43 results about "Clinical pathology" patented technology

The invention relates to a pathological section intelligent auxiliary differential diagnosis system based on multi-modal fusion, in particular to the field of clinical pathology, semantic unification of multi-modal data is achieved through meta-task construction and a cross-modal alignment technology, and transferable diagnostic knowledge is extracted by utilizing a meta-learning framework; the method combines a generative model and knowledge constraints to generate high-quality synthetic data, and finally fuses real and synthetic samples through a self-adaptive diagnosis mechanism, thereby remarkably improving the differential diagnosis capability of rare lesions, effectively solving the problem of model generalization in a training data scarcity scene, and improving the accuracy of model identification. And efficient and reliable intelligent auxiliary decision support is provided for clinical pathological diagnosis.

The invention belongs to the technical field of focus positioning, and particularly relates to a focus positioning method and system based on an infrared thermal imaging system. According to the method, temperature distribution, edge features, temperature gradient and time domain change information are integrated, and temperature and physiological compensation are combined, so that abnormal temperature information is more real and reliable, the accuracy of focus positioning is improved, the misdiagnosis rate is reduced, individualized data such as the body type of a patient, the blood perfusion rate and the environment temperature are introduced, and the accuracy of focus positioning is improved. The detection result can better adapt to individual differences, the pertinence and reliability of diagnosis are improved, abnormal features can be more accurately extracted by comparing an actual infrared image with a theoretical temperature distribution diagram, meanwhile, the credibility and clinical application value of focus positioning are improved in combination with a clinical pathology database, and the diagnosis accuracy is improved. The generated lesion probability thermodynamic diagram visually displays the core area and the diffusion boundary of the lesion, provides a visual reference basis, and is helpful for finding lesions earlier and formulating a reasonable treatment scheme.

The invention discloses an anti-soluble transferrin receptor monoclonal antibody as well as a preparation method and application thereof, and belongs to the technical field of antibody detection. According to the antibody, an L101-F760 amino acid sequence of human TfR protein is used as an immunogen, specific screening of B cells, reverse RNA (Ribonucleic Acid) into cDNA (Complementary Deoxyribose Nucleic Acid) through RT-PCR (Reverse Transcription-Polymerase Chain Reaction), then homologous recombination is performed to construct heavy and light chain vectors, and finally multiple rounds of screening are performed to obtain the purified sTfR monoclonal antibody, so that the sTfR protein can be specifically recognized, and the antibody has the advantages of high specificity and high sensitivity and can be widely applied to the field of immunotherapy. The natural sTfR protein in human serum can be recognized. The kit is applied to the detection and screening fields of immunohistochemistry, immunofluorescence chemistry, western blot, co-immunoprecipitation, indirect ELISA, flow cytometry and the like, especially has low background and no interference of non-specific binding impure protein during immunohistochemistry detection, can obtain more accurate detection and evaluation results, avoids interference of false positive and false negative, and has high detection accuracy. The method has great application value in the aspects of basic scientific research and clinical pathology detection.

The invention discloses a medical synthetic data analysis method and device based on causal reasoning and a medium. The method comprises the following steps: generating medical synthetic data through a generator created according to a real clinical data set; analyzing the real clinical data set and the medical synthetic data to obtain a first causal diagram and a second causal diagram, and comparing the two diagrams to identify an abnormal causal edge corresponding to the medical synthetic data; determining a path corresponding to the abnormal causal edge in the first causal graph, extracting risk factors and corresponding results in the path, and respectively calculating an average processing effect and an effect difference value of the risk factors on the results in the real clinical data set and the medical synthetic data; and determining an analysis result of the medical synthetic data according to the effect difference value. According to the method, the medical synthetic data quality can be deeply analyzed, the fidelity of the synthetic data to a real-world clinical pathology causal mechanism is verified, the specific link of causal distortion is accurately positioned, and clear and executable guidance is provided for the correction data generation process.

The invention relates to the field of image processing, in particular to a fundus disease image segmentation method and system based on multi-mode focus simulation. According to the method, for a simulation focus image of an OCT mode, image denoising, OCT layer information detection, layering abnormity judgment, RPE layer inclination angle determination, to-be-implanted OCT focus-free eye fundus image and focus implantation position determination, simulation focus size adjustment and simulation focus pixel value reconstruction are carried out in sequence, and an OCT multi-class synthetic eye fundus disease image and a segmentation label thereof are generated; and for the simulation focus image of the CFP mode, a CFP multi-class synthetic fundus disease image and a segmentation label thereof are generated by selecting a proper background image, adjusting the brightness of the background image, cutting off the edge of the simulation focus and reconstructing the pixel value of the simulation focus in sequence. Clinical pathology constraints are followed, fundus disease image data diversity is increased, meta-task enhancement is realized, and segmentation precision of a focus segmentation network is improved.

