Method and system for filling and verifying case report form

Abstract

The embodiment of the invention provides a case report form filling and checking method and system. The method comprises the following steps: acquiring case data; processing the case data by using an artificial intelligence model to determine first filling content corresponding to a first field in the case report form; and displaying the case report form filled with the first filling content and a first source identifier corresponding to the first filling content through the user terminal, wherein the first source identifier is configured to mark the first filling content and is determined based on the artificial intelligence model.

Description

Technical Field

[0001] This specification relates to the field of data extraction technology, and in particular to a method and system for filling and verifying a Case Report Form (CRF). Background Technology

[0002] When users (e.g., doctors or researchers) conduct research on a specific disease, they need to establish a disease-specific database. This database stores various data on patients corresponding to that disease (e.g., basic information, medical records, examination reports, medical orders, etc.). Due to the large amount of data in the database, the data for each patient needs to be organized into a structured Clinical Representation Flow (CRF) for subsequent statistical analysis. The CRF records key information extracted from the patient data that has a significant impact on the disease. Efficiently and accurately creating CRFs helps provide a high-quality data foundation for in-hospital research output and the development of precision medicine.

[0003] Therefore, it is necessary to provide a method and system for CRF filling and verification, so as to improve the accuracy of CRF filling while reducing the huge workload of CRF filling. Summary of the Invention

[0004] One embodiment of this specification provides a method for CRF filling and verification. The CRF filling and verification method includes: acquiring case data; processing the case data using an artificial intelligence model to determine first filling content corresponding to a first field in the CRF; and displaying the CRF filled with the first filling content and a first source identifier corresponding to the first filling content through a user terminal, wherein the first source identifier is configured to indicate that the first filling content is determined based on the artificial intelligence model.

[0005] In some embodiments, the method further includes: processing the case data using a preset mapping rule to determine the second fill content corresponding to the second field in the CRF; and further displaying the second fill content and the corresponding second source identifier through the user terminal, wherein the second source identifier is configured to mark the second fill content as determined based on the preset mapping rule.

[0006] In some embodiments, the method further includes: obtaining a modification instruction for the first fill content input by a user through the user terminal; and in response to the modification instruction, displaying the modified first fill content and a corresponding third source identifier through the user terminal, wherein the third source identifier is configured to mark that the modified first fill content was manually input by the user.

[0007] In some embodiments, the method further includes: displaying a tracing control corresponding to the first filled content through the user terminal; and, in response to a tracing instruction input by the user through the tracing control, displaying a reference basis corresponding to the first filled content through the user terminal.

[0008] In some embodiments, the method further includes: displaying the confidence level corresponding to the reference basis through the user terminal.

[0009] In some embodiments, the method further includes: displaying a view original text control corresponding to the reference reference on the user terminal; and in response to a view original text command input by the user through the view original text control, displaying a reference document corresponding to the reference reference on the user terminal, and highlighting the reference reference therein.

[0010] In some embodiments, the method further includes: displaying a quality control control via the user terminal; and, in response to a quality control instruction input by the user via the quality control control, verifying the first filled content based on the form template corresponding to the CRF.

[0011] In some embodiments, the method further includes: displaying an initial case list via the user terminal; obtaining a case adjustment instruction input by the user via the user terminal; and determining a case list based on the case adjustment instruction and the initial case list.

[0012] In some embodiments, the method further includes: displaying the form filling progress and / or confidence level corresponding to each case data in the case list via the user terminal.

[0013] In some embodiments, the case data is related to a target topic of a target disease, and the fields in the CRF are determined based on a topic form template corresponding to the target topic. The topic form template is configured by the user based on a disease-specific form template corresponding to the target disease.

[0014] One embodiment of this specification provides a CRF filling and verification system. The CRF filling and verification system includes: an acquisition module configured to acquire case data; a determination module configured to process the case data using an artificial intelligence model to determine first fill content corresponding to a first field in the CRF; and an interaction module configured to display, via a user terminal, the CRF filled with the first fill content and a first source identifier corresponding to the first fill content, wherein the first source identifier is configured to indicate that the first fill content was determined based on the artificial intelligence model.

[0015] One embodiment of this specification provides a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes a CRF filling and verification method. Attached Figure Description

[0016] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein: Figure 1 This is a schematic diagram illustrating application scenarios of the CRF filling and verification system according to some embodiments of this specification; Figure 2 This is an exemplary block diagram of a CRF filling and verification system according to some embodiments of this specification; Figure 3 This is an exemplary flowchart of a CRF filling and verification method according to some embodiments of this specification; Figure 4 This is an exemplary schematic diagram of a task form template configuration interface according to some embodiments of this specification; Figure 5 This is an exemplary schematic diagram of a form template configuration interface for another subject matter, as shown in some embodiments of this specification. Figure 6 This is an exemplary schematic diagram of a CRF filling and verification interface according to other embodiments of this specification; Figure 7 This is an exemplary flowchart of a method for tracing a CRF according to other embodiments of this specification; Figure 8 This is an exemplary schematic diagram of another CRF filling and verification interface shown according to other embodiments of this specification; Figure 9 This is an exemplary flowchart of a method for determining a list of cases according to other embodiments of this specification; Figure 10 This is an exemplary schematic diagram of a case list interface shown according to other embodiments of this specification. Detailed Implementation

[0017] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0018] It should be understood that the terms “system,” “device,” “unit,” and / or “module” used herein are one way to distinguish different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0019] As indicated in this specification and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of expressly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0020] Flowcharts are used in this specification to illustrate the operations performed by the system according to embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0021] Disease-specific clinical radiographs (CRFs) record key clinical indicators for diseases of concern to physicians and are an important component of disease-specific dataset construction. Existing CRF completion technologies generally follow a "one disease, one treatment" approach, including data collection, data preprocessing, and data label extraction. Some existing technologies combine AI models (e.g., large-scale models) to automate CRF completion. However, CRFs completed using AI models cannot guarantee 100% accuracy and often require manual verification by users. Furthermore, current systems do not integrate patient data with the CRF on the same screen, requiring users to switch between multiple interfaces to find reference information, making it difficult to trace back to the original reference and resulting in low efficiency for manual verification. On the other hand, existing CRF templates are fixed and cannot meet the personalized research needs of different users in different diseases and research topics.

[0022] To enable users to conveniently and efficiently verify the AI-filled results of CRF, this manual provides methods and systems for CRF filling and verification, which can improve the accuracy of CRF filling while reducing the workload of CRF filling.

