Tumor diagnosis and treatment standardized quality control method and system based on domain-specific language

Abstract

The invention provides a tumor diagnosis and treatment standardized quality control method and system based on a domain-specific language. The method comprises the following steps: acquiring professional tumor quality control indexes; the semantics of the index is recognized, a semantic logic relation is determined based on the semantics, index input data needed by the index are calculated, and the semantic logic relation indicates a relational expression describing a quality control index result determined based on the index input data; an index calculation task is generated based on the semantic logic relationship, a task is selected from a plurality of predefined tasks set according to a predetermined medical rule based on the index input data so as to generate an index composite task, the task logic of the index composite task is a directed acyclic graph, and a final node of the directed acyclic graph is the index calculation task, the plurality of predefined tasks and the index calculation task are described by a specific language in the tumor field; and inputting the medical data into the index composite task to obtain a tumor professional quality control index result.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and more specifically, to a method, system, device and medium for standardized quality control of tumor diagnosis and treatment based on a domain-specific language. Background Art

[0002] The quality control indicator system is an important part of the medical quality management and control system. Given that tumor diseases are major diseases that seriously threaten life and health, building a scientific and standardized medical quality control indicator system for tumor diseases is of great significance to strengthening scientific and refined medical quality management, promoting continuous improvement of medical quality, and improving the survival rate and quality of life of tumor patients.

[0003] There are a large number of tumor quality control indicators in the tumor quality control indicator system. These indicators will continue to change and update with the development of medical knowledge and technology, and the definition and calculation methods of these indicators are relatively complex. Therefore, tumor quality control indicators not only require professional medical technicians to handle, but also require medical technicians to analyze, calculate and count massive medical data one by one to obtain different tumor quality control indicators, which will consume a lot of labor costs and time, which is not only inefficient but may also cause errors or deviations due to manual statistics. Therefore, in order to solve the above technical problems, a fast, accurate and efficient standardized quality control method for tumor diagnosis and treatment is needed. Summary of the invention

[0004] In view of the above problems, the present invention provides a standardized quality control method, system, device and medium for tumor diagnosis and treatment based on a domain-specific language. The method first obtains tumor professional quality control indicators; determines the required one or more indicator input data and the semantic logical relationship of calculating the indicator result based on the indicator input data by identifying the semantics of the tumor professional quality control indicators; then generates a quality control indicator calculation task based on the semantic logical relationship and selects one or more tasks from multiple predefined tasks set according to predetermined medical rules based on the indicator input data, thereby generating a tumor professional quality control indicator composite task whose task logic is a directed acyclic graph and the final node of the directed acyclic graph is a quality control indicator calculation task, wherein the multiple predefined tasks and the quality control indicator calculation task are described by a tumor domain-specific language. In this way, a large amount of medical data can be input into the tumor professional quality control indicator composite task to quickly, accurately and efficiently obtain the tumor professional quality control indicator results without the need for processing by medical professionals, and significantly improves the efficiency of the diagnosis and treatment process and patient safety, which helps to strengthen scientific and refined medical quality management, promote continuous improvement of medical quality, and improve the survival rate of tumor patients and the quality of life of tumor patients.

[0005] According to one aspect of the present invention, a standardized quality control method for tumor diagnosis and treatment based on a domain-specific language is provided. The method includes obtaining a tumor-specific quality control indicator; identifying the semantics of the tumor-specific quality control indicator, and determining a semantic logical relationship and one or more indicator input data required to calculate the tumor-specific quality control indicator based on the semantics, wherein the semantic logical relationship indicates a relationship describing a determination of a tumor-specific quality control indicator result based on one or more indicator input data; generating a quality control indicator calculation task based on the semantic logical relationship, and selecting one or more tasks from a plurality of predefined tasks set according to predetermined medical rules based on the one or more indicator input data to generate a tumor-specific quality control indicator composite task, wherein the task logic of the tumor-specific quality control indicator composite task is a directed acyclic graph, and the final node of the directed acyclic graph is a quality control indicator calculation task, wherein the plurality of predefined tasks and the quality control indicator calculation task are described by a tumor-specific language; and inputting medical data into the tumor-specific quality control indicator composite task to obtain a tumor-specific quality control indicator result.

[0006] According to some embodiments of the present invention, in the above method, the tumor field-specific language defines input-output rules for the multiple predefined tasks, wherein the input-output rules indicate obtaining target information from target data related to medical information according to the predetermined medical rules, and the target information is associated with the one or more indicator input data.

[0007] According to some embodiments of the present invention, in the above method, the input-output rule indication of obtaining target information from target data related to medical information according to the predetermined medical rule includes: identifying keywords in the target data related to the medical information based on a regular expression generated by the predetermined medical rule; and filtering the target information from the keywords according to one or more indicator input data required for the oncology professional quality control indicator.

[0008] According to some embodiments of the present invention, in the above method, the predetermined medical rules include one or more of the following items: examination items, radiotherapy, diagnosis name, operation name, pathological diagnosis, standardized management of generic drug names, standardized management of sites, and standardized management of differentiation degree; wherein, identifying keywords in the target data related to the medical information based on regular expressions generated by predetermined medical rules includes: determining content containing specific topics from the target data related to the medical information based on regular expressions generated by predetermined medical rules, and determining keywords in the target data related to the medical information based on the content containing specific topics.

[0009] According to some embodiments of the present invention, in the above method, the target data related to medical information is obtained from structured information, wherein the structured information includes diagnosis structured information or treatment structured information, the diagnosis structured information includes diagnosis ID, diagnosis name, diagnosis date and diagnosis description, and the treatment structured information includes treatment ID, treatment name, treatment date and treatment description.

[0010] According to some embodiments of the present invention, in the above method, the tumor domain specific language defines one or more data structures for the multiple predefined tasks, and the one or more data structures are defined based on the following formula: Data Structure = T ( F('field name', field information), F('field name', Invoking function) ) Among them, Data Structure indicates the data type, T indicates the function used to create a table of the data structure, F indicates the function used to define the field, field name indicates the field name, field name indicates the information of the field, and Invokingfunction indicates the calling field that calls the structured information.

