Osteoarthritis intra-cartilage ossification image recognition method

Abstract

The invention relates to the technical field of medical data detection, in particular to an osteoarthritis intra-cartilage ossification image recognition method, which realizes accurate recognition of osteoarthritis intra-cartilage ossification. The method comprises the following steps: acquiring a cell sample, detecting degeneration, hypertrophy and IL-33 expression of cartilage cells through reverse transcription-polymerase chain reaction, performing hypertrophy induction on the cells, and comparing ossification conditions. Fluid shear force equipment is further utilized to simulate an in-vivo mechanical environment, the cartilage cell hypertrophy degree is evaluated, and multi-dimensional data are obtained through pathological sections, immunohistochemistry and tomography. And finally, acquiring an articular cartilage image of the old people by utilizing a 9.4 T magnetic resonance imaging technology, and obtaining an articular cartilage and subchondral ossification analysis result of the old people in combination with the trained osteoarthritis intra-cartilage ossification image analysis model. According to the method, the accuracy and efficiency of osteoarthritis identification are improved, and powerful support is provided for early prevention and treatment of osteoarthritis.

Description

Technical Field

[0001] The present invention relates to the technical field of medical data detection, and in particular to an image recognition method for endochondral ossification in osteoarthritis. Background Art

[0002] Osteoarthritis is the most common disabling joint disease in the elderly. Its pathological changes include cartilage degeneration, synovial inflammation, and abnormal bone remodeling of subchondral bone. However, there is currently no specific treatment for osteoarthritis. More and more studies have shown that abnormal mechanical stimulation is closely related to the occurrence of osteoarthritis, but the specific pathogenesis is still unclear.

[0003] During normal bone growth and development, chondrocyte hypertrophy is a necessary transient state that eventually forms bone through the process of endochondral ossification. However, in OA, chondrocytes begin to differentiate into a hypertrophic phenotype, inducing a series of events that ultimately lead to cartilage degeneration. For this reason, an imaging method for identifying endochondral ossification in osteoarthritis has been developed to identify and discover the causes of osteoarthritis. Summary of the invention

[0004] The purpose of the present invention is to provide an image recognition method for endochondral ossification in osteoarthritis to solve the problem that it is currently impossible to identify the phenotype of chondrocytes beginning to differentiate into hypertrophic phenotypes, which induces a series of events that ultimately lead to the cause of cartilage degeneration.

[0005] To achieve the above object, a method for image recognition of endochondral ossification in osteoarthritis comprises the following steps:

[0006] Step S1: using a magnetic resonance imaging device to obtain image data of articular cartilage;

[0007] Step S2: preprocessing the image data obtained in step S1 to obtain preprocessed data;

[0008] Step S3: inputting the preprocessed data obtained in step S2 into the osteoarthritis endochondral ossification image analysis model to obtain the analysis results of articular cartilage and subchondral ossification;

[0009] Step S4: extract key features from the analysis results obtained in step S3 to obtain recognition results of endochondral ossification in osteoarthritis.

[0010] Preferably, the preprocessing in step S2 comprises the following steps:

[0011] Step S21: removing noise from the image data;

[0012] Step S22: enhancing the details of the articular cartilage by an algorithm;

[0013] Step S23: Segment the articular cartilage region from the entire image to obtain pre-processed articular cartilage image data.

[0014] Preferably, the construction of the osteoarthritis endochondral ossification imaging analysis model in step S3 comprises the following steps:

[0015] Step S31: obtaining summary data including chondrocytes, cartilage, articular cartilage and subchondral bone;

[0016] The summary data of the chondrocytes include gene expression levels, cell morphology analysis, ossification marker detection results, and average gray value of cartilage in the medial tibial plateau weight-bearing area; the ossification marker detection results include alkaline phosphatase activity and osteocalcin content;

[0017] Step S32: Clean the summary data obtained in step S31, remove outliers, fill missing values, and perform standardization and normalization processing, and finally integrate them into a complete data set;

[0018] Step S33: Divide the data set obtained in step S32 into a training set, a validation set and a test set;

[0019] Step S34: Select a model constructed by a machine learning or deep learning algorithm, train the model using a training set, adjust the model parameters using a validation set, and use a test set to evaluate the performance of the model, including accuracy and recall indicators. Optimize the model based on the evaluation results to ultimately obtain an osteoarthritis endochondral ossification imaging analysis model.

[0020] Select a suitable machine learning or deep learning model for training, use the training set to train the model, adjust the model parameters through the validation set, use the test set to evaluate the performance of the model, including accuracy and recall rate indicators, and optimize the model based on the evaluation results.

[0021] Further preferably, the chondrocytes in step S31 include one or a combination of normal mouse chondrocytes, aged mouse chondrocytes, senescent chondrocytes obtained by continuous in vitro passage of aged mouse chondrocytes, and IL-33 knocked-out normal mouse chondrocytes.

[0022] Regarding the specific method of obtaining chondrocytes, the server receives the experimental design parameters input by the researchers according to the research objectives and experimental requirements. The experimental design parameters include the experimental goal of exploring the endochondral ossification mechanism of osteoarthritis, the required chondrocytes, the aging state of cell characteristics, and the specific requirements for the animal model;

[0023] Using the received experimental design parameters, the server queries the internal database to find animal models and cell types that meet the conditions. Based on the database query results, the server performs data analysis to determine the cell sources that meet the experimental requirements;

[0024] The server pushes the cell sources that meet the experimental requirements to the terminal device of the researcher. The server waits for the researcher's feedback. When receiving the researcher's confirmation, the server will record the decision and begin to prepare for the next experimental steps, including arranging the collection and cultivation of chondrocytes.

[0025] More preferably, the acquisition of the summary data of normal mouse chondrocytes in step S31 comprises the following steps:

[0026] Step A01: obtaining normal mouse chondrocytes;

[0027] Step A02: Inducing hypertrophy of the normal mouse chondrocytes obtained in step A01, obtaining ossification marker detection results and hypertrophy-related cell morphology analysis, and integrating them into the summary data of step S31;

[0028] Step A03: Use a magnetic resonance device to image the normal mouse chondrocytes obtained in step A01 to obtain image data, analyze the average grayscale value of the cartilage in the medial tibial plateau weight-bearing area based on the image data, and integrate it into the summary data of step S31.

[0029] More preferably, the acquisition of summary data of cartilage cells of elderly mice in step S31 comprises the following steps:

[0030] Step B01: injecting rIL-33 into the joint cavity of aged mice, and obtaining chondrocytes of aged mice after the injection;

[0031] Step B02: Perform pathological sectioning on the aged mouse chondrocytes obtained in step B02 to detect chondrocyte hypertrophy, and immunohistochemically analyze the expression of chondrocyte ossification markers to obtain ossification marker detection results and hypertrophy-related cell morphology analysis, and integrate them into the summary data of step S31;

[0032] Step B03: Use a magnetic resonance device to image the cartilage cells of the aged mice obtained in step B02 to obtain image data, analyze the average gray value of the cartilage in the weight-bearing area of ​​the medial tibial plateau based on the image data, and integrate it into the summary data of step S31.

[0033] Intra-articular injection of rIL-33, wild-type aged mice were selected for pretreatment, and the experimenters were instructed to perform intra-articular injection of rIL-33 on wild-type aged mice;

[0034] Pathological sections and immunohistochemical analysis: Collect joint samples at an appropriate time after injection, fix, dehydrate, embed, etc., make pathological sections of articular cartilage, and perform immunohistochemical staining on the sections to detect the expression of articular cartilage hypertrophy and ossification markers;

[0035] The server observed and recorded the staining results through a microscope, and analyzed the effects of rIL-33 injection on articular cartilage, as well as the changes in the expression of hypertrophy and ossification markers.

[0036] More preferably, the acquisition of summary data of aged chondrocytes obtained by continuous in vitro passage of aged mouse chondrocytes in step S31 comprises the following steps:

[0037] Step C01: Obtaining chondrocytes from aged mice;

[0038] Step C02: the aged rat chondrocytes obtained in step C01 are subjected to continuous in vitro passage to obtain senescent chondrocytes;

[0039] Step C03: Detecting the aged chondrocytes obtained in step C02 by reverse transcription-polymerase chain reaction to obtain the results of degeneration-related ossification marker detection, hypertrophy-related cell morphology analysis, and IL-33-related gene expression levels, and integrating them into the summary data of step S31;

[0040] For reverse transcription-polymerase chain reaction and detection, the server obtains RNA samples of aged chondrocytes prepared by the experimenter, controls the experimental equipment to reverse transcribe the RNA, converts the RNA into cDNA, performs polymerase chain reaction (PCR), and amplifies specific gene fragments, including genes related to chondrocyte degeneration, hypertrophy, and IL-33 expression. The server receives the PCR result data, performs quantitative analysis, and determines the expression level of the relevant genes;

[0041] Step C04: stimulating the aged chondrocytes obtained in step C02 in vitro with IL-33 recombinant protein to obtain hypertrophy-related cell morphology analysis and ossification marker detection results, and integrating them into the summary data of step S31;

[0042] IL-33 recombinant protein stimulates in vitro. The server detects the quality and activity of IL-33 recombinant protein. The server sends control commands to the experimental equipment to control the experimental equipment, add IL-33 recombinant protein to the culture environment of aged chondrocytes, and stimulate in vitro. The response of chondrocytes under the stimulation of IL-33 recombinant protein is monitored and recorded in real time, including cell morphology and proliferation.