The invention provides a clinical application level pathology large model construction method and system based on weak supervised learning. The method comprises the steps of performing weak supervised training on each historical clinical pathology image, adjusting an upstream learning parameter of a large model according to a training result, constructing an upstream task of the large model, and performing segmentation decoding on each weak supervised training result, determining a segmentation result, adjusting downstream learning parameters of the large model, constructing a large model downstream task, obtaining upstream fine features and downstream fine features, constructing an upstream and downstream cooperative task of the large model, and generating a pathological large model in combination with the large model upstream task and the large model downstream task. And inputting the current clinical pathological image into the large pathological model for pathological information identification, obtaining a plurality of pathological identification tags corresponding to the patient, and generating a pathological report of the patient, thereby reducing the calculation complexity when processing a high-resolution image, optimizing a collaboration mechanism among different tasks in the large model, and improving the efficiency of the large model. The method can be widely applied to pathological recognition technologies.

A system and method for generating a healthcare report from a pathology test report are provided. In one form, the system includes at least one processing device configured to: extract pathology data from the pathology test report using a machine learning model; provide, on a data interface, and to a user, the pathology data for review by the user; receive, on the data interface, and from the user, corrected pathology data; provide at least part of the corrected pathology data to the machine learning model to further improve the machine learning model; retrieve auxiliary content according to the corrected pathology data; and generate the healthcare report from the corrected pathology data and the auxiliary content.

The application relates to a whole-body organ metabolism network analysis method, device, equipment and storage medium. The method comprises the following steps: collecting a PET image of a subject, obtaining SUV images, Ki images and TAC data according to the PET image, and collecting physiological information, clinical pathology information and daily habit information of the subject; metabolism characteristics are extracted based on the PET images, SUV images and Ki images corresponding to the region of interest; the TAC data of each organ is curve-fitted by using the PET image, and the metabolism disturbance under different organs is represented by using the fitting residual error; a whole-body organ metabolism network is constructed according to the physiological information, clinical pathology information, daily habit information, metabolism characteristics and metabolism disturbance; and the whole-body organ metabolism of the subject is analyzed through the whole-body organ metabolism network. The application avoids individual differences, can more comprehensively capture the metabolism correlation among whole-body organs of a specific disease, and is helpful for early diagnosis, accurate evaluation and individualized treatment of clinical diseases.

The present invention relates to the application of toxicogenomics methods to the detection and / or prediction of histopathological abnormalities in human or animal subjects based on clinical pathology data. It has been observed that reliable results can be obtained even in the absence of image data. A computer-implemented method for predicting the presence of a histopathological abnormality in an organ of a human or animal subject based on clinical pathology data comprises receiving clinical pathology data obtained from the human or animal subject, applying an analytical model to the clinical pathology data, the analytical model being configured to output a result indicative of the likelihood of the presence of the histopathological abnormality in the organ of the human or animal subject, and outputting the result.

The application relates to the technical field of medical data processing, and discloses a knee joint three-dimensional movement analysis platform for clinical auxiliary diagnosis, which comprises the following steps: a data processing module adopts Magic software to collect knee joint three-dimensional movement original data, carries out coordinate correction and model processing, and generates a model group. A model analysis module imports the model group into a C3D clinical window, realizes real-time visualization of a three-dimensional movement track, calculates biomechanical parameters, and evaluates clinical pathology correlation. A man-machine interaction control module provides customized interaction logic and a graphical interface, supports parameter input, analysis process control, real-time monitoring, and automatic generation of a diagnosis report. A cloud publishing module converts the diagnosis report and the interface into a PHP dynamic webpage, deploys the webpage through HTTP, supports SSL encryption, multi-user concurrent access, and cross-platform remote calling. The application supports real-time visualization of a knee joint three-dimensional movement track, and improves data reliability.