[0011] According to some embodiments of the present invention, in the above method, identifying the semantics of the oncology professional quality control indicator includes: determining the semantics in the oncology professional quality control indicator based on a large language model, wherein the large language model is pre-trained based on pre-trained text data so that the large language model has the ability to identify medical terms in text data, wherein the types of the pre-trained text data include one or more of basic medical knowledge, medical interpretations, research papers, medical records, and medical records.

[0012] According to another aspect of the present invention, a standardized quality control system for tumor diagnosis and treatment based on a domain-specific language is provided. The system includes: an indicator acquisition unit configured to acquire a tumor-specific quality control indicator; an indicator analysis unit configured to identify the semantics of the tumor-specific quality control indicator, and determine a semantic logical relationship based on the semantics and one or more indicator input data required to calculate the tumor-specific quality control indicator, wherein the semantic logical relationship indicates a relationship describing a relationship for determining a tumor-specific quality control indicator result based on one or more indicator input data; a task generation unit configured to generate a quality control indicator calculation task based on the semantic logical relationship, and select one or more tasks from a plurality of predefined tasks set according to a predetermined medical rule based on the one or more indicator input data to generate a tumor-specific quality control indicator composite task, wherein the task logic of the tumor-specific quality control indicator composite task is a directed acyclic graph, and the final node of the directed acyclic graph is a quality control indicator calculation task, wherein the plurality of predefined tasks and the quality control indicator calculation task are described by a tumor-specific language; and an indicator calculation unit configured to input medical information into the tumor-specific quality control indicator composite task to obtain a tumor-specific quality control indicator result.

[0013] According to some embodiments of the present invention, in the above-mentioned system, the tumor field-specific language defines input-output rules for the multiple predefined tasks, wherein the input-output rules indicate obtaining target information from target data related to medical information according to the predetermined medical rules, and the target information is associated with the one or more indicator input data.

[0014] According to some embodiments of the present invention, in the above-mentioned system, the input-output rule indicates that obtaining target information from target data related to medical information according to the predetermined medical rule includes: identifying keywords in the target data related to medical information based on a regular expression generated by the predetermined medical rule; and filtering the target information from the keywords according to one or more indicator input data required for the oncology professional quality control indicators.

[0015] According to some embodiments of the present invention, in the above-mentioned system, the predetermined medical rules include one or more of the following items: examination items, radiotherapy, diagnosis name, operation name, pathological diagnosis, standardized management of generic drug names, standardized management of sites, and standardized management of differentiation degree; wherein, identifying keywords in the target data related to the medical information based on regular expressions generated by predetermined medical rules includes: determining content containing specific topics from the target data related to the medical information based on regular expressions generated by predetermined medical rules, and determining keywords in the target data related to the medical information based on the content containing specific topics.

[0016] According to some embodiments of the present invention, in the above-mentioned system, the target data related to medical information is obtained from structured information, wherein the structured information includes diagnosis structured information or treatment structured information, the diagnosis structured information includes diagnosis ID, diagnosis name, diagnosis date and diagnosis description, and the treatment structured information includes treatment ID, treatment name, treatment date and treatment description.

[0017] According to some embodiments of the present invention, in the above system, the tumor domain specific language defines one or more data structures for the multiple predefined tasks, and the one or more data structures are defined based on the following formula: Data Structure = T ( F('field name', field information), F('field name', Invoking function) ) Among them, Data Structure indicates the data type, T indicates the function used to create a table of the data structure, F indicates the function used to define the field, field name indicates the field name, field name indicates the information of the field, and Invokingfunction indicates the calling field that calls the structured information.

[0018] According to some embodiments of the present invention, in the above-mentioned system, in order to identify the semantics of the oncology professional quality control indicator, the indicator analysis unit is further configured to: determine the semantics in the oncology professional quality control indicator based on a large language model, wherein the large language model is pre-trained based on pre-trained text data so that the large language model has the ability to identify medical terms in text data, wherein the types of the pre-trained text data include one or more of basic medical knowledge, medical interpretations, research papers, medical records, and medical records.

[0019] According to another aspect of the present invention, there is provided an electronic device, comprising: a processor; and a memory, wherein the memory stores computer-readable code, and when the computer-readable code is executed by the processor, the aforementioned standardized quality control method for tumor diagnosis and treatment based on a domain-specific language is implemented.

[0020] According to another aspect of the present invention, a non-transitory computer-readable storage medium is provided, storing computer-readable instructions, wherein when the computer-readable instructions are executed by a processor, the aforementioned standardized quality control method for tumor diagnosis and treatment based on a domain-specific language is implemented.

[0021] Therefore, according to the standardized quality control method, system, device and medium for tumor diagnosis and treatment based on domain-specific language of the embodiment of the present invention, first, the tumor professional quality control index is obtained; by identifying the semantics of the tumor professional quality control index, the required one or more indicator input data and the semantic logical relationship of calculating the indicator result based on the indicator input data are determined; then the quality control indicator calculation task is generated based on the semantic logical relationship and one or more tasks are selected from multiple predefined tasks set according to predetermined medical rules based on the indicator input data, thereby generating a tumor professional quality control indicator composite task whose task logic is a directed acyclic graph and the final node of the directed acyclic graph is the quality control indicator calculation task, wherein the multiple predefined tasks and the quality control indicator calculation task are described by the tumor field-specific language, so that a large amount of medical data can be input into the tumor professional quality control indicator composite task to quickly, accurately and efficiently obtain the tumor professional quality control indicator results without the need for processing by medical professionals, and significantly improve the efficiency of the diagnosis and treatment process and patient safety, which helps to strengthen scientific and refined medical quality management, promote continuous improvement of medical quality, and improve the survival rate of tumor patients and the quality of life of tumor patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some exemplary embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without creative work.

[0023] Figure 1 A flowchart of a method for standardized quality control of tumor diagnosis and treatment based on a domain-specific language according to some embodiments of the present invention is shown; Figure 2 A schematic diagram of a composite task of oncology professional quality control indicators according to some embodiments of the present invention is shown; Figure 3A block diagram of a tumor diagnosis and treatment standardized quality control system based on a domain specific language according to some embodiments of the present invention is shown; Figure 4 A structural diagram of an electronic device according to some embodiments of the present invention is shown. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings of the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the described embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] Unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. "First", "second" and similar words used in the present invention do not represent any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing in front of the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly. In order to keep the following description of the embodiments of the present invention clear and concise, the present invention omits the detailed description of some known functions and known components.