[0023] Figure 1 This is a schematic diagram illustrating an application scenario of a CRF filling and verification system according to some embodiments of this specification. In some embodiments, the application scenario 100 of the CRF filling and verification system may include a CRF filling and verification system 110, a network 120, a hospital information system 130, and a user terminal 140.

[0024] CRF Population and Verification System 110 refers to a system used for populating and verifying CRFs. The CRF Population and Verification System 110 is configured to collect case data (e.g., basic information, medical records, examination reports, medical orders, etc.) from the hospital information system 130, allowing users to create a disease-specific dataset for a particular disease, and automatically process the case data in the disease-specific dataset through an artificial intelligence model to populate the corresponding CRF.

[0025] In some embodiments, the CRF filling and verification system 110 can be connected to the hospital information system 130 and the user terminal 140 via the network 120.

[0026] In some embodiments, processing and storage devices (not shown) may be integrated or installed on the CRF filling and verification system 110.

[0027] The processing device can be used to process data related to the CRF filling and verification system 110. For example, the processing device can acquire medical case data. The processing device can use an artificial intelligence model to process the medical case data to determine the first fill content corresponding to the first field in the CRF. Furthermore, the processing device can display the CRF filled with the first fill content and the first source identifier corresponding to the first fill content through a user terminal.

[0028] The processing device can be a single server or a group of servers. The server group can be centralized or distributed. In some embodiments, the processing device can be local or remote. In some embodiments, the processing device can be implemented on a cloud platform. By way of example only, the cloud platform can include private cloud, public cloud, hybrid cloud, community cloud, distributed cloud, internal cloud, multi-tiered cloud, etc., or any combination thereof.

[0029] The storage device may store data, instructions, and / or any other information. In some embodiments, the storage device may store data and / or instructions related to the CRF filling and verification system. For example, the storage device may store medical record data. As another example, the storage device may store instructions for determining the first fill content corresponding to a first field in the CRF.

[0030] In some embodiments, the storage device may be connected to network 120 to communicate with one or more other components of application scenario 100 of the CRF fill and verification system (e.g., hospital information system 130 and / or user terminal 140, etc.). One or more components of application scenario 100 of the CRF fill and verification system can access data or instructions stored in the storage device via network 120. In some embodiments, the storage device may be part of a processing device.

[0031] Network 120 may include any suitable wired or wireless network that facilitates the exchange of information and / or data. For example, CRF fill-in and verification system 110 and hospital information system 130 may transmit information and / or data via network 120. As another example, CRF fill-in and verification system 110 and user terminal 140 may transmit information and / or data via network 120.

[0032] The hospital information system 130 refers to the management system for hospital-related information.

[0033] In some embodiments, the hospital information system 130 can be configured to store and manage patient medical records. These records may include basic patient information, medical history, examination reports, and medical orders.

[0034] In some embodiments, the number of hospital information systems 130 can be multiple. For example, such as Figure 1 As shown, the hospital information system 130 may include a first hospital information system 130-1, a second hospital information system 130-2, and a third hospital information system 130-3, etc., corresponding to different hospitals. In some embodiments, the CRF filling and verification system 110 can be connected to each of the multiple hospital information systems 130 via a network 120.

[0035] In some embodiments, the user terminal 140 may include a mobile device 140-1, a tablet computer 140-2, a laptop computer 140-3, and a virtual reality device 140-4, or any combination thereof. In some embodiments, the processing device may display CRF-related content to the user through the user terminal 140. For example, the processing device may display a CRF filled with first fill content and a first source identifier corresponding to the first fill content to the user through the user terminal 140. As another example, the processing device may further display second fill content and a corresponding second source identifier to the user through the user terminal 140.

[0036] It should be noted that the application scenarios are provided for illustrative purposes only and are not intended to limit the scope of the invention. Those skilled in the art will recognize that various modifications or variations can be made based on the description provided. For example, the application scenarios can be implemented on other devices to achieve similar or different functions. However, these variations and modifications will not depart from the scope of the invention.

[0037] Figure 2 This is an exemplary block diagram of a CRF filling and verification system according to some embodiments of this specification.

[0038] In some embodiments, the CRF filling and verification system 110 may include an acquisition module 210, a determination module 220, and an interaction module 230.

[0039] The acquisition module 210 can be configured to acquire case data.

[0040] The determination module 220 can be configured to process case data using an artificial intelligence model to determine the first fill content corresponding to the first field in the CRF.

[0041] In some embodiments, the determining module 220 may be further configured to process case data using preset mapping rules to determine the second fill content corresponding to the second field in the CRF.

[0042] The interaction module 230 can be configured to display, via a user terminal, a CRF filled with the first fill content and a first source identifier corresponding to the first fill content, wherein the first source identifier is configured to mark that the first fill content is determined based on an artificial intelligence model.

[0043] In some embodiments, the interaction module 230 may be further configured to display the second fill content and the corresponding second source identifier through a user terminal, wherein the second source identifier is configured to mark the second fill content as determined based on a preset mapping rule.

[0044] In some embodiments, the interaction module 230 may be further configured to acquire a modification instruction for the first fill content input by the user through a user terminal; and in response to the modification instruction, display the modified first fill content and the corresponding third source identifier through the user terminal, wherein the third source identifier is configured to mark that the modified first fill content was manually input by the user.

[0045] In some embodiments, the interaction module 230 may be further configured to display the tracing control corresponding to the first filled content through the user terminal; and in response to the tracing instruction input by the user through the tracing control, to display the reference basis corresponding to the first filled content through the user terminal.

[0046] In some embodiments, the interaction module 230 may be further configured to display the confidence level corresponding to the reference basis through a user terminal.

[0047] In some embodiments, the interaction module 230 may be further configured to display the original text viewing control corresponding to the reference reference through the user terminal; and in response to the user's command to view the original text through the original text viewing control, to display the reference document corresponding to the reference reference through the user terminal, and to highlight the reference reference therein.

[0048] In some embodiments, the interaction module 230 may be further configured to display quality control controls via a user terminal; and in response to a quality control instruction input by the user via the quality control controls, to verify the first filled content based on the form template corresponding to the CRF.

[0049] In some embodiments, the interaction module 230 may be further configured to display an initial case list via a user terminal; obtain case adjustment instructions input by the user via the user terminal; and determine a case list based on the case adjustment instructions and the initial case list.

[0050] In some embodiments, the interaction module 230 may be further configured to display the form filling progress and / or confidence level corresponding to each case data in the case list via a user terminal.

