Clinical test subject management system and method based on block chain

Abstract

The invention relates to a block chain-based clinical trial subject management system and method, and relates to the technical field of clinical trials, and the system comprises a management system constructed based on a block chain network and a plurality of edge computing nodes deployed in a medical institution. The edge computing node processes and encrypts the case information of the subject and then transmits the case information to the block chain network for storage, also can generate a feature query condition to screen a target subject, obtains index data of the target subject to determine test information, follow-up questionnaires and nursing information, and predicts follow-up time to generate a follow-up plan; the block chain network can start the smart contract template to screen the target subject based on the feature query condition. According to the method, the technical effects that the data security of the subjects is guaranteed by utilizing the block chain technology, and the clinical test subjects are efficiently screened, precisely supervised and scientifically managed in a follow-up visit manner are achieved.

Description

Technical Field

[0001] The present invention relates to the field of clinical trial technology, and in particular to a blockchain-based clinical trial subject management system and method. Background Art

[0002] A clinical trial project refers to a scientific research project designed to verify the safety and effectiveness of new drugs, vaccines, treatments or diagnostic tools in humans. Clinical trial projects are an indispensable part of medical research. They are crucial to promoting the advancement of medical science, improving patient treatment outcomes and ensuring the safety of medical products.

[0003] Subjects are individuals who participate in clinical trials. They play a crucial role in clinical trials, and having subjects who meet the test requirements is crucial to their success.

[0004] Currently, commonly used methods include manually recording and managing subject medical information. Staff members need to manually collect, organize, and analyze each subject's medical records, which involves reams of paperwork and tedious manual work. Furthermore, subject screening often relies on the physician's experience and subjective judgment, who determines whether the trial meets the requirements by reviewing medical records and conducting face-to-face interviews with the subject. During the trial, subject supervision and follow-up often utilize traditional telephone communication or regular outpatient checkups, with medical staff inquiring about the subject's physical condition and trial progress.

[0005] However, these existing management methods have significant flaws. Manual recording and management of case information is prone to errors and omissions, is inefficient, and makes it difficult to quickly and accurately obtain and process large amounts of case data. Relying on the subjective experience of doctors to screen subjects can lead to biased and inconsistent screening results, affecting the quality and efficiency of trials. Traditional monitoring and follow-up methods lack real-time and accuracy, making it difficult to obtain timely information on subjects' vital signs and behaviors, making it difficult to promptly identify and address problems that arise during the trial. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to provide a blockchain-based clinical trial subject management system and method, aiming to solve at least one of the above technical problems.

[0007] The technical solution of the present invention to solve the above technical problems is as follows: In the first aspect, the present application provides an artificial intelligence-based clinical trial subject management system, which adopts the following technical solutions: A blockchain-based clinical trial subject management system, comprising a blockchain network and multiple edge computing nodes, wherein the edge computing nodes are deployed in various medical institutions; a smart contract template is deployed on the blockchain network, and the smart contract template includes a screening contract; Each of the edge computing nodes includes: a data processing module, configured to obtain case information of each subject in the medical institution, the case information including structured data and unstructured data; perform feature extraction on the case information of each subject to obtain feature information; determine structured feature information of the subject based on the feature information and a preset mapping method; perform homomorphic encryption on the structured feature information of each subject to obtain multiple encrypted feature information; and send the multiple encrypted feature information to the blockchain network; a subject screening module, configured to obtain a target trial protocol, generate a feature query condition based on the target trial protocol, and send the feature query condition to the blockchain network to screen a plurality of target subjects corresponding to the target trial protocol based on the blockchain network; A subject monitoring module is configured to obtain, for each target subject, indicator data of the target subject, the indicator data representing vital sign data; determine test information based on the indicator data, the test information including normal test information or abnormal test information; and determine a follow-up questionnaire and nursing information for the target subject based on the test information; A follow-up management module is configured to obtain, for each target subject, the behavior information and test requirement information of the target subject, predict the follow-up time based on the behavior information and test requirement information of the target subject, generate a follow-up plan based on the follow-up time and test information, and push a follow-up questionnaire to the target subject; The blockchain network is used to store the encrypted characteristic information of the subjects of each medical institution, and to activate the corresponding smart contract template based on the characteristic query conditions to screen multiple target subjects corresponding to the target trial plan.

[0008] The beneficial effects of the present invention are as follows: the present invention constructs a full-process closed-loop management system for clinical trial subject management through the collaborative architecture of blockchain and edge computing nodes: edge computing nodes realize privacy processing and distributed storage of case information based on homomorphic encryption technology, breaking down data barriers between medical institutions while ensuring data security; the blockchain network automatically performs feature matching, inclusion and exclusion criteria verification and quota allocation through smart contract templates, improving screening accuracy and reducing the risk of human intervention; the follow-up management module relies on real-time analysis of subject behavior data and trial progress to generate personalized follow-up plans and push dynamic questionnaires; the subject supervision module links nursing staff information distribution through real-time monitoring and abnormal warning of vital signs indicators, forming a rapid response mechanism for trial safety assurance. The overall solution achieves breakthroughs in privacy protection, process automation, dynamic adaptability and compliance, providing an efficient, safe and intelligent integrated solution for multi-center clinical trials.

[0009] On the basis of the above technical solution, the present invention can also be improved as follows.

