Pathological diagnosis method and system based on intelligent platform

Abstract

The invention discloses a pathological diagnosis method and system based on an intelligent platform, and the method comprises the steps: obtaining slice information, carrying out the AI pathological prediction through the intelligent platform according to the slice information, and obtaining a prediction result; generating a diagnosis request according to the prediction result, and obtaining a comprehensive diagnosis result of a pathological doctor in the pathological information system based on the diagnosis request; generating a diagnosis report according to the comprehensive diagnosis result, and performing intelligent auditing on the diagnosis report; and after the examination and verification are correct, synchronizing the diagnosis report to a pathological information system for recording and storing. Information intercommunication between the two systems can be realized, so that pathologists can diagnose on an intelligent platform, diagnosis timeliness and accuracy are improved, effective reference data are provided for diagnosis of the pathologists, practicability, reliability and diagnosis efficiency are improved, and data circulation efficiency is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of pathological diagnosis, and in particular to a pathological diagnosis method and system based on an intelligent platform. Background Art

[0002] At present, in the wave of digital medicine, pathology, as the core part of medical diagnosis, is undergoing unprecedented changes. Traditional pathological diagnosis relies on pathologists to observe and analyze pathological sections under a microscope. Although this method has a long history, it has many limitations, such as strong subjectivity, low efficiency, susceptibility to fatigue and experience, etc. With the rapid development of computer technology and artificial intelligence, digital pathology came into being. It digitizes pathological sections so that pathological images can be processed and analyzed by computer systems, thereby greatly improving the accuracy and efficiency of pathological diagnosis. The development of digital pathology has not only changed the way of pathological diagnosis, but also brought new opportunities for pathological research. Through digitized pathological sections, a large number of pathological images can be quickly browsed, analyzed and stored, which is of great significance to the research and education of pathology.

[0003] The development of digital pathology, especially its combination with artificial intelligence technology, has provided new possibilities for the automation and intelligence of pathological diagnosis. The powerful capabilities of AI technology in image recognition, pattern analysis and data processing have made the automated analysis of pathological images a reality. This technological advancement has not only improved the efficiency of pathological diagnosis, but more importantly, it is expected to reduce the subjectivity of diagnosis and improve the accuracy and repeatability of diagnosis. However, despite the broad application prospects of digital pathology and AI technology in pathological diagnosis, current research and practice have shown that the existing digital pathology slide cancer area prediction system still faces many challenges in terms of automation process, data processing efficiency, and inter-system compatibility. These challenges not only limit the further development of technology, but also affect the quality and efficiency of pathological diagnosis. In the traditional pathological diagnosis process, the workflow of the pathological information system and the intelligent pathology system is separated. The existing system often requires pathologists to manually upload slide images, manually obtain clinical information, and switch between multiple systems to complete the diagnosis process. These cumbersome manual operations are not only time-consuming and labor-intensive, but also prone to errors, affecting the timeliness and accuracy of diagnosis. At the same time, the two systems do not generate direct information flow, and cannot effectively integrate the capabilities of AI into the overall diagnosis process. Reduced practicality. Summary of the invention

[0004] In response to the above-mentioned problems, the present invention provides a pathological diagnosis method and system based on an intelligent platform to solve the problem mentioned in the background technology that in the traditional pathological diagnosis process, the workflow of the pathological information system and the intelligent pathology system are separated. The existing system often requires pathologists to manually upload slice images, manually obtain clinical information, and switch between multiple systems to complete the diagnosis process. These cumbersome manual operations are not only time-consuming and labor-intensive, but also prone to errors, affecting the timeliness and accuracy of the diagnosis. At the same time, the two systems do not generate direct information flow, and the capabilities of AI cannot be effectively integrated into the overall diagnosis process. The problem of reduced practicality.

[0005] A pathological diagnosis method based on an intelligent platform comprises the following steps:

[0006] Obtain slice information, perform AI pathology prediction through the intelligent platform based on the slice information, and obtain prediction results;

[0007] Generate a diagnosis request according to the prediction result, and obtain a comprehensive diagnosis result of a pathologist in a pathology information system based on the diagnosis request;

[0008] Generate a diagnostic report based on the comprehensive diagnostic results and conduct intelligent review of the diagnostic report;

[0009] After verification, the diagnosis report will be synchronized to the pathology information system for recording and storage.

[0010] Preferably, before obtaining the slice information, performing AI pathology prediction through an intelligent platform according to the slice information, and obtaining the prediction result, the method further includes:

[0011] Obtain the user's login account and password on the intelligent platform and verify them, and determine whether the user has the right to use according to the verification results;

[0012] If so, generate a service token and bind it to the user's login account;

[0013] Configure operations on the user's login account based on preset access control policies.