The invention discloses a curative effect prediction method based on multi-organ metastasis genome data, and belongs to the technical field of medical models, and the method specifically comprises the following steps: collecting clinical pathological characteristics, multi-organ metastasis genome data and a treatment scheme of a breast cancer patient, and recording a metastasis part and a load state; dimensionality reduction is conducted on high-dimensional genome data through a regularization algorithm, feature importance is evaluated in combination with a nonlinear model, and clinical, treatment and genome features related to treatment response are screened out; inputting the screened features into a machine learning and deep learning framework, randomly dividing a training set and a test set in a layered manner, optimizing hyper-parameters through cross validation, and constructing a classic machine learning set model and a deep learning model based on an attention mechanism; disturbing test queue treatment scheme data, evaluating the consistency of model recommendation and an actual scheme, and verifying the prediction capability and clinical practicability of the model; according to the method, multi-dimensional data are integrated, and the curative effect prediction accuracy of the metastatic breast cancer is improved.

The invention belongs to the technical field of intelligent drug delivery control, and particularly discloses a colon tumor drug delivery method and system based on AI control, and the method comprises the steps: collecting and preprocessing whole genome sequencing, radiomics, clinical pathology and historical drug treatment response data of a patient, extracting tumor image features, and carrying out drug delivery. The method comprises the following steps: screening oxaliplatin response related gene mutation markers by combining Lasso regression with a Cox model, forming feature vectors by t-SNE dimensionality reduction clinical pathological data, training a model through a gradient boosting tree and deep neural network fusion algorithm, taking oxaliplatin response probability as output, and recommending an FOLFOX scheme or an FULV scheme according to a model result. According to the method, precision and individuation of chemotherapy are achieved, unnecessary toxic and side effects are reduced, the life quality and lifetime of a patient are improved, and the problems that in the prior art, dependence on experience, single data dimension, no dynamic adjustment mechanism and poor stability are solved. Therefore, the problems of low response rate of oxaliplatin and obvious side effect are solved.

The invention provides a method, a device, equipment and a medium for predicting the expression level of renal cancer HIF-2alpha. The method comprises the following steps: S10, acquiring clinical pathological data and an abdominal enhanced arterial phase image of a patient with renal clear cell carcinoma; s20, sketching a three-dimensional region of interest of the tumor through the abdominal enhanced arterial phase image; s30, extracting image radiomics characteristics from the tumor three-dimensional region of interest, screening the image radiomics characteristics through Z-score standardization processing, stability test, an mRMR algorithm and an LASSO regression algorithm in sequence to obtain screened characteristics, and calculating to obtain a radiomics score through the screened characteristics; s40, screening out independent clinical predictive factors related to HIF-2alpha expression from the clinical pathology data, wherein the independent clinical predictive factors comprise WHO / ISUP classification and clinical T classification; and inputting the independent clinical prediction factor and the radiomics score into a joint prediction model to obtain a prediction score, and evaluating the probability of HIF-2alpha expression by using the prediction score.

The invention discloses an endometrial cancer typing system based on a multi-dimensional molecular marker, and the system comprises a data input module which is used for receiving multi-dimensional molecular marker state information which is inputted by a user and is related to an endometrial cancer sample, and the information contains genomics, apparent and expressive data and clinical pathology data; the data processing module is configured to sequentially execute multi-level typing logic based on the state information, and the logic comprises the following steps: typing based on POLE gene mutation state preferentially, then typing based on mismatch repair function and p53 function state in sequence aiming at an unshaped sample, and finally carrying out joint layering based on prognosis association indexes; and the typing result output module is used for outputting the final molecular subtype. Through multi-dimensional data cross validation and dynamic layering, the problems that an existing typing technology is single in dimension and poor in clinical applicability are solved, accurate and efficient typing of endometrial cancer is achieved, and reliable technical support is provided for individualized diagnosis and treatment.

The present invention relates to a method for determining a treatment for a cancer patient. The method includes measuring a variety of clinical pathological marker concentrations for gradient enhanced machine learning analysis.

The invention provides a system for predicting the applicability of triple negative breast cancer neoadjuvant immunotherapy. Specifically, the system comprises at least one detection unit comprising determining a multi-modal feature of a subject to be detected, the multi-modal feature comprising, or consisting of, a clinical pathological feature, a copy number variation feature, a plasma protein feature, and a somatic mutation feature; the system further comprises at least one data analysis unit and at least one judgment unit. The system is used for predicting new auxiliary immunotherapy of the triple negative breast cancer, and has excellent sensitivity and specificity.