[0026] The present invention uses a flow chart to illustrate the steps of the method according to an embodiment of the present invention. It should be understood that the preceding or following steps are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously. At the same time, other operations may also be added to these processes, or one or more steps may be removed from these processes.

[0027] In the specification and drawings of the present invention, elements are described in singular or plural form, depending on the embodiment. However, the singular and plural forms are appropriately selected for the proposed situation only for the convenience of explanation and are not intended to limit the present invention thereto. Therefore, the singular form may include the plural form, and the plural form may also include the singular form, unless the context clearly indicates otherwise.

[0028] The standardized quality control method, system, device and medium for tumor diagnosis and treatment based on domain-specific language provided by the present invention will be described in detail below with reference to the accompanying drawings.

[0029] First embodiment

[0030] The present invention Figure 1 A flow chart of a method for standardized quality control of tumor diagnosis and treatment based on domain-specific language according to some embodiments of the present invention is shown. Figure 1 The standardized quality control method for tumor diagnosis and treatment based on domain-specific language of the present invention is described in detail.

[0031] First, in step S102, a tumor professional quality control index may be obtained. The tumor professional quality control index may be, for example, a tumor professional quality control index published by the National Health Commission, or a specific tumor professional quality control index set according to the needs of a medical institution.

[0032] In one example, the oncology professional quality control indicators may include, for example, the TNM staging diagnosis rate of breast cancer patients before the first treatment, which is defined as the proportion of breast cancer patients who completed clinical TNM staging diagnosis before the first treatment to the total number of breast cancer patients who received the first treatment. The first treatment refers to surgery, radiotherapy, chemotherapy, targeted therapy, endocrine therapy, immunotherapy, etc. for the tumor, and does not include puncture, biopsy, examination and other diagnostic and treatment measures taken to clarify the diagnosis or condition. The significance of calculating this indicator is that a comprehensive evaluation of the condition before treatment is the basis for standardized tumor treatment.

[0033] In one example, oncology quality control indicators can also include other different indicators for breast cancer, such as the rate of slight compliance with clinical TNM staging evaluation before the first treatment of breast cancer patients, the rate of pathological diagnosis before non-surgical treatment of breast cancer patients, the rate of sentinel lymph node biopsy in patients with early breast cancer, the rate of radiotherapy after modified radical mastectomy in patients with breast cancer, the rate of neoadjuvant therapy in patients with locally advanced breast cancer, the rate of systemic treatment for the first treatment of patients with advanced metastatic breast cancer, the rate of adjuvant endocrine therapy after surgery in patients with hormone receptor-positive breast cancer, the rate of HER-2 positivity in breast cancer patients receiving anti-HER-2 targeted therapy, the rate of adequate axillary lymph node dissection, and the rate of surgical site infection of type I incision for breast cancer.

[0034] In one example, the professional quality control indicators for tumors may also include professional quality control indicators for other types of tumors, such as lung cancer, liver cancer, cervical cancer, etc. Obtaining multiple professional quality control indicators for each tumor type in various types of tumors and building a scientific and standardized medical quality control indicator system for tumor diseases is of great significance to strengthening scientific and refined medical quality management, promoting continuous improvement of medical quality, and improving the survival rate and quality of life of tumor patients.

[0035] However, these indicators will continue to change and update with the development of medical knowledge and technology, and the definition and calculation methods of these indicators are relatively complex. Therefore, tumor quality control indicators not only need to be processed by professional medical technicians, but also require medical technicians to analyze, calculate and count massive medical data one by one to obtain different tumor quality control indicators, which will consume a lot of labor costs and time, not only inefficient but also may cause errors or deviations due to manual statistics.

[0036] In order to solve such problems, according to the standardized quality control method for tumor diagnosis and treatment based on domain-specific language of the present invention, in step S104, the semantics of the tumor professional quality control indicator can be identified, and based on the semantics, the semantic logical relationship and one or more indicator input data required to calculate the tumor professional quality control indicator can be determined, wherein the semantic logical relationship indicates a relationship that describes the result of determining the tumor professional quality control indicator based on one or more indicator input data.

[0037] According to one embodiment of the present invention, in order to identify the semantics of the tumor professional quality control indicator, the semantics in the tumor professional quality control indicator can be determined based on a large language model, wherein the large language model can be pre-trained based on pre-trained text data, so that the large language model has the ability to identify medical terms in text data, wherein the types of pre-trained text data include one or more of basic medical knowledge, medical explanations, research papers, medical records, and medical records. Training based on multiple types of pre-trained texts can enable the large language model to fully understand the vocabulary that may appear in different contexts, ensuring that the model has a deep understanding of the terminology.

[0038] According to an example of the present invention, the semantics of the quality control indicators of oncology can also be identified based on various machine learning models with natural language processing functions. Compared with machine learning models with natural language processing functions, large language models can access and utilize cutting-edge medical knowledge in a timely manner, so as to better identify the semantics of the quality control indicators of oncology.

[0039] Continuing with the aforementioned tumor professional quality control indicator of TNM staging diagnosis rate before the first treatment of breast cancer patients as an example, the formula for calculating this indicator is: Clinical TNM staging diagnosis rate of breast cancer patients before first treatment

[0040] According to the formula of the indicator, one or more indicator input data required for calculating the oncology professional quality control indicator can be determined, including: whether the patient is a breast cancer patient, whether the patient is receiving the first treatment and the time of receiving the first treatment, whether the patient has a TNM staging diagnosis and the time of TNM staging diagnosis. It should be noted that the above description is the indicator input data that needs to be determined for one patient's data. In order to determine the indicator content, obtaining the medical data of all patients will help the accuracy of the indicator results.

[0041] In addition, according to the formula of the indicator, the semantic logical relationship can be determined as (the patient is a breast cancer patient AND the patient has received the first treatment AND the patient has received a TNM staging diagnosis AND the time when the patient has received the TNM staging diagnosis is earlier than the time when the patient received the first treatment) and (the patient is a breast cancer patient AND the patient has received the first treatment). In addition to expressing the semantic logical relationship through logical operations such as AND or NOT on the indicator input data, the semantic logical relationship can also be expressed in an appropriate manner by processing the indicator input data based on the processed indicator input data.