[0014] Preferably, the step of obtaining the slice information, performing AI pathology prediction through an intelligent platform according to the slice information, and obtaining the prediction result includes:

[0015] Acquire slice increment files, scan the slice increment files with a high-resolution scanner, and acquire slice digital images;

[0016] The digital images of the slices are encrypted and uploaded to the intelligent platform through the HTTPS protocol;

[0017] The digital image of the slice is recognized through the optical character recognition technology of the intelligent platform, and the slice number and pathology number are obtained according to the recognition results;

[0018] Relevant clinical information of the slice digital image is retrieved, and the relevant clinical information, slice number and pathology number are integrated to generate slice information.

[0019] The artificial intelligence algorithm of the intelligent platform is used to predict cancer areas and disease types based on deep learning technology to obtain prediction probabilities;

[0020] The disease types and their predicted probabilities are displayed in the form of a heat map.

[0021] Preferably, generating a diagnosis request according to the prediction result, and obtaining a comprehensive diagnosis result of a pathologist in the pathology information system based on the diagnosis request includes:

[0022] Obtaining a list of online pathologists provided by pathology information and selecting a target pathologist, and generating a diagnosis request for the target pathologist;

[0023] Obtain the login token, verify the permissions of the login token, and issue a request to call the diagnostic interface after the verification passes;

[0024] Upload the prediction results and slice digital images to the diagnosis interface for the target pathologist to make an online diagnosis;

[0025] The diagnostic data is received, and a comprehensive diagnostic result of a target pathologist is obtained based on the diagnostic data.

[0026] Preferably, generating a diagnostic report based on the comprehensive diagnostic results and conducting intelligent review of the diagnostic report include:

[0027] Obtain the online diagnostic editing content of pathologists on the intelligent platform to generate slice-level reports and pathology-level reports;

[0028] Integrate the slide-level report and the pathology-level report to generate a diagnosis report, and determine the report type of the diagnosis report;

[0029] Select audit indicators and audit methods according to the report type, conduct a formal audit on the diagnostic report according to the audit indicators and audit methods, and obtain the first audit result;

[0030] Retrieving a preset pathological diagnosis model, and using the preset pathological diagnosis model to review key information of the diagnosis result to obtain a second review result;

[0031] Whether the diagnosis report is qualified is determined based on the first audit result and the second audit result, and a warning reminder is issued if it is unqualified.

[0032] Preferably, after the review is correct, the diagnosis report is synchronized to the pathology information system for recording and storage, including:

[0033] Determine the forms and fields used to record and store diagnostic reports in the pathology information system, and determine the storage data type of the diagnostic report based on the forms and fields;

[0034] Write the API interface of pathology information provision and intelligent platform according to the storage data type of diagnostic reports in the pathology information system;

[0035] Select an appropriate file upload tool to upload the diagnostic report to the pathology information system through the API interface;

[0036] The uploaded diagnostic report is processed by data escape and decryption, and the processed data is obtained and converted into a data format suitable for pathological information import for recording and storage.

[0037] A pathological diagnosis system based on an intelligent platform, the system comprising:

[0038] The prediction module is used to obtain slice information, perform AI pathology prediction based on the slice information through the intelligent platform, and obtain the prediction results;

[0039] An acquisition module, used to generate a diagnosis request according to the prediction result, and acquire a comprehensive diagnosis result of a pathologist in the pathology information system based on the diagnosis request;

[0040] A generation module is used to generate a diagnosis report based on the comprehensive diagnosis results and to conduct intelligent review of the diagnosis report;

[0041] The synchronization module is used to synchronize the diagnosis report to the pathology information system for recording and storage after verification.

[0042] Preferably, the prediction module comprises:

[0043] A scanning submodule is used to obtain a slice increment file, scan the slice increment file using a high-resolution scanner, and obtain a slice digital image;

[0044] The first uploading submodule is used to encrypt and upload the slice digital image to the intelligent platform through the HTTPS protocol;

[0045] The recognition submodule is used to recognize the slice digital image through the optical character recognition technology of the intelligent platform, and obtain the slice number and pathology number according to the recognition result;

[0046] The integration submodule is used to retrieve the relevant clinical information of the slice digital image, and integrate the relevant clinical information, slice number and pathology number to generate slice information.

[0047] The prediction submodule is used to predict cancer areas and disease types based on deep learning technology through the artificial intelligence algorithm of the intelligent platform to obtain prediction probabilities;

[0048] The display submodule is used to display the disease types and their predicted probabilities in the form of a heat map.

