A medical information multi-agent expert thinking chain collaborative reasoning method and system

Abstract

The present invention discloses a collaborative reasoning method and system for a multi-agent expert thinking chain of medical information. First, a sub-expert agent and a summarizer agent are constructed based on a large language model to process medical information and output the processing conclusion of the medical information. Secondly, an evaluator agent is constructed to perform medical information conflict evaluation on the outputs of the sub-expert agent and the summarizer agent. When the conclusion output of the summarizer agent exceeds the medical conflict threshold, the multi-agent debate mechanism is activated to output whether the processing conclusion is credible. Then a coordinator agent is constructed to select the sub-expert agent and the summarizer agent to be called and generate multiple forward reasoning thinking chains. Finally, the evaluator agent evaluates whether the output of the forward reasoning thinking chain exceeds the medical conflict threshold and iteratively updates the output processing conclusion. The present invention can resolve complex problems in medical information processing problems and significantly improve the accuracy and interpretability of reasoning.

Description

Technical Field

[0001] The present invention belongs to the field of medical information processing, and specifically relates to a medical information multi-agent expert thinking chain collaborative reasoning method and system based on a large language model. Background Art

[0002] The medical field is plagued by a vast amount of information resources, including literature, clinical guidelines, and case reports. Efficiently extracting valuable information from this vast amount of data and performing efficient and explainable medical reasoning on complex issues within medical information remain challenging. Traditional AI methods based on rules or small models, lacking the ability to understand context, often fail to accurately parse unstructured medical text or ambiguously expressed meanings. The sole use of large language models for medical information processing also has limitations. Directly using large language models to analyze medical information from multiple perspectives is difficult. When faced with complex medical information reasoning scenarios that require comprehensive analysis of information from diverse sources to reach conclusions, directly using large language models can be difficult to address, including missing key information analysis, low fidelity in the reasoning process, homogeneous output results, and output illusions. Summary of the Invention

[0003] In response to the above problems, the present invention proposes a medical information multi-agent expert thinking chain collaborative reasoning method and system, focusing on using advanced large language model reasoning technology to improve the accuracy and interpretability of medical information processing. This method combines the powerful natural language processing capabilities of the large language model with the collaborative working advantages of the multi-agent system. Multiple sub-expert agents with different medical professional backgrounds are designed to collaborate with each other and jointly participate in the processing of medical information and the reasoning tasks of complex medical problems. By simulating the working mode of multidisciplinary medical teams in the real world, the present invention can more comprehensively consider various complex factors in medical information processing, thereby achieving more accurate and interpretable processing of complex medical problems. In addition, the application of the large language model enables the system to understand and process a large amount of medical literature and clinical cases, further enhancing its applicability and reliability in medical information processing. The present invention is particularly suitable for complex medical information reasoning scenarios that require the integration of multiple information sources and professional knowledge, and aims to promote the development of the field of accurate processing of medical information and interpretable reasoning of complex medical problems.

[0004] In order to solve the above technical problems, the technical solution of the present invention is:

[0005] A medical information multi-agent expert thinking chain collaborative reasoning method includes the following steps:

[0006] S1. Build sub-expert agents based on the large language model. Each sub-expert agent can call upon an external medical knowledge base to process medical information and output results. These sub-expert agents include a laboratory examination analysis sub-expert agent, an imaging report analysis sub-expert agent, and a medical record and history analysis sub-expert agent.

[0007] S2. Construct a summarizer agent based on the large language model. The summarizer agent receives the output results of each sub-expert agent and uses the text summarization and generalization capabilities of the large language model to obtain the processing conclusions of the medical information.

[0008] S3. Build an evaluator agent based on the large language model. The evaluator agent extracts the temporal state of the medical information to be processed and reconstructs the temporal features of the medical indicators in the medical information based on the temporal state. Leveraging the contextual understanding and text generation capabilities of the large language model, the evaluator agent generates a medical conflict judgment strategy based on the initial values ​​of the temporal features of the medical indicators through prompt word engineering. The evaluator agent then evaluates the outputs of each sub-expert agent and summarizer agent for medical information conflicts and determines whether they exceed the set medical conflict threshold.

[0009] S4. Leveraging the contextual understanding capabilities of the large language model, the evaluator agent iterates on its medical conflict judgment strategy. Based on the temporal characteristics of medical indicators, the evaluator agent splits the medical information into N chronologically ordered medical information nodes, each storing medical information over a period of time. Each medical information node serves as both an input and a verification value for the evaluator agent's medical conflict judgment strategy. The strategy is adjusted based on the deviation between the evaluator agent's generated medical conflict judgment and the verification value, until the evaluator agent completes verification of all medical information nodes and generates a final medical conflict judgment strategy.

[0010] S5. When the Evaluator Agent determines that the Summarizer Agent's conclusion exceeds the set medical conflict threshold, the multi-agent debate mechanism is activated to determine whether the conclusion is credible. The multi-agent debate mechanism then re-inputs the conclusion output to each sub-expert agent using a prompt word template, requiring each sub-expert agent to evaluate the output of the other sub-expert agents. Leveraging the contextual understanding and generation capabilities of the large language model, the multi-agent debate mechanism conducts up to t rounds of dynamic peer-to-peer questioning and debate between the sub-expert agents. The debate outputs of each sub-expert agent are then re-inputted to the Summarizer Agent to generate a new conclusion output.

[0011] S6. Construct a coordinator agent based on a large language model. The role function prompt word template of the coordinator agent stipulates that the coordinator agent can obtain the name list of all sub-expert agents and the name list of uncalled sub-expert agents, and can call the output results of all sub-expert agents.