[0043] Detect chondrocyte hypertrophy and ossification. Use existing image processing technology to analyze the morphological changes of chondrocytes to evaluate whether hypertrophy occurs. Detect the expression levels of ossification-related indicators in chondrocytes, including alkaline phosphatase activity and osteocalcin content. Record and analyze the test results to evaluate the effect of IL-33 on chondrocyte hypertrophy and ossification.

[0044] Step C05, using a magnetic resonance device to image the aged chondrocytes obtained in step C02, obtaining image data, analyzing the average gray value of the cartilage in the medial tibial plateau weight-bearing area based on the image data, and integrating it into the summary data of step S31.

[0045] More preferably, the acquisition of summary data of IL-33 knockout normal mouse chondrocytes in step S31 comprises the following steps:

[0046] Step D01: Obtain IL-33 knockout normal mouse chondrocytes;

[0047] Step D02: Inducing hypertrophy of the IL-33 knockout normal mouse chondrocytes obtained in step D01, obtaining ossification marker detection results and hypertrophy-related cell morphology analysis, and integrating them into the summary data of step S31;

[0048] Acquisition and analysis of normal and IL-33 knockout chondrocytes: Acquisition of normal IL-33 expressing chondrocytes and IL-33 knockout chondrocytes, respectively, and comparative analysis of these IL-33 and IL-33 knockout chondrocytes, including comparison of morphology, proliferation ability, hypertrophy and ossification tendency;

[0049] Based on the comparative analysis results, the mechanism of action of IL-33 in chondrocyte degeneration, hypertrophy and ossification was deduced;

[0050] Hypertrophy induction treatment: the server determines the reagents used to guide the experimenter to induce chondrocyte hypertrophy, and the hypertrophy induction treatment is performed on normal and IL-33 knockout chondrocytes. The hypertrophy induction treatment conditions of normal and IL-33 knockout chondrocytes need to be consistent;

[0051] Comparison of chondrocyte hypertrophy and ossification: observe and record the morphological changes of the two types of chondrocytes after induction under a microscope to evaluate the degree of hypertrophy; detect ossification markers in IL-33 and IL-33 knockout chondrocytes to quantify the degree of ossification, compare the differences in hypertrophy and ossification between normal and IL-33 knockout chondrocytes, and analyze the effect of IL-33 on chondrocyte hypertrophy;

[0052] Step D03: Use a magnetic resonance device to image the IL-33 knockout normal mouse chondrocytes obtained in step D01 to obtain image data, analyze the average grayscale value of the cartilage in the medial tibial plateau weight-bearing area based on the image data, and integrate it into the summary data of step S31.

[0053] Further preferably, the acquisition of summary data of cartilage articular cartilage and subchondral bone in step S31 comprises the following steps:

[0054] Step E01: obtaining chondrocytes from normal mice and IL-33 knockout aged mice respectively;

[0055] Step E02: using a fluid shear force device to apply a predetermined fluid shear force to the chondrocytes obtained in step E01, simulating the mechanical environment of joint movement in vivo, and obtaining the degree of cell morphological change;

[0056] Step E03: specifying a running exercise plan according to the degree of cell morphological changes obtained in step E02, and allowing normal mice and IL-33 knockout aged mice to complete the exercise plan;

[0057] Step E04: imaging the normal mice and the IL-33 knockout aged mice that completed the running exercise in step E03 respectively by micro-computed tomography technology to obtain image data, analyzing the results of changes in articular cartilage and subchondral bone according to the image data, and integrating them into the summary data of step S31;

[0058] Step E05: Obtain chondrocytes from normal mice and chondrocytes from IL-33 knockout aged mice after the running exercise in step E03, culture and multiply them, perform pathological section detection on chondrocyte hypertrophy, perform immunohistochemical analysis on the expression of chondrocyte ossification markers, obtain ossification marker detection results, hypertrophy-related cell morphology analysis, and proliferation capacity, and integrate them into the summary data of step S31.

[0059] Under the control of the server, a fluid shear force device is used to apply a predetermined fluid shear force to the two cells to simulate the mechanical environment of the joint movement in the body;

[0060] The response of chondrocytes to fluid shear force is monitored and recorded in real time. The server observes the changes in cell morphology through a microscope to evaluate the hypertrophy of chondrocytes.

[0061] The treated cells were tested for ossification-related markers to obtain the test results of ossification-related markers of the cells, and the differences in hypertrophy and ossification between wild-type mice and IL-33 knockout cells were compared.

[0062] Under the guidance of the server, a running exercise plan was formulated for the mice, including exercise intensity, time and frequency, and the exercise status of the mice was monitored in real time. After the exercise plan was completed, joint samples of the mice were collected, pathological sections of the joints were made, and immunohistochemical staining was performed to detect the histological changes of articular cartilage and subchondral bone;

[0063] Use micro-computed tomography to perform high-resolution imaging of joint samples to obtain detailed structural information of articular cartilage and subchondral bone;

[0064] Integrate the data of pathological sections, immunohistochemistry and tomography to analyze the changes in articular cartilage and subchondral bone of wild-type mice and IL-33 knockout mice after exercise, and obtain the analysis results

[0065] Preferably, in step S4, the key features include the size, shape and distribution of the ossification area, and the extracted key features are matched in the knowledge base to determine the recognition result of osteoarthritis endochondral ossification.

[0066] By removing noise from the articular cartilage images of the elderly, enhancing the details of the articular cartilage through an algorithm, and segmenting the articular cartilage area from the entire image, the preprocessed articular cartilage image is obtained;

[0067] Substitute the pre-processed articular cartilage images into the trained osteoarthritis endochondral ossification image analysis model, analyze the results inferred by the model, and preliminarily judge the articular cartilage and subchondral ossification of the elderly;

[0068] Key features are extracted from the preliminary judgment of the articular cartilage and subchondral ossification in the elderly. The key features include the size, shape, and distribution of the ossification area. The extracted features are matched in the knowledge base to determine the recognition results of intrachondral ossification in osteoarthritis in the elderly.

[0069] The specific innovative solutions of the present invention are as follows:

[0070] Experimental preparation, determining the cell source, selecting wild-type aged mice as the cell source to ensure the universality and reliability of the experimental results. Obtaining chondrocytes, obtaining chondrocytes from wild-type aged mice, and obtaining senescent chondrocytes through continuous in vitro subculture.

[0071] Genetic testing and analysis, using reverse transcription-polymerase chain reaction (PCR) technology, detects chondrocyte degeneration, hypertrophy and IL-33 expression. This step is designed to capture the initial signs that chondrocytes begin to differentiate into a hypertrophic phenotype. Analyze PCR results and compare the differences in gene expression between normal and degenerated chondrocytes, with a special focus on changes in IL-33 expression to explore its role in chondrocyte hypertrophy and ossification.

[0072] Cell hypertrophy induction and mechanical simulation: normal and IL-33 knockout chondrocytes are subjected to hypertrophy induction treatment, and specific induction conditions are applied to induce chondrocytes to differentiate into hypertrophic phenotypes. A fluid shear force device is used to apply a predetermined fluid shear force to chondrocytes to simulate the mechanical environment during joint movement in vivo. The response of chondrocytes under mechanical stimulation, especially the changes in the degree of hypertrophy, is observed and recorded.

[0073] Multi-dimensional data integration and analysis: collect and integrate multi-dimensional data such as gene expression levels, cell morphology analysis, and ossification marker detection. These data will provide comprehensive information on the state of chondrocytes. Use machine learning or deep learning models to deeply analyze the data to discover potential connections and patterns between data, and further reveal the cause of cartilage degeneration.

[0074] Magnetic resonance imaging and model analysis, using 9.4T magnetic resonance imaging to image the articular cartilage of the elderly and obtain high-resolution articular cartilage images. The acquired articular cartilage images are preprocessed to improve the image quality and analysis accuracy. The preprocessed articular cartilage images are substituted into the trained osteoarthritis endochondral ossification image analysis model, and the model is used to conduct in-depth analysis of the images to obtain the analysis results of articular cartilage and subchondral ossification in the elderly.

[0075] Interpretation and application of results: Based on the model analysis results, combined with clinical information and medical history, the health status of the articular cartilage of the elderly is evaluated. Based on the evaluation results, personalized osteoarthritis prevention and treatment recommendations are provided to the elderly to delay disease progression and improve quality of life.