[0042] After determining the semantic logical relationship and one or more indicator input data required to calculate the oncology professional quality control indicator, in step S106, a quality control indicator calculation task can be generated based on the semantic logical relationship, and based on the one or more indicator input data, one or more tasks are selected from multiple predefined tasks set according to predetermined medical rules to generate a composite task of oncology professional quality control indicators, wherein the task logic of the composite task of oncology professional quality control indicators is a directed acyclic graph (DAG), and the final node of the directed acyclic graph is a quality control indicator calculation task, wherein the multiple predefined tasks and the quality control indicator calculation task are described by a specific language for the oncology field. Using a specific language for the oncology field, even non-technical personnel can easily define and adjust treatment quality standards, which increases the flexibility and user-friendliness of the platform and allows medical professionals to customize the calculation logic of quality control indicators in a low-code or no-code manner.

[0043] According to one embodiment of the present invention, a tumor field-specific language defines input-output rules for multiple predefined tasks, wherein the input-output rules indicate obtaining target information from target data related to medical information according to predetermined medical rules, and the target information is associated with one or more indicator input data.

[0044] In one example, the input-output rules may specify processing rules for medical record text, which are used to identify and extract key information from the medical text.

[0045] In one example, the input-output rule may specify identifying the main diagnosis in the medical record text. For example, it may be defined based on the following formula: R_PRIMARY_DIAGNOSES = R( D( diagnoses_from_medical_record_front_page=DIAGNOSIS_LIST, diagnoses_from_discharge_summary=DIAGNOSIS_LIST, diagnoses_from_current_illness_history=DIAGNOSIS_LIST, diagnoses_from_clinical_course=DIAGNOSIS_LIST, context=VISIT, ), D( primary_diagnoses=DIAGNOSIS_LIST, ) ) Among them, R_PRIMARY_DIAGNOSES is the rule name; R indicates the function used to create rules, and D indicates the function used to create definitions; diagnoses_from_medical_record_front_page indicates the list of diagnoses extracted from the front page of the medical record; diagnoses_from_discharge_summary indicates the list of diagnoses extracted from the discharge summary; diagnoses_from_current_illness_history indicates the list of diagnoses extracted from the current illness history; diagnoses_from_clinical_course indicates the list of diagnoses extracted from the clinical course record; context: indicates context information, such as a patient's VISIT information.

[0046] Through the diagnostic information from the above multiple different sources, key information can be comprehensively obtained.

[0047] In addition, primary_diagnoses=DIAGNOSIS_LIST indicates a list of primary diagnoses, which is a comprehensive list based on the different diagnostic information provided in the input. In one example, based on analyzing and comparing the diagnosis lists of various input sources, it is possible to determine which diagnoses are critical and which diagnoses are the main manifestations of the disease. This process may involve prioritizing data, removing duplicate diagnoses, and determining which diagnoses are the most critical based on clinical significance.

[0048] In one example, the implementation of the R_PRIMARY_DIAGNOSES rule can utilize a large language model to parse and understand medical texts, while also combining medical expertise to assist in determining the primary diagnosis. According to an example of the present invention, a large language model can be utilized to identify keywords in target data related to medical information (such as the multiple diagnosis lists above) based on regular expressions generated using predetermined medical rules; and target information can be filtered from the keywords based on one or more indicator input data required for oncology professional quality control indicators. Determining regular expressions based on predetermined medical rules can more accurately extract keywords from medical text data.

[0049] It should be noted that the above-mentioned input and output rules for obtaining the diagnosis list are only an example of the present application. According to the different requirements of the indicator input data, other input and output rules can also be set. Therefore, according to one embodiment of the present invention, the predetermined medical rules may include one or more of the following: examination items, radiotherapy, diagnosis name, operation name, pathological diagnosis, standardized management of generic drug names, standardized management of parts, and standardized management of differentiation degree. Those skilled in the art can also set other predetermined medical rules according to needs.

[0050] In addition, identifying keywords in target data related to medical information based on regular expressions generated by predetermined medical rules may include: determining content containing specific topics from target data related to medical information based on regular expressions generated by predetermined medical rules, and determining keywords in target data related to medical information based on content containing specific topics. According to an example of the present invention, a specific topic can be determined based on one or more indicator input data. For example, when the indicator input data is whether the patient is a breast cancer patient, the specific topic may include diagnosis results, tumor type, or other topics that may include content related to the indicator input data. In one example, the category of the keyword topic can be, for example, breast cancer or liver cancer. The keyword topic can also be classified in other classification methods, or can be classified according to subcategories under a certain category (for example, according to a subtype of a certain type of cancer, etc.).

[0051] According to one embodiment of the present invention, target data related to medical information may be obtained from structured information, wherein the structured information may include diagnosis structured information or treatment structured information, the diagnosis structured information may include diagnosis ID, diagnosis name, diagnosis date and diagnosis description, and the treatment structured information may include treatment ID, treatment name, treatment date and treatment description.

[0052] In one example, the structured information may be defined based on the following formula: DIAGNOSIS = T( F('id', int), # diagnosis ID F('name'), # diagnosis name F('date', date), # diagnosis date F('description'), # diagnostic description # May also include other fields, such as diagnosis category, related symptoms, etc. ) In this example, the DIAGNOSIS type defines the basic fields contained in a diagnosis record: diagnosis ID (id), diagnosis name (name), diagnosis date (date), and diagnosis description (description), where T indicates the function used to create the table of the data structure, and F indicates the function used to define the fields.

[0053] In one example, structured information may be obtained through original unprocessed medical data, i.e., medical data to be input into the composite task of tumor professional quality control indicators described later. The above definition of structured information helps to extract important information from unstructured medical data and convert it into a structured format. Original unprocessed medical data may include unstructured text data such as medical records, diagnostic reports, and doctor's notes. In one example, medical data may also be medical imaging data. In one example, a large language model may be used to analyze and identify the content in the original unprocessed medical data to generate structured text data that meets specific requirements (e.g., having a specific format or presented in a specific field). Traditional methods rely on pipelines composed of multiple algorithm modules when processing unstructured medical data (such as medical records and diagnostic reports), which are highly complex, inefficient, and prone to errors. The present invention uses large language model technology to replace traditional processing pipelines to automatically extract key information from these texts, greatly reducing the complexity of the system and improving the efficiency and accuracy of data processing.