[0049] Preferably, the acquisition module includes:

[0050] A first acquisition submodule is used to acquire a list of online pathologists provided by pathology information and select a target pathologist, and generate a diagnosis request for the target pathologist;

[0051] The request submodule is used to obtain the login token, verify the permissions of the login token, and issue a request to call the diagnostic interface after the verification is passed;

[0052] The diagnosis submodule is used to upload the prediction results and the slice digital images to the diagnosis interface for the target pathologist to make an online diagnosis;

[0053] A receiving submodule, used to receive diagnostic data and obtain a comprehensive diagnostic result of a target pathologist based on the diagnostic data;

[0054] The generating module comprises:

[0055] The second acquisition submodule is used to obtain the online diagnosis editing content of the pathologist on the intelligent platform to generate a slice-level report and a pathology-level report;

[0056] A generation submodule, for integrating the slice-level report and the pathology-level report to generate a diagnosis report, and determining a report type of the diagnosis report;

[0057] The first audit submodule is used to select audit indicators and audit methods according to the report type, conduct a formal audit on the diagnostic report according to the audit indicators and audit methods, and obtain a first audit result;

[0058] The second review submodule is used to retrieve a preset pathological diagnosis model, use the preset pathological diagnosis model to review the key information of the diagnosis result, and obtain a second review result;

[0059] The first determination submodule is used to determine whether the diagnosis report is qualified according to the first audit result and the second audit result and to issue a warning reminder if it is unqualified.

[0060] Preferably, the synchronization module includes:

[0061] The second determination submodule is used to determine the form and field used to record and store the diagnosis report in the pathology information system, and determine the storage data type of the diagnosis report according to the form and field;

[0062] Writing submodules for writing API interfaces for pathology information provision and intelligent platforms according to the storage data types of diagnostic reports in the pathology information system;

[0063] The second upload submodule is used to select a suitable file upload tool to upload the diagnosis report to the pathology information system through the API interface;

[0064] The conversion submodule is used to perform data escape and decryption processing on the uploaded diagnostic report, obtain the processed data, and convert the processed data into a data format suitable for pathological information import for recording and storage.

[0065] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0066] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0068] Figure 1 A workflow diagram of a pathological diagnosis method based on an intelligent platform provided by the present invention;

[0069] Figure 2 Another workflow diagram of a pathological diagnosis method based on an intelligent platform provided by the present invention;

[0070] Figure 3 A schematic diagram of the structure of a pathological diagnosis system based on an intelligent platform provided by the present invention;

[0071] Figure 4 A schematic diagram of the structure of a synchronization module in a pathology diagnosis system based on an intelligent platform provided by the present invention;

[0072] Figure 5 The workflow diagrams for the pathology information system and the intelligent pathology system respectively;

[0073] Figure 6 The present invention provides a workflow diagram of an intelligent pathology diagnosis platform created by a pathology diagnosis method based on an intelligent platform. DETAILED DESCRIPTION

[0074] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0075] At present, in the wave of digital medicine, pathology, as the core part of medical diagnosis, is undergoing unprecedented changes. Traditional pathological diagnosis relies on pathologists to observe and analyze pathological sections under a microscope. Although this method has a long history, it has many limitations, such as strong subjectivity, low efficiency, susceptibility to fatigue and experience, etc. With the rapid development of computer technology and artificial intelligence, digital pathology came into being. It digitizes pathological sections so that pathological images can be processed and analyzed by computer systems, thereby greatly improving the accuracy and efficiency of pathological diagnosis. The development of digital pathology has not only changed the way of pathological diagnosis, but also brought new opportunities for pathological research. Through digitized pathological sections, a large number of pathological images can be quickly browsed, analyzed and stored, which is of great significance to the research and education of pathology.

[0076] The development of digital pathology, especially its combination with artificial intelligence technology, has provided new possibilities for the automation and intelligence of pathological diagnosis. The powerful capabilities of AI technology in image recognition, pattern analysis and data processing have made the automated analysis of pathological images a reality. This technological advancement has not only improved the efficiency of pathological diagnosis, but more importantly, it is expected to reduce the subjectivity of diagnosis and improve the accuracy and repeatability of diagnosis. However, despite the broad application prospects of digital pathology and AI technology in pathological diagnosis, current research and practice have shown that the existing digital pathology slide cancer area prediction system still faces many challenges in terms of automation process, data processing efficiency, and inter-system compatibility. These challenges not only limit the further development of technology, but also affect the quality and efficiency of pathological diagnosis. In the traditional pathological diagnosis process, the workflow of the pathological information system and the intelligent pathology system is separated. The existing system often requires pathologists to manually upload slide images, manually obtain clinical information, and switch between multiple systems to complete the diagnosis process. These cumbersome manual operations are not only time-consuming and labor-intensive, but also prone to errors, affecting the timeliness and accuracy of diagnosis. At the same time, the two systems do not generate direct information flow, and cannot effectively integrate the capabilities of AI into the overall diagnosis process. Reduced practicality. In order to solve the above problems, this embodiment discloses a pathological diagnosis method based on an intelligent platform.

[0077] A pathological diagnosis method based on an intelligent platform, such as Figure 1As shown, the following steps are included:

[0078] Step S101, obtaining slice information, and performing AI pathology prediction through an intelligent platform according to the slice information to obtain prediction results;

[0079] Step S102: Generate a diagnosis request according to the prediction result, and obtain a comprehensive diagnosis result of a pathologist in the pathology information system based on the diagnosis request;

[0080] Step S103: Generate a diagnosis report based on the comprehensive diagnosis result, and perform intelligent review on the diagnosis report;

[0081] Step S104: After verification, the diagnosis report is synchronized to the pathology information system for recording and storage.