[0012] S7. Leveraging the decision-making and planning capabilities of the large language model, the coordinator agent autonomously selects multiple sub-expert agents and summarizer agents to invoke based on the input medical information to be processed and plans the invocation sequence. The forward reasoning chain tasks of each defined sub-expert agent and summarizer agent are sequentially invoked in the order planned by the coordinator agent, constructing a multi-agent forward reasoning chain for medical information processing. The coordinator agent can simultaneously plan the invocation of multiple sub-expert agents, generating multiple forward reasoning chains for medical information processing.

[0013] S8. The names and outputs of all sub-expert agents and summarizer agents in the forward reasoning thinking chain will be added to the backward reflective agent stack in the order of the coordinator agent's call. After the forward reasoning thinking chain completes the processing of medical information and gives the forward reasoning thinking chain output, the coordinator agent calls the evaluator agent to evaluate whether the forward reasoning thinking chain output exceeds the set medical conflict threshold.

[0014] S9. When the evaluator agent evaluates that the output of the forward reasoning chain exceeds the set medical conflict threshold, the backward reflection mechanism is activated and the output processing conclusion is iteratively updated until it is lower than the medical conflict threshold. The evaluator agent will evaluate the output of each sub-expert agent and summarizer agent in the stack one by one in the order of the backward reflection agent stack to see if it exceeds the set medical conflict threshold. After finding the agent that exceeds the medical conflict threshold, it will generate targeted modification suggestions for each sub-expert agent based on the final medical conflict judgment strategy, and require the agent that exceeds the medical conflict threshold to iteratively output the result until it passes the medical conflict judgment evaluation of the evaluator agent or reaches the maximum number of output iterations.

[0015] Preferably, in step S1, the method for constructing a sub-expert agent based on the large language model is:

[0016] In the collaborative reasoning framework of the multi-agent expert thinking chain of medical information, each sub-expert agent is constructed based on the prompt word engineering of the large language model into a test and examination analysis sub-expert agent, an imaging report analysis sub-expert agent, and a medical record and history analysis sub-expert agent. Each sub-expert agent is constructed through the following four steps:

[0017] S1-1. The prompt word engineering based on the large language model defines the forward reasoning thinking chain task in each sub-expert intelligent agent. The forward reasoning thinking chain task includes the name of the sub-expert intelligent agent and the functional prompt word template. The output content, output format and basic specifications of the forward reasoning thinking chain are clearly defined in the functional prompt word template of each sub-expert intelligent agent.

[0018] S1-2. Each sub-expert agent is allowed to autonomously access an external medical knowledge base before outputting its forward reasoning chain and perform a top-k search of the knowledge base based on the input medical information to be processed. The retrieved text is provided to the sub-expert agent as a reference for medical information processing. This operation is autonomously selected by the sub-expert agent and depends on the complexity of the medical information to be processed and the relevance of the external medical knowledge base.

[0019] S1-3. During the forward reasoning chain output process, the sub-expert agent can also autonomously select and call upon the forward reasoning chain outputs of other sub-expert agents as reference text for medical information processing, and provide critical medical judgments and detailed recommendations. This is intended to make communication between the sub-expert agents more constructive and to allow for mutual verification of the medical information reasoning process and results.

[0020] The large language model-based prompt word engineering defines a backward reflection mechanism template for each sub-expert agent. This template includes the original medical information to be processed, the output set of other sub-expert agents, the summarized processing output of the summarizer agent, and the textual description of medical conflicts exceeding the threshold generated by the evaluator agent. It also defines instructions for the sub-expert agent to iterate its own output. When the backward reflection mechanism is activated, the sub-expert agent regenerates the medical information processing results based on its own backward reflection mechanism template.

[0021] Preferably, in step S2, the method for constructing a summarizer agent based on the large language model is:

[0022] Based on the large language model's prompt word project, a prompt word template for the summarizer agent's role function was constructed. The summarizer agent was able to access the output of each sub-expert agent. The prompt word project configured the summarizer agent to focus on key summary points in medical information processing, including the extraction of medical professional terms, the extraction and organization of professional term relationships, and a strict upper limit on the output word count. Based on these prompt word project functional settings, the large language model's text summarization capabilities were leveraged to obtain conclusions on the processing of medical information.

[0023] Preferably, in step S3, the method for constructing an evaluator agent based on a large language model is:

[0024] Based on the large language model's prompt word engineering, a role function prompt word template is constructed for the evaluator agent. The evaluator agent first extracts the temporal state of the medical information to be processed according to the time tags in the medical information and reconstructs the temporal characteristics of the medical indicators in the medical information based on the temporal state. Leveraging the contextual understanding and text generation capabilities of the large language model, prompt word engineering is used to enable the large language model to generate the evaluator agent's initial medical conflict judgment strategy based on the initial value of the temporal state. The initial medical conflict judgment strategy is customized for each sub-expert agent and summarizer agent. The prompt word engineering constrains the initial medical conflict judgment strategy to set a medical conflict threshold for each sub-expert agent's analysis domain. The evaluator agent then evaluates the medical information conflict based on the output of each sub-expert agent and summarizer agent based on the initial medical conflict judgment strategy and determines whether it exceeds the set medical conflict threshold.

[0025] Preferably, in step S4, the method for the assessor agent to iterate the medical conflict judgment strategy is:

[0026] Leveraging the contextual understanding capabilities of a large language model, the evaluator agent's medical conflict judgment strategy is iterated. The evaluator agent splits medical information into N chronologically ordered medical information nodes based on the temporal characteristics of medical indicators. Each medical information node stores medical information over a time span. Each medical information node serves as both an input and a validation value for the evaluator's medical conflict judgment strategy. Based on the deviation between the evaluator's generated medical conflict judgment and the validation value, the medical conflict judgment strategies of each sub-expert agent and summarizer agent are adjusted. Adjustments to the medical conflict judgment strategy are iteratively constrained through prompt word engineering using the large language model. The deviation between the medical conflict judgment and the validation value is analyzed based on factors such as long- and short-term medical temporal dependencies, multi-variable relationships, medical feature selection methods, and data sparsity. New medical conflict judgment strategies are iteratively generated until the evaluator agent completes validation of all medical information nodes and generates a final medical conflict judgment strategy.