[0051] For more details on the acquisition module 210, the determination module 220, and the interaction module 230, please refer to the relevant descriptions below.

[0052] It should be understood that Figure 2 The system and its modules shown can be implemented in various ways. It should be noted that the above description of the CRF filling and verification system and its modules is for convenience only and should not limit this specification to the scope of the illustrated embodiments. It is understood that those skilled in the art, after understanding the principles of the system, may arbitrarily combine the various modules or construct subsystems connected to other modules without departing from these principles. In some embodiments, Figure 2 The acquisition module 210, determination module 220, and interaction module 230 disclosed herein can be different modules within a single system, or a single module can implement the functions of two or more of the aforementioned modules. For example, the modules can share a single storage module, or each module can have its own separate storage module. Such variations are all within the scope of protection of this specification.

[0053] Figure 3 This is an exemplary flowchart illustrating a CRF filling and verification method according to some embodiments of this specification. Figure 3 As shown, process 300 includes the following steps. In some embodiments, Figure 3 One or more operations in the process 300 shown can be performed Figure 1 The CRF filling and verification system shown is implemented in application scenario 100. For example, Figure 3 The process 300 shown can be stored in the form of instructions in a storage device and invoked and / or executed by the CRF population and verification system 110.

[0054] Step 310: Acquire case data. In some embodiments, step 310 may be performed by the acquisition module 210.

[0055] Case data refers to case-related data that needs to be populated in a CRF (Clinical Request for Comments). For example, case data may include basic information, medical records, examination reports, and medical orders.

[0056] In some embodiments, the case data are related to a target disease. In some embodiments, the case data are related to a target topic of the target disease.

[0057] The target disease refers to the disease that the user is currently researching. For example, the target disease could include chronic sinusitis, type 2 diabetes, lung cancer, etc.

[0058] A target research topic refers to a research project established by a user to investigate a specific scientific question related to a target disease. For example, a target research topic could include a comparative study of the efficacy of nasal corticosteroid A versus antihistamine B in the treatment of allergic chronic sinusitis, a study on the impact of SGLT-2 inhibitor C on cardiovascular risk in patients with type 2 diabetes, or a study on the efficacy of targeted drug D as first-line treatment for EGFR-mutant non-small cell lung cancer.

[0059] In some embodiments, case data can be a single record. For example, a user selects a patient's case data on a CRF population and verification system and specifies that CRF population be performed on that patient; in this case, the case data is a single record.

[0060] In some embodiments, case data can be multiple records. For example, if a user selects to perform CRF imputation on a target disease, the case data includes multiple case records from the disease-specific dataset corresponding to the target disease. As another example, if a user selects to perform CRF imputation on a target topic of the target disease, the case data includes multiple case records from the disease-specific dataset related to the target topic.

[0061] A disease-specific dataset refers to a collection of case data corresponding to the target disease in a user's research. For more information on disease-specific datasets, please refer to [link to relevant documentation]. Figure 7 And its corresponding description.

[0062] In some embodiments, the processing device can filter out suitable case data from the hospital information system based on user-defined research objectives, such as selected target diseases, target topics, time periods (e.g., the last 3 years, the last 1 year, the last 6 months), and patient characteristics (e.g., male, female).

[0063] Step 320 involves processing the case data using an artificial intelligence model to determine the first fill content corresponding to the first field in the CRF. In some embodiments, step 320 may be performed by the determination module 220.

[0064] An artificial intelligence model refers to a language model used to determine the first padding content corresponding to certain specific fields (i.e., the first field) in a CRF. For example, such as... Figure 5 As shown, the analysis model (i.e., the artificial intelligence model) is DeepSeek.

[0065] A language model (LM) is a machine learning model trained using deep learning techniques and computing resources. Primarily designed for natural language processing, it can also be adapted to handle other forms of data. For example, language models can include those based on pre-trained and fine-tuned patterns (e.g., BERT (Bidirectional Encoder Representation from Transformers)) and dialogue-based language models (e.g., Generative Pre-Trained Transformer (GPT), Deepseek).

[0066] Among them, language models based on a pre-training and fine-tuning model refer to language models that include two stages of processing: general feature extraction and task output. BERT is a representative of this type of model. The two stages of this type of model can be trained separately; the general feature extraction part can be pre-trained, and the task output part can be trained (fine-tuned) based on the application task.

[0067] The language model based on the pre-training and fine-tuning model employs a pre-training + fine-tuning training mode. Pre-training is performed on a massive corpus, during which the model learns vocabulary, syntax, and semantic features. For specific tasks, only a smaller amount of data is needed for fine-tuning. Furthermore, this pre-training and fine-tuning model is a bidirectional language model, possessing contextual information. Therefore, in some embodiments, the processing device can generate an artificial intelligence model based on hospital filling rules through data collection, model fine-tuning, and model distillation to improve field matching accuracy.

[0068] A field is the most basic, indivisible unit of data in a CRF. A field is used to store a single piece of information of a specific type. For example, ... Figure 4 As shown, the fields can include name, date and time, number, medical insurance type, preoperative total IgE, etc.

[0069] In some embodiments, the processing device may determine the fields in the CRF based on the form template corresponding to the CRF.

[0070] A form template is a standardized, structured document framework or table corresponding to a form. A form template records the names of the fields that the form must include, the field types (e.g., single-choice, multiple-choice, single-line text, date, number, etc.), and the rules that the content to be filled in for each field must meet (e.g., whether it is required, maximum length, options, question visibility logic, etc.).

[0071] In some embodiments, before processing medical record data and populating the CRF, users can set the form template corresponding to the CRF through a user terminal. For example, the CRF population and verification system allows users to select or configure various form templates, including disease-specific form templates and research project form templates. In some embodiments, when the case data corresponds to a target research project, the fields in the CRF are determined based on the research project form template corresponding to the target research project. The research project form template is configured by the user based on the disease-specific form template corresponding to the target disease.

[0072] Disease-specific form templates include general disease-specific form templates and personalized disease-specific form templates. General disease-specific form templates are provided by the CRF filling and verification system and correspond to various diseases (e.g., chronic sinusitis, lung cancer), containing the core indicators for each disease. Personalized disease-specific form templates are form templates specifically designed by users for a particular disease; for example, they can be modified from general disease-specific form templates to generate personalized disease-specific form templates.