[0082] In this embodiment, the slice information comes from the intelligent pathology system. By interacting and integrating the information of the two systems, cross-system diagnosis can be achieved, which improves the information flow of data between systems and improves diagnostic efficiency.

[0083] The working principle of the above technical solution is: obtain slice information, perform AI pathology prediction through an intelligent platform based on the slice information, and obtain the prediction results; generate a diagnosis request based on the prediction results, and obtain the comprehensive diagnosis results of the pathologist in the pathology information system based on the diagnosis request; generate a diagnosis report based on the comprehensive diagnosis results, and perform intelligent review of the diagnosis report; after the review is correct, synchronize the diagnosis report to the pathology information system for recording and storage.

[0084] The beneficial effects of the above technical solution are: information exchange between the two systems can be achieved through comprehensive diagnosis of the AI ​​prediction results of the intelligent platform by the pathologist in the pathology information system, so that the pathologist can make a diagnosis on the intelligent platform, which improves the timeliness and accuracy of the diagnosis, provides effective reference data for the diagnosis of the pathologist, improves the practicality, reliability and diagnostic efficiency, and further, by synchronously storing and recording the diagnostic report, information exchange between the two systems can be further achieved, ensuring the efficiency of data circulation, further improving the practicality, and solving the problem mentioned in the prior art that the two systems do not generate direct information flow and cannot effectively integrate the capabilities of AI into the overall diagnostic process. The problem of reduced practicality.

[0085] In this embodiment, before obtaining slice information, performing AI pathology prediction through an intelligent platform based on the slice information, and obtaining the prediction result, the following is also included:

[0086] Obtain image information of the lesion tissue, and extract the imaging features of the lesion tissue based on the image information;

[0087] Perform image scoring on the image information according to the radiomics features, and determine the expression level index of the image information according to the scoring results;

[0088] The eligibility of the slice digital image is judged according to the expression level index of the image information;

[0089] If the slice digital image is unqualified, an image unqualified reminder will be issued.

[0090] The beneficial effects of the above technical solution are: by judging the eligibility of the slice digital image, the data quality of the slice information can be guaranteed, which lays the foundation for subsequent pathology prediction and improves practicality and reliability.

[0091] In one embodiment, before obtaining slice information, performing AI pathology prediction through an intelligent platform based on the slice information, and obtaining a prediction result, the method further includes:

[0092] Obtain the user's login account and password on the intelligent platform and verify them, and determine whether the user has the right to use according to the verification results;

[0093] If so, generate a service token and bind it to the user's login account;

[0094] Configure operations on the user's login account based on preset access control policies.

[0095] The beneficial effects of the above technical solution are: ensuring that users do not need to log in again after the token expires, improving the account usage experience, and further, configuring the user's login account through preset access control policies can ensure the rationality of the user's operation.

[0096] In one embodiment, the step of obtaining slice information, performing AI pathology prediction through an intelligent platform according to the slice information, and obtaining prediction results includes:

[0097] Acquire slice increment files, scan the slice increment files with a high-resolution scanner, and acquire slice digital images;

[0098] The digital images of the slices are encrypted and uploaded to the intelligent platform through the HTTPS protocol;

[0099] The digital image of the slice is recognized through the optical character recognition technology of the intelligent platform, and the slice number and pathology number are obtained according to the recognition results;

[0100] Relevant clinical information of the slice digital image is retrieved, and the relevant clinical information, slice number and pathology number are integrated to generate slice information.

[0101] The artificial intelligence algorithm of the intelligent platform is used to predict cancer areas and disease types based on deep learning technology to obtain prediction probabilities;

[0102] The disease types and their predicted probabilities are displayed in the form of a heat map.

[0103] The beneficial effects of the above technical solution are as follows: by acquiring digital slice images and uploading them to the intelligent platform through the HTTPS protocol for encryption, no human intervention is required, thus ensuring the timeliness and integrity of the data; the slice digital images are identified through optical character recognition technology to obtain the slice number and pathology number, which are integrated with the relevant clinical information of the slice digital images to generate slice information, which helps doctors better understand and analyze the patient's condition, thereby making more accurate diagnosis and treatment decisions. At the same time, it can also improve the hospital's work efficiency and reduce the error rate. Furthermore, the cancer area and disease typing are predicted, the predicted probability is obtained and displayed in the form of a heat map, which can quickly and accurately predict the patient's disease type and its probability, and the prediction results are more intuitive and easy to understand.

[0104] In one embodiment, Figure 2 As shown, the generating of a diagnosis request according to the prediction result and obtaining the comprehensive diagnosis result of the pathologist in the pathology information system based on the diagnosis request include:

[0105] Step S201: obtaining a list of online pathologists providing pathological information, selecting a target pathologist, and generating a diagnosis request for the target pathologist;

[0106] Step S202: Obtain a login token, perform authority verification on the login token, and issue a request to call a diagnostic interface after passing the verification;

[0107] Step S203, uploading the prediction results and the slice digital images to the diagnosis interface for the target pathologist to perform online diagnosis;

[0108] Step S204: receiving the diagnosis data, and obtaining the comprehensive diagnosis result of the target pathologist according to the diagnosis data.