[0076] Compared with the prior art, this application has the following beneficial effects:

[0077] Beneficial effects of the present invention:

[0078] 1) Improve identification accuracy: By quantitatively detecting the expression of genes or proteins in samples, combined with 9.4T magnetic resonance imaging technology, the identification results of endochondral ossification in osteoarthritis can be more accurately identified. This comprehensive identification method has higher sensitivity and specificity than traditional methods, which is helpful for early detection and intervention of osteoarthritis.

[0079] 2) Deepen understanding of the disease: By inducing hypertrophy of normal and IL-33 knockout chondrocytes and comparing the hypertrophy and ossification of chondrocytes, the present invention helps to gain a deeper understanding of the pathogenesis of osteoarthritis and the process of endochondral ossification.

[0080] 3) Innovative research methods: The present invention uses advanced 9.4T magnetic resonance imaging technology to image articular cartilage, as well as innovative methods such as applying fluid shear force to wild-type mice and IL-33 knockout aged chondrocytes, providing new perspectives and methods for the study of osteoarthritis.

[0081] 4) Promote drug development: By accurately identifying the cause of endochondral ossification in osteoarthritis, the present invention is expected to provide new targets and ideas for drug development, thereby promoting the development of specific drugs for osteoarthritis.

[0082] 5) Clinical application value: The osteoarthritis endochondral ossification image recognition method of the present invention has potential clinical application value. By identifying the cause of osteoarthritis at an early stage, it can help doctors develop more personalized treatment plans and improve the quality of life of patients.

[0083] In summary, the present invention improves the accuracy of osteoarthritis identification by comprehensively using quantitative detection, magnetic resonance imaging, fluid shear force experiments and other means, deepens the understanding of the disease, and provides new ideas and methods for drug development and clinical treatment. DETAILED DESCRIPTION

[0084] The technical solutions in the embodiments of the present invention are described clearly and completely. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0085] In this application, wild-type mice are used as the main experimental material to obtain the corresponding chondrocytes. Wild-type mice refer to mice that have not been genetically modified and can age naturally. Such mice are usually used in scientific experiments as a control or a model for studying changes in organisms during natural aging. Source: Wild-type mice generally come from experimental animal centers or specialized experimental animal breeding facilities. These facilities have strict control over environmental conditions to ensure the health of the animals. Depending on the specific needs of the experiment, the experimenter will select wild-type mice of the appropriate age.

[0086] Senescent chondrocytes obtained by continuous in vitro passage of aged mouse chondrocytes are obtained by continuous in vitro passage of wild-type aged mouse chondrocytes.

[0087] Steps for obtaining chondrocytes and obtaining senescent chondrocytes by continuous passage in vitro:

[0088] Obtaining chondrocytes from wild-type aged mice:

[0089] Experimental preparation: Prepare surgical instruments, culture dishes, culture medium, etc., and ensure that all materials are sterile.

[0090] Animal anesthesia and dissection: After anesthetizing wild-type elderly mice, their joints are exposed through surgery, and areas rich in cartilage tissue such as the knee joint are often selected.

[0091] Chondrocyte Isolation: Under sterile conditions, cartilage tissue was carefully dissected and converted into a single-cell suspension using enzymatic digestion.

[0092] Cell collection: Chondrocytes are collected by centrifugation or other methods, and then washed and resuspended in culture medium.

[0093] Obtaining Senescent Chondrocytes by Continuous Passaging in Vitro:

[0094] Cell culture: Chondrocytes are inoculated in a culture medium containing necessary nutrients and placed in a constant temperature, constant humidity, and sterile incubator.

[0095] Subculture: After the cells grow to a certain density, they are subcultured. Use trypsin or other methods to dissociate the cells from the culture dish and inoculate them into new culture dishes in proportion.

[0096] Serial passaging: Repeated passaging process. As the number of passaging increases, chondrocytes will gradually show aging characteristics.

[0097] Identification of senescent chondrocytes: The senescent state of chondrocytes is confirmed through morphological observation, growth curve analysis, and gene expression detection. Senescent cells usually have characteristics such as morphological changes and decreased proliferation ability.

[0098] Normal mouse chondrocytes are derived from the articular cartilage of normal wild-type mice.

[0099] Steps to obtain normal mouse chondrocytes:

[0100] Source: Normal mouse chondrocytes can be isolated from the articular cartilage of healthy wild-type young mice.

[0101] Isolation method: A method similar to that used to obtain senescent chondrocytes is used, including anesthesia, dissection, cartilage tissue stripping, and enzymatic digestion to separate cells.

[0102] Culture: The isolated normal chondrocytes are inoculated in an appropriate culture medium for culture and expansion.

[0103] The aged mouse chondrocytes were derived from the articular cartilage of normally aged wild-type aged mice.

[0104] Steps for obtaining chondrocytes from aged mice:

[0105] Source: Rat chondrocytes can be isolated from the articular cartilage of aged wild-type rats. The remaining steps are the same as those for obtaining normal rat chondrocytes.

[0106] IL-33 knockout normal mouse chondrocytes are obtained by knocking out the IL-33 gene based on normal mouse chondrocytes.

[0107] Steps for obtaining IL-33 knockout normal mouse chondrocytes:

[0108] Gene editing technology: Use CRISPR-Cas9 and other gene editing technologies to specifically knock out the IL-33 gene.

[0109] Design gRNA: Based on the sequence of the IL-33 gene, design a specific gRNA (guide RNA) to guide the Cas9 protein to cut the IL-33 gene.

[0110] Transfection and screening: The designed gRNA and Cas9 protein expression vector are transfected into chondrocytes, and then cells with successful knockout of the IL-33 gene are screened out through drug screening or other methods.

[0111] Verification: Use sequencing, RT-PCR or Western Blot to verify whether the IL-33 gene has been successfully knocked out.

[0112] This example further describes the method for obtaining senescent chondrocytes obtained by continuous passage of chondrocytes of aged mice in vitro:

[0113] Cell source: Wild-type aged mice were selected as the cell source in this experiment because they can more truly reflect the changes in cells during the natural aging process.

[0114] Collection method: First, wild-type aged mice were anesthetized and dissected to expose the joints. Then, the cartilage tissue was carefully peeled off and converted into a single cell suspension by enzymatic digestion. Finally, chondrocytes were collected by centrifugation to prepare for subsequent in vitro culture.

[0115] Cell passaging:

[0116] Passaging method: The collected chondrocytes are inoculated in a culture medium containing appropriate nutrients for in vitro culture. When the cells grow to a certain density, the passaging operation is performed, that is, the cells are dissociated from the bottom of the culture dish using digestive enzymes such as trypsin, and then inoculated into a new culture dish in a certain proportion for continued culture.

[0117] Objective: To obtain aged chondrocytes by simulating the natural aging process of cells in vivo with age through serial passage.

[0118] Taking aged chondrocytes obtained by continuous in vitro passage of aged mouse chondrocytes as an example, the specific method for detecting gene expression of chondrocytes in this embodiment is as follows:

[0119] Detection method: Reverse transcription-polymerase chain reaction (RT-PCR) technology is used to detect the degeneration, hypertrophy and IL-33 expression of aged chondrocytes. The specific steps include extracting RNA samples from chondrocytes, reverse transcription to obtain cDNA, and then amplifying specific gene fragments through PCR, and finally performing quantitative analysis.

[0120] The senescent chondrocytes obtained by continuous passage of chondrocytes of aged mice in vitro, the specific method of protein stimulation in this example is:

[0121] Stimulating substances: IL-33 recombinant protein (rIL-33) was used to stimulate senescent chondrocytes cultured in vitro.

[0122] Experimental steps: First, test the quality and activity of IL-33 recombinant protein to ensure that it meets the experimental requirements. Then add it to the culture environment of chondrocytes to monitor and record the response of chondrocytes to IL-33 stimulation in real time.

[0123] The specific gene editing method of this embodiment is to obtain aged chondrocytes from aged rat chondrocytes by continuous passage in vitro:

[0124] Methods: The IL-33 gene was knocked out in chondrocytes using gene editing technology (such as the CRISPR-Cas9 system) to obtain IL-33 knockout (IL-33 knockout) chondrocytes.

[0125] Objective: To investigate the specific role of IL-33 in the hypertrophy and ossification process by comparing the differences between normal chondrocytes and IL-33 knockout chondrocytes.

[0126] Method for inducing hypertrophy of normal mouse chondrocytes and IL-33 knockout normal mouse chondrocytes:

[0127] Induction treatment: The normal and IL-33 knockout chondrocytes were subjected to the same hypertrophy induction treatment to observe and compare the differences in the hypertrophy process between the two.

[0128] Evaluation method: The degree of chondrocyte hypertrophy was evaluated by observing and recording the morphological changes of chondrocytes under a microscope.

[0129] Intra-articular injection of rIL-33 into aged mouse chondrocytes:

[0130] Experimental subjects: Healthy wild-type aged mice were selected as experimental subjects.

[0131] Injection method: rIL-33 was injected intra-articularly into aged mice to study the effect of rIL-33 on articular cartilage in vivo.

