[0037] See figure 1.
[0038] The hardware device of the present invention is mainly composed of a microscope, a camera, a desktop computer and a printer. The microscope uses a transmission biological microscope, the camera uses a high-definition digital camera, the computer uses a desktop computer, and the printer uses a color laser printer. The camera is installed on the upper part of the microscope and is connected to the microscope through the microscope camera interface to collect images of the field of view observed under the microscope; the camera is connected to the USB port of the computer through a data cable, and the pathological slice image observed by the camera is transmitted to the computer; The printer is connected to the computer through a USB port, and the DNA test report generated by the analysis software is printed out on the printer.
[0039] See attached Figure 2-4.
[0040] Tumor cell DNA content detection methods include five steps: length calibration, background calibration, normal diploid cell measurement, tumor cell measurement, and DNA report generation: First, place an approved light-transmitting glass slide scale on the microscope stage , Through the camera, take the transparent glass slide scale image to the computer, hold down the left mouse button and drag on the scale image, the analysis software calculates the current microscope based on the actual value of the drag scale and the number of pixels dragged The actual value of each pixel of the collected image under the magnification is completed, and the length calibration is completed; then the patient specimen section is placed on the microscope stage, and the current field of view is adjusted to the blank position without cells in the section by adjusting the microscope field of view. Adjust the brightness of the microscope so that the peak value of the gray histogram of the microscope field of view is between 220 and 230. Collect and save the current field of view image to complete the background calibration operation; then, select the field of view with typical normal diploid cells under the microscope, Collect the image of the patient’s normal diploid cells into the computer, divide the nucleus from the background through image segmentation, and calculate the average optical density value of the normal diploid cell nucleus; then move the X-axis and Y-axis of the microscope, Select multiple fields of view with diseased cells, collect and save the images of each field separately, perform image segmentation on the nucleus of each field, and segment the nucleus from the background according to the difference between the color gray value of the nucleus and the background, and calculate the segmentation The average optical density value and DNA index DI of each nucleus in the nucleus generate DNA histograms and scatter plots; finally, according to the test results, a DNA test report is generated to form diagnostic recommendations.
[0041] The specific process is: length calibration, background calibration, normal diploid cell measurement, tumor cell measurement, and DNA report generation in five steps: first collect the scale image and perform length calibration; collect the background image and measure the gray value of the background image pixels ;Acquire images of normal diploid cells, perform image segmentation on the images of diploid cells, measure the gray value of diploid cell nuclei, calculate the average optical density value of diploid cell nuclei; collect tumor cell images, and image tumor cells Segmentation, measuring the optical density value of each tumor cell nucleus pixel point, calculating the DI value and area of each tumor cell nucleus; generating a DNA histogram and generating a scatter plot, counting the number of various cells according to DI classification, and generating a DNA detection report.
[0042] Examples of using this method:
[0043] 1. Length calibration
[0044] A general scale usually has 100 graduations, and the length of each graduation is 10 microns, and the total graduation length is 1000 microns. When using the image analysis method to measure the length and area of the target to be measured, it is necessary to calibrate the length in advance. The length calibration is to place an approved transparent glass slide scale on the microscope stage and adjust the X-axis and Y-axis of the microscope. Axis, so that the scale is located in the center of the field of view, while rotating the camera to ensure that the scale is parallel to the upper edge of the image in the captured scale image, and the scale image is captured to the computer through the camera. Press and hold the left mouse button on the scale image Drag, enter the drag scale length, the analysis software will automatically record the actual value of each pixel of the image collected under the current microscope magnification, enter the scale name and save it. When performing length calibration, it is necessary to collect images for each magnification of the microscope and calibrate separately. Before length measurement, the required scale name must also be selected in advance.
[0045] 2. Background calibration
[0046] Background image calibration is to correct deviations caused by non-uniform illumination, (usually the center of the microscope's field of view is the brightest), changes in the camera's response to the fixed pattern, and dust in the camera or light path.