[0073] Furthermore, for a target disease, users may create multiple topics, each focusing on different information. The CRF filling and verification system allows users to adjust general or personalized disease-specific form templates to configure personalized topic form templates. For example, such as... Figure 4 As shown, users can add or delete fields according to the needs of the target topic through the project form template configuration interface (for example, a radiology user can add the field "clarity of image lesion boundaries") to configure the project form template. The project form configuration interface can display a "Reference Dictionary" control, allowing users to query and reuse existing configurations from the corresponding disease-specific form template. When the user clicks the "Reference Dictionary" control, a reference dictionary will pop up on the interface, displaying the disease-specific form template, where the user can select the fields and corresponding configurations to retain.

[0074] In some embodiments, as the research progresses, the user's research focus may change, and the user can adjust the research form template. For example, the user can add or delete indicators that were originally set in the research form template (e.g., it is found during the research that the "clarity of image lesion boundaries" indicator needs to be added). After the research form template is updated, the processing device can automatically reprocess the case data and update the CRF.

[0075] In some embodiments, the processing device may determine the fields in the CRF based on the topic form template corresponding to the target topic. For example, the processing device may use the fields contained in the topic form template corresponding to the target topic as the fields in the CRF.

[0076] The first field refers to the field whose content is determined using an artificial intelligence model.

[0077] The first fill content refers to the fill content corresponding to the first field determined by the artificial intelligence model. For example, the first fill content is a conclusion or suggestion derived by the artificial intelligence model through analysis, reasoning, and calculation based on case data and other prior knowledge (e.g., medical manuals, rules, guidelines). This is just an example. Figure 6 As shown, "Right maxillary sinus soft tissue density shadow filling" can be the first field, and the corresponding "Yes" can be the corresponding first filling content.

[0078] In some embodiments, the processing device can input case data into an artificial intelligence model and output the first fill content corresponding to the first field in the CRF through the artificial intelligence model.

[0079] Step 330: Display the CRF filled with the first fill content and the first source identifier corresponding to the first fill content through the user terminal. In some embodiments, step 330 can be performed by the interaction module 230.

[0080] The first source identifier is used to mark that the first fill content is determined based on an artificial intelligence model. For example, the first source identifier can be the text identifier "AI Fill" or "AI Inference," etc. For instance, as shown... Figure 6 As shown, for the first field "soft tissue density shadow filling in the right maxillary sinus", the first filling content "is" is determined based on the artificial intelligence model. Therefore, the first source identifier "AI filling" corresponding to the first filling content is displayed on the user terminal.

[0081] In some embodiments of this specification, artificial intelligence models are used to implement automatic CRF filling and batch processing functions. This reduces the workload of users manually filling in CRFs, avoids errors caused by manual filling, and improves the efficiency and accuracy of CRF filling. Furthermore, by displaying the source identifiers corresponding to each filled content, users can easily understand how the filled content was determined and quickly locate the content in the CRF that requires manual verification, thus improving verification efficiency.

[0082] It should be noted that the above description of the methods for filling in and verifying the process CRF is for illustrative purposes only and does not limit the scope of this specification. Those skilled in the art can make various modifications and changes to the methods for filling in and verifying the process CRF under the guidance of this specification. However, these modifications and changes are still within the scope of this specification.

[0083] In some embodiments, the processing device can display the CRF and other information on the same screen through a user terminal, making it convenient for the user to view other information. For example, the user terminal can display a first interface area, a second interface area, and a third interface area on the same screen.

[0084] The first interface area is configured to display medical information. For example, such as... Figure 6 As shown, the first interface area on the left displays the patient's medical history timeline. The timeline is a complete record of all a patient's medical-related events (e.g., outpatient visits, hospital admissions, examinations, surgeries, etc.) arranged chronologically. The timeline supports zooming and click-to-target. Zooming allows users to dynamically adjust the time range displayed on the timeline view; click-to-target allows users to trigger corresponding actions by directly clicking on a specific event on the timeline (e.g., an icon, a bar, or a piece of text), such as quickly locating and displaying the event's complete details.

[0085] The second interface area is configured to display the document or source information selected by the user. For example, Figure 6 As shown, in response to a user clicking "Sinus CT" in the first interface area, the processing device can display the original "Sinus CT Report" in the second interface area. Source information refers to information related to the reference data corresponding to the first filled content. More information about source information can be found in [link to relevant documentation]. Figure 7 And its corresponding description.

[0086] The third interface area is configured to display CRF. For example, as Figure 6 As shown, the third interface area on the right displays the CRF, including the fields, fill content, source identifier, etc.

[0087] In some embodiments of this specification, a first interface area, a second interface area, and a third interface area are displayed on the user terminal, achieving simultaneous screen interaction among the three. The integrated display of clinical data and CRF, along with the timeline interaction design, reduces the frequency of interface switching and improves the efficiency of users in obtaining reference information.

[0088] In some embodiments, process 300 may further include steps 340-350. It should be noted that steps 320 and 340 can be executed simultaneously, and steps 330 and 350 can be executed simultaneously.

[0089] Step 340 involves processing the case data using preset mapping rules to determine the second fill content corresponding to the second field in the CRF. In some embodiments, step 340 may be performed by the determination module 220.

[0090] Preset mapping rules refer to predefined, automated correspondences. Preset mapping rules specify how to automatically extract or match the second field's corresponding fill content based on case data.

[0091] The second field refers to the field whose content is determined using preset mapping rules. For example, the second field can be name, age, etc.

[0092] Understandably, the first field is a judgment field, and the second field is a fact field. That is, the first field requires judgment, rather than being directly extracted from case data; it is subjective, inferential, uncertain, and its accuracy is questionable. The second field is a fact field directly extracted from case data; it represents objectively existing, indisputable facts or fixed identifiers. For a given field, there are two processing methods: using an artificial intelligence model and using preset mapping rules. Different fields will have pre-defined processing methods. For example, a field requiring judgment is identified as the first field and processed using an artificial intelligence model. Conversely, a field directly extracted from case data is identified as the second field and processed using preset mapping rules. In some embodiments, the CRF also includes a third field, which requires user input. More information about the third field can be found in the description of step 370 below.

[0093] The second fill content refers to the fill content determined using preset mapping rules. For example, such as... Figure 6 As shown, "Examination Items" can be the second field, and the corresponding "Sinus CT Plain Scan + 3D" can be the second fill content.

[0094] In some embodiments, the processing device can find a completely matching field in the case data based on a preset mapping rule, and use the corresponding original text as the second fill content, thereby obtaining the second fill content corresponding to the second field in the CRF.