[0010] Furthermore, the subject monitoring module is used to determine the follow-up questionnaire of the target subject based on the trial information, specifically to: Determine basic questionnaire information based on the trial information of the target subject, wherein the basic questionnaire information includes target subject identity information, trial stage, trial type, clinical indicators, behavioral characteristics information, and risk level; Performing keyword feature extraction on the basic information of the questionnaire to obtain multiple keywords of the basic information of the questionnaire; Determining the questionnaire topic and content based on the number and preset ratio of each of the keywords; Determining the questionnaire type based on the target subject's identity information, wherein the target subject's identity information includes age, behavioral ability, and cognitive ability, and the questionnaire types include paper questionnaires with different font sizes and word explanation levels and voice questionnaires with different word explanation levels; Based on the questionnaire type, determining a plurality of recommended packaging templates; determining a first weight of each recommended packaging template based on a usage frequency of each recommended packaging template; determining a second weight of each of the recommended packaging templates based on the target subject's historical usage count of each of the recommended packaging templates; determining a third weight of each of the recommended packaging templates based on a dropout rate of a questionnaire corresponding to each of the recommended packaging templates; Determining a disease response coefficient based on the test information of the target subject, and determining a fourth weight of each of the recommended packaging templates according to the disease response coefficient; determining a target packaging template based on the first weight, the second weight, the third weight, and the fourth weight of each of the recommended packaging templates; A follow-up questionnaire is generated based on the questionnaire subject, the questionnaire content, the questionnaire type and the target packaging template.

[0011] The beneficial effects of adopting the above-mentioned further scheme are: first, based on the trial information of the target subjects (such as trial stage, clinical indicators, and risk level), keywords are extracted and questionnaire topics and content are generated to ensure that the questionnaire content is highly relevant to the trial progress; second, based on the subject's identity information (age, behavioral ability, cognitive ability), the paper or voice questionnaire type is automatically selected, and the text size and degree of word explanation are adjusted to solve the subject's comprehension difficulties caused by the single format of traditional questionnaires; third, through the dynamic weight calculation of the recommended packaging template (comprehensive frequency of use, historical number of uses, dropout rate, and disease response coefficient), the subject's preferences and trial needs are accurately matched to reduce the risk of dropout during questionnaire filling; the final follow-up questionnaire is both targeted in content, adaptable in form, and reliable in execution, which significantly improves subject compliance, ensures the integrity and accuracy of trial data collection, and provides high-quality data support for the analysis of clinical trial results.

[0012] Furthermore, the subject supervision module is used to determine the target subject nursing information based on the test information, specifically to: Determining the category level of the target subject based on the test information and a preset K-means clustering model; Determining nursing needs based on the category level of the target subject, wherein the nursing needs include nursing frequency; Based on the historical nursing records and trial stage of the target subject, predicting the nursing time and nursing duration of the target subject in a target time interval; Matching multiple managers at a current medical institution based on the location of the target subject; Based on a preset matching rule and multiple managers of the current medical institution, the caregiver of the target subject is determined, and based on the care time, care duration and caregiver, the care information is determined.

[0013] The beneficial effects of adopting the above-mentioned further scheme are: by classifying the trial information of the target subjects (such as clinical indicators, risk level, and trial stage) based on the K-means clustering model, dynamically adjusting the nursing frequency in combination with historical nursing records, and solving the resource waste or demand mismatch problems caused by traditional fixed-frequency nursing; by predicting the nursing time and duration of the target time interval, quickly matching the management personnel of the current medical institution in combination with the real-time location of the subjects to ensure the timeliness of nursing response; through preset matching rules, integrating the expertise, workload and nursing needs of managers, optimizing the allocation of nursing personnel, and improving the professionalism and efficiency of nursing services; the nursing information finally generated has the dual advantages of personalized demand satisfaction and optimal resource allocation, significantly improving the quality of subject care, reducing the risks caused by improper care during the trial, and providing strong guarantees for the safety of clinical trials.

[0014] Furthermore, the subject supervision module determines the caregiver of the target subject based on preset matching rules and multiple managers of the current medical institution, including: Obtaining the current subject information corresponding to each of the managers, the current subject information including the location, nursing time, and nursing duration of the subject to be processed, and the location, nursing time, and nursing duration of the subject currently being managed; Based on the timeliness information of each manager in the historical time period and the current subject information corresponding to each manager, the multiple managers are screened to obtain multiple first managers; Based on the nursing time and nursing duration of the target subject and the current subject information corresponding to each of the first managers, the plurality of managers are screened to obtain a plurality of second managers; determining a distance coefficient for each second manager based on the location of the target subject and current subject information corresponding to each second manager; determining a first priority value for each of the second managers based on the distance coefficient of each of the second managers; determining, based on the historical nursing information of each second manager, the degree of matching between each second manager and the trial type and trial phase of the target subject; determining a second priority value for each second manager based on a degree of matching between each second manager and the trial type and trial phase of the target subject; determining a third priority value for each second manager based on the number of mistakes made by each second manager in a historical time period; determining a fourth priority value for each second manager based on a degree of matching between each second manager and the subject; A caregiver for the target subject is determined based on the first priority value, the second priority value, the third priority value, and the fourth priority value of each of the second managers.

[0015] The benefits of adopting this further approach include: By combining managers' historical timeliness information with current task loads to screen available personnel, resource conflicts and overloads are avoided, improving basic allocation efficiency; by using the distance coefficient to match nursing staff to the nearest location, nursing response time is significantly shortened, and service professionalism is ensured by combining expertise matching based on trial type and stage; by introducing dual priority values ​​based on the number of historical errors and the degree of subject matching, operational risk is reduced; and finally, a weighted decision-making model that integrates these four priority values ​​achieves an optimal balance between timeliness, expertise, safety, and user experience. This addresses the subjectivity and inefficiency of traditional manual allocation and achieves dynamic optimization of nursing resources.