[0027] Preferably, in step S5, the method of the multi-agent debate mechanism based on the large language model is:

[0028] S5-1. The evaluator agent determines whether the treatment conclusion exceeds the set medical conflict threshold based on the medical conflict assessment method. If the set medical conflict threshold is exceeded, the multi-agent debate mechanism is activated. The context understanding ability of the large language model is used to disassemble the core medical information processing contradiction problem to be solved, and the iterative limit of the debate round is specified to build a meta-prompt template framework for the multi-agent debate mechanism. There are N sub-agents in the multi-agent debate mechanism. Participate in the debate framework. In each debate iteration, speak in order and express your own views based on the previous debate history H, that is, D i (H) = h.

[0029] S5-2. The evaluator agent combines all historical debate information to perform a medical conflict assessment for each round of debate. It then uses the medical conflict threshold to determine consensus and the debate termination mechanism. Specifically, after all sub-expert agents have completed their debates, the evaluator agent determines whether a correct output result was obtained. If an output result that does not exceed the medical conflict threshold within the specified number of rounds is not obtained, the debate result is marked as untrusted in the output set. The coordinator agent and sub-expert agents are informed of the untrustworthiness of the output result when attempting to use it as a reference in the medical information reasoning process. Any forward reasoning chain within the multi-agent debate mechanism that exceeds the evaluator agent's medical conflict threshold will inevitably activate the backward reflection mechanism.

[0030] Preferably, in step S6, the method for constructing a coordinator agent based on the large language model is:

[0031] Based on the large language model prompt word project, a role function prompt word template for the coordinator agent is constructed. This template specifies that the coordinator agent has access to a list of all sub-expert agents, including a list of uncalled sub-expert agents, and can call the output of all sub-expert agents. After initializing the coordinator agent, leveraging the decision-making and planning capabilities of the large language model, the coordinator agent autonomously selects and calls each sub-expert agent and summarizer agent based on the input medical information to be processed. As each sub-expert agent outputs, the coordinator agent synchronously updates the sub-expert agent's medical information processing output and integrates the prompt words with the previous outputs of other sub-expert agents to generate a new output set. This process not only ensures that each decision made by the coordinator agent is based on the latest and most comprehensive information, but also gradually optimizes the performance of the entire system through continuous information accumulation. This, in particular, provides more accurate and interpretable medical reasoning processes and results when faced with complex medical information reasoning problems.

[0032] Preferably, in step S7, the method for constructing a multi-agent medical information forward reasoning thinking chain based on a large language model is:

[0033] Leveraging the decision-making planning capabilities of the large language model, the coordinator intelligence plans the call sequence based on the new output result set, i.e., forward reasoning chain. Forward reasoning chain is a sequential decision-making process in which the set of sub-expert agents is regarded as an action space, the medical information to be processed is defined as P, and a set of predefined sub-expert agents is defined as Where n is the total number of sub-expert agents, φ i represents the configuration of the i-th sub-expert agent, Denotes the i-th sub-expert agent. Each sub-expert agent is associated with an optional knowledge base and a prompt template. The forward reasoning chain tasks of the defined sub-expert agents and summarizer agent are sequentially invoked in the order planned by the coordinator agent, constructing a multi-agent forward reasoning chain for medical information processing. The coordinator agent can simultaneously plan the invocation of multiple sub-expert agents, generating multiple forward reasoning chains for medical information processing.

[0034] Preferably, in step S8, the method for evaluating the output result of the forward reasoning thinking chain based on the large language model is:

[0035] The names and outputs of all sub-expert agents and summarizers in the forward reasoning chain are added to the backward reflection agent stack in the order in which they were called by the coordinator. After the forward reasoning chain completes its medical information processing and generates its output, the coordinator calls the evaluator to assess whether the forward reasoning chain output exceeds the set medical conflict threshold. When multiple forward reasoning chains exist simultaneously, the evaluator prioritizes the evaluation of all forward reasoning chain outputs. If any forward reasoning chain output does not exceed the set medical conflict threshold, that forward reasoning chain output is directly used as the final output. If all forward reasoning chain outputs exceed the set medical conflict threshold, the backward reflection mechanism is activated for each forward reasoning chain one by one. Forward reasoning chains containing untrusted results in the S4 multi-agent debate are placed last among all forward reasoning chains, and the backward reflection mechanism is activated.

[0036] Preferably, in step S9, the method for constructing a backward reflection mechanism based on a large language model is:

[0037] When the evaluator agent assesses that the output of a forward reasoning chain exceeds the set medical conflict threshold, the backward reflection mechanism of that forward reasoning chain is activated. The evaluator agent will evaluate the outputs of each sub-expert agent and summarizer agent in the S8 backward reflection agent stack, one by one, to see if they exceed the set medical conflict threshold. After identifying a sub-expert agent that exceeds the medical conflict threshold, the evaluator agent generates targeted modification suggestions for the sub-expert agent based on the final medical conflict judgment strategy. The sub-expert agent that exceeds the medical conflict threshold is required to iteratively output its results according to the final medical conflict judgment strategy until it passes the evaluator agent's medical conflict judgment assessment or reaches the maximum number of output iterations. The sub-expert agent that exceeds the medical conflict threshold will iterate its output results using the defined sub-expert agent backward reflection mechanism template until it passes the evaluator agent's medical conflict threshold assessment or reaches the maximum number of output iterations. When the backward reflection mechanism is activated for a forward reasoning chain, the backward reflection mechanisms of all other forward reasoning chains will enter a blocked state. When the output result of the backward reflection mechanism of a forward reasoning thinking chain is lower than the medical conflict threshold of the evaluator agent within the maximum number of iterations, the output result is taken as the final output result.