The invention belongs to the field of surgical instruments, and provides a surgical minimally invasive surgery endoscope artificial intelligence system which comprises a control box, a mechanical arm, a visual sensor, an inner needle assembly and an air bag. The air bag is used for being placed in an operation opening, a minimally invasive hole or a natural hole of a human body in a closed state and expanding by transmitting air pressure to support and open organs or tissues; the visual sensor comprises a CMOS sensor, a light path assembly, a light filter, a light field and a prism assembly, the visual sensor is combined with the endoscope or the endoscope, and the visual sensor is driven by the mechanical arm to enter a human body along a surgical opening, a minimally invasive hole or a natural hole of the human body so as to capture images of organ tissues in the body; the control box comprises a feature processing unit, a mechanical arm driving control unit, an inner needle control unit and an air bag control unit, the feature processing unit is used for segmenting, recognizing and classifying captured images, the mechanical arm driving control unit is used for driving the mechanical arm to move, and the inner needle control unit is used for controlling the position, angle and depth of an inner needle assembly or a catheter; the airbag control unit is used for controlling air pressure of airbags; according to the method, the time of characteristic analysis of clinical pathology can be shortened, and the reliability is improved.

The invention discloses application of solamargine in preparation of a medicine for preventing and / or treating cholestatic liver diseases, and belongs to the field of biological medicines. By simulating a classical Mdr2- / -mouse model of clinical cholestasis pathology, the solamargine is researched and proved to treat cholestatic liver injury by improving hepatocyte injury, bile duct dysfunction and bile metabolism disorder, and the solamargine shows good application potential in clinical treatment of cholestatic liver diseases.

The utility model relates to the technical field of clinical pathological diagnosis, in particular to a pathological specimen cutting knife for clinical pathological diagnosis, which comprises a knife body arranged in a cutting groove on a cutting box, a baffle plate and two push plates, the knife body comprises a knife back and a knife edge, the baffle plate is fixed on one side of the knife back, and the two push plates are respectively fixed on two sides of the bottom of the knife back. Two positioning blocks are fixed at one end in the cutting groove, and residues after specimen cutting are prevented from falling into the cutting groove through a baffle; the pushing device is located on one side of the cutting groove, and the pushing device abuts against the knife back; after a specimen is cut by the blade, specimen residues slide to the outer side of the cutting groove through the baffle, and when the cutter body is replaced through the pushing device, the push plate at the bottom of the cutter back pushes out the residues in the cutting groove, so that the situation that insertion of subsequent bodies is affected due to the fact that the specimen residues of the cutter fall into the cutting groove is prevented.

The invention relates to the technical field of medical image processing. The invention discloses a habitat imaging-based gastric cancer neoadjuvant chemotherapy curative effect prediction method and a habitat imaging-based gastric cancer neoadjuvant chemotherapy curative effect prediction system, and the method comprises the following steps: S1, obtaining pre-treatment CT and clinical pathology information, and marking the maximum tumor area and upper and lower layers; s2, dividing the marked region into habitat sub-regions with different biological characteristics through habitat imaging; and S3, performing prognosis analysis on the habitat subregion, screening independent prognosis features related to survival, and generating a habitat subregion image. And S4, constructing a joint attention model, inputting images of the marked area and the habitat subarea, predicting the curative effect and the total lifetime, and obtaining a survival score. And S5, integrating clinical pathological information and survival scores, performing prognosis analysis, constructing a column graph, and evaluating correlation. After a CT image label before treatment is obtained, habitat imaging is used for dividing subareas, related images are generated, the model is input in a combined mode, tumor heterogeneity is revealed, and the curative effect is predicted.

The invention provides biomarkers, methods involving their use, and a diagnostic kit that enables the distinction between muscle-invasive bladder cancer (MIBC) and non-muscle- invasive bladder cancer (NMIBC) which represents a critical step in the diagnosis and treatment of bladder cancer. This kit, based on widely used and cost-effective immunohistochemistry (IHC) techniques in pathology practice, comprises an effective biomarker panel for distinguishing between MIBC and NMIBC. This invention is suitable for routine use in clinical pathology practice, enabling better guidance of patient treatment processes and more accurate determination of disease progression.