[0016] Furthermore, the follow-up management module is specifically used to: Obtaining location information of the target subject; Predicting the location of the target subject based on the positioning information of the target subject and the historical behavior data of the target subject; Predicting an initial follow-up time based on the target subject's behavioral information, test requirement information, the target subject's historical behavioral data, and an LSTM model; adjusting the initial follow-up time based on the predicted location of the target subject to obtain a target follow-up time for the target subject; A follow-up plan is generated based on the target follow-up time and the trial information, and the follow-up plan is sent to the target subject.

[0017] The beneficial effects of adopting the above further scheme are: combining the real-time positioning information of the target subject with historical behavioral data to predict his or her future location, thus solving the problem of inconvenience or lateness of the subject caused by the traditional fixed follow-up location; based on the analysis of behavioral information and experimental requirements by the LSTM model, the initial follow-up time is dynamically predicted to improve the accuracy of the time arrangement; the initial time is adaptively adjusted according to the predicted location to ensure that the follow-up plan is highly consistent with the subject's actual activity trajectory, avoiding the invalidation of the plan due to location changes.

[0018] Furthermore, the blockchain network is used to activate the corresponding smart contract template based on the feature query condition to screen multiple target subjects corresponding to the target trial plan, specifically for: Obtaining a number directory from a preset number directory smart contract, the number directory including a plurality of numbers, each of the numbers corresponding to a plurality of data names, the data names representing case names, and each of the data names being associated with at least one first keyword set; Determining a matching number for each subject and the number of matching fields corresponding to the matching number based on the encrypted feature information and the number directory of each subject; Determining a second keyword set based on the feature query condition; Matching the second keyword set with the first keyword set associated with each of the data names, and calculating a plurality of matching values; determining a plurality of candidate subjects based on a matching number of each subject, a number of matching fields corresponding to the matching number, and a plurality of matching degree values; Screening multiple candidate subjects based on the inclusion and exclusion criteria of the target trial protocol to generate a target subject list, wherein the inclusion and exclusion criteria represent conditions that must be met by the subjects and conditions that must be excluded; After the target subject list is verified by the node consensus, it is stored in the target subject smart contract and the data association relationship in the number directory smart contract is updated to obtain multiple target subjects.

[0019] The beneficial effects of adopting the above-mentioned further scheme are: realizing structured association and privacy-protected storage of multi-center case data based on the number directory smart contract, solving the data silos and privacy leakage risks of traditional centralized databases; improving the accuracy and efficiency of candidate subject screening through multi-dimensional matching of encrypted feature information and keyword sets (matching number, number of fields, matching value); smart contracts automatically execute inclusion and exclusion criteria verification to ensure that the screening results strictly comply with the requirements of the trial plan and reduce human operation errors; the target subject list and number directory update mechanism, which are recorded by node consensus, not only ensure process compliance, but also realize dynamic optimization of data association relationships.

[0020] Furthermore, the blockchain network is used to determine multiple candidate subjects based on the matching number of each subject, the number of matching fields corresponding to the matching number, and multiple matching degree values, specifically for: Determining a target number in the target number catalog based on the plurality of matching degree values; determining a plurality of initial subjects based on the matching number of each subject, the number of matching fields corresponding to the matching number, and the target number; Calculating a weighted matching degree for each of the initial subjects based on the number of matching fields for each of the initial subjects, the medication time difference, and the semantic similarity between the case text of each of the initial subjects and the feature query condition, wherein the medication time difference represents the difference between the historical medication time and the time of the target trial regimen; Based on the weighted matching scores of each of the initial subjects, a plurality of candidate subjects are determined.

[0021] The beneficial effects of adopting the above-mentioned further scheme are: through a dynamic determination mechanism based on the target number, combined with the matching number and the number of fields, the initial subject group that is highly relevant to the feature query conditions can be quickly locked, thereby improving the basic screening efficiency; the medication time difference is introduced as a key parameter, and by quantifying the time correlation between historical medication and the test plan, the impact of drug interference on the test results is avoided, and the scientific nature of the screening is significantly improved; by integrating the semantic similarity analysis of case text and query conditions, the semantic accuracy of feature association is ensured; through the weighted fusion of the number of matching fields, medication time difference, and semantic similarity, the objectivity and comprehensiveness of the candidate subject ranking are achieved.

[0022] Secondly, this application provides a blockchain-based clinical trial subject management method, which adopts the following technical solutions: A blockchain-based clinical trial subject management method, comprising: The edge computing node obtains case information of each subject in the medical institution, the case information including structured data and unstructured data; performs feature extraction on the case information of each subject to obtain feature information, and determines structured feature information of the subject based on the feature information and a preset mapping method; performs homomorphic encryption on the structured feature information of each subject to obtain multiple encrypted feature information; and sends the multiple encrypted feature information to the blockchain network; The edge computing node obtains a target test plan, generates a feature query condition based on the target test plan, and sends the feature query condition to the blockchain network to screen multiple target subjects corresponding to the target test plan based on the blockchain network; The edge computing node obtains, for each target subject, indicator data of the target subject, wherein the indicator data represents vital sign data; determines test information based on the indicator data, wherein the test information includes normal test information or abnormal test information; and determines a follow-up questionnaire and caregiver information of the target subject based on the test information; The edge computing node obtains the behavior information and test requirement information of each target subject, predicts the follow-up time based on the behavior information and test requirement information of the target subject, generates a follow-up plan based on the follow-up time and test information, and pushes a follow-up questionnaire to the target subject; The blockchain network stores the encrypted characteristic information of the subjects of each of the medical institutions, and activates the corresponding smart contract template based on the characteristic query conditions. The smart contract template includes a dynamic matching contract, an informed consent contract and a dynamic quota contract.

[0023] In a third aspect, the present application provides an electronic device, which adopts the following technical solution: An electronic device comprising a memory and a processor, wherein the memory stores a computer program capable of being loaded by the processor and executing a blockchain-based clinical trial subject management method as described in any one of the first aspects.