[0132] Pathological analysis:

[0133] Sample Collection: Collect joint samples at appropriate times after injection for subsequent analysis.

[0134] Processing method: The collected joint samples are fixed, dehydrated, embedded, and made into pathological sections.

[0135] Analysis method: Immunohistochemical staining was performed on the sections to detect the expression of articular cartilage hypertrophy and ossification markers. The effects of rIL-33 injection on articular cartilage were evaluated by observing and analyzing the staining results.

[0136] The articular cartilage was imaged using 9.4T magnetic resonance imaging, and the average grayscale value of the cartilage in the weight-bearing area of ​​the medial tibial plateau was analyzed. The sample selection method for imaging the articular cartilage using 9.4T magnetic resonance imaging, the specific operation process of magnetic resonance imaging, and the analysis method of the average grayscale value of the cartilage in the weight-bearing area of ​​the medial tibial plateau are as follows:

[0137] Sample selection method:

[0138] Sample source: Articular cartilage of wild-type aged mice that had undergone a series of experimental treatments was selected as the sample. The chondrocytes of these aged mice included senescent chondrocytes that had been previously obtained by continuous passage in vitro and subsequently treated with IL-33.

[0139] Sample preparation: Before magnetic resonance imaging, the joints of these elderly mice need to be properly fixed and prepared to ensure sample stability and image clarity during imaging.

[0140] The specific operation process of magnetic resonance imaging:

[0141] Equipment used: 9.4T magnetic resonance imaging equipment is used to image articular cartilage. This equipment has high-resolution imaging capabilities, which helps to accurately observe the fine structure of cartilage.

[0142] Set imaging parameters: According to the specific needs of the experiment, set appropriate imaging parameters, such as scanning sequence, layer thickness, layer spacing, etc., to ensure the quality of imaging and the accuracy of data.

[0143] Perform imaging: Place the prepared sample in the MRI device for imaging. During the imaging process, the device scans the sample according to preset parameters and generates corresponding image data.

[0144] Obtain image data: After imaging is completed, export image data from the device for subsequent analysis and processing.

[0145] Analysis method of the average gray value of cartilage in the weight-bearing area of ​​the medial tibial plateau:

[0146] Image data preprocessing: The acquired image data is first preprocessed, including noise removal, contrast enhancement, etc., to improve the image quality and the accuracy of subsequent analysis.

[0147] Precisely select the area: In the pre-processed image, accurately select the area of ​​cartilage in the weight-bearing area of ​​the medial tibial plateau. This usually requires the use of professional medical image processing software, through the setting of thresholds, edge detection and other technical means to achieve.

[0148] Calculate grayscale value: Calculate the average grayscale value of cartilage in the selected area. Grayscale value is an indicator that reflects the brightness of the image. In magnetic resonance imaging, different tissues have different grayscale values, so it can be used to distinguish different tissue types.

[0149] Result analysis and comparison: The calculated average gray value is compared with the standard value or the control group data to evaluate the health status of the cartilage and possible changes. The change of gray value can reflect the pathological process of cartilage degeneration, hypertrophy and ossification.

[0150] Fluid shear force was applied to chondrocytes of wild-type mice and IL-33 knockout aged chondrocytes to detect chondrocyte hypertrophy and ossification. Wild-type mice and IL-33 knockout aged mice were subjected to running exercise. The changes in articular cartilage and subchondral bone were analyzed by pathological sections, immunohistochemistry and micro-computed tomography technology to obtain analysis data on changes in articular cartilage and subchondral bone.

[0151] The data generated in the above detection and data analysis process are summarized to obtain a data set, and the model is trained through the data set to obtain a trained osteoarthritis endochondral ossification imaging analysis model; the specific steps of running exercise of wild-type mice and IL-33 knockout aged mice, and subsequent pathological section and immunohistochemical analysis are as follows:

[0152] Running exercise process:

[0153] Experimental preparation: First, select wild-type mice and IL-33 knockout aged mice to ensure that their physiological status and age are similar for control experiments.

[0154] Exercise plan: Next, develop a running exercise plan for the mice, which includes determining the appropriate exercise intensity, duration, and frequency to meet the needs of the experiment.

[0155] Real-time monitoring: During exercise, mice were placed in a treadmill and their movement status was monitored in real time by monitoring equipment to ensure the accurate execution of the exercise plan.

[0156] Sample collection: After the running program, collect joint samples from mice in time for subsequent analysis.

[0157] Obtaining pathological sections:

[0158] Sample processing: The collected joint samples were first fixed to maintain their morphological structure.

[0159] Dehydration and embedding: Fixed samples are dehydrated and embedded in paraffin in preparation for sectioning.

[0160] Sectioning: Use a microtome to cut the embedded tissue into thin slices, maintaining the integrity and continuity of the slices.

[0161] Sticking and baking: The cut tissue sections are attached to the slide and baked to enhance adhesion.

[0162] Immunohistochemical analysis:

[0163] Pretreatment: The sections are first dewaxed and hydrated to restore the water-soluble state, and then antigen retrieval is performed to expose the antigenic sites.

[0164] Blocking and blocking: After blocking endogenous peroxidase, use blocking solution to block non-specific binding sites.

[0165] Antibody incubation: Use specific primary antibody and secondary antibody for incubation in sequence to allow the antibody to specifically bind to the antigen.

[0166] Color development and observation: The location of the antigen is displayed through color development reaction, and finally the slides are counterstained and sealed to make permanent sections for microscopic observation and analysis.

[0167] These steps ensure the accuracy and repeatability of the experiment and provide an important experimental basis for studying endochondral ossification in osteoarthritis.

[0168] In order to obtain analytical data on changes in articular cartilage and subchondral bone, the key parameters and steps involved include:

[0169] Applying fluid shear stress to wild-type mouse and IL-33 knockout aged chondrocytes:

[0170] Specialized equipment is used to simulate the mechanical environment of joint movement in the body.

[0171] Under the control of the server, a preset fluid shear force is applied to the two cells.

[0172] The response of chondrocytes to fluid shear stress was monitored and recorded in real time, and the degree of hypertrophy was evaluated by observing their morphological changes under a microscope.

[0173] Detection of chondrocyte hypertrophy and ossification:

[0174] Ossification-related markers were detected in the treated cells to compare the differences in hypertrophy and ossification between wild-type mice and IL-33 knockout cells.

[0175] The changes in cell morphology and the detection results of ossification markers are combined to comprehensively evaluate the status of chondrocytes.

[0176] Running exercise for wild-type mice and IL-33 knockout aged mice:

[0177] Under the guidance of the server, a specific running exercise plan is formulated for the mice, which involves setting the exercise intensity, time and frequency.

[0178] Monitor the movement status of mice in real time to ensure that the movement plan is accurately executed.

[0179] After the exercise program, joint samples from mice were collected for subsequent analysis.

[0180] Analysis using pathological sections, immunohistochemistry and micro-computed tomography:

[0181] Pathological sections of the joints were made to observe the histological changes of the articular cartilage and subchondral bone.

[0182] Immunohistochemical staining was performed on pathological sections to detect the expression of articular cartilage hypertrophy and ossification markers.

[0183] Micro-computed tomography was used to perform high-resolution imaging of joint samples, obtaining detailed structural information.

[0184] Obtain analysis data of changes in articular cartilage and subchondral bone:

[0185] Integrate data obtained from pathology slides, immunohistochemistry, and tomography.

[0186] Through comparative analysis, the differences and characteristics of changes in articular cartilage and subchondral bone between wild-type mice and IL-33 knockout mice were revealed.

[0187] In summary, by simulating the in vivo mechanical environment, evaluating cell status, formulating and executing mouse exercise plans, and combining a variety of technical means for comprehensive analysis, we ultimately obtained comprehensive and in-depth analysis data.

[0188] There are specific reasons and purposes for selecting this data, building specific models, and using the algorithms involved.

[0189] Reasons for selecting these data:

[0190] Comprehensive reflection of the pathological process: The selected data can comprehensively reflect the degeneration, hypertrophy and ossification process of chondrocytes, which is the key to understanding the pathological mechanism of osteoarthritis.

[0191] Simulating real environment: By using chondrocytes from aged mice and simulating the in vivo mechanical environment, the data is closer to the real condition of human osteoarthritis patients, improving the accuracy and reliability of the research.

[0192] High-resolution imaging data: The data provided by 9.4T magnetic resonance imaging and micro-computed tomography technology have high resolution and can capture detailed structural information of articular cartilage and subchondral bone, providing rich samples for image recognition models.

[0193] Construction of specific model:

[0194] Data integration: First, the diversified data collected in the aforementioned experiments, including gene expression, cell morphology, and ossification marker detection, were summarized to form a comprehensive data set.

[0195] Data preprocessing: Clean, standardize, and normalize the data to ensure data quality and consistency in preparation for model training.