[0109] The beneficial effects of the above technical solution are: selecting a target pathologist and generating a diagnosis request for the target pathologist, obtaining a login token and performing authority verification, which can ensure that a suitable doctor is selected and the persuasiveness of the diagnosis result is guaranteed. At the same time, it can effectively reduce the risk of malicious attacks and protect the privacy and security of the patient's health data. Furthermore, the prediction results and slice digital images are provided to the target pathologist for online diagnosis, and the comprehensive diagnosis results of the target pathologist are obtained. The intuitive user interface is called to allow the doctor to easily view and operate the slices and reports, which greatly improves the efficiency of pathological diagnosis and provides more convenient services. In addition, comprehensive analysis based on the diagnostic data can obtain more accurate diagnostic results and improve the quality of medical services.

[0110] In one embodiment, generating a diagnosis report according to the comprehensive diagnosis result and performing intelligent review on the diagnosis report includes:

[0111] Obtain the online diagnostic editing content of pathologists on the intelligent platform to generate slice-level reports and pathology-level reports;

[0112] Integrate the slide-level report and the pathology-level report to generate a diagnosis report, and determine the report type of the diagnosis report;

[0113] Select audit indicators and audit methods according to the report type, conduct a formal audit on the diagnostic report according to the audit indicators and audit methods, and obtain the first audit result;

[0114] Retrieving a preset pathological diagnosis model, and using the preset pathological diagnosis model to review key information of the diagnosis result to obtain a second review result;

[0115] Whether the diagnosis report is qualified is determined based on the first audit result and the second audit result, and a warning reminder is issued if it is unqualified.

[0116] The beneficial effects of the above technical solution are: integrating slice-level reports and pathology-level reports to generate a diagnosis report and a report type, selecting audit indicators and audit methods according to the report type, conducting a formal audit of the diagnosis report, and using a preset pathology diagnosis model to audit the key information of the diagnosis results, which can ensure the accuracy and effectiveness of the diagnosis results and avoid inaccurate diagnosis results caused by formal problems and information problems.

[0117] In one embodiment, after the review is correct, the diagnosis report is synchronized to the pathology information system for recording and storage, including:

[0118] Determine the forms and fields used to record and store diagnostic reports in the pathology information system, and determine the storage data type of the diagnostic report based on the forms and fields;

[0119] Write the API interface of pathology information provision and intelligent platform according to the storage data type of diagnostic reports in the pathology information system;

[0120] Select an appropriate file upload tool to upload the diagnostic report to the pathology information system through the API interface;

[0121] The uploaded diagnostic report is processed by data escape and decryption, and the processed data is obtained and converted into a data format suitable for pathological information import for recording and storage.

[0122] The beneficial effects of the above technical solution are: the pathology information system determines the data type for storing diagnostic reports by designing forms and fields, writes an API interface for external platforms to call, selects an applicable file upload tool, uses the API interface to upload diagnostic reports to the system, and performs data escape and decryption to obtain a data format suitable for import. It can eliminate non-standardized terms, abbreviations and privacy information, making subsequent data analysis and processing more convenient and accurate, which helps to improve doctors' work efficiency and accuracy, while also protecting patients' privacy and sensitive information.

[0123] In one embodiment, this embodiment also discloses a pathological diagnosis system based on an intelligent platform, such as Figure 3 As shown, the system includes:

[0124] The prediction module 301 is used to obtain slice information, perform AI pathology prediction through an intelligent platform based on the slice information, and obtain prediction results;

[0125] An acquisition module 302 is used to generate a diagnosis request according to the prediction result, and acquire a comprehensive diagnosis result of a pathologist in the pathology information system based on the diagnosis request;

[0126] A generating module 303 is used to generate a diagnostic report according to the comprehensive diagnostic result and to perform intelligent review on the diagnostic report;

[0127] The synchronization module 304 is used to synchronize the diagnosis report to the pathology information system for recording and storage after verification.

[0128] The working principle of the above technical solution is: first, the slice information is obtained through the prediction module, and AI pathology prediction is performed through the intelligent platform according to the slice information to obtain the prediction results; secondly, the acquisition module is used to generate a diagnosis request according to the prediction result, and the comprehensive diagnosis result of the pathologist in the pathology information system is obtained based on the diagnosis request; then, the generation module is used to generate a diagnosis report according to the comprehensive diagnosis result, and the diagnosis report is intelligently reviewed; finally, the synchronization module is used to synchronize the diagnosis report to the pathology information system for recording and storage after the review is correct.