[0054] In addition, traditional methods face integration and standardization challenges when processing medical data from different sources. When processing data from different sources (such as different hospitals and departments), large language models can ensure the consistency and comparability of data, have strong robustness, and provide a basis for accurate analysis.

[0055] According to an example of the present invention, the medical data may be medical text data after data cleaning and NLP preprocessing to prevent unwanted information in the unprocessed medical text data from interfering with the recognition and analysis process. According to an embodiment of the present invention, the medical text data associated with the user may also be cleaned to filter out garbled characters and special characters in the medical text data; and then the filtered medical text data may be obtained to input into a large language model for further analysis and recognition.

[0056] According to one embodiment of the present invention, the tumor domain specific language defines one or more data structures for multiple predefined tasks. The one or more data structures may be defined based on the following formula: Data Structure = T ( F('field name', field information), F('field name', Invoking function) ) Among them, Data Structure indicates the data type, T indicates the function used to create a table of the data structure, F indicates the function used to define the field, field name indicates the field name, field name indicates the information of the field, and Invokingfunction indicates the calling field that calls the structured information.

[0057] In one example, the medical data structure may include hospital records (HOSPITAL_ADMISSION_RECORD), discharge summaries (DISCHARGE_SUMMARY), clinical records (CLINICAL_COURSE), etc. For example, the data structure of a clinical record may include fields such as date, type, and description. An example of the data structure of a clinical record is as follows: HOSPITAL_ADMISSION_RECORD = T( F('admission_date', date), # Admission date F('initial_diagnosis', DIAGNOSIS), # Initial diagnosis F('patient_info', PATIENT_INFO), # Patient basic information # May also include other fields, such as reason for admission, attending physician, etc. ) In this example, HOSPITAL_ADMISSION_RECORD is a structured type that contains fields such as admission date (admission_date), initial diagnosis (initial_diagnosis, which refers to the structured information DIAGNOSIS of the diagnosis type), and patient basic information (patient_info, which refers to the patient information type PATIENT_INFO). In addition, T indicates a function for creating a table of data structures, and F indicates a function for defining fields.

[0058] After defining the medical data structure, data structure, and input and output rules based on the domain-specific language, tasks (or task blocks) can be further defined. Tasks can be considered as independent functional units, each of which is responsible for different types of operations such as specific data extraction, conversion, and analysis. Each task block has clear inputs and outputs.

[0059] For example, to obtain the main diagnostic information, you can define the following task: @task( R_PRIMARY_DIAGNOSES ) def extract_primary_diagnoses(self, input_data_dict): # Implement the R_PRIMARY_DIAGNOSES input and output rules, that is, determine the patient's primary diagnosis ... In this task, the purpose is only to obtain diagnostic information according to the input-output rule R_PRIMARY_DIAGNOSES described above. Each task may have one or more input-output rules according to a predetermined medical rule setting, and use different data structures for input, output, and intermediate processing.

[0060] According to an example of the present invention, a specific task can be implemented based on a large language model, and natural language processing technology can be used to efficiently process the task. The following are two example tasks for extracting diagnostic information and recommending treatment plans using a large language model: Task Block: Extracting Diagnostic Information @task( R_DIAGNOSIS # Extract diagnostics ) def extract_diagnosis_with_llm(self, input_data_dict): # Extract diagnostic information using a large language model self._check_pause_and_cancel() medical_text = input_data_dict['medical_text'] prediction =llm(system_prompt="Extract diagnostic information:", user_query=medical_text) return {'diagnosis': prediction} Task Block: Recommend Treatment Plan @task( R_RECOMMEND_TREATMENT # Treatment recommendation ) def recommend_treatment_with_llm(self, input_data_dict): # Use a large language model to recommend treatment options self._check_pause_and_cancel() diagnosis_text = input_data_dict['diagnosis_text'] prediction =llm(system_prompt="Recommended treatment plan:", user_query=diagnosis_text) return {'treatment_plan': prediction} The large language model used in this article (including the large language model for identifying the semantics of quality control indicators in the field of oncology and the large language model for performing tasks) is a large language model specifically used in the field of oncology. The model is built on the basis of the LLAMA2 platform and has been continuously trained on the diagnosis and treatment data of approximately 15 million cancer patients and customized fine-tuned specifically for medical tasks. These tasks include but are not limited to diagnostic analysis, treatment recommendations, and case studies, ensuring that the model can provide accurate and effective support and advice when dealing with problems in these specific areas. The large language model for oncology not only relies on deep learning and big data training, but also combines rich external knowledge resources. External knowledge resources include comprehensive knowledge bases, the latest medical guidelines, the content of authoritative textbooks, and the latest clinical experimental studies. The integration of these resources enables the large language model to access and utilize cutting-edge medical knowledge in a timely manner, thereby providing the latest and most accurate guidance in tumor diagnosis and treatment.

[0061] By combining large language model technology with domain-specific languages, the efficiency, accuracy, and personalization of the tumor diagnosis and treatment process have been greatly improved, solving many challenges in traditional tumor diagnosis and treatment, providing patients with higher-quality medical services, and significantly improving patients' treatment effects and overall medical service quality. Through this efficient, accurate, and easy-to-use quality control solution, the present invention not only improves patient safety and treatment effects, but also provides medical institutions with more advanced tumor diagnosis and treatment management tools.

[0062] According to an example of the present invention, the task is usually predefined according to a predetermined medical rule. In some cases, a new task can also be constructed according to one or more new indicator input and output data.

[0063] According to an example of the present invention, a quality control indicator calculation task is generated based on the semantic logical relationship, and based on the one or more indicator input data, one or more tasks are selected from a plurality of predefined tasks set according to predetermined medical rules to generate a composite task of quality control indicators for oncology. The composite task of quality control indicator calculation task and the oncology quality control indicator generated by the selected one or more tasks contains a set of task blocks. The inputs and outputs between task blocks define the dependencies between them. The output of one task block can become the input of another task block. The dependencies between task blocks form a directed acyclic graph. This dependency ensures that data flows correctly between task blocks and that tasks are executed in a predetermined order, thereby pointing to the final node quality control indicator calculation task to calculate the oncology quality control indicator results.