[0038] A medical information multi-agent expert thinking chain collaborative reasoning system, including the following modules:

[0039] The conclusion output module constructs a sub-expert agent and a summarizer agent based on the large language model. The sub-expert agent calls the external medical knowledge base to process the medical information, and the summarizer agent uses the text summarization and generalization capabilities of the large language model to output the processing conclusion of the medical information.

[0040] The evaluation module constructs an evaluator agent based on the large language model, performs medical information conflict evaluation on the outputs of each sub-expert agent and summarizer agent, determines whether the medical conflict threshold is exceeded, and generates a medical conflict judgment strategy.

[0041] Coordination module: builds a coordinator agent based on a large language model, autonomously selects multiple sub-expert agents and summarizer agents to be called based on the input medical information to be processed, plans the calling order, and generates multiple forward reasoning thinking chains.

[0042] Furthermore, the evaluation module also includes: when it is judged that the conclusion output of the summarizer intelligent agent exceeds the medical conflict threshold, the multi-agent debate mechanism is activated to output whether the processing conclusion is credible; the evaluator intelligent agent evaluates whether the output of the forward reasoning thinking chain exceeds the medical conflict threshold, the backward reflection mechanism is activated, and the output processing conclusion is iteratively updated.

[0043] Beneficial effects of the present invention: The medical information multi-agent expert thinking chain collaborative reasoning method provided by the present invention can achieve the same and more flexible medical information reasoning and decision-making functions as the decision-making agent without any model parameter training. In the field of medical information processing, the expert selection strategy constructed by the present invention can efficiently and accurately call the expert knowledge in the relevant field according to the type of medical information, so that the large language model can effectively call and integrate the professional knowledge from multiple sub-expert agents. In the construction of the medical information reasoning framework, the present invention provides a forward reasoning thinking chain and a backward reflection mechanism. This method can parse complex problems in medical information processing problems, and realize the evaluation of the intermediate process of reasoning through the multi-agent debate mechanism, which can significantly improve the accuracy and explainability of reasoning. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0045] Figure 1 Flow chart for the implementation of the present invention;

[0046] Figure 2 This is a diagram of the collaborative reasoning framework of multi-agent expert thinking chain in medical information. DETAILED DESCRIPTION

[0047] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features therein can be combined with each other. In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc. is based on the orientation or positional relationship shown in the accompanying drawings, which is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first", "second", etc. may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, the meaning of "multiple" is two or more.

[0048] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; mechanical connections or electrical connections; direct connections or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0049] The present invention provides a medical information multi-agent expert thinking chain collaborative reasoning method based on a large language model. Figure 1 and Figure 2 As shown, the following steps are included:

[0050] S1: In the medical multi-agent collaborative reasoning framework, each sub-expert agent is assigned a medical information processing role, namely the test report analysis sub-agent, the imaging report analysis sub-agent, and the medical record analysis sub-agent. They are constructed based on the following four steps, with the main approach being:

[0051] S1-1. The prompt word engineering based on the large language model defines the forward reasoning thinking chain task in each sub-expert intelligent agent. The forward reasoning thinking chain task includes the name of the sub-expert intelligent agent and the functional prompt word template. The output content, output format and basic specifications of the forward reasoning thinking chain are clearly defined in the functional prompt word template of each sub-expert intelligent agent.

[0052] S1-2. Each sub-expert agent is allowed to autonomously access an external medical knowledge base before outputting its forward reasoning chain and perform a top-k search of the knowledge base based on the input medical information to be processed. The retrieved text is provided to the sub-expert agent as a reference for medical information processing. This operation is autonomously selected by the sub-expert agent and depends on the complexity of the medical information to be processed and the relevance of the external medical knowledge base.

[0053] S1-3. During the forward reasoning chain output process, the sub-expert agent can also autonomously call upon the forward reasoning chain output of other sub-expert agents as a reference for medical information processing, and provide critical medical judgments and detailed recommendations. This is intended to facilitate more constructive communication between sub-expert agents and allow for mutual verification of medical information reasoning processes and results.

[0054] S1-4. The large language model-based prompt word engineering defines a backward reflection mechanism template for each sub-expert agent. This template includes the original medical information to be processed, the output set of other sub-expert agents, the summarized processing output of the summarizer agent, and the textual description of medical conflicts exceeding the threshold generated by the evaluator agent. It also defines instructions for the sub-expert agent to iterate its own output. When the backward reflection mechanism is activated, the sub-expert agent regenerates the medical information processing results based on its own backward reflection mechanism template.

[0055] The following is a specific design template for the prompt words of the medical examination report analysis sub-expert agent based on the large language model:

[0056] "You are an expert in analyzing medical test reports. Based on the provided medical text and test analysis reports, you provide professional medical information opinions.

[0057] The following external medical information and guidelines can be used as reference:

[0058] Based on the problem description (Pr), provide your opinion on how to process medical information to help solve this problem. The judgments (C) given by other medical experts are as follows. Please refer to them carefully.