[0129] The beneficial effects of the above technical solution are as follows: through the comprehensive diagnosis of the AI ​​prediction results of the intelligent platform by the pathologist in the pathology information system, information exchange between the two systems can be realized, so that the pathologist can make a diagnosis on the intelligent platform, which improves the timeliness and accuracy of the diagnosis, provides effective reference data for the pathologist's diagnosis, improves the practicality and reliability as well as the efficiency of the diagnosis, and further, by synchronously storing and recording the diagnosis report, information exchange between the two systems can be further realized, which ensures the efficiency of data circulation and further improves the practicality.

[0130] In one embodiment, the prediction module includes:

[0131] A scanning submodule is used to obtain a slice increment file, scan the slice increment file using a high-resolution scanner, and obtain a slice digital image;

[0132] The first uploading submodule is used to encrypt and upload the slice digital image to the intelligent platform through the HTTPS protocol;

[0133] The recognition submodule is used to recognize the slice digital image through the optical character recognition technology of the intelligent platform, and obtain the slice number and pathology number according to the recognition result;

[0134] The integration submodule is used to retrieve the relevant clinical information of the slice digital image, and integrate the relevant clinical information, slice number and pathology number to generate slice information.

[0135] The prediction submodule is used to predict cancer areas and disease types based on deep learning technology through the artificial intelligence algorithm of the intelligent platform to obtain prediction probabilities;

[0136] The display submodule is used to display the disease types and their predicted probabilities in the form of a heat map.

[0137] The beneficial effects of the above technical solution are as follows: by acquiring digital slice images and uploading them to the intelligent platform through the HTTPS protocol for encryption, no human intervention is required, thus ensuring the timeliness and integrity of the data; the slice digital images are identified through optical character recognition technology to obtain the slice number and pathology number, which are integrated with the relevant clinical information of the slice digital images to generate slice information, which helps doctors better understand and analyze the patient's condition, thereby making more accurate diagnosis and treatment decisions. At the same time, it can also improve the hospital's work efficiency and reduce the error rate. Furthermore, the cancer area and disease typing are predicted, the predicted probability is obtained and displayed in the form of a heat map, which can quickly and accurately predict the patient's disease type and its probability, and the prediction results are more intuitive and easy to understand.

[0138] In one embodiment, the acquisition module includes:

[0139] A first acquisition submodule is used to acquire a list of online pathologists provided by pathology information and select a target pathologist, and generate a diagnosis request for the target pathologist;

[0140] The request submodule is used to obtain the login token, verify the permissions of the login token, and issue a request to call the diagnostic interface after the verification is passed;

[0141] The diagnosis submodule is used to upload the prediction results and the slice digital images to the diagnosis interface for the target pathologist to make an online diagnosis;

[0142] A receiving submodule, used to receive diagnostic data and obtain a comprehensive diagnostic result of a target pathologist based on the diagnostic data;

[0143] The generating module comprises:

[0144] The second acquisition submodule is used to obtain the online diagnosis editing content of the pathologist on the intelligent platform to generate a slice-level report and a pathology-level report;

[0145] A generation submodule, for integrating the slice-level report and the pathology-level report to generate a diagnosis report, and determining a report type of the diagnosis report;

[0146] The first audit submodule is used to select audit indicators and audit methods according to the report type, conduct a formal audit on the diagnostic report according to the audit indicators and audit methods, and obtain a first audit result;

[0147] The second review submodule is used to retrieve a preset pathological diagnosis model, use the preset pathological diagnosis model to review the key information of the diagnosis result, and obtain a second review result;

[0148] The first determination submodule is used to determine whether the diagnosis report is qualified according to the first audit result and the second audit result and to issue a warning reminder if it is unqualified.

[0149] The beneficial effects of the above technical solution are: selecting a target pathologist and generating a diagnosis request for the target pathologist, obtaining a login token and performing authority verification, which can ensure that a suitable doctor is selected and the persuasiveness of the diagnosis result is guaranteed. At the same time, it can effectively reduce the risk of malicious attacks and protect the privacy and security of the patient's health data. Further, the prediction results and slice digital images are provided to the target pathologist for online diagnosis, and the comprehensive diagnosis results of the target pathologist are obtained. The intuitive user interface is called to allow doctors to easily view and operate slices and reports, which greatly improves the efficiency of pathological diagnosis and provides more convenient services. In addition, comprehensive analysis based on diagnostic data can obtain more accurate diagnostic results and improve the quality of medical services. Integrating slice-level reports and pathology-level reports to generate diagnostic reports and report types, selecting audit indicators and audit methods according to report types, conducting formal audits on diagnostic reports, and using preset pathological diagnosis models to audit key information of the content of diagnostic results can ensure the accuracy and effectiveness of diagnostic results and avoid inaccurate diagnostic results caused by formal problems and information problems.