[0024] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium storing a computer program capable of being loaded by a processor and executing a blockchain-based clinical trial subject management method as described in any one of the first aspects.

[0025] Additional aspects and advantages of the present application will be given in part in the following description, which will become apparent from the following description, or will be learned through practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 A schematic diagram of the structure of a blockchain-based clinical trial subject management system provided by one embodiment of the present invention; Figure 2 A flowchart of a blockchain-based clinical trial subject management method provided in accordance with one embodiment of the present invention; Figure 3 This is a structural diagram of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0027] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0028] In this document, the term "and / or" simply describes a relationship between related objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " in this document, unless otherwise specified, generally indicates an "or" relationship between the related objects.

[0029] like Figure 1 As shown, a blockchain-based clinical trial subject management system 100 includes a blockchain network 102 and multiple edge computing nodes 101, wherein the edge computing nodes 101 are deployed in various medical institutions, and a smart contract template is deployed on the blockchain network 102, wherein the smart contract template includes a screening contract; Each of the edge computing nodes 101 includes: a data processing module, configured to obtain case information of each subject in the medical institution, the case information including structured data and unstructured data; perform feature extraction on the case information of each subject to obtain feature information; determine structured feature information of the subject based on the feature information and a preset mapping method; perform homomorphic encryption on the structured feature information of each subject to obtain multiple encrypted feature information; and send the multiple encrypted feature information to the blockchain network; a subject screening module, configured to obtain a target trial protocol, generate a feature query condition based on the target trial protocol, and send the feature query condition to the blockchain network to screen a plurality of target subjects corresponding to the target trial protocol based on the blockchain network; A subject monitoring module is configured to obtain, for each target subject, indicator data of the target subject, the indicator data representing vital sign data; determine test information based on the indicator data, the test information including normal test information or abnormal test information; and determine a follow-up questionnaire and nursing information for the target subject based on the test information; A follow-up management module is configured to obtain, for each target subject, the behavior information and test requirement information of the target subject, predict the follow-up time based on the behavior information and test requirement information of the target subject, generate a follow-up plan based on the follow-up time and test information, and push a follow-up questionnaire to the target subject; The blockchain network is used to store the encrypted characteristic information of the subjects of each medical institution, and to activate the corresponding smart contract template based on the characteristic query conditions to screen multiple target subjects corresponding to the target trial plan.

[0030] The present invention constructs a full-process closed-loop management system for clinical trial subject management through the collaborative architecture of blockchain and edge computing nodes: edge computing nodes implement privacy processing and distributed storage of case information based on homomorphic encryption technology, breaking down data barriers between medical institutions while ensuring data security; the blockchain network automatically performs feature matching, inclusion and exclusion criteria verification, and quota allocation through smart contract templates, improving screening accuracy and reducing the risk of human intervention; the follow-up management module relies on real-time analysis of subject behavior data and trial progress to generate personalized follow-up plans and push dynamic questionnaires; the subject supervision module uses real-time monitoring and abnormal warning of vital signs indicators to link nursing staff information distribution, forming a rapid response mechanism for trial safety assurance. The overall solution achieves breakthroughs in privacy protection, process automation, dynamic adaptability, and compliance, providing an efficient, secure, and intelligent integrated solution for multi-center clinical trials.

[0031] In the embodiment of the present application, the data processing module is specifically used to: The subject's case information is obtained through the medical institution information system interface or electronic medical record system (EMR). Structured data includes directly parseable fields such as age, gender, laboratory test result values, and diagnostic codes; unstructured data includes unstructured text or image data such as imaging examination report text and pathological slide descriptions.

[0032] For structured data, directly extract key fields as feature information; For unstructured text data, natural language processing technology is used to perform entity recognition and relationship extraction, extracting features such as symptoms, medication records, and previous medical history; For unstructured image data, a convolutional neural network model is used to extract features such as lesion location and size.

[0033] Then, based on the feature information and the preset mapping method, structured feature information is generated. This process uses a predefined rule engine or machine learning model (such as random forest) to achieve feature standardization. For example, the "headache" symptom extracted from the text is mapped to a standard medical terminology code; Finally, the structured feature information is encrypted using a homomorphic encryption algorithm to generate encrypted feature information, which is then uploaded to the blockchain.

[0034] Optionally, the subject monitoring module is configured to determine the follow-up questionnaire of the target subject based on the trial information, specifically to: Based on the trial information of the target subject, determining basic questionnaire information, the basic questionnaire information includes the target subject's identity information, trial stage, trial type, clinical indicators, behavioral characteristics, and risk level. Behavioral characteristics can reflect whether the subject takes medication on time and follows dietary restrictions; Performing keyword feature extraction on the basic information of the questionnaire to obtain multiple keywords of the basic information of the questionnaire; Determining the questionnaire topic and content based on the number and preset ratio of each of the keywords; Determining the questionnaire type based on the target subject's identity information, wherein the target subject's identity information includes age, behavioral ability, and cognitive ability, and the questionnaire types include paper questionnaires with different font sizes and word explanation levels and voice questionnaires with different word explanation levels; Based on the questionnaire type, determining a plurality of recommended packaging templates; determining a first weight of each recommended packaging template based on a usage frequency of each recommended packaging template; determining a second weight of each of the recommended packaging templates based on the target subject's historical usage count of each of the recommended packaging templates; determining a third weight of each of the recommended packaging templates based on a dropout rate of a questionnaire corresponding to each of the recommended packaging templates; Determining a disease response coefficient based on the test information of the target subject, and determining a fourth weight of each of the recommended packaging templates according to the disease response coefficient; determining a target packaging template based on the first weight, the second weight, the third weight, and the fourth weight of each of the recommended packaging templates; A follow-up questionnaire is generated based on the questionnaire subject, the questionnaire content, the questionnaire type and the target packaging template.