[0059] Your professional analysis based on the medical examination report is:

[0060] S2: The main approach to building a summarizer agent based on a large language model is:

[0061] The role of the Summarizer sub-agent is to summarize the summaries of all medical judgments provided by the selected sub-expert agents, carefully analyze the medical judgments, and generate a final answer. The prompt template of the Summarizer sub-agent may vary depending on the specific type of final answer required. Below is an example of a prompt template for the Summarizer sub-agent when the goal is to obtain a procedure:

[0062] The following is a specific design template for the prompt words of the summarizer sub-agent based on the large language model:

[0063] You need to summarize all the medical information judgments and opinions of the expert and generate a summary that incorporates the insights and recommendations provided by the selected expert. The questions related to medical information are as follows:

[0064] The expert's opinion and medical judgment C are as follows:

[0065] Based on professional methods of medical information processing, the opinions of experts are efficiently summarized:"

[0066] S3: The main approach to building an evaluator agent based on a large language model is:

[0067] Based on the large language model prompt word engineering, a role function prompt word template for the evaluator agent is constructed. The evaluator agent can call the external medical knowledge base, and based on the external medical knowledge base and the set medical conflict judgment method, perform medical information conflict evaluation on the outputs of each sub-expert agent and the summarizer agent and judge whether the medical conflict threshold is exceeded.

[0068] The following is a specific prompt word design template for the evaluator sub-agent based on a large language model:

[0069] You are a medical informatics expert. You will be responsible for analyzing the core medical information in a text passage and comparing it with entries in an external medical knowledge base to assess any conflicts in the medical information.

[0070] You will receive a text containing medical information. Please first extract the time status of the medical information based on the time tag in the information and output it strictly in JSON format. Use this time status as the key value and the medical information under the corresponding time tag as the value to organize and reconstruct the medical information.

[0071] The following are the medical expert agents you will evaluate: sub-expert agent E;

[0072] Please generate initial medical conflict judgment strategies for each of their medical sub-fields;

[0073] Your output is:"

[0074] S4: The evaluator agent iterates the medical conflict judgment strategy based on the large language model. The main approach is:

[0075] The evaluator agent splits medical information into N chronologically ordered medical information nodes based on the temporal characteristics of medical indicators. Each medical information node stores medical information over a period of time. Each medical information node serves as the input and verification value for the evaluator's medical conflict judgment strategy. The medical conflict judgment strategies for each sub-expert agent and the summarizer agent are adjusted based on the deviation between the evaluator's generated medical conflict judgment and the verification value.

[0076] The following is a specific prompt word design template for the evaluator sub-agent iterative medical conflict judgment strategy:

[0077] "You are a medical information evaluation expert, skilled at assessing the degree of medical conflict in medical texts. Now you are required to make a medical conflict judgment based on the analysis of sub-expert agent E. The medical conflict judgment strategy JP you previously used n The judgment J at the previous medical information node is as follows:

[0078] If there is a deviation from the verification value V, please analyze the cause of the deviation from the aspects of long-term and short-term medical time dependence, multi-medical variable relationship, medical feature selection method and data sparsity, and iteratively generate a new medical conflict judgment strategy JP n+1 ”

[0079] S5: Build a multi-agent medical debate mechanism based on a large language model. The main approach is:

[0080] Construct a system meta-prompt template. Use meta-prompts to analyze the core medical information conflicts to be resolved, the sub-expert agents participating in the debate mechanism, debate round iteration limits, and other requirements. The high-level arbitrator agent will combine all debate history information to arbitrate the results of each round of debate. It is given a judgment mechanism to determine whether consensus has been reached and terminate the debate. That is, in this iteration, after all experts have completed the debate, the high-level arbitrator agent determines whether the correct solution can be obtained; or if a medical consensus with sufficient confidence cannot be reached within the limited rounds, the debate result will be placed at a low priority in the forward reasoning chain. The following is a specific prompt word design template for the arbitrator agent in the medical professional debate mechanism:

[0081] You are the arbitrator of this debate. This debate on medical information processing will feature the following experts. They will present their arguments and engage in a lively discussion. First, based on the content and historical information from each round of debate, please analyze and describe the core disagreements and medical conflicts. Then, combine the opinions of other expert agents and your medical expertise to form a comprehensive judgment.

[0082] The following are external medical information and guidelines that can be used as reference.

[0083] Other medical experts gave the following judgment C.

[0084] What is your analysis of the conflicts and contradictions in this medical information?

[0085] S6: The main approach of the language model construction coordinator agent is:

[0086] The coordinator agent's strategy is modeled, and sub-expert agents are selected based on the agent's strategy function. This sub-expert agent selection strategy can be translated into the design of a prompt template, and the optimal strategy is achieved through prompt word engineering. The coordinator high-level agent is the core of the entire reasoning framework and requires more complex prompt design than other sub-expert agents. The following is a specific prompt word design template for the coordinator high-level agent:

[0087] You are the leader and coordinator of a team of medical information processing experts. Now, you need to coordinate all the experts you manage so that they can work together to solve the problem. Next, you will be given specific medical texts and various medical information. First, you need to coordinate entity information recognition, entity relationship extraction, and implicit relationship extraction of medical texts. After that, your task is to select the experts you think are most suitable to seek professional insights and suggestions from various medical departments. The description of the problem is presented as follows:

[0088] Based on the capabilities of different experts and the current state of the problem-solving process, you need to decide which expert to consult next. The capabilities of the experts are described as follows:

[0089] Experts who have expressed their opinions include: EL You must select from the list of experts above. Please note that you will need to complete the entire workflow in the remaining Rs steps. Now, carefully consider your choice and provide your reasons.