[0196] Model selection and training: Use machine learning or deep learning algorithms to build models, such as convolutional neural networks (CNN), support vector machines (SVM), or random forests. The model is trained using the training set, the validation set adjusts the parameters, and the test set is used to evaluate the performance.

[0197] Algorithms involved:

[0198] RT-PCR: Used to detect gene expression levels, it is a commonly used technique in molecular biology.

[0199] Image processing technology: used to analyze the morphological changes of chondrocytes and evaluate their hypertrophy degree, which involves algorithms such as image segmentation and feature extraction.

[0200] Machine learning / deep learning: Select appropriate algorithms for model training based on data characteristics and analysis requirements to achieve accurate identification of endochondral ossification in osteoarthritis.

[0201] The articular cartilage of the elderly was imaged using 9.4T magnetic resonance imaging to obtain images of the articular cartilage of the elderly, and the images of the articular cartilage of the elderly were preprocessed. The preprocessed articular cartilage of the elderly was substituted into the trained osteoarthritis intrachondral ossification image analysis model to obtain analysis results of the articular cartilage and subchondral ossification in the elderly, and feature extraction was performed on the analysis results of intrachondral ossification in osteoarthritis to obtain the cause of intrachondral ossification in osteoarthritis in the elderly.

[0202] In order to define and image endochondral ossification in the elderly, the following key data are required:

[0203] Basic information for seniors:

[0204] Age: Ensure that the subjects have reached the geriatric criteria, usually defined as 65 years and above, which is an important basis for judging the elderly population.

[0205] Gender: Collecting gender data facilitates statistical analysis to understand the prevalence and distribution of OA in different genders.

[0206] Weight and height: These two data can be used to calculate BMI (body mass index), thereby exploring the potential impact of factors such as obesity on osteoarthritis.

[0207] Medical history: Detailed record of previous history of joint diseases, surgical history, medication history, etc., which will help eliminate other interfering factors that may affect the research results.

[0208] Articular cartilage imaging data:

[0209] High-resolution images of articular cartilage in elderly people were obtained using 9.4T magnetic resonance imaging technology. This technology can provide clearer details of the cartilage and the underlying bone structure.

[0210] The images obtained must include adequate anatomical detail, particularly areas commonly affected by osteoarthritis, such as the medial tibial plateau weight-bearing area.

[0211] Image preprocessing data:

[0212] The preprocessing steps include removing image noise and enhancing the details of articular cartilage through specific algorithms to improve image quality and ensure the accuracy of subsequent analysis.

[0213] It is also necessary to accurately segment the articular cartilage area and extract the cartilage part from the entire image in preparation for analysis.

[0214] Osteoarthritis endochondral ossification imaging analysis model:

[0215] This model needs to be trained in advance with a diverse data set (such as gene expression data, cell morphology analysis, ossification marker detection results, etc.).

[0216] A well-trained model should be able to accurately identify the ossification of articular cartilage and subchondral bone and provide quantitative analysis results.

[0217] Feature extraction of ossification analysis results:

[0218] Key features such as the size, shape and distribution of the ossified areas are extracted from the results of the model analysis.

[0219] The extracted features will be used for knowledge base matching to determine the specific causes of endochondral ossification in osteoarthritis.

[0220] Clinical information combined with medical history analysis:

[0221] Finally, the analysis results of the model are considered comprehensively with the clinical information and medical history data of the elderly.

[0222] Based on the results of the comprehensive assessment, osteoarthritis prevention and treatment recommendations are tailored for each elderly person.

[0223] The technical solution of the present invention follows the following steps in the process of making image recognition and judgment to obtain the cause of endochondral ossification of osteoarthritis in the elderly:

[0224] First, we prepare the data and train the model. This includes in-depth analysis of changes in articular cartilage and subchondral bone in specific types of mice through a series of experimental techniques, such as fluid shear stress experiments, pathological sections, immunohistochemistry, and micro-computed tomography, and the collection of relevant data. These data are integrated with other multi-dimensional data such as gene expression levels and cell morphology analysis to form a comprehensive data set. Subsequently, we use these data sets to train machine learning or deep learning models to obtain an image analysis model specifically for endochondral ossification in osteoarthritis.

[0225] Next, we used 9.4T magnetic resonance imaging technology to perform high-resolution imaging of the articular cartilage of the elderly. The acquired images will undergo a series of preprocessing steps, such as noise removal, detail enhancement, and articular cartilage area segmentation, to improve image quality and ensure the accuracy of analysis.

[0226] The pre-processed articular cartilage images will be substituted into the previously trained osteoarthritis endochondral ossification image analysis model. The model will conduct in-depth analysis of these images and make a preliminary judgment on the articular cartilage and subchondral ossification of the elderly.

[0227] After obtaining the preliminary judgment results, we will further extract key features, such as the size, shape and distribution of the ossification area. These features will be used for matching in the knowledge base to ultimately determine the specific cause of endochondral ossification in osteoarthritis in the elderly.

[0228] Finally, combining the model's analysis results, clinical information, and medical history, we provide a comprehensive assessment of the health of the elderly's articular cartilage. Based on this assessment, we are able to provide each elderly person with personalized osteoarthritis prevention and treatment recommendations, aiming to delay disease progression and improve their quality of life.

[0229] In general, the technical solution of the present invention achieves accurate judgment of the cause of endochondral ossification of osteoarthritis in the elderly and provision of personalized treatment recommendations through the comprehensive application of multi-dimensional data and advanced machine learning or deep learning models.

[0230] Embodiment 1

[0231] This embodiment discloses a method for image recognition of endochondral ossification of osteoarthritis, comprising the following steps: Step S1: using a 9.4T magnetic resonance device to obtain image data of articular cartilage of the elderly;

[0232] Step S2: preprocessing the elderly image data obtained in step S1 to obtain preprocessed data;

[0233] Step S3: inputting the preprocessed data obtained in step S2 into the osteoarthritis endochondral ossification image analysis model to obtain the analysis results of articular cartilage and subchondral ossification;

[0234] Step S4: extract key features from the analysis results obtained in step S3 to obtain recognition results of endochondral ossification in osteoarthritis.

[0235] The preprocessing in step S2 includes the following steps:

[0236] Step S21: removing noise from the image data;

[0237] Step S22: enhancing the details of the articular cartilage by an algorithm;

[0238] Step S23: Segment the articular cartilage region from the entire image to obtain pre-processed articular cartilage image data.

[0239] In step S4, the key features including the size, shape, and distribution of the ossified area are combined with the extracted key features and matched in the knowledge base to determine the recognition result of osteoarthritis endochondral ossification.

[0240] By removing noise from the articular cartilage images of the elderly, enhancing the details of the articular cartilage through an algorithm, and segmenting the articular cartilage area from the entire image, the preprocessed articular cartilage image is obtained;

[0241] Substitute the pre-processed articular cartilage images into the trained osteoarthritis endochondral ossification image analysis model, analyze the results inferred by the model, and preliminarily judge the articular cartilage and subchondral ossification of the elderly;

[0242] Key features are extracted from the preliminary judgment of the articular cartilage and subchondral ossification in the elderly. The key features include the size, shape, and distribution of the ossification area. The extracted features are matched in the knowledge base to determine the recognition results of intrachondral ossification in osteoarthritis in the elderly.

[0243] The process of noise removal, detail enhancement, model inference result analysis, and key feature extraction and matching for articular cartilage images of the elderly is as follows:

[0244] Remove Noise:

[0245] Technical description: Filters such as Gaussian filters, median filters or bilateral filters are usually used to reduce or eliminate random noise in the image, thereby smoothing the image while preserving edge information as much as possible.

[0246] Implementation details: The algorithm automatically detects the noise level in the image and selects appropriate filter parameters for denoising accordingly.

[0247] Detail Enhancement:

[0248] Description of the technique: To improve visibility and contrast of the articular cartilage area, techniques such as image sharpening and edge enhancement are used.

[0249] Implementation details: Specific algorithms include Laplacian operator, Sobel operator, etc. to detect and enhance the edges in the image, while histogram equalization and other techniques are used to improve the overall contrast.

[0250] Model inference results analysis:

[0251] Method description: The preprocessed articular cartilage images will be substituted into the trained osteoarthritis endochondral ossification image analysis model.

[0252] Analysis process: involves interpreting the model output, which includes information such as the identified ossification areas and their extent, and validating and evaluating it in combination with clinical knowledge.

[0253] Key feature extraction and matching:

[0254] Characterization: Key features include size, shape, and distribution of ossified areas.

[0255] Extraction methods: Image processing algorithms, such as morphological operations and region analysis, are used to ensure accurate and repeatable extraction of these features.

[0256] Matching process and cause determination: These key features are then matched with a knowledge base containing information related to known causes of osteoarthritis endochondral ossification, combined with clinical information and medical history to ultimately determine the specific cause. This step requires the assistance of medical experts for verification.