[0035] In this embodiment, identity information is extracted from the trial information, such as age, behavioral ability, cognitive ability assessment results, trial stage, trial type, clinical indicators, behavioral characteristics, and risk level. Clinical indicators include the white blood cell count in the most recent blood test, and risk levels such as high risk are defined as a history of severe adverse reactions.

[0036] Afterwards, the TF-IDF algorithm was used to extract keywords from the basic information of the questionnaire. For example, “white blood cell count,” “stage II,” and “high risk” were extracted as core keywords, and the questionnaire theme and content framework were constructed according to the preset ratio.

[0037] Afterwards, the questionnaire type was selected based on the participant's identity information. If the participant was ≥75 years old or had a cognitive ability score ≤24, a paper questionnaire with a large font and a vocabulary explanation level of "basic medical terms + daily terms" was preferred. If the subject has visual impairment or meets the smartphone proficiency standard, a voice questionnaire will be provided.

[0038] Afterwards, recommended packaging templates are screened from the preset template library, and the weights of each template are calculated. The first weight is based on the frequency of template use, such as the proportion of total system usage in the past three months. The second weight is the historical usage times of the target subject, such as the subject chose the graphic version in the past three follow-ups. The third weight is the template dropout rate, such as the process-guided version has a dropout rate of only 5%. The fourth weight is adjusted by the disease response coefficient. Finally, a linear weighted model is used to determine the target template.

[0039] Finally, the questionnaire topic and content are entered into the target template to generate a follow-up questionnaire, which is then sent to the subjects via the medical institution app or SMS. This accurately matches subject preferences with trial requirements and reduces the risk of dropouts midway through the questionnaire.

[0040] Optionally, the subject supervision module is configured to determine the target subject nursing information based on the trial information, specifically to: Determining the category level of the target subject based on the test information and a preset K-means clustering model; Determining nursing needs based on the category level of the target subject, wherein the nursing needs include nursing frequency; Based on the historical nursing records and trial stage of the target subject, predicting the nursing time and nursing duration of the target subject in a target time interval; Matching multiple managers at a current medical institution based on the location of the target subject; Based on a preset matching rule and multiple managers of the current medical institution, the caregiver of the target subject is determined, and based on the care time, care duration and caregiver, the care information is determined.

[0041] In an embodiment of the present application, a feature vector is constructed based on the test information (such as clinical indicators, risk level, and test stage), and a preset K-means clustering model is input to divide the subjects into different categories and levels. The levels can be a high-risk group or a stable group, and the category levels are directly related to the nursing needs.

[0042] A time series prediction model is used to predict the duration and duration of care for the target time interval, combining the subject's historical care records with the current trial phase. The subject's real-time location is obtained through the subject's location tracking module (GPS or indoor Bluetooth positioning). The patient is matched with the closest management personnel within the current medical institution. This management personnel's information includes current workload (number of assigned subjects) and areas of expertise (e.g., oncology care, chronic disease management).

[0043] In the embodiment of the present application, the subject supervision module determines the caregiver of the target subject based on the preset matching rules and multiple managers of the current medical institution, including: Obtaining the current subject information corresponding to each of the managers, the current subject information including the location, nursing time, and nursing duration of the subject to be processed, and the location, nursing time, and nursing duration of the subject currently being managed; Based on the timeliness information of each manager in the historical time period and the current subject information corresponding to each manager, the multiple managers are screened to obtain multiple first managers; Based on the nursing time and nursing duration of the target subject and the current subject information corresponding to each of the first managers, the plurality of first managers are screened to obtain a plurality of second managers; determining a distance coefficient for each second manager based on the location of the target subject and current subject information corresponding to each second manager; determining a first priority value for each of the second managers based on the distance coefficient of each of the second managers; determining, based on the historical nursing information of each second manager, the degree of matching between each second manager and the trial type and trial phase of the target subject; determining a second priority value for each second manager based on a degree of matching between each second manager and the trial type and trial phase of the target subject; determining a third priority value for each second manager based on the number of mistakes made by each second manager in a historical time period; determining a fourth priority value for each second manager based on a degree of matching between each second manager and the subject; A caregiver for the target subject is determined based on the first priority value, the second priority value, the third priority value, and the fourth priority value of each of the second managers.

[0044] By combining managers' historical timeliness information with current task loads to screen available personnel, resource conflicts and overloads are avoided, improving basic allocation efficiency. The proximity coefficient is used to match nursing staff to the nearest location, significantly shortening nursing response times. Expertise is combined with trial type and stage to ensure service professionalism. Operational risk is reduced by introducing dual priority values ​​based on the number of historical errors and the degree of subject matching. Finally, a weighted decision-making model that integrates these four priority values ​​achieves an optimal balance between timeliness, expertise, safety, and user experience. This addresses the subjectivity and inefficiency of traditional manual allocation and enables dynamic optimization of nursing resources.

[0045] Optional follow-up management module, specifically used for: Obtaining location information of the target subject; Predicting the location of the target subject based on the positioning information of the target subject and the historical behavior data of the target subject; Predicting an initial follow-up time based on the target subject's behavioral information, test requirement information, the target subject's historical behavioral data, and an LSTM model; adjusting the initial follow-up time based on the predicted location of the target subject to obtain a target follow-up time for the target subject; A follow-up plan is generated based on the target follow-up time and the trial information, and the follow-up plan is sent to the target subject.