[0090] S7: The main approach to constructing a forward reasoning chain for medical information based on a large language model is:

[0091] Forward medical information reasoning thinking construction. In this stage, the coordinator high-level intelligent body will select sub-expert intelligent bodies in turn to make medical judgments and express opinions. Forward reasoning thinking construction is expressed as a sequential decision-making process, in which the expert set is regarded as the action space, and the input problem description is defined as P, and a set of predefined experts is defined as Where n is the total number of sub-expert agents, φ i Denote the i-th sub-expert agent. Each sub-expert agent is associated with an optional knowledge base and a prompt template. Denote the set of medical judgments at the t-th reasoning step as Ct, and define the state in the following formula:

[0092] S8: The main approach to evaluating the output of the forward reasoning chain based on the large language model is:

[0093] The names and outputs of all sub-expert agents and summarizers in the forward reasoning chain are added to the backward reflection agent stack in the order in which they were called by the coordinator. After the forward reasoning chain completes its medical information processing and generates its output, the coordinator calls the evaluator to assess whether the forward reasoning chain output exceeds the set medical conflict threshold. When multiple forward reasoning chains exist simultaneously, the evaluator prioritizes the evaluation of all forward reasoning chain outputs. If any forward reasoning chain output does not exceed the set medical conflict threshold, that forward reasoning chain output is directly used as the final output. If all forward reasoning chain outputs exceed the set medical conflict threshold, the backward reflection mechanism is activated for each forward reasoning chain one by one. Forward reasoning chains containing untrusted results in the S5 multi-agent debate are placed last among all forward reasoning chains, and the backward reflection mechanism is activated.

[0094] S9: The main approaches to building a backward reflection mechanism for medical information based on a large language model are:

[0095] The backward reflection reasoning mechanism in the expert thinking chain enables the system to utilize external feedback and adjust the collaboration between sub-expert agents and evaluate the results of problem solving based on their feedback. The trajectory of the sub-expert agents selected in sequence is defined as where i t represents the index of the expert selected in step t. The backward reflective reasoning process receives external feedback r raw Initially, this feedback is usually provided by the program execution environment. This process can be expressed as The initial signal is obtained from the evaluation of the original external feedback (r0,sr0)←evaluate(r raw ), where r0 is a Boolean signal indicating whether the backward reflection process needs to continue, and sr0 represents the natural language summary of the feedback, which is used to locate the error in the backward reflection process. If it is correct, r0 is set to false and the whole reasoning process is terminated. Otherwise, the expert thinking chain starts the backward self-reflection reasoning mechanism to update the output.

[0096] After the backward reflection mechanism obtains the output of the forward reasoning, an evaluator sub-agent collects feedback and converts it into natural language feedback to evaluate its correctness. If the solution is considered incorrect, the system enters the backward reflection phase. In this phase, the sub-expert agents are removed from the stack and iteratively questioned in reverse order. After finding the agent that exceeds the medical conflict threshold, targeted modification suggestions are generated for each sub-expert agent based on the final medical conflict judgment strategy. The iterative improvement of the forward thinking construction and backward reflection steps is repeated until a satisfactory solution is obtained or the maximum number of iterations T is reached. The output includes the reasoning process and results of the decomposition and analysis of complex medical information problems.

[0097] A medical information multi-agent expert thinking chain collaborative reasoning system, including a conclusion output module, an evaluation module, and a coordination module:

[0098] The conclusion output module constructs a sub-expert agent and a summarizer agent based on the large language model. The sub-expert agent calls the external medical knowledge base to process the medical information, and the summarizer agent uses the text summarization and generalization capabilities of the large language model to output the processing conclusion of the medical information.

[0099] The evaluation module constructs an evaluator agent based on the large language model, performs medical information conflict evaluation on the outputs of each sub-expert agent and summarizer agent, determines whether the medical conflict threshold is exceeded, and generates a medical conflict judgment strategy.

[0100] Coordination module: builds a coordinator agent based on a large language model, autonomously selects multiple sub-expert agents and summarizer agents to be called based on the input medical information to be processed, plans the calling order, and generates multiple forward reasoning thinking chains.

[0101] Furthermore, the evaluation module also includes: when it is judged that the conclusion output of the summarizer intelligent agent exceeds the medical conflict threshold, the multi-agent debate mechanism is activated to output whether the processing conclusion is credible; the evaluator intelligent agent evaluates whether the output of the forward reasoning thinking chain exceeds the medical conflict threshold, the backward reflection mechanism is activated, and the output processing conclusion is iteratively updated.

[0102] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited to the described embodiments. It will be apparent to those skilled in the art that various changes, modifications, substitutions, and variations of these embodiments, including components, without departing from the principles and spirit of the present invention are still within the scope of protection of the present invention.

Claims

1. A medical information multi-agent expert thinking chain collaborative reasoning method, characterized by: The following steps are involved: S1. Build sub-expert agents and summarizer agents based on the large language model. The sub-expert agents process the medical information, and the summarizer agent outputs the processing conclusion of the medical information. S2. Build an evaluator agent based on the large language model to evaluate medical information conflicts based on the outputs of each sub-expert agent and the summarizer agent, determine whether the medical conflict threshold is exceeded, and generate the final medical conflict judgment strategy through iteration. S3. When the evaluator agent determines that the conclusion output of the summarizer agent exceeds the set medical conflict threshold, the multi-agent debate mechanism is activated to output whether the conclusion is credible; The multi-agent debate mechanism re-inputs the conclusion output to each sub-expert agent through the prompt word template, requiring each sub-expert agent to evaluate the output of other sub-expert agents. Based on the contextual understanding and generation capabilities of the large language model, a point-to-point dynamic questioning and debate is carried out between sub-expert agents for up to t rounds. The debate output of each sub-expert agent is re-inputted to the summarizer agent to generate a new conclusion output. S4. Build a coordinator agent based on the large language model, select the sub-expert agents and summarizer agents to be called, plan the calling sequence, and generate multiple forward reasoning chains; S5. The evaluator agent evaluates whether the output of the forward reasoning chain exceeds the medical conflict threshold and iteratively updates the output processing conclusion as follows: The names and outputs of all sub-expert agents and summarizer agents in the forward reasoning thinking chain are added to the backward reflective agent stack in the order in which the coordinator agent calls them. After the forward reasoning thinking chain completes the processing of medical information and gives the forward reasoning thinking chain output, the coordinator agent calls the evaluator agent to evaluate whether the forward reasoning thinking chain output exceeds the set medical conflict threshold; When the evaluator agent evaluates that the output of the forward reasoning thinking chain exceeds the set medical conflict threshold, the backward reflection mechanism is activated, and the output processing conclusion is iteratively updated until it is lower than the medical conflict threshold; the evaluator agent will evaluate the output of each sub-expert agent and summarizer agent in the stack one by one in the order of the backward reflection agent stack to see whether it exceeds the set medical conflict threshold. After finding the sub-expert agent that exceeds the medical conflict threshold, it will generate targeted modification opinions for each sub-expert agent based on the final medical conflict judgment strategy, and require the sub-expert agent that exceeds the medical conflict threshold to iteratively output the results until it can pass the medical conflict judgment evaluation of the evaluator agent or reaches the maximum number of output iterations.

