the Sobel operator is used to calculate its derivatives in the X direction (horizontal) and Y direction (vertical), respectively, resulting in two matrices: Gx_r and Gy_r;

[0085] Similarly, for the G channel image, we obtain Gx_g and Gy_g;

[0086] For the B-channel image, we obtain Gx_b and Gy_b;

[0087] For Gx_r and Gy_r of the R channel, calculate the position of each pixel. gradient magnitude The formula is:

[0088] ;

[0089] in, and These represent the pixel positions on the R channel image. The gradient values ​​in the horizontal (x-direction) and vertical (y-direction) directions;

[0090] Iterate through each pixel in the R channel to generate a gradient magnitude map of the R channel.

[0091] Similarly, the gradient magnitude maps for the G and B channels are calculated.

[0092] Specifically, the method for merging and deduplicating candidate cell nucleus contours extracted by the two methods includes the following steps:

[0093] Traverse the candidate cell nuclei contours extracted by the two methods, calculate the center point coordinates and area of ​​each candidate contour in the image, and save the center point coordinates and area of ​​the first candidate contour.

[0094] Starting from the second candidate contour, calculate the distance between the center point of the new contour and the center point of the saved contour, and determine whether the distance is less than the set threshold.

[0095] If the distance is less than the set threshold, the new contour is considered to be a duplicate contour of the saved contour, and only the contour with the larger area is retained.

[0096] If the distance is greater than the set threshold, the new contour is considered not to be a repeating contour, and the center point coordinates and area of ​​the new contour are directly saved.

[0097] Repeat the above steps until all candidate contours have been traversed, resulting in a merged and deduplicated set of cell nucleus contours.

[0098] Due to the high complexity of cell images, different cell nuclei may exhibit different morphological features and contrasts during imaging. Some cell nuclei show significant grayscale differences from the pericarp region, which can be effectively extracted using grayscale thresholding. Other cell nuclei have blurred boundaries or insufficient grayscale, but exhibit prominent changes in gradient (i.e., edges). In these cases, gradient segmentation can better capture the edge information of the cell nuclei.

[0099] This invention employs a strategy combining grayscale binarization and gradient amplitude methods for cell nucleus segmentation. Through merging and deduplication mechanisms, it significantly improves the comprehensiveness and accuracy of cell nucleus identification. The two methods complement each other, effectively handling cell nuclei with different contrasts and boundary clarity, overcoming the limitations of single methods, reducing missed detections and false detections, and providing a reliable guarantee for achieving high-throughput automated analysis.

[0100] Specifically, step S5 includes the following steps:

[0101] S51 merges and fills all the cell nucleus outlines to generate a mask for the cell nucleus region;

[0102] S52, calculate the integral value of the light absorption entropy in the cell nucleus region;

[0103] S53, calculate the difference between the integral value of the light absorption entropy in the cell nucleus region and the standard reference value to obtain the cell DNA content.

[0104] Further, in step S51, the method for generating the mask of the cell nucleus region is as follows: merging all intersecting cell nucleus contours into a new contour, setting the grayscale value of pixels within the contour to 255, and setting the grayscale value of pixels outside the contour to 0, thus obtaining the mask of the cell nucleus region, as shown below. Figure 7 As shown, Figure 7 The white area in the middle is the cell nucleus.

[0105] Furthermore, in step S52, the formula for calculating the integral value of the light absorption entropy of the cell nucleus region is as follows:

[0106] ;

[0107] Wherein, OAEI is the integral value of the light absorption entropy of a certain nuclear region. for The light absorption entropy at the location, For the mask at position The value of , within the cell nucleus region, Outside the cell nucleus region, ; For position pixel grayscale values, This represents the background grayscale value outside the cell nucleus region.

[0108] The standard reference value is the average OAEI of a certain number (usually 1000 or 2000) of cells whose nuclear area is less than a certain threshold and whose nuclear roundness is greater than a certain threshold.

[0109] The final quantitative analysis results of cellular DNA content are as follows: Figure 8 As shown, Figure 8In the histogram, the horizontal axis represents the DNA index (DI), where DI = 1 indicates normal diploid cells, DI = 2 indicates tetraploid cells, and DI > 2.5 indicates aneuploid cells, which is a strong indicator of malignancy. The vertical axis of the histogram represents the number of cell nuclei corresponding to each DNA index, and the vertical axis of the scatter plot represents the number of cells corresponding to each DNA index. As can be seen from the figure, the sample is mainly composed of normal diploid cells, but a certain number of tetraploid cells and a small number of clearly aneuploid cells are present. This distribution pattern strongly suggests the presence of DNA ploidy abnormal cells in the sample, requiring further clinical examination.

[0110] This invention utilizes the integral value of light absorption entropy for DNA quantitative analysis. This method converts grayscale information into DNA content based on the principle of light absorption, which has clear physical meaning and a good linear relationship. The algorithm is scientific and reliable, and can output stable and objective quantitative results, significantly reducing the error caused by subjective human judgment, and providing repeatable quantitative evidence for early cancer diagnosis and efficacy evaluation.