[0046] In the embodiments of the present application, real-time location information is obtained through the GPS module built into the smart terminal carried by the target subject. Based on historical behavioral data, a time-space sequence model is constructed. The current location information is combined with the prediction of future location distribution. The behavioral information, test requirement information, and historical behavioral data are input into a pre-trained LSTM model to output the initial follow-up time. The time is adjusted according to the predicted location: if the predicted location is local, the original time is maintained; if the predicted location is not local, the initial follow-up time is adjusted to obtain the target follow-up time for the target subject.

[0047] Optionally, the blockchain network is used to activate a corresponding smart contract template based on a feature query condition to screen multiple target subjects corresponding to the target trial protocol, specifically for: Obtaining a number directory from a preset number directory smart contract, the number directory including a plurality of numbers, each of the numbers corresponding to a plurality of data names, the data names representing case names, and each of the data names being associated with at least one first keyword set; Determining a matching number for each subject and the number of matching fields corresponding to the matching number based on the encrypted feature information and the number directory of each subject; Determining a second keyword set based on the feature query condition; Matching the second keyword set with the first keyword set associated with each of the data names, and calculating a plurality of matching values; determining a plurality of candidate subjects based on a matching number of each subject, a number of matching fields corresponding to the matching number, and a plurality of matching degree values; Screening multiple candidate subjects based on the inclusion and exclusion criteria of the target trial protocol to generate a target subject list, wherein the inclusion and exclusion criteria represent conditions that must be met by the subjects and conditions that must be excluded; After the target subject list is verified by the node consensus, it is stored in the target subject smart contract and the data association relationship in the number directory smart contract is updated to obtain multiple target subjects.

[0048] In an embodiment of the present application, by calling a preset number directory smart contract, for example, an on-chain structured table that stores case data indexes, each number corresponds to multiple data names such as "Case-001" and "Case-002", and the data names are associated with a first keyword set such as "breast cancer" and "HER2 positive", the full number directory is obtained.

[0049] Afterwards, a ciphertext match is performed between the encrypted feature information uploaded by the subject supervision module and the number directory. The matching number and the corresponding number of matching fields are determined (for example, "Case-001" matches 3 fields) by calculating the number of field overlaps between each subject feature and the data name corresponding to the number (for example, "45 years old" matches the "Age = 40-50 years old" field in "Case-001").

[0050] Afterwards, based on the feature query conditions, the first keyword set is extracted. For example, the first keyword set includes "breast cancer", "HER2 positive", "untargeted therapy", etc., and semantic matching is performed with the second keyword set associated with each data name, and the TF-IDF algorithm is used to calculate the matching value.

[0051] Afterwards, candidate subjects are screened based on the number of matching number fields and matching value.

[0052] Smart contracts automatically validate inclusion and exclusion criteria to generate a target subject list. Finally, after blockchain node consensus verification and the adoption of the PBFT consensus algorithm, the target subject list is stored in the target subject smart contract, and the data association in the number directory smart contract is updated to ensure that enrolled cases are excluded from subsequent screening.

[0053] In an embodiment of the present application, the blockchain network is used to determine multiple candidate subjects based on the matching number of each subject, the number of matching fields corresponding to the matching number, and multiple matching degree values, specifically for: Determining a target number in the target number catalog based on the plurality of matching degree values; determining a plurality of initial subjects based on the matching number of each subject, the number of matching fields corresponding to the matching number, and the target number; Calculating a weighted matching degree for each of the initial subjects based on the number of matching fields for each of the initial subjects, the medication time difference, and the semantic similarity between the case text of each of the initial subjects and the feature query condition, wherein the medication time difference represents the difference between the historical medication time and the time of the target trial regimen; Based on the weighted matching scores of each of the initial subjects, a plurality of candidate subjects are determined.

[0054] Through a dynamic determination mechanism based on the target number, combined with the matching number and the number of fields, the initial group of subjects highly correlated with the feature query conditions can be quickly locked in, thereby improving the efficiency of basic screening; by introducing the medication time difference as a key parameter, by quantifying the temporal correlation between historical medication and the trial plan, the impact of drug interference on the trial results can be avoided, significantly improving the scientific nature of the screening; by integrating the semantic similarity analysis of case text and query conditions, the semantic accuracy of feature association is ensured; through the weighted fusion of the number of matching fields, medication time difference, and semantic similarity, the objectivity and comprehensiveness of candidate subject ranking are achieved.

[0055] Figure 2 A flowchart of a blockchain-based clinical trial subject management method is shown.

[0056] like Figure 2 As shown in FIG, a clinical trial subject management method based on blockchain mainly includes: S201, the edge computing node obtains case information of each subject in the medical institution, the case information including structured data and unstructured data; for each subject's case information, performs feature extraction on the case information to obtain feature information, and determines the subject's structured feature information based on the feature information and a preset mapping method; performs homomorphic encryption on the structured feature information of each subject to obtain multiple encrypted feature information; and sends the multiple encrypted feature information to the blockchain network; S202: The edge computing node obtains a target test protocol, generates a feature query condition based on the target test protocol, and sends the feature query condition to the blockchain network to screen multiple target subjects corresponding to the target test protocol based on the blockchain network; S203: The edge computing node obtains, for each target subject, indicator data of the target subject, wherein the indicator data represents vital sign data; determines test information based on the indicator data, wherein the test information includes normal test information or abnormal test information; and determines a follow-up questionnaire and nursing staff information of the target subject based on the test information. S204: The edge computing node obtains the behavior information and test requirement information of each target subject, predicts the follow-up time based on the behavior information and test requirement information of the target subject, generates a follow-up plan based on the follow-up time and test information, and pushes the follow-up questionnaire to the target subject; S205, the blockchain network stores the encrypted feature information of the subjects of each of the medical institutions, and activates the corresponding smart contract template based on the feature query conditions. The smart contract template includes a dynamic matching contract, an informed consent contract, and a dynamic quota contract.