[0095] Step 350: Display the second filled content and the corresponding second source identifier through the user terminal. In some embodiments, step 340 may be performed by the interaction module 230.

[0096] The second source identifier is used to mark that the second fill content is determined based on a preset mapping rule. For example, the second source identifier can be a text identifier such as "original text mapping" or "mapping determined". For example, such as... Figure 6As shown, for the second field "Examination Items", the second fill content "Sinus CT Plain Scan + 3D" is determined based on the preset mapping rules. Therefore, the second source identifier "Original Text Mapping" corresponding to the second fill content is displayed on the user terminal.

[0097] In some embodiments of this specification, by displaying the second source identifier corresponding to the second fill content, users can quickly locate the second fill content in the CRF that needs to be manually verified, thereby improving verification efficiency.

[0098] In some embodiments, process 300 may further include steps 360-370.

[0099] Step 360: Obtain the user's modification instruction for any fill content input through the user terminal. In some embodiments, step 360 may be executed by the interaction module 230.

[0100] Arbitrary fill content refers to the fill content corresponding to any field. For example, arbitrary fill content can include the first fill content, the second fill content, etc.

[0101] A modification instruction is a request to modify any fill content. For example, a modification instruction may include a modification instruction for the first fill content, a modification instruction for the second fill content, and so on.

[0102] In some embodiments, when a user verifies any of the fillable content and finds it to be incorrect, they can issue a modification instruction for that fillable content. For example, such as... Figure 6 As shown, when the first filling content corresponding to the first field "left maxillary sinus soft tissue density shadow filling" is "none", the user can issue the command "change the first filling content of the first field 'left maxillary sinus soft tissue density shadow filling' from 'none' to 'yes'".

[0103] In some embodiments, the processing device can receive, via a network, a user's instruction to modify any fill content input through a user terminal.

[0104] Step 370: In response to the modification instruction, the modified fill content and the corresponding third-source identifier are displayed on the user terminal. In some embodiments, step 370 may be performed by the interaction module 230.

[0105] Modified fill content refers to fill content that has been modified by the user. For example, modified fill content may include modified first fill content, modified second fill content, etc.

[0106] The third-source identifier is used to indicate that the modified fill-in content was manually entered by the user. For example, the third-source identifier can be the text identifier "manually filled in" or "manually modified." For instance, such as... Figure 6 As shown, for the field "soft tissue density shadow filling in the left maxillary sinus," the modified filling content "Yes" was manually entered by the user. Therefore, the user terminal displays the third-source identifier "Manually filled" corresponding to the modified filling content. In some embodiments, the CRF may also include a third field. The third field is a field that specifies the filling content to be filled by the user. The third filling content corresponding to the third field refers to the filling content filled by the user. After the user fills in the third filling content, the processing device can further display the third filling content and the corresponding third-source identifier through the user terminal.

[0107] Understandably, when a CRF includes a third field, any filler content can also include third filler content. The modification command is an instruction to modify the third filler content entered by the user.

[0108] In some embodiments of this specification, users are allowed to modify any fill content, and the corresponding third fill content is displayed at the same time. This makes it convenient for users to quickly locate the fill content they have modified, avoids users repeatedly checking the fill content of the same field, and improves the efficiency of CRF verification.

[0109] In some embodiments, the processing device can display a confirmation control corresponding to any fill content through a user terminal; and in response to a confirmation command input by the user through the confirmation control, the processing device can display a confirmed indicator corresponding to any fill content through a user terminal.

[0110] A confirmation control is a control used to confirm the correctness of any filled-in content. For example, a confirmation control can be a confirmation button.

[0111] In some embodiments, the confirmation control may be always displayed, or it may be displayed only when the mouse moves over any area containing the fill content or clicks on any area containing the fill content.

[0112] In some embodiments, since the first field is a judgment field and the second field is a fact field, the processing device may only display the confirmation control corresponding to the first filled content through the user terminal. In some embodiments, the processing device may also display the confirmation controls corresponding to the first filled content, the second filled content, and / or the third filled content respectively through the user terminal.

[0113] A confirmation instruction is a request to confirm any filled-in content. For example, a confirmation instruction may include a confirmation instruction for the first filled-in content, a confirmation instruction for the second filled-in content, a confirmation instruction for the third filled-in content, and so on.

[0114] A "confirmed" flag is a marker configured to indicate that any fill-in content has been confirmed by the user. For example, the confirmed flag can be the text identifier "Confirmed" or "Confirmed without error." In some embodiments, once the user confirms any fill-in content, its source identifier is hidden, and only the confirmed flag is displayed.

[0115] In some embodiments of this specification, displaying a confirmation indicator can prevent users from repeatedly confirming the same filled content, thereby improving verification efficiency.

[0116] In some embodiments, the processing device can display the tracing controls and references corresponding to the first filled content via a user terminal. More information regarding the display of tracing controls and references can be found at [link to relevant documentation]. Figure 7 And its corresponding description.

[0117] In some embodiments, the processing device may further display quality control controls through a user terminal; and in response to quality control instructions input by the user through the quality control controls, verify any filled content based on the form template corresponding to the CRF.

[0118] Quality control controls are controls used to confirm the execution of quality control procedures. Quality control refers to the process of evaluating whether the filled content conforms to standards and specifications. For example, a quality control control can be a quality control button.

[0119] In some embodiments, the quality control controls may be always displayed, or they may only be displayed when the mouse moves over or clicks on any area containing the fill content. In some embodiments, quality control controls can be set for the entire CRF, or separate quality control controls can be set for different fill contents.

[0120] A quality control instruction is a request to perform quality control on all or any of the fill content in a CRF. For example, a quality control instruction may include a quality control instruction for the first fill content, a quality control instruction for the second fill content, a quality control instruction for the third fill content, and so on. Alternatively, a quality control instruction may include a quality control instruction for all fill content in the CRF.

[0121] In some embodiments, the processing device can verify all or any filled content in the CRF based on the form template corresponding to the CRF. Since users configure the form template and set rules and attributes that the displayed content must meet (e.g., age 1-99, date format YYYY-MM-DD, integer length 8, decimal length 3, etc.), the processing device can verify any filled content based on the form template corresponding to the CRF. For example, when the age is entered as 130, the processing device can recognize that it exceeds the range of 1-99 and prompt the error through a dialog box, voice, and / or highlighting.

[0122] In some embodiments, the processing device may also verify the arbitrary fill content based on the system's default field rules.