[0257] Embodiment 2

[0258] This embodiment discloses a method for constructing an imaging analysis model of endochondral ossification of osteoarthritis, comprising the following steps:

[0259] Step S31: obtaining summary data including chondrocytes, cartilage, articular cartilage and subchondral bone;

[0260] The summary data of the chondrocytes include gene expression levels, cell morphology analysis, ossification marker detection results, and average gray value of cartilage in the medial tibial plateau weight-bearing area; the ossification marker detection results include alkaline phosphatase activity and osteocalcin content;

[0261] Step S32: Clean the summary data obtained in step S31, remove outliers, fill missing values, and perform standardization and normalization processing, and finally integrate them into a complete data set;

[0262] Step S33: Divide the data set obtained in step S32 into a training set, a validation set and a test set;

[0263] Step S34: Select a model constructed by a machine learning or deep learning algorithm, train the model using a training set, adjust the model parameters using a validation set, and use a test set to evaluate the performance of the model, including accuracy and recall indicators. Optimize the model based on the evaluation results to ultimately obtain an osteoarthritis endochondral ossification imaging analysis model.

[0264] Select a suitable machine learning or deep learning model for training, use the training set to train the model, adjust the model parameters through the validation set, use the test set to evaluate the performance of the model, including accuracy and recall rate indicators, and optimize the model based on the evaluation results.

[0265] The specific training and verification technical solutions are as follows:

[0266] Training technology program:

[0267] Data integration and preprocessing: First, we integrated the data collected from steps S2, S3, and S4, which covered gene expression levels, cell morphology analysis, and ossification marker detection results. Then, we cleaned the data, removed outliers, filled missing values, and completed standardization and normalization.

[0268] Dataset segmentation: The preprocessed data will be integrated into a complete dataset and divided into training set, validation set and test set according to the requirements of model training.

[0269] Model selection and training: We will select a suitable machine learning or deep learning model for training. Commonly used models include convolutional neural networks (CNN), support vector machines (SVM) or random forests. The specific model to be selected will be determined based on the characteristics of the data and the complexity of the problem. Afterwards, the model is trained using the training set.

[0270] Parameter tuning: During training, we use the validation set to adjust the model parameters, such as the learning rate, regularization coefficient, or number of network layers, to enhance the model's performance on unknown data.

[0271] Verification technology solution:

[0272] Model effect verification: Use the validation set to test the effect of the trained model to evaluate the performance of the model when processing independent data sets.

[0273] Performance evaluation: Use the test set to comprehensively evaluate the performance of the model, mainly examining indicators such as accuracy and recall. Accuracy shows the correct proportion of model predictions, while recall reflects the model's ability to find all positive examples. Based on these evaluation results, we will make necessary optimizations to the model.

[0274] Result interpretation and optimization: In-depth analysis of the model's specific performance on the training set, validation set, and test set to identify and address possible overfitting or underfitting issues. Based on these analyses, further optimize the model to improve its prediction accuracy and generalization ability.

[0275] Embodiment 3

[0276] This example specifically describes the specific method for obtaining chondrocytes:

[0277] The server receives the experimental design parameters input by researchers based on the research objectives and experimental requirements. The experimental design parameters include the experimental goal of exploring the endochondral ossification mechanism of osteoarthritis, the required chondrocytes, the aging state of cell characteristics, and the specific requirements for the animal model;

[0278] Using the received experimental design parameters, the server queries the internal database to find animal models and cell types that meet the conditions. Based on the database query results, the server performs data analysis to determine the cell sources that meet the experimental requirements;

[0279] The server pushes the cell sources that meet the experimental requirements to the terminal device of the researcher. The server waits for the researcher's feedback. When receiving the researcher's confirmation, the server will record the decision and begin to prepare for the next experimental steps, including arranging the collection and cultivation of chondrocytes.

[0280] In this embodiment, the process of the server performing data analysis to determine the cell source that meets the experimental requirements is meticulous and critical. The following is a detailed description of this process:

[0281] First, the server will receive the experimental design parameters entered by researchers based on the research objectives and experimental requirements. These parameters provide the server with a basis for screening and evaluation, including experimental objectives (such as exploring the mechanism of endochondral ossification in osteoarthritis), required chondrocytes, cell characteristics, aging status, and specific requirements for animal models.

[0282] Then, using these parameters, the server queries its internal database for matching animal models and cell types. This database is a rich repository of detailed information on a variety of animal models and cell types, including species, age, genetic background, health status, etc.

[0283] After obtaining the query results, the server will conduct in-depth data analysis. This process involves several important links:

[0284] Compare experimental requirements: The server will compare the query results with the experimental design parameters one by one in order to screen out those animal models and cell types that fully meet the experimental requirements.

[0285] Evaluate cell status: Based on the specific requirements of the experiment for the aging state of cell characteristics, the server will further evaluate whether the candidate cells are exactly in the aging stage required by the experiment. This step is crucial to ensure the effectiveness and accuracy of the experiment.

[0286] Consider availability and feasibility: When filtering out multiple options that meet the criteria, the server will also consider the availability of cell sources and the feasibility of experimental operations. This includes evaluating factors such as the ease of obtaining cells, cost-effectiveness, and the reproducibility of experimental operations.

[0287] Through this series of comprehensive analyses, the server can ultimately determine the cell source that best meets the experimental needs. This decision-making process not only relies on rigorous data comparison and evaluation, but also combines the actual feasibility and operability of the experiment, thereby ensuring the smooth progress of the experiment and the achievement of the goal.

[0288] Finally, the server will push the selected cell source information to the researchers and wait for their feedback. Once the researchers confirm, the server will start to make full preparations for the next experimental steps, including coordinating laboratory resources, arranging cell collection time, and preparing necessary culture media.

[0289] Example 4 Normal

[0290] This embodiment discloses that the acquisition of summary data of normal mouse chondrocytes in step S31 includes the following steps:

[0291] Step A01: obtaining normal mouse chondrocytes;

[0292] Step A02: Inducing hypertrophy of the normal mouse chondrocytes obtained in step A01, obtaining ossification marker detection results and hypertrophy-related cell morphology analysis, and integrating them into the summary data of step S31;

[0293] Step A03: Use a magnetic resonance device to image the normal mouse chondrocytes obtained in step A01 to obtain image data, analyze the average grayscale value of the cartilage in the medial tibial plateau weight-bearing area based on the image data, and integrate it into the summary data of step S31.

[0294] Example 5 Aging

[0295] This embodiment discloses that the acquisition of summary data of cartilage cells of elderly mice in step S31 includes the following steps:

[0296] Step B01: injecting rIL-33 into the joint cavity of aged mice, and obtaining chondrocytes of aged mice after the injection;

[0297] Step B02: Perform pathological sectioning on the aged mouse chondrocytes obtained in step B02 to detect chondrocyte hypertrophy, and immunohistochemically analyze the expression of chondrocyte ossification markers to obtain ossification marker detection results and hypertrophy-related cell morphology analysis, and integrate them into the summary data of step S31;

[0298] Step B03: Use a magnetic resonance device to image the cartilage cells of the aged mice obtained in step B02 to obtain image data, analyze the average gray value of the cartilage in the weight-bearing area of ​​the medial tibial plateau based on the image data, and integrate it into the summary data of step S31.

[0299] Intra-articular injection of rIL-33, wild-type aged mice were selected for pretreatment, and the experimenters were instructed to perform intra-articular injection of rIL-33 on wild-type aged mice;

[0300] Pathological sections and immunohistochemical analysis: Collect joint samples at an appropriate time after injection, fix, dehydrate, embed, etc., make pathological sections of articular cartilage, and perform immunohistochemical staining on the sections to detect the expression of articular cartilage hypertrophy and ossification markers;

[0301] The server observed and recorded the staining results through a microscope, and analyzed the effects of rIL-33 injection on articular cartilage, as well as the changes in the expression of hypertrophy and ossification markers.

[0302] Example 6 Passaging

[0303] This embodiment discloses that the acquisition of summary data of aged chondrocytes obtained by continuous in vitro passage of aged mouse chondrocytes in step S31 includes the following steps:

[0304] Step C01: Obtaining chondrocytes from aged mice;

[0305] Step C02: the aged rat chondrocytes obtained in step C01 are subjected to continuous in vitro passage to obtain senescent chondrocytes;

[0306] Step C03: Detecting the aged chondrocytes obtained in step C02 by reverse transcription-polymerase chain reaction to obtain the results of degeneration-related ossification marker detection, hypertrophy-related cell morphology analysis, and IL-33-related gene expression levels, and integrating them into the summary data of step S31;

[0307] For reverse transcription-polymerase chain reaction and detection, the server obtains RNA samples of aged chondrocytes prepared by the experimenter, controls the experimental equipment to reverse transcribe the RNA, converts the RNA into cDNA, performs polymerase chain reaction (PCR), and amplifies specific gene fragments, including genes related to chondrocyte degeneration, hypertrophy, and IL-33 expression. The server receives the PCR result data, performs quantitative analysis, and determines the expression level of the relevant genes;

[0308] Step C04: stimulating the aged chondrocytes obtained in step C02 in vitro with IL-33 recombinant protein to obtain hypertrophy-related cell morphology analysis and ossification marker detection results, and integrating them into the summary data of step S31;

[0309] IL-33 recombinant protein stimulates in vitro. The server detects the quality and activity of IL-33 recombinant protein. The server sends control commands to the experimental equipment to control the experimental equipment, add IL-33 recombinant protein to the culture environment of aged chondrocytes, and stimulate in vitro. The response of chondrocytes under the stimulation of IL-33 recombinant protein is monitored and recorded in real time, including cell morphology and proliferation.