[0057] Figure 3 This is a structural block diagram of an electronic device 300 according to an embodiment of the present application.

[0058] like Figure 3 As shown, the electronic device 300 includes a processor 301 and a memory 302 , and may further include one or more of an information input / information output (I / O) interface 303 , a communication component 304 , and a communication bus 305 .

[0059] The processor 301 is used to control the overall operation of the electronic device 300 to complete all or part of the steps in the blockchain-based clinical trial subject management method described above. The memory 302 is used to store various types of data to support the operation of the electronic device 300. Such data may include, for example, instructions for any application or method operating on the electronic device 300, as well as application-related data. The memory 302 may be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as one or more of static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0060] The I / O interface 303 provides an interface between the processor 301 and other interface modules, which may be a keyboard, a mouse, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 304 is used to test wired or wireless communication between the electronic device 300 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more thereof, may include: a Wi-Fi component, a Bluetooth component, and an NFC component.

[0061] Communication bus 305 may include a path for transmitting information between the aforementioned components. Communication bus 305 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, for example. Communication bus 305 may be divided into an address bus, a data bus, a control bus, and the like.

[0062] The electronic device 300 can be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components, and is used to execute a blockchain-based clinical trial subject management method given in the above embodiment.

[0063] The computer-readable storage medium provided in an embodiment of the present application is introduced below. The computer-readable storage medium described below and the blockchain-based clinical trial subject management method described above can be referenced to each other.

[0064] The present application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the above-mentioned blockchain-based clinical trial subject management method are implemented.

[0065] The computer-readable storage medium may include: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc., which can store program codes.

[0066] The terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0067] The above description is merely a preferred embodiment of the present application and an illustration of the technical principles employed. Those skilled in the art should understand that the scope of application involved in this application is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the aforementioned application concept. For example, a technical solution formed by replacing the above-mentioned features with (but not limited to) technical features with similar functions applied for in this application.

Claims

1. A blockchain-based clinical trial subject management system, characterized by: It includes a blockchain network and multiple edge computing nodes, wherein the edge computing nodes are deployed in various medical institutions, and a smart contract template is deployed on the blockchain network, wherein the smart contract template includes a screening contract; Each of the edge computing nodes includes: a data processing module, configured to obtain case information of each subject in the medical institution, the case information including structured data and unstructured data; perform feature extraction on the case information of each subject to obtain feature information; determine structured feature information of the subject based on the feature information and a preset mapping method; perform homomorphic encryption on the structured feature information of each subject to obtain multiple encrypted feature information; and send the multiple encrypted feature information to the blockchain network; a subject screening module, configured to obtain a target trial protocol, generate a feature query condition based on the target trial protocol, and send the feature query condition to the blockchain network to screen a plurality of target subjects corresponding to the target trial protocol based on the blockchain network; A subject monitoring module is configured to obtain, for each target subject, indicator data of the target subject, the indicator data representing vital sign data; determine test information based on the indicator data, the test information including normal test information or abnormal test information; and determine a follow-up questionnaire and nursing information for the target subject based on the test information; A follow-up management module is configured to obtain, for each target subject, the behavior information and test requirement information of the target subject, predict the follow-up time based on the behavior information and test requirement information of the target subject, generate a follow-up plan based on the follow-up time and test information, and push a follow-up questionnaire to the target subject; The blockchain network is used to store the encrypted characteristic information of the subjects of each medical institution, and to activate the corresponding smart contract template based on the characteristic query conditions to screen multiple target subjects corresponding to the target trial plan.

2. A blockchain-based clinical trial subject management system according to claim 1, characterized in that: The subject monitoring module is configured to determine the follow-up questionnaire of the target subject based on the trial information, specifically to: Determine basic questionnaire information based on the trial information of the target subject, wherein the basic questionnaire information includes target subject identity information, trial stage, trial type, clinical indicators, behavioral characteristics information, and risk level; Performing keyword feature extraction on the basic information of the questionnaire to obtain multiple keywords of the basic information of the questionnaire; Determining the questionnaire topic and content based on the number and preset ratio of each of the keywords; Determining the questionnaire type based on the target subject's identity information, wherein the target subject's identity information includes age, behavioral ability, and cognitive ability, and the questionnaire types include paper questionnaires with different font sizes and word explanation levels and voice questionnaires with different word explanation levels; Based on the questionnaire type, determining a plurality of recommended packaging templates; determining a first weight of each recommended packaging template based on a usage frequency of each recommended packaging template; determining a second weight of each of the recommended packaging templates based on the target subject's historical usage count of each of the recommended packaging templates; determining a third weight of each of the recommended packaging templates based on a dropout rate of a questionnaire corresponding to each of the recommended packaging templates; Determining a disease response coefficient based on the test information of the target subject, and determining a fourth weight of each of the recommended packaging templates according to the disease response coefficient; determining a target packaging template based on the first weight, the second weight, the third weight, and the fourth weight of each of the recommended packaging templates; A follow-up questionnaire is generated based on the questionnaire subject, the questionnaire content, the questionnaire type and the target packaging template.

3. A blockchain-based clinical trial subject management system according to claim 2, characterized in that: The subject supervision module is used to determine the target subject nursing information based on the test information, specifically to: Determining the category level of the target subject based on the test information and a preset K-means clustering model; Determining nursing needs based on the category level of the target subject, wherein the nursing needs include nursing frequency; Based on the historical nursing records and trial stage of the target subject, predicting the nursing time and nursing duration of the target subject in a target time interval; Matching multiple managers at a current medical institution based on the location of the target subject; Based on a preset matching rule and multiple managers of the current medical institution, the caregiver of the target subject is determined, and based on the care time, care duration and caregiver, the care information is determined.