[0123] In some embodiments of this specification, by checking all or any of the fill content in the CRF, it is possible to avoid fill content that does not conform to preset rules, thereby improving the accuracy of the fill content.

[0124] In some embodiments, the processing device may determine a case list based on case adjustment instructions and an initial case list, and automatically execute the method in process 300 for each case data entry. For more information on displaying references, please refer to... Figure 7 And its corresponding description.

[0125] Figure 7 This is an exemplary flowchart illustrating a method for tracing a CRF according to other embodiments of this specification. Figure 7 As shown, process 700 includes the following steps. In some embodiments, Figure 7 One or more operations in the process 700 shown can be performed Figure 1 The CRF filling and verification system shown is implemented in application scenario 100. For example, Figure 7 The illustrated process 700 can be stored in a storage device as instructions and invoked and / or executed by the CRF population and verification system 110. In some embodiments, process 700 can be executed by the interaction module 230.

[0126] Step 710: The source control corresponding to the first filled content is further displayed through the user terminal.

[0127] A source tracing control is a control used to trigger the display of the reference data corresponding to the first piece of content to be filled. For example, such as... Figure 8 As shown, the tracing control can be a tracing button.

[0128] In some embodiments, the tracing control may be displayed continuously, or it may be displayed only when the mouse moves over the area where the field is located, or when the area where the field is located is clicked.

[0129] In some embodiments, in response to detecting that a user has selected a first field, the processing device can display a traceability control corresponding to the first filled content via a user terminal. For example, such as Figure 8 As shown, when the user moves the mouse to the area of ​​the first field "Right Maxillary Sinus Soft Tissue Density Shadow Filling", the processing device can display the traceability button corresponding to the first filling content through the user terminal.

[0130] In some embodiments of this specification, the traceability control corresponding to the first fill content is displayed only when the user selects the first field, which can keep the interface simple and avoid affecting the user's viewing of the CRF.

[0131] Step 720: Receive the tracing instruction input by the user through the tracing control.

[0132] A tracing instruction is a request to display the reference information corresponding to the first filled content. In response to a tracing instruction entered by the user through the tracing control, step 730 can be executed.

[0133] Step 730: Display the reference information corresponding to the first filled content through the user terminal.

[0134] Reference material refers to the text content that the AI ​​model consults when determining the first field to be filled. For example, reference material is a portion of the text content in a reference file that relates to the first field. Reference files can include examination reports, medical orders, medical images, etc.

[0135] In some embodiments, the processing device may display at least a portion of the reference file corresponding to the first fill content through a user terminal, and highlight the reference basis therein.

[0136] At least a portion of the reference document may include key text fragments or tags extracted from the reference document (e.g., the section containing the references). For example, the processing device can display at least a portion of the reference document corresponding to the first filled content in a second interface area via a user terminal, and highlight the specific references corresponding to the first filled content within that portion of the reference document. The highlighting may include bolding, underlining, italics, changing text color, changing background color, etc. For example, as... Figure 8 As shown, the processing device can display part of the "sinus CT plain scan + 3D" examination report, and bold the reference evidence "bilateral maxillary sinuses" and "soft tissue density shadow filling is seen in both".

[0137] In some embodiments of this specification, the reference references are highlighted to improve their intuitiveness and save users time in locating key information.

[0138] In some embodiments, the reference may include statistical charts of key indicators related to the target disease corresponding to the case data. For example, the processing device may provide a "trend chart" for quantitative indicators requiring long-term monitoring (e.g., blood glucose, liver and kidney function, tumor markers), and annotate normal ranges and key events (e.g., postoperative day 3). For example, as... Figure 8 As shown, statistical charts for key indicators can include a "sinus VAS score trend chart." For example, the processing device can provide a "Gantt chart" for medication history to show medication duration, dosage changes, and drug overlap. For another example, the processing device can use a "summary chart" for related content across multiple indicators (e.g., scoring items).

[0139] In some embodiments, the processing device can generate statistical charts based on multiple reference documents. For example, the processing device can extract key indicators from multiple reference documents and create statistical charts from the extracted key indicators. In some embodiments, users can click on a specific value on the statistical chart to display the reference basis corresponding to that specific value.

[0140] In some embodiments of this specification, statistical charts of key indicators are displayed to enhance the intuitiveness of information. At the same time, providing visual display formats such as charts caters to the usage habits of different research roles and lowers the operational threshold.

[0141] In some embodiments, steps 740 and / or 750 may also be performed in response to a traceability instruction input by a user through a traceability control. Steps 740 and / or 750 may be performed synchronously with step 730.

[0142] Step 740: Display the confidence level corresponding to the reference basis through the user terminal.

[0143] Confidence level refers to the reliability of a reference. For example, it can display a confidence level for each reference. Or, it can display the overall confidence level for all references (e.g., the average confidence level). Confidence levels can be provided by artificial intelligence models. Confidence levels can be expressed in the form of scores or rankings.

[0144] In some embodiments, the first filled content corresponds to multiple references, which are displayed in order of their confidence levels. For example, the multiple references can be displayed in descending order of their confidence levels. Thus, users can quickly determine the accuracy of the first filled content based on the first few references with higher confidence levels.

[0145] In some embodiments, in response to the total confidence level corresponding to the reference being less than a preset threshold, the processing device further displays a warning indicator via the user terminal. The preset threshold is a pre-set threshold used to determine whether the total confidence level meets the requirements. A warning indicator is an indicator that serves to warn of a low total confidence level; for example, a warning indicator could be an exclamation mark or a "low confidence" indicator. By displaying a warning indicator, the user is reminded that the initially filled content is questionable and requires close verification. This reduces data errors and improves the validity of CRF data.

[0146] The above-mentioned form template-based verification based on CRF is a format check. This section focuses on content verification based on confidence level. By adopting a dual quality control mechanism of format and content, the qualification of CRF data can be further improved.

[0147] In some embodiments, steps 750-760 may also be performed in response to a traceability instruction entered by the user through the traceability control.

[0148] Step 750: Display the corresponding "View Original Text" control on the user terminal.

[0149] The "View Original Text" control is a control used to trigger the display of the original text of the reference document containing the cited reference. In some embodiments, the "View Original Text" control may be always displayed, or it may be displayed only when the mouse moves over or clicks on the area containing the reference. For example, ... Figure 8 As shown, the processing device can display a "View Original Text" button near the references on the user terminal.

[0150] Step 760: In response to the user's command to view the original text via the view original text control, the reference document corresponding to the reference is displayed on the user terminal, and the reference is highlighted in it.