[0150] In one embodiment, Figure 4 As shown, the synchronization module 304 includes:

[0151] The second determination submodule 3041 is used to determine the form and field used to record and store the diagnosis report in the pathology information system, and determine the storage data type of the diagnosis report according to the form and field;

[0152] A writing submodule 3042 is used to write an API interface of the pathology information provision and intelligent platform according to the storage data type of the diagnosis report in the pathology information system;

[0153] The second uploading submodule 3043 is used to select a suitable file uploading tool to upload the diagnosis report to the pathology information system through the API interface;

[0154] The conversion submodule 3044 is used to perform data escape and decryption processing on the uploaded diagnosis report, obtain the processed data, and convert the processed data into a data format suitable for pathological information to be imported for recording and storage.

[0155] The beneficial effects of the above technical solution are: the pathology information system determines the data type for storing diagnostic reports by designing forms and fields, writes an API interface for external platforms to call, selects an applicable file upload tool, uses the API interface to upload diagnostic reports to the system, and performs data escape and decryption to obtain a data format suitable for import. It can eliminate non-standardized terms, abbreviations and privacy information, making subsequent data analysis and processing more convenient and accurate, which helps to improve doctors' work efficiency and accuracy, while also protecting patients' privacy and sensitive information.

[0156] In one embodiment, this embodiment discloses an intelligent pathology diagnosis platform, including:

[0157] In the traditional pathology diagnosis process, the workflows of the pathology information system and the intelligent pathology system are separated. Figure 5 As shown in the figure, during the sampling and slide preparation process, the pathologist maintains the basic information of the specimen through the pathology information system. After the solid slices are digitally scanned, they are uploaded to the intelligent pathology system for AI prediction. The pathologist refers to the AI ​​prediction results and writes and issues the report through the pathology information system. In this process, there is no direct information flow between the two systems, and it is impossible to effectively integrate the capabilities of AI into the overall diagnostic process.

[0158] The intelligent pathology diagnosis platform of the present invention can build an information highway between the pathology information system and the intelligent pathology system. This enables the intelligent pathology system to obtain basic information of the specimen from the pathology information system, and at the same time, the full scan digital slices and prediction results can be called by the latter, thereby seamlessly synchronizing the intelligent report. Figure 6 As shown in the figure, pathologists do not need to frequently switch between two systems during the diagnosis process, forming an integrated diagnostic process

[0159] Those skilled in the art should understand that the first and second in the present invention merely refer to different application stages.

[0160] Other embodiments of the present disclosure will be readily apparent to those skilled in the art after considering the specification and practicing the disclosure disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are to be considered exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.

[0161] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A pathological diagnosis method based on an intelligent platform, characterized in that: The following steps are involved: Obtain slice information, perform AI pathology prediction through the intelligent platform based on the slice information, and obtain prediction results; Generate a diagnosis request according to the prediction result, and obtain a comprehensive diagnosis result of a pathologist in a pathology information system based on the diagnosis request; Generate a diagnostic report based on the comprehensive diagnostic results and conduct intelligent review of the diagnostic report; After verification, the diagnosis report will be synchronized to the pathology information system for recording and storage.

2. The pathological diagnosis method based on the intelligent platform according to claim 1, characterized in that: Before obtaining the slice information, performing AI pathology prediction through the intelligent platform according to the slice information, and obtaining the prediction result, the method further includes: Obtain the user's login account and password on the intelligent platform and verify them, and determine whether the user has the right to use according to the verification result; If so, generate a service token and bind it to the user's login account; Configure operations on the user's login account based on preset access control policies.

3. The pathological diagnosis method based on the intelligent platform according to claim 1, characterized in that: The step of obtaining the slice information, performing AI pathology prediction through an intelligent platform according to the slice information, and obtaining the prediction result includes: Acquire slice increment files, scan the slice increment files with a high-resolution scanner, and acquire slice digital images; The digital images of the slices are encrypted and uploaded to the intelligent platform through the HTTPS protocol; The digital image of the slice is recognized through the optical character recognition technology of the intelligent platform, and the slice number and pathology number are obtained according to the recognition results; Retrieving relevant clinical information of the slice digital image, integrating the relevant clinical information, slice number and pathology number to generate slice information; The artificial intelligence algorithm of the intelligent platform is used to predict cancer areas and disease types based on deep learning technology to obtain prediction probabilities; The disease types and their predicted probabilities are displayed in the form of a heat map.

4. The pathological diagnosis method based on the intelligent platform according to claim 1, characterized in that: The generating of a diagnosis request according to the prediction result and obtaining a comprehensive diagnosis result of a pathologist in the pathology information system based on the diagnosis request includes: Obtaining a list of online pathologists provided by pathology information and selecting a target pathologist, and generating a diagnosis request for the target pathologist; Obtain the login token, verify the permissions of the login token, and issue a request to call the diagnostic interface after the verification passes; Upload the prediction results and slice digital images to the diagnosis interface for the target pathologist to make an online diagnosis; The diagnostic data is received, and a comprehensive diagnostic result of a target pathologist is obtained based on the diagnostic data.