2. A medical information multi-agent expert thinking chain collaborative reasoning method according to claim 1, characterized in that: The specific implementation process of step S1 is as follows: S1-1. Based on the large language model prompt word project, a forward reasoning thought chain task is defined in each sub-expert agent. The forward reasoning thought chain task includes the name of the sub-expert agent and a functional prompt word template. The output content of the forward reasoning thought chain is clearly defined in the functional prompt word template of each sub-expert agent. S1-2. Each sub-expert agent autonomously accesses an external medical knowledge base before outputting the forward reasoning thought chain and performs a Top-k search on the knowledge base based on the input medical information to be processed. The retrieved text will be provided to the sub-expert agent as a reference text for medical information processing; S1-3. During the forward reasoning thought chain output process of the sub-expert agent, the sub-expert agent can also autonomously choose to call the forward reasoning thought chain output results of other sub-expert agents that have already output as reference text for medical information processing and provide medical judgments and suggestions; S1-4. Based on the large language model prompt word project, a role function prompt word template of the summarizer agent is constructed. The summarizer agent can obtain the output of each sub-expert agent. The summarizer agent is set by the prompt word project to focus on the key points of summary in medical information processing, including the extraction of medical professional terms, the extraction and sorting of professional term relationships.

3. A medical information multi-agent expert thinking chain collaborative reasoning method according to claim 2, characterized in that: The specific implementation process of the evaluator agent in step S2 is as follows: Based on the large language model prompt word engineering, a role function prompt word template for the evaluator agent is constructed. The evaluator agent first extracts the temporal state of the medical information to be processed according to the time tags in the medical information. Then, using the large language model through prompt word engineering, the large language model generates the evaluator agent's initial medical conflict judgment strategy based on the initial value of the temporal state. The initial medical conflict judgment strategy constrains each sub-expert agent and summarizer agent through prompt word engineering, and sets a medical conflict threshold for the analysis field of each sub-expert agent; then the evaluator agent evaluates the medical information conflict based on the output of each sub-expert agent and summarizer agent based on the initial medical conflict judgment strategy and determines whether it exceeds the set medical conflict threshold.

4. A medical information multi-agent expert thinking chain collaborative reasoning method according to claim 3, characterized in that: The specific implementation of the final medical conflict judgment strategy is as follows: The contextual understanding capability of the large language model is used to iterate the medical conflict judgment strategy of the evaluator agent. The evaluator agent splits the medical information into N chronologically ordered medical information nodes based on the temporal characteristics of medical indicators. Each medical information node stores medical information within a time span. Each medical information node is used as the input value and verification value of the evaluator agent's medical conflict judgment strategy in turn. Based on the deviation between the medical conflict judgment generated by the evaluator agent and the verification value, the medical conflict judgment strategy for each sub-expert agent and summarizer agent is adjusted; medicine The adjustment of the conflict judgment strategy is iteratively constrained through the prompt word engineering of the large language model. The deviation between the medical conflict judgment and the verification value is analyzed based on the long- and short-term medical time dependencies, the relationship between multiple medical variables, the medical feature selection method and data sparsity, and a new medical conflict judgment strategy is iteratively generated until the evaluator agent completes the verification of all medical information nodes and generates the final medical conflict judgment strategy.

5. A medical information multi-agent expert thinking chain collaborative reasoning method according to claim 4, characterized in that: The specific implementation process of the multi-agent debate mechanism is as follows: S3-1. The evaluator agent determines whether the treatment conclusion exceeds the set medical conflict threshold based on the medical conflict assessment method. If the set medical conflict threshold is exceeded, the multi-agent debate mechanism is activated; Leveraging the contextual understanding capabilities of large language models, we break down the paradoxical issues in medical information processing, set iterative limits on debate rounds, and construct a meta-prompt template framework for multi-agent debate mechanisms. There are N sub-expert agents in the multi-agent debate mechanism Participate in the debate framework; in each debate iteration, speak in order and express your own views based on the previous debate history H, that is, ; S3-2. The evaluator agent combines all historical debate information to conduct a medical conflict assessment for each round of debate results, and judges the consensus reached and the judgment mechanism for terminating the debate based on the medical conflict threshold. That is, in this round of debate, after all sub-expert agents have completed the debate, the evaluator agent decides whether the correct output result can be obtained; or if the output result that does not exceed the medical conflict threshold is not obtained in the limited rounds, the debate result is set as an untrusted result in the output set, and the coordinator agent and the sub-expert agent are informed that the output result is untrustworthy when trying to call it as a reference for the medical information reasoning process; the forward reasoning thinking chain contained in the multi-agent debate mechanism that exceeds the output result of the evaluator agent's medical conflict threshold will inevitably activate the backward reflection mechanism.