[0151] The "View Original" command is a request to view the original text.

[0152] In some embodiments, the processing device can display the reference document corresponding to the reference basis in the second interface area via a user terminal, and highlight the specific reference basis corresponding to the first filled content in the reference document. The highlighting can include bolding, underlining, italics, changing the text color, changing the background color, etc. For example, ... Figure 6 As shown, the processing device can display the reference document "Sinus CT Report" and highlight the reference evidence "Bilateral Maxillary Sinus" and "Soft tissue density shadow filling is seen in both cases" in bold.

[0153] In some embodiments of this specification, when a user enters a command to view the original text through the view original text control, the reference document corresponding to the reference is displayed on the user terminal, thus realizing a direct association between the CRF and the reference document.

[0154] In some embodiments, users can determine the validity of the first filled content based on reference criteria. If a problem exists, the user can input a modification command for the first filled content through the user terminal. At this time, the first filled content will be updated to the modified first filled content, and the first source identifier will be automatically updated to the third source identifier. If no problem exists, the user can click the "Verification Passed" button. After clicking, the "Verification Passed" button will turn gray or disappear, and a "Verification Passed" identifier will be further displayed near the first source identifier (for example, the "Verification Passed" identifier can be √).

[0155] In some embodiments, the second fill content determined using preset mapping rules can also be verified by the user. For example, when the mouse moves over the second field, a traceability button and a verification pass button are displayed. The specific verification method is similar to that for the first field, as described above and in related descriptions.

[0156] Figure 9 This is an exemplary flowchart of a method for determining a list of cases according to other embodiments of this specification. Figure 9 As shown, process 900 includes the following steps. In some embodiments, Figure 9 One or more operations in the process shown in 900 can be performed Figure 1 The CRF filling and verification system shown is implemented in application scenario 100. For example, Figure 9 The illustrated process 900 can be stored in a storage device as instructions and invoked and / or executed by the CRF population and verification system 110. In some embodiments, process 900 can be executed by the interaction module 230.

[0157] In some embodiments, the processing device may display an initial case list via a user terminal; obtain case adjustment instructions input by the user via the user terminal; and determine a case list based on the case adjustment instructions and the initial case list, and automatically execute the CRF filling and verification methods disclosed in this specification (such as process 300 and process 700) for each case data therein.

[0158] Step 910: Display the initial case list via the user terminal.

[0159] In some embodiments, users can set research objectives through a user terminal. The processing device can then filter suitable case data from a medical information system based on these objectives and present it to the user as an initial case list for confirmation. Research objectives may include one or more of the following: the target disease, the target topic, the target timeframe, and required patient characteristics (e.g., male or female). For example, the user terminal may display a list of disease-specific datasets for the user to select, with each dataset corresponding to a specific disease. The user can select the disease-specific dataset corresponding to the disease they plan to study. After clicking on a specific disease-specific dataset, the user terminal can display an initial case list corresponding to the case data within that dataset.

[0160] The case list presents a list of case-related data that needs to be populated and verified using CRF. For example... Figure 10As shown, the case list can display patient information such as name, patient ID, mobile phone number, and national ID number. The initial case list refers to the list of cases initially selected based on the research objectives set by the user. For example, the initial case list can be a patient cohort list or an examination cohort list. The patient cohort list and the examination cohort list correspond to different information organization dimensions. The patient cohort list is based on individual patients, with one record corresponding to one patient; the examination cohort list is based on a single examination (e.g., CT imaging), with one record corresponding to one examination.

[0161] In some embodiments, the fields contained in the CRF corresponding to the patient queue list and the examination queue list can be the same or different, and the user can configure them himself.

[0162] In some embodiments, the initial case list may be empty.

[0163] Step 920: Obtain the case adjustment instructions input by the user through the user terminal.

[0164] A case adjustment instruction is a request to modify the initial case list. For example, case adjustment instructions may include instructions to delete cases, instructions to add cases, etc.

[0165] In some embodiments, such as Figure 10 As shown, the interface displaying the initial case list includes delete and add controls. If a user finds at least one case in the initial case list unsuitable, they can select that case and click the delete control to delete it. To add other cases, the user can click the add control, which will redirect them to the case search interface. In this interface, users can search for new case data using criteria such as time, hospital, region, and disease, and then add the desired cases to the initial case list.

[0166] Step 930: Determine the case list based on the case adjustment instructions and the initial case list.

[0167] The case list refers to the final confirmed list of cases.

[0168] In some embodiments, the processing device may determine a final case list by adding and / or deleting cases based on case adjustment instructions, building upon an initial case list.

[0169] Once the case list is determined, the processing device can automatically execute the CRF fill and verification methods disclosed in this specification for each case data entry. For example, the cases in the case list can be sorted, and a corresponding fill task list can be generated, executing the fill tasks sequentially or in batches. Alternatively, after the case list is determined, the user needs to select a form fill control on the side-by-side list interface to issue a form fill command. In response to the form fill command, the processing device can automatically execute the CRF fill and verification methods disclosed in this specification for each case data entry in the case list.

[0170] In some embodiments of this specification, defining a case list allows users to quickly and easily query case data, providing an accurate and comprehensive data foundation for subsequent research. After the medical record data is defined, the system automatically begins a batch CRF filling task, simplifying user operations and improving efficiency.

[0171] In some embodiments, during the form filling process, the processing device can display the form filling progress and / or confidence level corresponding to each case data in the case list through a user terminal.

[0172] Form population progress refers to the population progress of the CRF corresponding to the case data. For example, ... Figure 10 As shown, patient A's form completion progress can be 80%.

[0173] The confidence level corresponding to case data refers to the overall confidence level of the references to all filler content (including the first and second filler content) of the CRF corresponding to the case data, reflecting the overall reliability of the CRF. In some embodiments, in response to the confidence level corresponding to case data being less than a preset threshold, the processing device can display a warning icon through the user terminal to prompt the user to conduct a focused review of the CRF of the case data.

[0174] In some embodiments of this specification, displaying the form filling progress in the case list allows users to easily understand the data processing progress. Displaying the confidence level in the case list allows users to quickly locate case data that requires manual verification.

[0175] In some embodiments, when a user is researching a target topic, the cases in their case list belong to a disease-specific dataset and may have previously undergone CRF filling operations. The processing device can verify whether any cases in the case list have corresponding previously filled historical CRFs. The processing device can identify common fields in the form template and historical CRFs, and directly fill (i.e., backfill) the content corresponding to these common fields into the current target topic's CRF, which can improve filling efficiency and reduce the time doctors spend filling out forms. More information about form templates can be found in [link to relevant documentation]. Figure 3 And its corresponding description.

