Serous cavity effusion sample property identification method and system based on data statistics

Abstract

The invention discloses a serous cavity effusion sample property identification method and system based on data statistics, and the method comprises the following steps: S1, preparing a serous cavity effusion sample slide, the operation comprising manual slide pushing and dyeing; s2, checking the prepared slide, wherein the checking content comprises whether cell distribution is uniform or not and whether dyeing is clear or not, and re-flaking the sample of which the flaking quality does not meet the item checking requirement; s3, observing the image of the slide under the microscope, judging the type and number of cells according to the serous cavity effusion cell morphological map, and recording the result; s4, carrying out statistical analysis on a recording result obtained in the step S3, and calculating the proportion of the total number of cancer cells in all the visual fields to the total number of nucleated cells in all the visual fields; and S5, judging the sample property according to the ratio and the number of the total number of cancer cells in the visual field. The invention provides a clear judgment rule and an implementation method of the sample property, and provides a clear judgment rule for the result in the gray region, so that the accuracy of the final result is ensured.

Description

Technical Field

[0001] This invention relates to the field of medical microscopy, and more particularly to a method and system for identifying the properties of serous cavity effusion samples based on statistical data. Background Technology

[0002] Cytological examination of serous cavity effusions plays an important role in the early diagnosis of tumors and the assessment of clinical efficacy. The vast majority of malignant serous cavity effusions (mainly referring to pleural effusion, ascites, and pericardial effusion) are caused by serous metastasis of malignant tumors, with a minority caused by pleural and peritoneal mesothelioma. Malignant serous cavity effusion is a prominent clinical manifestation of malignant tumor invasion and metastasis; almost all malignant tumors can present with malignant serous cavity effusion. Lung cancer, breast cancer, liver cancer, ovarian cancer, gastrointestinal cancer, and lymphoma are common causes of malignant serous cavity effusion.

[0003] Currently, serous cavity effusion samples are prepared through artificial staining and then subjected to microscopic cellular morphology observation by professional laboratory physicians to provide a qualitative diagnosis. However, due to limitations in the experience and skill of the operators, the diagnostic results are somewhat subjective and may introduce some bias. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for identifying the properties of serous cavity effusion samples based on statistical data, so as to solve the problems mentioned in the background art.

[0005] To achieve the above-mentioned objective, one aspect of the present invention provides a method for identifying the properties of serous cavity effusion samples based on statistical data, comprising the following steps:

[0006] Step S1: Prepare a slide for a sample of serous cavity effusion. The operation includes manual slide preparation and staining.

[0007] Step S2: Inspect the prepared slides. The inspection includes checking whether the cell distribution is uniform and whether the staining is clear. For samples whose slide quality does not meet the inspection requirements, re-prepare the slides.

[0008] Step S3: Observe the slide image under the microscope, determine the cell type and number based on the morphological atlas of cells in the serous cavity effusion, and record the results;

[0009] Step S4: Statistically analyze the recorded results obtained in step S3 and calculate the proportion of the total number of cancer cells in all fields of view to the total number of nucleated cells in all fields of view.

[0010] Step S5: Determine the nature of the sample based on the proportion and number of cancer cells in the field of view.

[0011] Furthermore, in step S3, the observation field of view shall be no less than 100 fields of view.

[0012] Furthermore, in step S3, the number of nucleated cells included in the observation field of view for statistical analysis is no less than 1000.

[0013] Furthermore, in step S3, the types and numbers of nucleated cells in the microscopic field of view are accurately identified and counted. When cancer cells have identifiable nucleolar features, the number of nucleoli is counted.

[0014] Furthermore, in step S5, the rules for determining the nature of the sample are as follows:

[0015] When the percentage of cancer cells is greater than or equal to 3.1%, the sample is considered a cancerous sample.

[0016] When the percentage of cancer cells is less than 2.9%, the sample is considered an inflammatory sample.

[0017] When the percentage of cancer cells is within the range of 3 ± 0.1%, the number of cancer cell nucleoli is further counted. If the number of cancer cell nucleoli is greater than 3, the sample is considered cancerous; otherwise, it is considered inflammatory.

[0018] Furthermore, in step S1, an automatic dyeing machine is used for automatic dyeing.

[0019] Furthermore, in step S2, an automatic microscopic image acquisition device is used to inspect the prepared glass slide.

[0020] Furthermore, in step S3, a proven and effective computer image algorithm or artificial intelligence network model is used to assist in the interpretation of cell types and quantities under the microscope and obtain the result data through computer software.

[0021] Another aspect of the present invention provides a system for identifying the properties of serous cavity effusion samples based on statistical data, comprising a preparation module, an inspection module, an observation module, a statistical module, and a judgment module, wherein:

[0022] The preparation module is used to prepare slides for serous cavity effusion samples, and the operations include manual slide preparation and staining.

[0023] The inspection module is used to inspect the prepared slides. The inspection includes checking whether the cell distribution is uniform and whether the staining is clear. Samples whose slide quality does not meet the inspection requirements of the project will be re-prepared.