[0047] Before analyzing and testing each patient specimen slice, background calibration is required. The specific steps are: place the patient specimen on the microscope stage, first select the field of view where the cells can be observed, adjust the focus of the microscope to make the cells in the clearest position, and then move the X axis and Y axis of the microscope stage to The microscope field of view is adjusted to the blank position where there are no cells in the slice. By adjusting the microscope brightness, the analysis software can obtain the pixel gray value of the current field of view image, and generate the gray value of the image to the gray histogram, the peak value of the gray histogram Located between 220 and 230, then collect the current field of view image and save it as a background image. When the gray value histogram peak value is greater than 230, reduce the microscope brightness, when the gray value histogram peak value is less than 220, increase the microscope brightness.
[0048] 3. Measurement of normal diploid cells
[0049] Select the field of view with normal diploid cells under the microscope, then use the camera to collect images of the patient's normal diploid cells into the computer, select image segmentation, segment the nucleus from the background, and then use the mouse in the segmented image , Select the normal diploid cell, and calculate the optical density value od_2c(i,j) of each pixel of all normal diploid cell nuclei. The calculation formula is:
[0050]
[0051] Among them, grey_2c(i,j) is the gray value of the pixel, and grey_back(i,j) is the gray value of the same position of the background image.
[0052] After calculating the optical density value of each pixel of normal diploid cell nucleus, calculate its average optical density value mean_od_2c, the calculation formula is as follows:
[0053]
[0054] Where N is the total number of pixels of the normal diploid cell nucleus segmented, and R is the segmented cell nucleus area.
[0055] 4. Tumor cell measurement
[0056] Move the X-axis and Y-axis of the microscope, select at least five fields with diseased cells, collect and save the images of each field, and then analyze and measure the cells in each field. The specific steps are as follows: First, use image segmentation technology to separate the nucleus from the background according to the difference between the color gray value of the nucleus and the background, and then use the same method of calculating the pixel optical density as a normal diploid cell to calculate the segmented cell nucleus. The optical density value od_cell of each pixel point, and then the average optical density value mean_od_cell of each cell nucleus is calculated, and finally the DNA index (DI) of each cell is calculated. The calculation formula of the DI value is:
[0057]
[0058] After calculating the DI value of each cell in all collected images, a DNA histogram is generated. The abscissa of the DNA histogram is the DI value and the ordinate is the number of cells. The horizontal axis ranges from 0 to 5, and each interval is 0.1. Count the number of cells in each DI value interval and use it as the vertical axis value to generate a DNA histogram. The DNA histogram of this example is attached image 3 Shown.
[0059] The scatter chart is mainly used to show the distribution of cells with different DI values. The abscissa is the DI value, and the ordinate is the nucleus area value, and its unit is square micrometers. When calculating the DI value of each cell, the specific area of each cell nucleus is measured according to the length calibration. The value range of the horizontal axis of the scatter chart is 0-5, and each interval is 0.1. For each cell According to its DI value and area, a dot is calibrated in the scatter chart. When all cells are calibrated, the distribution of cell DI value and area can be clearly seen. The DNA scatter diagram of this example is attached Figure 4 Shown.
[0060] 5. DNA report generation
[0061] The DNA test report is an effective reference basis for doctors to make pathological diagnosis of patients. After the above tests are completed, the analysis system will automatically import the test results into the test report, including DNA histograms, scatter plots, and test results (including: total number of cells, normal two Ploidy cells, normal hyperplasia or suspected diseased cells, diseased cells, total number of cells, etc.), while importing basic patient information, and generating diagnostic recommendations based on the measurement results and the peak distribution of DNA histograms. The criteria for determining diagnostic recommendations are as follows:
[0062] 1) Normal: Normal diploid cells are predominantly (DI value is 1), and no aneuploid cells and peaks of aneuploid cells are seen.
[0063] 2) When DI is between 1 and 2, it is mostly HPV infected cells or inflammatory cells.
[0064] 3) Abnormal, biopsy is recommended: when DI value>2.5, or when the number of cells between diploid and tetraploid exceeds 10% of the total number of cells tested.
[0065] 4) Tumor cells: DI value ≥ 4.5.