[0176] In some embodiments, a user can click on a case in the case list (e.g., by clicking...). Figure 10 The “Patient 360” control shown can jump to the CRF display interface.

[0177] In some embodiments, a computer-readable storage medium stores computer instructions, and when a computer reads the computer instructions in the storage medium, the computer runs the CRF filling and verification method.

[0178] The beneficial effects that the embodiments of this specification may bring include, but are not limited to: (1) Using artificial intelligence models to realize the automatic filling and batch processing functions of CRF can reduce the workload of users manually filling CRF, while avoiding errors caused by manual filling, and improving the efficiency and accuracy of CRF filling. In addition, by displaying the source identifiers corresponding to each filling content, it is convenient for users to understand the determination method of the filling content, quickly locate the filling content in the CRF that needs to be manually checked, and improve the checking efficiency. (2) Using the source association function between CRF fields and original clinical documents can help users quickly check the basis of the content filled by the artificial intelligence model. (3) By determining the case list, it is convenient for users to quickly and easily query case data, providing an accurate and comprehensive data foundation for subsequent research. After the medical record data is determined, the system automatically starts the batch filling task of CRF, simplifying user operation and improving efficiency.

[0179] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are suggested in this specification and therefore remain within the spirit and scope of the exemplary embodiments described herein.

[0180] Furthermore, this specification uses specific terms to describe embodiments thereof. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Moreover, certain features, structures, or characteristics in one or more embodiments of this specification can be appropriately combined.

[0181] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this specification are not intended to limit the order of the processes and methods described herein. Although various examples have been discussed in the foregoing disclosure of some embodiments of the invention that are currently considered useful, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments; rather, the claims are intended to cover all modifications and equivalent combinations that conform to the spirit and scope of the embodiments described herein. For example, while the system components described above can be implemented using hardware devices, they can also be implemented solely using software solutions, such as installing the described system on existing servers or mobile devices.

[0182] Similarly, it should be noted that, in order to simplify the description disclosed herein and thus aid in the understanding of one or more embodiments of the invention, the foregoing description of embodiments in this specification may sometimes combine multiple features into a single embodiment, drawing, or description thereof. However, this method of disclosure does not imply that the subject matter of this specification requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of a single embodiment disclosed above.

[0183] In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of embodiments are modified in some examples with the terms "approximately," "approximately," or "generally." Unless otherwise stated, "approximately," "approximately," or "generally" indicates that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may be changed depending on the characteristics required by individual embodiments. In some embodiments, numerical parameters should take into account specified significant digits and employ a general method of digit reservation. Although the numerical ranges and parameters used to confirm their breadth of range in some embodiments of this specification are approximate values, in specific embodiments, such values ​​are set as precisely as feasible.

[0184] For each patent, patent application, patent application publication, and other material such as articles, books, specifications, publications, and documents referenced in this specification, the entire contents of which are incorporated herein by reference. This excludes historical application documents that are inconsistent with or conflict with the content of this specification, as well as documents that limit the broadest scope of the claims in this specification (currently or subsequently appended to this specification). It should be noted that in the event of any inconsistency or conflict between the descriptions, definitions, and / or terminology used in the supplementary materials to this specification and the content of this specification, the descriptions, definitions, and / or terminology used in this specification shall prevail.

[0185] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and are considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. A method for filling out and verifying a case report form, characterized in that, include: Obtain case data; The case data is processed using an artificial intelligence model to determine the first fill-in content corresponding to the first field in the case report form; as well as The case report form filled with the first filled content and the first source identifier corresponding to the first filled content are displayed on the user terminal. The first source identifier is configured to mark that the first filled content is determined based on the artificial intelligence model.

2. The method according to claim 1, characterized in that, The method further includes: The case data is processed using preset mapping rules to determine the second fill content corresponding to the second field in the case report form; and The second fill content and the corresponding second source identifier are further displayed through the user terminal. The second source identifier is configured to mark the second fill content as determined based on the preset mapping rule.

3. The method according to claim 1, characterized in that, The method further includes: Obtain the user's modification instruction for the first filled content input through the user terminal; and In response to the modification instruction, the modified first fill content and the corresponding third source identifier are displayed on the user terminal. The third source identifier is configured to mark that the modified first fill content was manually entered by the user.

4. The method according to claim 1, characterized in that, The method further includes: The user terminal further displays the source control corresponding to the first filled content; and In response to the tracing command input by the user through the tracing control, the reference basis corresponding to the first filled content is displayed on the user terminal.

5. The method according to claim 4, characterized in that, The method further includes: The confidence level corresponding to the reference criterion is displayed on the user terminal.

6. The method according to claim 4, characterized in that, The method further includes: The user terminal displays the original text view control corresponding to the reference; and In response to the user's command to view the original text via the view original text control, the reference document corresponding to the reference is displayed on the user terminal, and the reference is highlighted therein.

7. The method according to claim 1, characterized in that, The method further includes: The quality control controls are further displayed via the user terminal; and In response to the quality control instruction input by the user through the quality control control, the first filled content is checked based on the form template corresponding to the case report form.

8. The method according to claim 1, characterized in that, The method further includes: The initial case list is displayed via the user terminal; Obtain the case adjustment instructions input by the user through the user terminal; and Based on the case adjustment instructions and the initial case list, a case list is determined.

9. The method according to claim 8, characterized in that, The method further includes: The user terminal displays the form filling progress and / or confidence level for each case data entry in the case list.

10. The method according to claim 1, characterized in that, The case data is related to the target topic of the target disease. The fields in the case report form are determined based on the topic form template corresponding to the target topic. The topic form template is configured by the user based on the disease-specific form template corresponding to the target disease.

11. A system for filling out and verifying case report forms, characterized in that, include: The acquisition module is configured to acquire case data; The determination module is configured to process the case data using an artificial intelligence model to determine the first fill content corresponding to the first field in the case report form; as well as The interaction module is configured to display, via a user terminal, the case report form filled with the first filled content and a first source identifier corresponding to the first filled content, wherein the first source identifier is configured to mark that the first filled content is determined based on the artificial intelligence model.

12. A computer-readable storage medium storing computer instructions, wherein when a computer reads the computer instructions in the storage medium, the computer executes the method for filling and verifying a case report form as described in claims 1-10.

Images