[0064] The composite task of oncology quality control indicators can be regarded as a special task that contains multiple internal tasks. The input of the composite task of oncology quality control indicators is the external input of its internal tasks, which are not generated by any internal tasks. The output of the task is the collection of the outputs of all its internal tasks.

[0065] like Figure 2 As shown, in one example of the present invention, in order to calculate the quality control index of postoperative complications, a composite task of oncology professional quality control index can be generated based on the following method.

[0066] # Task 1: Extract diagnostic information @task( R( D(raw_text=RAW_TEXT), D(diagnoses=DIAGNOSIS_LIST) ) ) def ExtractDiagnosisTask(input_data): # Extraction logic... return diagnosis_data # Task 2: Extract treatment information @task( R( D(diagnoses=DIAGNOSIS_LIST), D(treatments=TREATMENT_LIST) ) ) def ExtractTreatmentTask(input_data): # Extraction logic... return treatment_data # Task 3: Identify complications @task( R( D(treatments=TREATMENT_LIST), D(complications=SYMPTOM_LIST) ) ) def IdentifyComplicationsTask(input_data): # Determine the logic... return complications_data # Task 4: Calculate quality control metrics @task( R( D(complications=SYMPTOM_LIST, treatments=TREATMENT_LIST), D(quality_control_metrics=QUALITY_CONTROL_METRIC_EVALUATION_LIST) ) ) def CalculateQualityControlMetricsTask(input_data): # Calculation logic... return quality_control_metrics # Create tasks tasks = ComposedTask("QualityControlComplicationsTask", ExtractDiagnosisTask, ExtractTreatmentTask, IdentifyComplicationsTask, CalculateQualityControlMetricsTask ) In this example, the oncology quality control indicator composite task QualityControlComplicationsTask includes three predefined tasks and a quality control indicator calculation task. The oncology quality control indicator composite task QualityControlComplicationsTask extracts diagnosis information from medical records, then extracts treatment information based on this information, then determines related complications, and finally calculates quality control indicators. Each task block in the oncology quality control indicator composite task QualityControlComplicationsTask receives the output of the previous task block as input and passes its output to the next task block to ensure the continuity and correctness of the data flow. In some examples, tasks may not be executed in a sequential manner, but in a directed acyclic graph as described above (for example, the input of a task depends on the output of the previous two tasks), so that the final data flows to the quality control indicator calculation task.

[0067] Finally, in step S108, the medical data may be input into the oncology professional quality control index composite task to obtain the oncology professional quality control index result.

[0068] Continuing with the previous example, for the oncology quality control indicator composite task QualityControlComplicationsTask, the task can be executed through a domain-specific language and the indicator results can be obtained.

[0069] # Execute composite tasks quality_control_results = tasks.execute(inputs) # Print results print("Quality Control Results:", quality_control_results) According to an example of the present invention, a thread pool can also be used to concurrently execute different task blocks for different medical data. This method allows tasks to run in parallel, thereby improving overall execution efficiency and response speed.

[0070] Combination of the above Figure 1The invention describes in detail a standardized quality control method for tumor diagnosis and treatment based on a domain-specific language, which can first obtain tumor professional quality control indicators; determine the required one or more indicator input data and the semantic logical relationship for calculating the indicator results based on the indicator input data by identifying the semantics of the tumor professional quality control indicators; then generate a quality control indicator calculation task based on the semantic logical relationship and select one or more tasks from multiple predefined tasks set according to predetermined medical rules based on the indicator input data, thereby generating a tumor professional quality control indicator composite task whose task logic is a directed acyclic graph and the final node of the directed acyclic graph is a quality control indicator calculation task, wherein the multiple predefined tasks and the quality control indicator calculation task are described by a tumor domain-specific language, so that a large amount of medical data can be input into the tumor professional quality control indicator composite task to quickly, accurately and efficiently obtain the tumor professional quality control indicator results without the need for processing by medical professionals, and significantly improve the efficiency of the diagnosis and treatment process and patient safety, which is helpful to strengthen scientific and refined medical quality management, promote continuous improvement of medical quality, and improve the survival rate of tumor patients and the quality of life of tumor patients.

[0071] Second embodiment

[0072] In addition to providing the above-mentioned tumor diagnosis and treatment standardized quality control method based on domain-specific language, the present invention also provides a tumor diagnosis and treatment standardized quality control system based on domain-specific language. Figure 3 This is described in detail.

[0073] Figure 3 FIG. 4 is a block diagram of a tumor diagnosis and treatment standardized quality control system based on a domain specific language according to some embodiments of the present invention. Figure 3 As shown, the standardized quality control system 300 for tumor diagnosis and treatment based on domain-specific language described in the present invention may include an indicator acquisition unit 310, an indicator analysis unit 320, a task generation unit 330 and an indicator calculation unit 340.

[0074] According to some embodiments of the present invention, the indicator acquisition unit 310 may be configured to acquire a tumor-specific quality control indicator. The indicator analysis unit 320 may be configured to identify the semantics of the tumor-specific quality control indicator, and determine a semantic logical relationship and one or more indicator input data required to calculate the tumor-specific quality control indicator based on the semantics, wherein the semantic logical relationship indicates a relationship describing a determination of a tumor-specific quality control indicator result based on one or more indicator input data. The task generation unit 330 may be configured to generate a quality control indicator calculation task based on the semantic logical relationship, and select one or more tasks from a plurality of predefined tasks set according to a predetermined medical rule based on the one or more indicator input data to generate a tumor-specific quality control indicator composite task, wherein the task logic of the tumor-specific quality control indicator composite task is a directed acyclic graph, and the final node of the directed acyclic graph is a quality control indicator calculation task, wherein the plurality of predefined tasks and the quality control indicator calculation task are described in a tumor-specific language. The indicator calculation unit 340 may be configured to input medical information into the tumor-specific quality control indicator composite task to obtain a tumor-specific quality control indicator result.