The utility model relates to an automatic embedding device of a tissue dehydrator, which is suitable for the fields of medical clinical pathology, cytology examination and biological research. The equipment integrates tissue dehydration and embedding functions, mainly comprises a wax dipping cylinder assembly, an embedding wax cylinder assembly, a transmission mechanism assembly and a fixed supporting structure, and can fully automatically complete dehydration and embedding treatment of a biological tissue sample. Through the driving of the motor, the embedding wax cylinder cover can rotate by 90 degrees to be automatically opened and closed, so that the automatic in-out operation of the specimen basket frame is realized, the manual intervention is reduced, and the working efficiency is improved. Besides, the reagent volatilization is effectively reduced, the environmental pollution and the operation risk are reduced, and meanwhile, the problem of confusion of sample marks is avoided, so that the safety and the accuracy of operation are improved. According to the invention, the operation process is simplified, the working efficiency of a pathology laboratory is improved, and the method is suitable for wide biomedical application scenarios.

The invention relates to the technical field of medical imaging omics, in particular to a pancreatic cancer early recurrence prediction method and system based on an interpretable machine model, and the method comprises the following steps: extracting tumor internal and external imaging omics features from a CT image; performing single-factor analysis and multivariable logistic regression analysis in the body composition parameters and the clinical pathology data to obtain clinical features; six groups of classifier models are constructed based on tumor internal and external radiomics characteristics through six machine learning algorithms, and a radiomics model is obtained according to model performance comparison; and constructing a clinical-radiomics combined model by using the clinical features and the radiomics model, and carrying out interpretable SHAP analysis on the clinical-radiomics combined model. According to the method, intratumoral-peritumoral CT image omics characteristics and body composition parameters are combined, a machine learning model for predicting the early recurrence risk after PDAC excision is constructed, interpretable SHAP analysis is incorporated, and the transparency of the machine learning model decision making process is enhanced.

The invention discloses a diabetic nephropathy prognosis and diagnosis system based on BOLD-MRI and multi-dimensional parameters, and the system comprises an acquisition module which is used for acquiring clinical pathological indexes of a patient and acquiring medullary R2 * parameters through a BOLD-MRI semi-automatic 12-layer concentric object method; the training module is used for training an ESRD prognosis prediction model and a DN / NDRD differential diagnosis model based on clinical pathology indexes and medullary R2 * parameters; the weighting module is used for performing weighting calculation on the clinical pathology indexes and the medullary R2 * parameters based on the trained model, and determining the weight coefficient of each parameter; and the diagnosis prediction module comprises a prognosis prediction sub-module and a differential diagnosis sub-module, and is used for analyzing to-be-predicted clinical pathological indexes and medullary R2 * parameters based on the prediction diagnosis model to realize ESRD risk prediction and DN / NDRD differential diagnosis. According to the invention, clinical and BOLD parameters can be effectively integrated, so that the prediction model can more accurately identify diabetic nephropathy and carry out prognosis evaluation, thereby providing earlier intervention opportunities, enabling a diagnosis window to move forward, improving diagnosis performance and reducing unnecessary renal biopsy.

The invention provides an artificial-intelligence–based platform for automated analysis of Pap smear slides. The system integrates preprocessing, dual deep learning models (CNN-Transformer and EfficientNet-LSTM), and reporting modules to classify cytological images into diagnostic categories including NILM, ASC-US, LSIL, HSIL, and SCC. The platform includes interpretability outputs, secure data management, and active learning capabilities, enabling scalable, accurate, and transparent cervical-cancer screening across diverse laboratory settings. The platform is suitable for implementation in clinical pathology laboratories, telemedicine networks, and population-level screening programs. It allows rapid, reproducible, and objective evaluation of cytology slides, improving throughput while reducing reliance on manual slide examination. The invention is deployable on local servers, private clouds, or public cloud infrastructures and is adaptable to various cytology specimens. Adoption of this system enhances early detection of precancerous lesions, reduces inter-observer variability, and facilitates integration with laboratory information systems. It also supports continuous learning and adaptation to evolving imaging protocols. The invention provides multiple advantages over existing methods: (1) Dual-model architecture combining CNN-Transformer and EfficientNet- LSTM ensures robust classification across both dense and sparse cellular regions. (2)Preprocessing and adaptive tiling improve feature extraction from variable-quality images. (3)Interpretability modules enhance clinician trust and facilitate regulatory compliance. (4) Active learning allows continuous improvement based on expert feedback. (5) Modular deployment supports both local laboratory and cloud-based operation. (6) Data security features meet international standards for privacy and encryption. (7) High scalability enables screening of large slide volumes without compromising accuracy. The invention provides a clinically relevant, efficient, and technologically advanced solution for automated cervical-cancer screening, applicable in diverse healthcare environments and adaptable to future cytology domains.