[0024] The observation module is used to observe slide images under a microscope, determine cell type and number based on the morphological atlas of cells in the serous cavity effusion, and record the results;

[0025] The statistics module is used to statistically analyze the recorded results obtained in step S3 and calculate the proportion of the total number of cancer cells in all fields of view to the total number of nucleated cells in all fields of view.

[0026] The judgment module is used to determine the nature of a sample based on the proportion and number of cancer cells in the field of view.

[0027] Compared with existing technologies, this system and method have the following advantages:

[0028] Compared to the more subjective implementation of traditional microscopic examination, this technical solution provides clear rules and methods for determining the nature of samples, and proposes clear judgment rules for results in the gray area to ensure the accuracy of the final results. Attached Figure Description

[0029] Figure 1 This is a flowchart of a method for identifying the properties of serous cavity effusion samples based on statistical data.

[0030] Figure 2 This is a schematic diagram illustrating the working principle of a method for identifying the properties of serous cavity effusion samples based on statistical data.

[0031] Figure 3 This is a glass slide image of a sample of fluid accumulation in a serous cavity.

[0032] Figure 4 Example image of cancer cells and nucleoli. Detailed Implementation

[0033] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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 skilled in the art without creative effort are within the scope of protection of the present invention.

[0034] In this invention, "observation field of view" refers to the image content of the effusion sample in the serous cavity observed through the eyepiece under the objective lens of a 100X oil immersion optical microscope.

[0035] In this invention, "cancer cells" refers to cells and their aggregated structures that exhibit carcinogenic morphological characteristics based on cell morphology analysis.

[0036] In this invention, "cancer cell nucleolus" refers to the morphological features of the cell nucleolus that exist inside cancer cells. Based on existing knowledge of cell morphology, the clear visibility of cancer cell nucleoli is an important basis for determining the nature of cancer cells.

[0037] In this invention, "cancer sample" refers to a sample in which the test results report indicates the presence of cancer cells, and "inflammatory sample" refers to a sample in which the test results report indicates the absence of cancer cells.

[0038] like Figure 1 and Figure 2The diagram shown illustrates the method flowchart and working principle of this invention. This embodiment provides a method for identifying the properties of serous cavity effusion samples based on statistical data. The specific steps are as follows:

[0039] Step S1 involves preparing a glass slide from the serous cavity effusion sample, which includes manual slide preparation and staining.

[0040] Step S2 involves inspecting the prepared slides, specifically checking the uniformity of cell distribution and the clarity of staining. Samples whose slide quality does not meet the inspection requirements are re-prepared. Figure 3 The image shown is a glass slide containing a sample of fluid accumulation in the serous cavity.

[0041] Step S3: Microscopic image observation. Determine the cell type and number based on the morphological atlas of the serous cavity effusion and record the results. After the current field of view is interpreted, switch to different fields of view to continue interpretation until the conditions are met: a total of no less than 100 fields of view and a total number of nucleated cells of no less than 1000 are counted.

[0042] Step S4: Statistically analyze the recorded results obtained in step S3 and calculate the proportion of the total number of cancer cells in all fields of view to the total number of nucleated cells in all fields of view.

[0043] Step S5: Determine the nature of the sample based on the proportion and number of cancer cells in the field of view.

[0044] The judgment criteria are as follows:

[0045] When the percentage of cancer cells is greater than or equal to 3.1%, the sample is considered a cancerous sample.

[0046] When the percentage of cancer cells is less than 2.9%, the sample is considered an inflammatory sample.

[0047] When the percentage of cancer cells is within the range of 3 ± 0.1%, the number of cancer cell nucleoli is further counted. If the number of cancer cell nucleoli is greater than 3, the sample is considered cancerous; otherwise, it is considered inflammatory.

[0048] The nucleoli of cancer cells are defined and described by the morphological atlas of cells in the serous cavity effusion, such as... Figure 4 The image shown is a reference example of a cell nucleolus. Nucleolus morphology is crucial for determining cell characteristics. Only cells with clearly visible nucleolus features that conform to the definition of a morphological atlas of cells in serous cavity effusion can be included in the cell nucleolus count.

[0049] To further improve the accuracy and efficiency of this technology application, proven and effective computer graphics algorithms or artificial intelligence network models are employed to complete the process. Figure 1Step S3 in the workflow involves using computer software to interpret the types and numbers of cells under a microscope and obtain the results data.

[0050] Using computer algorithm software, implement the attached Figure 1 Step S5 in the workflow enables automatic data aggregation and analysis, and outputs the final judgment results of the samples.

[0051] By employing an automatic staining machine and an automatic microscopic image acquisition device, the attached... Figure 1 Steps S1 and S2 in the workflow reduce the deviation of manual operation, improve the consistency of sample preparation quality, and at the same time increase the degree of automation of project implementation and improve the work efficiency of inspection projects.