[0075] According to some embodiments of the present invention, the tumor field-specific language defines input-output rules for the multiple predefined tasks, wherein the input-output rules indicate obtaining target information from target data related to medical information according to the predetermined medical rules, and the target information is associated with the one or more indicator input data.

[0076] According to some embodiments of the present invention, the input-output rule indication that target information is obtained from target data related to medical information according to the predetermined medical rule includes: identifying keywords in the target data related to the medical information based on a regular expression generated by the predetermined medical rule; and filtering the target information from the keywords according to one or more indicator input data required for the oncology professional quality control indicator.

[0077] According to some embodiments of the present invention, the predetermined medical rules include one or more of the following items: examination items, radiotherapy, diagnosis name, operation name, pathological diagnosis, standardized management of generic drug names, standardized management of sites, and standardized management of differentiation degree; wherein, identifying keywords in the target data related to the medical information based on regular expressions generated by predetermined medical rules includes: determining content containing specific topics from the target data related to the medical information based on regular expressions generated by predetermined medical rules, and determining keywords in the target data related to the medical information based on the content containing specific topics.

[0078] According to some embodiments of the present invention, the target data related to medical information is obtained from structured information, wherein the structured information includes diagnosis structured information or treatment structured information, the diagnosis structured information includes diagnosis ID, diagnosis name, diagnosis date and diagnosis description, and the treatment structured information includes treatment ID, treatment name, treatment date and treatment description.

[0079] According to some embodiments of the present invention, the tumor domain specific language defines one or more data structures for the plurality of predefined tasks, and the one or more data structures are defined based on the following formula: Data Structure = T ( F('field name', field information), F('field name', Invoking function) ) Among them, Data Structure indicates the data type, T indicates the function used to create a table of the data structure, F indicates the function used to define the field, field name indicates the field name, field name indicates the information of the field, and Invokingfunction indicates the calling field that calls the structured information.

[0080] According to some embodiments of the present invention, in order to identify the semantics of the tumor professional quality control indicator, the indicator acquisition unit 310 is further configured to: The semantics in the tumor professional quality control indicator is determined based on a large language model, wherein the large language model is pre-trained based on pre-trained text data so that the large language model has the ability to identify medical terms in the text data, The types of the pre-training text data include one or more of basic medical knowledge, medical explanations, research papers, medical records, and medical records.

[0081] about Figure 3 Some specific details of the standardized quality control system for tumor diagnosis and treatment based on domain-specific language can also be referred to Figure 1 The content of the standardized quality control method for tumor diagnosis and treatment based on domain-specific language is shown in.

[0082] Figure 4 A structural diagram of an electronic device according to some embodiments of the present invention is shown.

[0083] See also Figure 4, the electronic device 400 may include a processor 401 and a memory 402. The processor 401 and the memory 402 may be connected via a bus 403. The electronic device 400 may be any type of portable device (such as a smart camera, a smart phone, a tablet computer, etc.) or any type of fixed device (such as a desktop computer, a server, etc.).

[0084] The processor 401 can perform various actions and processes according to the program stored in the memory 402. Specifically, the processor 401 can be an integrated circuit chip with signal processing capabilities. The above processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The disclosed methods, steps and logic block diagrams in the embodiments of the present invention can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc., which can be an X86 architecture or an ARM architecture.

[0085] The memory 402 stores computer executable instructions, and when the computer executable instructions are executed by the processor 401, the above-mentioned standardized quality control method for tumor diagnosis and treatment based on a domain-specific language is implemented. The memory 402 can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. The non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of exemplary but not limiting explanation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous connection dynamic random access memory (SLDRAM) and direct memory bus random access memory (DR RAM). It should be noted that the memory of the method described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0086] In addition, the standardized quality control method for tumor diagnosis and treatment based on domain-specific language according to the present invention can be recorded in a computer-readable recording medium. Specifically, according to the present invention, a computer-readable recording medium storing computer-executable instructions can be provided, and when the computer-executable instructions are executed by a processor, the processor can be prompted to execute the standardized quality control method for tumor diagnosis and treatment based on domain-specific language as described above.

[0087] It should be noted that the flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, a program segment, or a part of a code, and the module, program segment, or a part of a code contains at least one executable instruction for realizing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0088] In general, various example embodiments of the present invention may be implemented in hardware or dedicated circuits, software, firmware, logic, or any combination thereof. Certain aspects may be implemented in hardware, while other aspects may be implemented in firmware or software that may be executed by a controller, microprocessor, or other computing device. When various aspects of embodiments of the present invention are illustrated or described as block diagrams, flow charts, or using some other graphical representation, it will be understood that the blocks, devices, systems, techniques, or methods described herein may be implemented in hardware, software, firmware, dedicated circuits or logic, general purpose hardware or controllers or other computing devices, or some combination thereof as non-limiting examples.

[0089] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.

[0090] The above is an explanation of the present invention and should not be considered as a limitation thereof. Although several exemplary embodiments of the present invention have been described, it will be readily understood by those skilled in the art that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present invention. Therefore, all such modifications are intended to be included within the scope of the present invention as defined in the claims. It should be understood that the above is an explanation of the present invention and should not be considered as being limited to the specific embodiments disclosed, and modifications to the disclosed embodiments and other embodiments are intended to be included within the scope of the appended claims. The present invention is defined by the claims and their equivalents.

Claims

1. A standardized quality control method for tumor diagnosis and treatment based on domain-specific language, the method comprising: Obtain quality control indicators for oncology specialty; Identify the semantics of the oncology professional quality control indicator, and determine a semantic logical relationship and one or more indicator input data required to calculate the oncology professional quality control indicator based on the semantics, wherein the semantic logical relationship indicates a relational expression describing a result of determining the oncology professional quality control indicator based on the one or more indicator input data; A quality control indicator calculation task is generated based on the semantic logical relationship, and based on the one or more indicator input data, one or more tasks are selected from a plurality of predefined tasks set according to predetermined medical rules to generate a composite task of quality control indicators for oncology specialty, wherein the task logic of the composite task of quality control indicators for oncology specialty is a directed acyclic graph, and the final node of the directed acyclic graph is a quality control indicator calculation task, wherein the plurality of predefined tasks and the quality control indicator calculation task are described in a language specific to the oncology field; and The medical data is input into the oncology professional quality control indicator composite task to obtain the oncology professional quality control indicator result.

