Virtual patient emotion and physiology dynamic response method and system based on artificial intelligence

Abstract

The invention relates to the technical field of medical teaching assistance, and discloses a virtual patient emotion and physiology dynamic response method and system based on artificial intelligence, and the method comprises the steps: constructing an emotional state, physiological parameters and a digital twinborn body of a disease of a virtual patient; receiving external stimulus information of operation stimulus, speech stimulus and environment stimulus of the trainee in real time; basic disease information of a virtual patient is obtained, and dynamic physiological parameters are determined in combination with operation stimulation and the digital twins; determining a dynamic emotional state based on the external stimulus information, the dynamic physiological parameter, and the digital twin; performing multi-modal rendering on the dynamic physiological parameters and the dynamic emotional states and outputting the dynamic physiological parameters and the dynamic emotional states; and the practical trainee makes a clinical response according to the output result. According to the method, the physiological and emotional states under multiple stimuli are dynamically calculated by constructing the digital twins, and coupling feedback is realized, so that the problems of single virtual patient response, no dynamic interaction and limited practical training effect in the prior art are solved.

Description

Technical Field

[0001] This invention relates to the field of medical teaching aids, and more specifically, to a method and system for dynamic emotional and physiological responses of virtual patients based on artificial intelligence. Background Technology

[0002] In existing virtual medical training systems, the behavior patterns of virtual patients (VPs) are typically driven by pre-set scripts or simple state machines. Their physiological parameters (such as heart rate and blood pressure) and emotional responses (such as pain and anxiety) change in a singular, linear, and predictable manner.

[0003] With the deepening integration of metaverse technology and artificial intelligence in the field of medical simulation education, virtual nursing training has become an important way to cultivate nurses' clinical skills because it can overcome the time and space limitations of physical training and reduce resource consumption. Current training demands significantly higher levels of realism in virtual patient interaction, requiring simulation of the emotional fluctuations and physiological changes of real patients under multiple stimuli such as manipulation, speech, and environment. This is to support trainees in conducting clinically relevant skills training, driving the transformation of virtual training from "fixed script operation" to "dynamic interactive simulation," and meeting the needs of large-scale personalized nursing education.

[0004] In traditional virtual medical training systems, the physiological parameter changes and emotional responses of virtual patients largely rely on preset scripts or simple state machines, exhibiting singular, linear, and predictable characteristics. This makes them unable to respond to complex clinical scenarios with multiple superimposed stimuli. For example, when students perform pain stimulation operations, the patient always makes the same painful expression and preset groans. This rigid interaction fails to simulate the complexity and uncertainty of patients in real clinical situations, resulting in poor immersion and hindering the effective training of nursing students' clinical adaptability and humanistic care abilities. While some highly realistic simulators can provide limited physiological feedback, their emotional expression is weak, and they are difficult to deploy in a distributed manner within a virtual environment, leading to insufficient immersion and hindering the effective training of trainees' judgment and emotional care abilities in dynamic clinical situations, thus limiting the teaching effectiveness of virtual training.

[0005] Moreover, its high cost makes it impossible to deploy on a large scale, in a distributed and personalized manner in the metaverse environment.

[0006] Therefore, it is necessary to design a method for the dynamic emotional and physiological responses of virtual patients based on artificial intelligence, and to provide a method and system that enables virtual patients in the metaverse to generate realistic, dynamic, and unpredictable emotional and physiological responses, thereby improving the authenticity and effectiveness of nursing training. Summary of the Invention

[0007] In view of this, the present invention proposes a method and system for dynamic response of virtual patient emotions and physiology based on artificial intelligence. By constructing a digital twin to dynamically calculate the physiological and emotional states under multiple stimuli and realize coupled feedback, it solves the problems of single response, lack of dynamic interaction and limited training effect of virtual patients in the prior art.

[0008] In one aspect, this invention proposes a method for dynamic emotional and physiological responses of virtual patients based on artificial intelligence, comprising: Construct a digital twin of a virtual patient's emotional state, physiological parameters, and disease; It receives external stimulus information in real time, including operational stimuli, verbal stimuli, and environmental stimuli from trainees; Obtain basic disease information of the virtual patient, and determine dynamic physiological parameters by combining the operational stimuli and the digital twin; Dynamic emotional states are determined based on the external stimulus information, dynamic physiological parameters, and the digital twin. The dynamic physiological parameters and dynamic emotional states are rendered and output in a multimodal manner; Trainees respond clinically based on the output results.