6. A medical information multi-agent expert thinking chain collaborative reasoning method according to claim 5, characterized in that: The coordinator agent is specifically implemented as follows: A role function prompt word template for the coordinator agent is constructed based on the large language model. The role function prompt word template of the coordinator agent stipulates that the coordinator agent can obtain the name list of all sub-expert agents and the name list of uncalled sub-expert agents, and call the output results of all sub-expert agents; after initializing the coordinator agent, the coordinator agent autonomously chooses to call each sub-expert agent and the summarizer agent based on the input medical information to be processed; when each sub-expert agent outputs, the coordinator agent synchronously updates the medical information processing output result of the sub-expert agent, and integrates the prompt words of the output results of other sub-expert agents to obtain a new output result set.

7. A medical information multi-agent expert thinking chain collaborative reasoning method according to claim 6, characterized in that: The specific implementation process of generating multiple forward reasoning chains is as follows: Utilizing the decision-making planning capability of the large language model, the coordinator intelligence plans the calling sequence based on the new output result set, i.e., the forward reasoning chain. The forward reasoning chain is a sequential decision-making process in which the set of sub-expert agents is regarded as the action space, the medical information to be processed is defined as P, and a set of predefined sub-expert agents is defined as , where n is the total number of sub-expert agents, represents the configuration of the i-th sub-expert agent, represents the i-th sub-expert agent; each sub-expert agent is associated with a knowledge base and a prompt template; the forward reasoning thinking chain tasks of the defined sub-expert agents and summarizer agent are called in sequence according to the order planned by the coordinator agent, thus constructing a multi-agent medical information processing forward reasoning thinking chain; The coordinator agent simultaneously plans the calling plans of multiple sub-expert agents and generates multiple forward reasoning thinking chains to process medical information.

8. A medical information multi-agent expert thinking chain collaborative reasoning method according to claim 7, characterized in that: The specific implementation process of step S5 is as follows: The names and outputs of the sub-expert agents and summarizer agents in all forward reasoning chains are added to the backward reflective agent stack in the order of the coordinator agent's call. After the forward reasoning chain completes the processing of medical information and gives the forward reasoning chain output, the coordinator agent calls the evaluator agent to evaluate whether the forward reasoning chain output exceeds the set medical conflict threshold. When multiple forward reasoning chains exist at the same time, the evaluator agent takes priority in completing the evaluation of the output results of all forward reasoning chains. If the output of any forward reasoning chain does not exceed the set medical conflict threshold, the output result of the forward reasoning chain is directly used as the final output result. If the outputs of all forward reasoning chains exceed the set medical conflict threshold, the backward reflection mechanism will be activated for each forward reasoning chain one by one; the forward reasoning chain containing untrusted results in the multi-agent debate will be placed at the end of all forward reasoning chains to activate the backward reflection mechanism.

9. A medical information multi-agent expert thinking chain collaborative reasoning method according to claim 8, characterized in that: The specific implementation process of the backward reflection mechanism is as follows: The backward reflection mechanism template includes the original medical information to be processed, the output set of other sub-expert agents, the summary processing output of the summarizer agent, and the text description of the medical conflict exceeding the threshold generated by the evaluator agent. It also defines the instructions for the sub-expert agent to iterate its own output. When the backward reflection mechanism is activated, the sub-expert agent regenerates the medical information processing results based on its own backward reflection mechanism template. When the evaluator agent evaluates that the output result of a forward reasoning thinking chain exceeds the set medical conflict threshold, the backward reflection mechanism of the forward reasoning thinking chain is activated; the evaluator agent will evaluate whether the output of each sub-expert agent and summarizer agent in the stack exceeds the set medical conflict threshold one by one in the order of the backward reflection agent stack; after finding the sub-expert agent that exceeds the medical conflict threshold, it will generate targeted modification opinions for the sub-expert agent based on the final medical conflict judgment strategy, and require the sub-expert agent that exceeds the medical conflict threshold to iteratively output the result according to the final medical conflict judgment strategy until it can pass the medical conflict judgment evaluation of the evaluator agent or reaches the maximum number of output iterations; The sub-expert agent that exceeds the medical conflict threshold uses the defined sub-expert agent backward reflection mechanism template to iterate the output results until it passes the medical conflict threshold evaluation of the evaluator agent or reaches the maximum number of output iterations; When a forward reasoning thinking chain activates the backward reflection mechanism, the backward reflection mechanisms of all other forward reasoning thinking chains will enter a blocked state; when the backward reflection mechanism of a forward reasoning thinking chain outputs a result lower than the medical conflict threshold of the evaluator agent within the maximum number of iterations, the output result will be used as the final output result.

10. A medical information multi-agent expert thinking chain collaborative reasoning system, used to implement the reasoning method described in any one of claims 1 to 9, characterized in that: Includes the following modules: The conclusion output module builds sub-expert agents and summarizer agents based on the large language model. The sub-expert agents call the external medical knowledge base to process medical information, and the summarizer agent uses the text summarization and generalization capabilities of the large language model to output the processing conclusion of the medical information. The evaluation module builds an evaluator agent based on the large language model, evaluates the medical information conflicts of the outputs of each sub-expert agent and the summarizer agent, determines whether the medical conflict threshold is exceeded, and generates a medical conflict judgment strategy; When the conclusion output of the summarizer agent exceeds the medical conflict threshold, the multi-agent debate mechanism is activated to output whether the processing conclusion is credible. The evaluator agent evaluates whether the output of the forward reasoning thinking chain exceeds the medical conflict threshold, and the backward reflection mechanism is activated to iteratively update the output processing conclusion. Coordination module; A coordinator agent is constructed based on a large language model. Based on the input medical information to be processed, it autonomously selects multiple sub-expert agents and summarizer agents to be called and plans the calling order, generating multiple forward reasoning thinking chains.

Images