2. The method according to claim 1, wherein: The tumor domain specific language defines input-output rules for the multiple predefined tasks, wherein the input-output rules indicate obtaining target information from target data related to medical information according to the predetermined medical rules, and the target information is associated with the one or more indicator input data.

3. The method according to claim 2, wherein: The input-output rule indicates that obtaining target information from target data related to medical information according to the predetermined medical rule includes: identifying keywords in the target data related to the medical information based on a regular expression generated according to a predetermined medical rule; and The target information is filtered from the keywords according to one or more indicator input data required by the tumor professional quality control indicator.

4. The method according to claim 3, wherein: The predetermined medical rules include one or more of the following: examination items, radiotherapy, diagnosis name, operation name, pathological diagnosis, standardized management of generic drug names, standardized management of parts, and standardized management of differentiation degree; The method of identifying keywords in the target data related to the medical information based on a regular expression generated by a predetermined medical rule includes: determining content containing a specific topic from the target data related to the medical information based on a regular expression generated according to a predetermined medical rule, and Keywords in the target data related to the medical information are determined based on the content containing the specific subject.

5. The method according to claim 4, wherein: The target data related to medical information is obtained from structured information, wherein the structured information includes diagnosis structured information or treatment structured information, the diagnosis structured information includes diagnosis ID, diagnosis name, diagnosis date and diagnosis description, and the treatment structured information includes treatment ID, treatment name, treatment date and treatment description.

6. The method according to claim 5, wherein: The tumor domain specific language defines one or more data structures for the plurality of predefined tasks, and the one or more data structures are defined based on the following formula: Data Structure = T ( F('field name', field information), F('field name', Invoking function) ) Among them, Data Structure indicates the data type, T indicates the function used to create a table of the data structure, F indicates the function used to define the field, field name indicates the field name, field name indicates the information of the field, and Invokingfunction indicates the calling field that calls the structured information.

7. The method according to claim 1, wherein identifying the semantics of the oncology professional quality control indicator comprises: The semantics in the tumor professional quality control indicator is determined based on a large language model, wherein the large language model is pre-trained based on pre-trained text data so that the large language model has the ability to identify medical terms in the text data, The types of the pre-training text data include one or more of basic medical knowledge, medical explanations, research papers, medical records, and medical records.

8. A standardized quality control system for tumor diagnosis and treatment based on domain-specific language, the system comprising: An indicator acquisition unit, configured to acquire a tumor-specific quality control indicator; An indicator analysis unit is configured to identify the semantics of the oncology professional quality control indicator, and determine a semantic logical relationship and one or more indicator input data required to calculate the oncology professional quality control indicator based on the semantics, wherein the semantic logical relationship indicates a relational expression describing a result of determining the oncology professional quality control indicator based on the one or more indicator input data; a task generating unit configured to generate a quality control indicator calculation task based on the semantic logical relationship, and to select one or more tasks from a plurality of predefined tasks set according to predetermined medical rules based on the one or more indicator input data to generate a composite task of quality control indicators for oncology specialty, wherein the task logic of the composite task of quality control indicators for oncology specialty is a directed acyclic graph, and the final node of the directed acyclic graph is a quality control indicator calculation task, wherein the plurality of predefined tasks and the quality control indicator calculation task are described by a language specific to the oncology field; and The indicator calculation unit is configured to input medical information into the oncology professional quality control indicator composite task to obtain the oncology professional quality control indicator result.

9. The system according to claim 8, wherein: The tumor domain specific language defines input-output rules for the multiple predefined tasks, wherein the input-output rules indicate obtaining target information from target data related to medical information according to the predetermined medical rules, and the target information is associated with the one or more indicator input data.

10. The system according to claim 9, wherein: The input-output rule indicates that obtaining target information from target data related to medical information according to the predetermined medical rule includes: identifying keywords in the target data related to the medical information based on a regular expression generated according to a predetermined medical rule; and The target information is filtered from the keywords according to one or more indicator input data required by the tumor professional quality control indicator.

11. The system according to claim 10, wherein: The predetermined medical rules include one or more of the following: examination items, radiotherapy, diagnosis name, operation name, pathological diagnosis, standardized management of generic drug names, standardized management of parts, and standardized management of differentiation degree; The method of identifying keywords in the target data related to the medical information based on a regular expression generated by a predetermined medical rule includes: determining content containing a specific topic from the target data related to the medical information based on a regular expression generated according to a predetermined medical rule, and Keywords in the target data related to the medical information are determined based on the content containing the specific subject.

12. The system according to claim 11, wherein: The target data related to medical information is obtained from structured information, wherein the structured information includes diagnosis structured information or treatment structured information, the diagnosis structured information includes diagnosis ID, diagnosis name, diagnosis date and diagnosis description, and the treatment structured information includes treatment ID, treatment name, treatment date and treatment description.

13. The system according to claim 12, wherein: The tumor domain specific language defines one or more data structures for the plurality of predefined tasks, and the one or more data structures are defined based on the following formula: Data Structure = T ( F('field name', field information), F('field name', Invoking function) ) Among them, Data Structure indicates the data type, T indicates the function used to create a table of the data structure, F indicates the function used to define the field, field name indicates the field name, field name indicates the information of the field, and Invokingfunction indicates the calling field that calls the structured information.

14. The system according to claim 8, wherein: In order to identify the semantics of the tumor professional quality control indicator, the indicator analysis unit is further configured to: The semantics in the tumor professional quality control indicator is determined based on a large language model, wherein the large language model is pre-trained based on pre-trained text data so that the large language model has the ability to identify medical terms in the text data, The types of the pre-training text data include one or more of basic medical knowledge, medical explanations, research papers, medical records, and medical records.

15. An electronic device comprising: processor; as well as A memory, wherein the memory stores computer-readable codes, and when the computer-readable codes are executed by the processor, the standardized quality control method for tumor diagnosis and treatment based on a domain-specific language as described in any one of claims 1 to 7 is implemented.

16. A non-transitory computer-readable storage medium storing computer-readable instructions, wherein: When the computer-readable instructions are executed by a processor, the standardized quality control method for tumor diagnosis and treatment based on a domain-specific language as described in any one of claims 1 to 7 is implemented.

Images