The invention provides a detection kit for diagnosing colorectal cancer and application thereof, blood samples of 100 colorectal cancer patients diagnosed by clinical pathology and / or cytology are compared with blood samples of 100 healthy control people, a subject work mapping curve is described, and the curve and the area below the curve are calculated. The kit provided by the invention has the sensitivity of 100%, the specificity of 83.0%, the accuracy of 91.5% and the AUC of 0.911 when being used for detecting blood samples of colorectal cancer patients and healthy control people, so that the detection kit can be used for distinguishing the colorectal cancer patients and the healthy control people and has good accuracy and sensitivity, and therefore, the kit can be used for detecting the colorectal cancer patients and the healthy control people. The invention provides a kit which is based on blood metabolites and can be used for early colorectal cancer detection.

According to the detection kit for diagnosing the non-small cell lung cancer and the application, blood samples of 100 non-small cell lung cancer patients diagnosed by clinical pathology and / or cytology are compared with blood samples of 100 healthy control people, a subject working correspondence curve is depicted, the curve and the area below the curve are calculated, and the non-small cell lung cancer is diagnosed. The kit provided by the invention has the sensitivity of 99.0%, the specificity of 86.0%, the accuracy of 92.5% and the AUC of 0.940 when being used for detecting blood samples of non-small cell lung cancer patients and healthy control people, so that the detection kit can be used for distinguishing the non-small cell lung cancer patients and the healthy control people and has good accuracy and sensitivity, and therefore, the kit can be used for detecting blood samples of the non-small cell lung cancer patients and the healthy control people. The invention provides a kit which is based on blood metabolites and can be used for early non-small cell lung cancer detection.

The application belongs to the technical field of medical image processing and artificial intelligence, and discloses a colorectal cancer liver metastasis risk assessment method and system based on multi-modal fusion. The method comprises the following steps: obtaining preoperative abdominal computed tomography (CT) images and clinical pathological data of a patient; pre-processing the CT images, segmenting a three-dimensional region of interest, and extracting radiomics features, and standardizing the clinical data; inputting the radiomics features into a first prediction model to obtain an imaging score, and inputting the clinical data into a second prediction model to obtain a clinical score; based on a preset fusion strategy, the two scores are weighted and fused, and a comprehensive risk score of postoperative liver metastasis of the patient is output as auxiliary assessment information. The application can accurately assist in assessing the risk of colorectal cancer liver metastasis after surgery by non-invasively and efficiently integrating CT image quantitative features and clinical information, and provides an objective decision-making tool for clinicians to develop individualized diagnosis and treatment plans and follow-up strategies.

The invention discloses a liver pathology image microenvironment segmentation method and system, and the method comprises the steps: constructing and training a liver pathology image microenvironment segmentation model, and obtaining a trained liver pathology image microenvironment segmentation model; a to-be-recognized liver pathology image is input into the trained liver pathology image microenvironment segmentation model, a liver tissue microenvironment segmentation result is obtained, and the liver tissue microenvironment segmentation result is a liver pathology image obtained by conducting pixel-level labeling on multiple microenvironment elements; the microenvironment elements comprise a normal liver dyeing area, a light dyeing water change area, a normal cell nucleus area, a heterochromatic lesion area, lipid droplets, a lumen blank area and a deep dyeing cell nucleus area; the invention provides an accurate and comprehensive pathological microenvironment segmentation method for pathological diagnosis, provides a reliable automatic tool for quantitative analysis of clinical pathology, and can assist in comprehensively mastering distribution characteristics and evolution rules of pathological characteristics of liver tissues clinically.

The application provides a clinical application level pathological large model construction method and system based on weak supervision learning, comprising: respectively performing weak supervision training on each historical clinical pathology image, adjusting upstream learning parameters of a large model according to a training result and constructing an upstream task of the large model, respectively performing segmentation decoding on each weak supervision training result, determining downstream learning parameters of the large model according to a segmentation result and constructing a downstream task of the large model, acquiring upstream fine features and downstream fine features, constructing an upstream and downstream collaborative task of the large model, combining the upstream task of the large model and the downstream task of the large model to generate a pathological large model, inputting a current clinical pathology image into the pathological large model to perform pathological information recognition, obtaining a plurality of pathological recognition labels corresponding to a patient, and generating a pathological report of the patient, so that the computational complexity in processing a high-resolution image is reduced, the collaborative mechanism between different tasks in the large model is optimized, and the pathological recognition technology can be widely applied.