[0052] Experimental verification:

[0053] Two hundred samples of serous cavity effusion (pleural effusion / ascites) were collected, including: 1) 100 pathologically confirmed cancerous samples (lung cancer, ovarian cancer, gastric cancer, etc.), all of which were confirmed by histopathology as the gold standard; 2) 100 inflammatory samples (bacterial pleurisy, cirrhotic ascites, etc.), all of which were confirmed by review through the electronic medical record system. Each sample was independently examined by two different cytology examiners using a blind microscopy method. Each slide was examined in ≥100 fields of view according to the requirements of this invention, totaling ≥1000 nucleated cells. The following were recorded: total number of nucleated cells; number of cancer cells (cancer cells with nucleoli were separately marked); number of nucleoli in cancer cells (only clearly visible nucleoli were counted). The percentage of cancer cells in each sample was calculated (number of cancer cells / total number of nucleated cells × 100%). According to the rules of this invention, ≥3.1% was cancerous; ≤2.9% was inflammatory; 3±0.1% was based on the number of nucleoli (>3 for cancer, otherwise inflammatory). The identification results obtained according to this method are shown in the table below:

[0054] Sample type Result of this invention Compliance rate Gold Standard Cancerous samples (100 cases) 94 cases were cancerous, and 6 cases were in the gray zone (5 of which were ultimately diagnosed as cancerous). 99% （94+5）% Inflammatory samples (100 cases) 97 cases were inflammatory, and 3 cases were gray zone (1 of which was eventually diagnosed as cancerous). 98% （97+2）% Gray area samples (9 cases) All cases were accurately classified using the nucleolar number rule (7 cases were cancerous, 2 cases were inflammatory).

[0055] Key indicators include: sensitivity 99%, specificity 98%; and the cancer cell percentage threshold (3.1% vs 2.9%) effectively distinguishes between cancerous and inflammatory samples (AUC = 0.992).

[0056] Although embodiments of the 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 to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for identifying the properties of serous cavity effusion samples based on statistical data, characterized in that, Includes the following steps: Step S1: Prepare a slide for a sample of serous cavity effusion. The operation includes manual slide preparation and staining. Step S2: Inspect the prepared slides. The inspection includes checking whether the cell distribution is uniform and whether the staining is clear. For samples whose slide quality does not meet the inspection requirements, re-prepare the slides. Step S3: Observe the slide image under the microscope, determine the cell type and number based on the morphological atlas of cells in the serous cavity effusion, and record the results; Step S4: Statistically analyze the recorded results obtained in step S3 and calculate the proportion of the total number of cancer cells in all fields of view to the total number of nucleated cells in all fields of view. Step S5: Determine the nature of the sample based on the proportion and number of cancer cells in the field of view.

2. The method for identifying the properties of serous cavity effusion samples based on data statistics according to claim 1, characterized in that, In step S3, the observation field of view shall be no less than 100 fields of view.

3. The method for identifying the properties of serous cavity effusion samples based on data statistics according to claim 1, characterized in that, In step S3, the number of nucleated cells included in the observation field of view for statistical analysis shall not be less than 1,000.

4. The method for identifying the properties of serous cavity effusion samples based on data statistics according to claim 1, characterized in that, In step S3, the types and numbers of nucleated cells in the microscopic field of view are accurately identified and counted. When cancer cells have identifiable nucleolar features, the number of nucleoli is counted.

5. The method for identifying the properties of serous cavity effusion samples based on data statistics according to claim 1, characterized in that, In step S5, the rules for determining the properties of the sample are as follows: When the percentage of cancer cells is greater than or equal to 3.1%, the sample is considered a cancerous sample. When the percentage of cancer cells is less than 2.9%, the sample is considered an inflammatory sample. When the percentage of cancer cells is within the range of 3 ± 0.1%, the number of cancer cell nucleoli is further counted. If the number of cancer cell nucleoli is greater than 3, the sample is considered cancerous; otherwise, it is considered inflammatory.

6. The method for identifying the properties of serous cavity effusion samples based on data statistics according to claim 1, characterized in that, In step S1, an automatic dyeing machine is used for automatic dyeing.

7. The method for identifying the properties of serous cavity effusion samples based on data statistics according to claim 1, characterized in that, In step S2, an automatic microscopic image acquisition device is used to inspect the prepared glass slide.

8. The method for identifying the properties of serous cavity effusion samples based on data statistics according to claim 1, characterized in that, In step S3, a proven and effective computer image algorithm or artificial intelligence network model is used to assist in the interpretation of cell types and quantities under the microscope and obtain the result data through computer software.

9. A system for identifying the properties of serous cavity effusion samples based on statistical data, characterized in that, It includes a creation module, an inspection module, an observation module, a statistics module, and a judgment module, among which: The preparation module is used to prepare slides for serous cavity effusion samples, and the operations include manual slide preparation and staining. The inspection module is used to inspect the prepared slides. The inspection includes checking whether the cell distribution is uniform and whether the staining is clear. Samples whose slide quality does not meet the inspection requirements of the project will be re-prepared. The observation module is used to observe slide images under a microscope, determine cell type and number based on the morphological atlas of cells in the serous cavity effusion, and record the results; The statistics module is used to statistically analyze the recorded results obtained in step S3 and calculate the proportion of the total number of cancer cells in all fields of view to the total number of nucleated cells in all fields of view. The judgment module is used to determine the nature of a sample based on the proportion and number of cancer cells in the field of view.

Images