[0111] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for quantitative analysis of cellular DNA based on white light, characterized in that, Includes the following steps: S1, uses a color camera and a white light source to acquire RGB color images of cells; S2, use the dark averaging method to convert the RGB color image to grayscale to obtain a grayscale image; S3, a mask for the cell encapsulation region is generated using a threshold segmentation method; S4. All candidate cell nuclei in the grayscale image are extracted using grayscale binarization and gradient amplitude methods, respectively. The candidate cell nuclei extracted by the two methods are then merged and deduplicated to obtain a set of cell nuclei contours. The method for extracting all candidate cell nucleus contours in a grayscale image using grayscale binarization includes the following steps: The grayscale image of S2 is processed using the mask of S3. The grayscale values ​​outside the mask in step S2 are set to 0, while the grayscale values ​​inside the mask remain unchanged. The processed grayscale image is then subjected to Gaussian smoothing. The grayscale image after Gaussian smoothing was binarized by using multiple different grayscale thresholds in sequence, and candidate contours of cell nuclei were extracted from the binarized image using OpenCV. Each candidate contour is subjected to feature filtering to remove abnormal or duplicate contours, resulting in a set of candidate cell nuclei contours. The method for extracting all candidate cell nucleus contours in a grayscale image using the gradient magnitude method includes the following steps: The three channels of the RGB color image of S1 are preprocessed using the mask of S3. The gray values ​​outside the mask of the three channel images are set to 0, while the gray values ​​inside the mask remain unchanged. The gradient magnitude of each pixel in the three channels is calculated using the Sobel operator; Take the maximum gradient magnitude of the three channels as the comprehensive gradient magnitude of the pixel, and obtain a gradient magnitude map containing the comprehensive gradient magnitude of each pixel. Multiple different gradient magnitude thresholds were used to binarize the gradient magnitude map, and OpenCV was used to extract candidate contours of cell nuclei from the binarized image. Each candidate contour is subjected to feature filtering to remove abnormal or duplicate contours, resulting in a set of candidate cell nuclei contours. The method for merging and deduplicating candidate cell nucleus contours extracted by the two methods includes the following steps: Traverse the candidate cell nuclei contours extracted by the two methods, calculate the center point coordinates and area of ​​each candidate contour in the image, and save the center point coordinates and area of ​​the first candidate contour. Starting from the second candidate contour, calculate the distance between the center point of the new contour and the center point of the saved contour, and determine whether the distance is less than the set threshold. If the distance is less than the set threshold, the new contour is considered to be a duplicate contour of the saved contour, and only the contour with the larger area is retained. If the distance is greater than the set threshold, the new contour is considered not to be a repeating contour, and the center point coordinates and area of ​​the new contour are directly saved. Repeat the above steps until all candidate contours have been traversed, and obtain the merged and deduplicated set of cell nucleus contours. S5, within the cell nucleus contour region, uses the optical absorption entropy integral value algorithm to perform quantitative analysis of cellular DNA.

2. The method for quantitative analysis of cellular DNA based on white light as described in claim 1, characterized in that, Step S2 includes the following steps: S201: For a specific pixel in an RGB color image, extract the values ​​of the corresponding R, G, and B channels respectively. S202, remove the minimum value among the R, G, and B channel values, and take the average value of the remaining two channel values ​​as the gray value of the pixel. S203, using the methods of steps S201 and S202, perform grayscale processing on each pixel in the RGB color image to obtain a grayscale image.

3. The method for quantitative analysis of cellular DNA based on white light as described in claim 1, characterized in that, In step S3, a set threshold is used to perform binary segmentation on the grayscale image. The grayscale values ​​of pixels with grayscale values ​​greater than the set threshold are set to 255, and the grayscale values ​​of pixels with grayscale values ​​less than or equal to the set threshold are set to 0. The grayscale values ​​of the binary segmented image are then inverted. The grayscale values ​​of the cell capsule region with grayscale values ​​less than or equal to the set threshold are inverted to 255, and the grayscale values ​​of the background region with grayscale values ​​greater than the set threshold are inverted to 0, thus obtaining a mask for the cell capsule region.

4. The method for quantitative analysis of cellular DNA based on white light as described in claim 1, characterized in that, Step S5 includes the following steps: S51 merges and fills all the cell nucleus outlines to generate a mask for the cell nucleus region; S52, calculate the integral value of the light absorption entropy in the cell nucleus region; S53, calculate the difference between the integral value of the light absorption entropy in the cell nucleus region and the standard reference value to obtain the cell DNA content.

5. The method for quantitative analysis of cellular DNA based on white light as described in claim 4, characterized in that, In step S51, the method for generating the mask of the cell nucleus region is as follows: merge all intersecting cell nucleus contours into a new contour, set the gray value of the pixels inside the contour to 255, and set the gray value of the pixels outside the contour to 0, thereby obtaining the mask of the cell nucleus region.

6. The method for quantitative analysis of cellular DNA based on white light as described in claim 4, characterized in that, In step S52, the formula for calculating the integral value of the light absorption entropy in the cell nucleus region is as follows: ； Wherein, OAEI is the integral value of the light absorption entropy of a certain nuclear region. for The light absorption entropy at the location, For the mask at position The value of , within the cell nucleus region, Outside the cell nucleus region, ; For position pixel grayscale values, This represents the background grayscale value outside the cell nucleus region.

Images