[0310] Detect chondrocyte hypertrophy and ossification. Use existing image processing technology to analyze the morphological changes of chondrocytes to evaluate whether hypertrophy occurs. Detect the expression levels of ossification-related indicators in chondrocytes, including alkaline phosphatase activity and osteocalcin content. Record and analyze the test results to evaluate the effect of IL-33 on chondrocyte hypertrophy and ossification.

[0311] Image processing technology is used to analyze the morphological changes of chondrocytes to assess whether they are hypertrophic. The specific process includes the following steps:

[0312] Image acquisition: First, clear images of chondrocytes are captured using high-precision equipment such as a microscope.

[0313] Image preprocessing: The captured images are preprocessed, including denoising and contrast enhancement, to improve image quality and ensure clear visibility of cell features.

[0314] Feature extraction: Use advanced image processing techniques such as edge detection and morphological operations to accurately extract cell morphological parameters such as area, perimeter, and aspect ratio.

[0315] Morphological analysis: Based on the extracted morphological parameters and the preset cell hypertrophy standards (such as a significant increase in cell area or irregular changes in morphology), it is determined whether chondrocytes have hypertrophy.

[0316] Step C05, using a magnetic resonance device to image the aged chondrocytes obtained in step C02, obtaining image data, analyzing the average gray value of the cartilage in the medial tibial plateau weight-bearing area based on the image data, and integrating it into the summary data of step S31.

[0317] Example 7 Knockout

[0318] This embodiment discloses that the acquisition of summary data of IL-33 knockout normal mouse chondrocytes in step S31 includes the following steps:

[0319] Step D01: Obtain IL-33 knockout normal mouse chondrocytes;

[0320] Step D02: Inducing hypertrophy of the IL-33 knockout normal mouse chondrocytes obtained in step D01, obtaining ossification marker detection results and hypertrophy-related cell morphology analysis, and integrating them into the summary data of step S31;

[0321] Acquisition and analysis of normal and IL-33 knockout chondrocytes: Acquisition of normal IL-33 expressing chondrocytes and IL-33 knockout chondrocytes, respectively, and comparative analysis of these IL-33 and IL-33 knockout chondrocytes, including comparison of morphology, proliferation ability, hypertrophy and ossification tendency;

[0322] Based on the comparative analysis results, the mechanism of action of IL-33 in the degeneration, hypertrophy and ossification of chondrocytes is deduced; for hypertrophy induction treatment, the server determines the reagents used to guide the experimenters to induce chondrocyte hypertrophy, and hypertrophy induction treatment is performed on normal and IL-33 knockout chondrocytes. The hypertrophy induction treatment conditions of normal and IL-33 knockout chondrocytes need to be consistent;

[0323] Comparison of chondrocyte hypertrophy and ossification: observe and record the morphological changes of the two types of chondrocytes after induction under a microscope to evaluate the degree of hypertrophy; detect ossification markers in IL-33 and IL-33 knockout chondrocytes to quantify the degree of ossification, compare the differences in hypertrophy and ossification between normal and IL-33 knockout chondrocytes, and analyze the effect of IL-33 on chondrocyte hypertrophy;

[0324] Step D03: Use a magnetic resonance device to image the IL-33 knockout normal mouse chondrocytes obtained in step D01 to obtain image data, analyze the average grayscale value of the cartilage in the medial tibial plateau weight-bearing area based on the image data, and integrate it into the summary data of step S31.

[0325] Example 8 Cartilage

[0326] This embodiment discloses that the acquisition of the summary data of cartilage articular cartilage and subchondral bone in step S31 includes the following steps:

[0327] Step E01: obtaining chondrocytes from normal mice and IL-33 knockout aged mice respectively;

[0328] Step E02: using a fluid shear force device to apply a predetermined fluid shear force to the chondrocytes obtained in step E01, simulating the mechanical environment of joint movement in vivo, and obtaining the degree of cell morphological change;

[0329] Step E03: specifying a running exercise plan according to the degree of cell morphological changes obtained in step E02, and allowing normal mice and IL-33 knockout aged mice to complete the exercise plan;

[0330] Step E04: imaging the normal mice and the IL-33 knockout aged mice that completed the running exercise in step E03 respectively by micro-computed tomography technology to obtain image data, analyzing the results of changes in articular cartilage and subchondral bone according to the image data, and integrating them into the summary data of step S31;

[0331] Step E05: Obtain chondrocytes from normal mice and chondrocytes from IL-33 knockout aged mice after the running exercise in step E03, culture and multiply them, perform pathological section detection on chondrocyte hypertrophy, perform immunohistochemical analysis on the expression of chondrocyte ossification markers, obtain ossification marker detection results, hypertrophy-related cell morphology analysis, and proliferation capacity, and integrate them into the summary data of step S31.

[0332] Under the control of the server, a fluid shear force device is used to apply a predetermined fluid shear force to the two cells to simulate the mechanical environment of the joint movement in the body;

[0333] The response of chondrocytes to fluid shear force is monitored and recorded in real time. The server observes the changes in cell morphology through a microscope to evaluate the hypertrophy of chondrocytes.

[0334] The treated cells were tested for ossification-related markers to obtain the test results of ossification-related markers of the cells, and the differences in hypertrophy and ossification between wild-type mice and IL-33 knockout cells were compared.

[0335] Under the guidance of the server, a running exercise plan was formulated for the mice, including exercise intensity, time and frequency, and the movement status of the mice was monitored in real time. After the exercise plan was completed, the mice's joint samples were collected, pathological sections of the joints were made, and immunohistochemical staining was performed to detect the histological changes of the articular cartilage and subchondral bone; micro-computed tomography technology was used to perform high-resolution imaging of the joint samples to obtain detailed structural information of the articular cartilage and subchondral bone;

[0336] The data of pathological sections, immunohistochemistry and tomography were integrated to analyze the changes in articular cartilage and subchondral bone of wild-type mice and IL-33 knockout mice after exercise, and the analysis results were obtained.

[0337] The following techniques were used to compare and analyze normal and IL-33 knockout chondrocytes:

[0338] Statistical analysis: Statistical analysis was performed on multiple indicators of chondrocytes in the two groups, such as morphological parameters, proliferation capacity, hypertrophy and ossification-related indicators, etc. Statistical methods such as t-test and analysis of variance were used to determine whether these differences were statistically significant.

[0339] Biological analysis: In-depth study of biological data such as gene expression levels and protein expression levels to reveal the specific role of IL-33 in the physiological process of chondrocytes.

[0340] Based on these comparative analysis results, the mechanism of action of IL-33 in chondrocyte degeneration, hypertrophy and ossification can be systematically deduced:

[0341] Comparison of morphology and proliferation ability: Compare the morphological characteristics and proliferation ability of normal chondrocytes and IL-33 knockout chondrocytes to preliminarily determine the effect of IL-33 on chondrocyte growth and division.

[0342] Hypertrophy and ossification marker detection: By quantitatively detecting hypertrophy markers (such as the expression levels of certain specific genes) and ossification-related indicators (such as alkaline phosphatase activity and osteocalcin content) in chondrocytes, the degree of cell hypertrophy and ossification can be accurately assessed.

[0343] Comprehensive analysis: Finally, a comprehensive and in-depth analysis was conducted by combining the above test results and biological data. This may involve multiple aspects such as how IL-33 regulates the signaling pathway of chondrocytes, affects the synthesis of cartilage matrix, and has a potential impact on chondrocyte apoptosis.

[0344] Through the above comprehensive research and analysis, we can have a more comprehensive understanding of the specific role of IL-33 in the occurrence and development of osteoarthritis, and provide a scientific basis for future treatment strategies.

[0345] Embodiment 9

[0346] The present invention uses big data, artificial intelligence and knowledge base technology to accurately infer the identification results of endochondral ossification in osteoarthritis in the elderly. The following are the detailed steps:

[0347] Big Data Processing:

[0348] The present invention first integrates various data from the experimental and testing process, such as gene expression levels, cell morphology analysis, and ossification marker detection results, and performs preprocessing to ensure the quality and consistency of the data. Through steps such as cleaning, filling missing values, standardization, and normalization, a comprehensive and high-quality data set is constructed, providing a solid foundation for subsequent artificial intelligence model training.

[0349] Application of artificial intelligence models:

[0350] Select an appropriate machine learning or deep learning model for training, use the training set to train the model, and adjust the model parameters through the validation set to ensure its performance is optimal.