5. The pathological diagnosis method based on the intelligent platform according to claim 1, characterized in that: The generating of a diagnostic report based on the comprehensive diagnostic results and the intelligent review of the diagnostic report include: Obtain the online diagnostic editing content of pathologists on the intelligent platform to generate slice-level reports and pathology-level reports; Integrate the slide-level report and the pathology-level report to generate a diagnosis report, and determine the report type of the diagnosis report; Select audit indicators and audit methods according to the report type, conduct a formal audit on the diagnostic report according to the audit indicators and audit methods, and obtain the first audit result; Retrieving a preset pathological diagnosis model, and using the preset pathological diagnosis model to review key information of the diagnosis result to obtain a second review result; Whether the diagnosis report is qualified is determined based on the first audit result and the second audit result, and a warning reminder is issued if it is unqualified.

6. The pathological diagnosis method based on the intelligent platform according to claim 1, characterized in that: After the above review is correct, the diagnosis report will be synchronized to the pathology information system for recording and storage, including: Determine the forms and fields used to record and store diagnostic reports in the pathology information system, and determine the storage data type of the diagnostic report based on the forms and fields; Write the API interface of pathology information provision and intelligent platform according to the storage data type of diagnostic reports in the pathology information system; Select an appropriate file upload tool to upload the diagnostic report to the pathology information system through the API interface; The uploaded diagnostic report is processed by data escape and decryption, and the processed data is obtained and converted into a data format suitable for pathological information import for recording and storage.

7. A pathological diagnosis system based on an intelligent platform, characterized in that: The system includes: The prediction module is used to obtain slice information, perform AI pathology prediction based on the slice information through the intelligent platform, and obtain the prediction results; An acquisition module, used to generate a diagnosis request according to the prediction result, and acquire a comprehensive diagnosis result of a pathologist in the pathology information system based on the diagnosis request; A generation module is used to generate a diagnosis report based on the comprehensive diagnosis results and to conduct intelligent review of the diagnosis report; The synchronization module is used to synchronize the diagnosis report to the pathology information system for recording and storage after verification.

8. The pathological diagnosis system based on the intelligent platform according to claim 7, characterized in that: The prediction module comprises: A scanning submodule is used to obtain a slice increment file, scan the slice increment file using a high-resolution scanner, and obtain a slice digital image; The first uploading submodule is used to encrypt and upload the slice digital image to the intelligent platform through the HTTPS protocol; The recognition submodule is used to recognize the slice digital image through the optical character recognition technology of the intelligent platform, and obtain the slice number and pathology number according to the recognition result; An integration submodule is used to retrieve relevant clinical information of the slice digital image, and integrate the relevant clinical information, slice number and pathology number to generate slice information; The prediction submodule is used to predict cancer areas and disease types based on deep learning technology through the artificial intelligence algorithm of the intelligent platform to obtain prediction probabilities; The display submodule is used to display the disease types and their predicted probabilities in the form of a heat map.

9. The pathological diagnosis system based on the intelligent platform according to claim 7, characterized in that: The acquisition module comprises: A first acquisition submodule is used to acquire a list of online pathologists provided by pathology information and select a target pathologist, and generate a diagnosis request for the target pathologist; The request submodule is used to obtain the login token, verify the permissions of the login token, and issue a request to call the diagnostic interface after the verification is passed; The diagnosis submodule is used to upload the prediction results and the slice digital images to the diagnosis interface for the target pathologist to make an online diagnosis; A receiving submodule, used to receive diagnostic data and obtain a comprehensive diagnostic result of a target pathologist based on the diagnostic data; The generating module comprises: The second acquisition submodule is used to obtain the online diagnosis editing content of the pathologist on the intelligent platform to generate a slice-level report and a pathology-level report; A generation submodule, for integrating the slice-level report and the pathology-level report to generate a diagnosis report, and determining a report type of the diagnosis report; The first audit submodule is used to select audit indicators and audit methods according to the report type, conduct a formal audit on the diagnostic report according to the audit indicators and audit methods, and obtain a first audit result; The second review submodule is used to retrieve a preset pathological diagnosis model, use the preset pathological diagnosis model to review the key information of the diagnosis result, and obtain a second review result; The first determination submodule is used to determine whether the diagnosis report is qualified according to the first audit result and the second audit result and to issue a warning reminder if it is unqualified.

10. The pathological diagnosis system based on the intelligent platform according to claim 7, characterized in that: The synchronization module comprises: The second determination submodule is used to determine the form and field used to record and store the diagnosis report in the pathology information system, and determine the storage data type of the diagnosis report according to the form and field; Writing submodules for writing API interfaces for pathology information provision and intelligent platforms according to the storage data types of diagnostic reports in the pathology information system; The second upload submodule is used to select a suitable file upload tool to upload the diagnosis report to the pathology information system through the API interface; The conversion submodule is used to perform data escape and decryption processing on the uploaded diagnostic report, obtain the processed data, and convert the processed data into a data format suitable for pathological information import for recording and storage.

Images