4. A blockchain-based clinical trial subject management system according to claim 3, characterized in that: The subject supervision module determines the caregiver of the target subject based on preset matching rules and multiple managers of the current medical institution, including: Obtaining the current subject information corresponding to each of the managers, the current subject information including the location, nursing time, and nursing duration of the subject to be processed, and the location, nursing time, and nursing duration of the subject currently being managed; Based on the timeliness information of each manager in the historical time period and the current subject information corresponding to each manager, the multiple managers are screened to obtain multiple first managers; Based on the nursing time and nursing duration of the target subject and the current subject information corresponding to each of the first managers, the plurality of first managers are screened to obtain a plurality of second managers; determining a distance coefficient for each second manager based on the location of the target subject and current subject information corresponding to each second manager; determining a first priority value for each of the second managers based on the distance coefficient of each of the second managers; determining, based on the historical nursing information of each second manager, the degree of matching between each second manager and the trial type and trial phase of the target subject; determining a second priority value for each second manager based on a degree of matching between each second manager and the trial type and trial phase of the target subject; determining a third priority value for each second manager based on the number of mistakes made by each second manager in a historical time period; determining a fourth priority value for each second manager based on a degree of matching between each second manager and the subject; A caregiver for the target subject is determined based on the first priority value, the second priority value, the third priority value, and the fourth priority value of each of the second managers.

5. A blockchain-based clinical trial subject management system according to claim 1, characterized in that: The follow-up management module is specifically used to: Obtaining location information of the target subject; Predicting the location of the target subject based on the positioning information of the target subject and the historical behavior data of the target subject; Predicting an initial follow-up time based on the target subject's behavioral information, test requirement information, the target subject's historical behavioral data, and an LSTM model; adjusting the initial follow-up time based on the predicted location of the target subject to obtain a target follow-up time for the target subject; A follow-up plan is generated based on the target follow-up time and the trial information, and the follow-up plan is sent to the target subject.

6. A blockchain-based clinical trial subject management system according to claim 1, characterized in that: The blockchain network is used to activate the corresponding smart contract template based on the feature query condition to screen multiple target subjects corresponding to the target trial plan, specifically for: Obtaining a number directory from a preset number directory smart contract, the number directory including a plurality of numbers, each of the numbers corresponding to a plurality of data names, the data names representing case names, and each of the data names being associated with at least one first keyword set; Determining a matching number for each subject and the number of matching fields corresponding to the matching number based on the encrypted feature information and the number directory of each subject; Determining a second keyword set based on the feature query condition; Matching the second keyword set with the first keyword set associated with each of the data names, and calculating a plurality of matching values; determining a plurality of candidate subjects based on a matching number of each subject, a number of matching fields corresponding to the matching number, and a plurality of matching degree values; Screening multiple candidate subjects based on the inclusion and exclusion criteria of the target trial protocol to generate a target subject list, wherein the inclusion and exclusion criteria represent conditions that must be met by the subjects and conditions that must be excluded; After the target subject list is verified by the node consensus, it is stored in the target subject smart contract and the data association relationship in the number directory smart contract is updated to obtain multiple target subjects.

7. A blockchain-based clinical trial subject management system according to claim 6, characterized in that: The blockchain network is configured to determine multiple candidate subjects based on the matching number of each subject, the number of matching fields corresponding to the matching number, and multiple matching degree values, specifically for: Based on the multiple matching values, a target number is determined in the target number directory; determining a plurality of initial subjects based on the matching number of each subject, the number of matching fields corresponding to the matching number, and the target number; Calculating a weighted matching degree for each of the initial subjects based on the number of matching fields for each of the initial subjects, the medication time difference, and the semantic similarity between the case text of each of the initial subjects and the feature query condition, wherein the medication time difference represents the difference between the historical medication time and the time of the target trial regimen; Based on the weighted matching scores of each of the initial subjects, a plurality of candidate subjects are determined.

8. A blockchain-based clinical trial subject management method, characterized in that: include: The edge computing node obtains case information of each subject in the medical institution, the case information including structured data and unstructured data; performs feature extraction on the case information of each subject to obtain feature information, and determines structured feature information of the subject based on the feature information and a preset mapping method; performs homomorphic encryption on the structured feature information of each subject to obtain multiple encrypted feature information; and sends the multiple encrypted feature information to the blockchain network; The edge computing node obtains a target test plan, generates a feature query condition based on the target test plan, and sends the feature query condition to the blockchain network to screen multiple target subjects corresponding to the target test plan based on the blockchain network; The edge computing node obtains, for each target subject, indicator data of the target subject, wherein the indicator data represents vital sign data; determines test information based on the indicator data, wherein the test information includes normal test information or abnormal test information; and determines a follow-up questionnaire and caregiver information of the target subject based on the test information; The edge computing node obtains the behavior information and test requirement information of each target subject, predicts the follow-up time based on the behavior information and test requirement information of the target subject, generates a follow-up plan based on the follow-up time and test information, and pushes a follow-up questionnaire to the target subject; The blockchain network stores the encrypted characteristic information of the subjects of each medical institution, and activates the corresponding smart contract template based on the characteristic query conditions. The smart contract template includes a screening contract, a dynamic quota contract and an informed consent contract.

9. An electronic device, characterized in that: comprising a processor coupled to a memory; The processor is configured to execute the computer program stored in the memory, so that the electronic device performs the method according to claim 8.

10. A computer-readable storage medium, characterized in that A computer program is stored that can be loaded by a processor and executes a blockchain-based clinical trial subject management method as described in claim 8.

Images