[0351] The articular cartilage images of the elderly were preprocessed and substituted into the trained osteoarthritis endochondral ossification image analysis model to preliminarily determine the articular cartilage and subchondral ossification.

[0352] Extract key features from the model's preliminary judgment results, such as the size, shape, and distribution of the ossification area. The role of the knowledge base in etiology inference:

[0353] The extracted key features are matched with the data in the knowledge base, which contains rich knowledge and cases related to endochondral ossification in osteoarthritis.

[0354] Through this matching, combined with the extracted features and the information in the knowledge base, the etiology of endochondral ossification in osteoarthritis in the elderly can be accurately determined.

[0355] In general, the present invention achieves accurate inference of the cause of endochondral ossification in osteoarthritis in the elderly through the integration and processing of big data, the training and optimization of artificial intelligence models, and the auxiliary matching of knowledge bases. It not only improves the accuracy and efficiency of identification, but also provides important support for the early prevention and treatment of osteoarthritis.

[0356] In summary, the present invention successfully solves the problem of being unable to identify the phenotype of chondrocytes beginning to differentiate into hypertrophic phenotypes and the series of events induced that ultimately lead to the cause of cartilage degeneration through innovative detection technology and analysis methods. This provides strong support for the early prevention and treatment of osteoarthritis.

[0357] Through the above implementation scheme, the present invention aims to provide a comprehensive and accurate method for image recognition of endochondral ossification in osteoarthritis, providing strong support for the early prevention and treatment of osteoarthritis.

[0358] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations within the meaning and scope of the equivalent elements of the claims be included in the invention. Any marking in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A method for image recognition of endochondral ossification in osteoarthritis, characterized in that: The steps include: Step S1: using a magnetic resonance imaging device to obtain image data of articular cartilage; Step S2: preprocessing the image data obtained in step S1 to obtain preprocessed data; Step S3: inputting the preprocessed data obtained in step S2 into the osteoarthritis endochondral ossification image analysis model to obtain the analysis results of articular cartilage and subchondral ossification; Step S4: extract key features from the analysis results obtained in step S3 to obtain recognition results of endochondral ossification in osteoarthritis.

2. The method for image recognition of endochondral ossification in osteoarthritis according to claim 1, characterized in that: The preprocessing in step S2 includes the following steps: Step S21: removing noise from the image data; Step S22: enhancing the details of the articular cartilage by an algorithm; Step S23: Segment the articular cartilage region from the entire image to obtain pre-processed articular cartilage image data.

3. The method for image recognition of endochondral ossification in osteoarthritis according to claim 1, characterized in that: The construction of the osteoarthritis endochondral ossification imaging analysis model in step S3 includes the following steps: Step S31: obtaining summary data including chondrocytes, cartilage, articular cartilage and subchondral bone; The summary data of the chondrocytes include gene expression levels, cell morphology analysis, ossification marker detection results, and average gray value of cartilage in the medial tibial plateau weight-bearing area; the ossification marker detection results include alkaline phosphatase activity and osteocalcin content; Step S32: Clean the summary data obtained in step S31, remove outliers, fill missing values, and perform standardization and normalization processing, and finally integrate them into a complete data set; Step S33: Divide the data set obtained in step S32 into a training set, a validation set and a test set; Step S34: Select a model constructed by a machine learning or deep learning algorithm, train the model using a training set, adjust the model parameters using a validation set, and use a test set to evaluate the performance of the model, including accuracy and recall indicators. Optimize the model based on the evaluation results to ultimately obtain an osteoarthritis endochondral ossification imaging analysis model.

4. The method for image recognition of endochondral ossification in osteoarthritis according to claim 3, characterized in that: The chondrocytes in step S31 include one or a combination of normal mouse chondrocytes, aged mouse chondrocytes, senescent chondrocytes obtained by continuous in vitro passage of aged mouse chondrocytes, and IL-33 knocked-out normal mouse chondrocytes.

5. The method for image recognition of endochondral ossification in osteoarthritis according to claim 4, characterized in that: The acquisition of summary data of normal mouse chondrocytes in step S31 includes the following steps: Step A01: obtaining normal mouse chondrocytes; Step A02: Inducing hypertrophy of the normal mouse chondrocytes obtained in step A01, obtaining ossification marker detection results and hypertrophy-related cell morphology analysis, and integrating them into the summary data of step S31; Step A03: Use a magnetic resonance device to image the normal mouse chondrocytes obtained in step A01 to obtain image data, analyze the average grayscale value of the cartilage in the medial tibial plateau weight-bearing area based on the image data, and integrate it into the summary data of step S31.

6. The method for image recognition of endochondral ossification in osteoarthritis according to claim 4, characterized in that: The acquisition of summary data of cartilage cells of elderly mice in step S31 includes the following steps: Step B01: injecting rIL-33 into the joint cavity of aged mice, and obtaining chondrocytes of aged mice after the injection; Step B02: Perform pathological sectioning on the aged mouse chondrocytes obtained in step B02 to detect chondrocyte hypertrophy, and immunohistochemically analyze the expression of chondrocyte ossification markers to obtain ossification marker detection results and hypertrophy-related cell morphology analysis, and integrate them into the summary data of step S31; Step B03: Use a magnetic resonance device to image the cartilage cells of the aged mice obtained in step B02 to obtain image data, analyze the average gray value of the cartilage in the weight-bearing area of ​​the medial tibial plateau based on the image data, and integrate it into the summary data of step S31.

7. The method for image recognition of endochondral ossification in osteoarthritis according to claim 4, characterized in that: Step S31 of obtaining the summary data of senescent chondrocytes obtained by continuous in vitro passage of chondrocytes of middle-aged and elderly mice comprises the following steps: Step C01: Obtaining chondrocytes from aged mice; Step C02: the aged rat chondrocytes obtained in step C01 are subjected to continuous in vitro passage to obtain senescent chondrocytes; Step C03: Detecting the aged chondrocytes obtained in step C02 by reverse transcription-polymerase chain reaction to obtain the results of degeneration-related ossification marker detection, hypertrophy-related cell morphology analysis, and IL-33-related gene expression levels, and integrating them into the summary data of step S31; Step C04: stimulating the aged chondrocytes obtained in step C02 in vitro with IL-33 recombinant protein to obtain hypertrophy-related cell morphology analysis and ossification marker detection results, and integrating them into the summary data of step S31; Step C05, using a magnetic resonance device to image the aged chondrocytes obtained in step C02, obtaining image data, analyzing the average gray value of the cartilage in the medial tibial plateau weight-bearing area based on the image data, and integrating it into the summary data of step S31.

8. The method for image recognition of endochondral ossification in osteoarthritis according to claim 4, characterized in that: The acquisition of summary data of IL-33 knockout normal mouse chondrocytes in step S31 includes the following steps: Step D01: Obtain IL-33 knockout normal mouse chondrocytes; Step D02: Inducing hypertrophy of the IL-33 knockout normal mouse chondrocytes obtained in step D01, obtaining ossification marker detection results and hypertrophy-related cell morphology analysis, and integrating them into the summary data of step S31; Step D03: Use a magnetic resonance device to image the IL-33 knockout normal mouse chondrocytes obtained in step D01 to obtain image data, analyze the average grayscale value of the cartilage in the medial tibial plateau weight-bearing area based on the image data, and integrate it into the summary data of step S31.

9. The method for image recognition of endochondral ossification in osteoarthritis according to claim 3, characterized in that: The acquisition of the summary data of cartilage, articular cartilage and subchondral bone in step S31 includes the following steps: Step E01: obtaining chondrocytes from normal mice and IL-33 knockout aged mice respectively; Step E02: using a fluid shear force device to apply a predetermined fluid shear force to the chondrocytes obtained in step E01, simulating the mechanical environment of joint movement in vivo, and obtaining the degree of cell morphological change; Step E03: specifying a running exercise plan according to the degree of cell morphological changes obtained in step E02, and allowing normal mice and IL-33 knockout aged mice to complete the exercise plan; Step E04: imaging the normal mice and the IL-33 knockout aged mice that completed the running exercise in step E03 respectively by micro-computed tomography technology to obtain image data, analyzing the results of changes in articular cartilage and subchondral bone according to the image data, and integrating them into the summary data of step S31; Step E05: Obtain chondrocytes from normal mice and chondrocytes from IL-33 knockout aged mice after the running exercise in step E03, culture and multiply them, perform pathological section detection on chondrocyte hypertrophy, perform immunohistochemical analysis on the expression of chondrocyte ossification markers, obtain ossification marker detection results, hypertrophy-related cell morphology analysis, and proliferation capacity, and integrate them into the summary data of step S31.

10. The method for image recognition of endochondral ossification in osteoarthritis according to claim 1, characterized in that: In step S4, the key features include the size, shape, and distribution of the ossification area, and the extracted key features are matched in the knowledge base to determine the recognition result of osteoarthritis endochondral ossification.