[0009] Furthermore, the digital twin is also pre-configured with: intervention measures corresponding to the underlying disease, and physiological parameter adjustment schemes corresponding to the intervention measures; The determination of dynamic physiological parameters includes: Based on the pre-defined intervention measures corresponding to the screening of the aforementioned basic diseases; Based on the comparison results between the described stimuli and the preset intervention measures, a physiological parameter adjustment plan is determined; Acquire initial physiological parameters, and determine initial dynamic physiological parameters based on the physiological parameter adjustment scheme; If the operational stimulus matches the preset intervention measures, the initial dynamic physiological parameters are determined according to the physiological parameter adjustment plan and initial physiological parameters corresponding to the preset intervention measures. If the operational stimulus does not match the preset basic disease intervention measures, then the physiological parameters of the virtual patient are collected after a preset duration, and the physiological parameters after the preset duration are determined as the initial dynamic physiological parameters.

[0010] Furthermore, the preset intervention measures include a first preset operation intervention, a second preset operation intervention, and a third preset operation intervention; The initial dynamic physiological parameters are determined based on the physiological parameter adjustment plan and initial physiological parameters corresponding to the preset intervention measures, including: If the external stimulus matches the first preset operation intervention, then the dynamic physiological parameter is determined as the first dynamic physiological parameter; If the external stimulus matches the second preset operation intervention, then the dynamic physiological parameter is determined as the second dynamic physiological parameter; If the external stimulus matches the third preset intervention, then the dynamic physiological parameter is determined as the third dynamic physiological parameter.

[0011] Furthermore, the digital twin is also pre-defined with three emotional state types and their corresponding physiological parameter ranges; The determination of dynamic emotional state includes: Based on the comparison results between the dynamic physiological parameters after the operation stimulus and the physiological parameter ranges corresponding to the three preset emotional state types, the initial dynamic emotional state is determined. Based on the external stimulus information, determine whether it is a single operant stimulus; If so, then the initial dynamic emotional state is determined to be a dynamic emotional state; If not, the dynamic emotional state is determined by adjusting the order of the preset stimulus types. The preset stimulus type confirmation order is as follows: operational stimuli are the first confirmation order, language stimuli are the second confirmation order, and environmental stimuli are the third confirmation order.

[0012] Furthermore, the three preset emotional state types correspond to the following physiological parameter ranges: a first preset physiological parameter range, a second preset physiological parameter range, and a third preset physiological parameter range. Determining the initial dynamic emotional state includes: When the dynamic physiological parameters after the operation stimulus are within the first preset physiological parameter range, the initial dynamic emotional state is determined to be the first emotional state type. When the dynamic physiological parameters after the operation stimulus are within the range of the second preset physiological parameters, the initial dynamic emotional state is determined to be the second emotional state type. When the dynamic physiological parameters after the operation stimulus are within the range of the third preset physiological parameters, the initial dynamic emotional state is determined to be the third emotional state type.

[0013] Furthermore, the digital twin is also pre-defined with: two types of verbal stimuli and their corresponding emotional state change types, and two types of environmental stimuli and their corresponding emotional state change types; When determining the dynamic emotional state according to the preset stimulus type adjustment order, Based on the comparison results between the verbal stimulus and the preset verbal stimulus type, the first level of emotional state change type is determined; Based on the comparison results between the environmental stimuli and the preset environmental stimulus types, the type of change in the second-level emotional state is determined. Based on the integrated analysis of the initial dynamic emotional state, the first-level emotional state change type, and the second-level emotional state change type, the dynamic emotional state is determined.

[0014] Furthermore, determining the dynamic physiological parameters also includes: After determining the dynamic emotional state Based on the comparison results between the dynamic emotional state and the preset dynamic physiological parameter ranges corresponding to the three emotional state types, the dynamic physiological parameter range is determined. If the initial dynamic physiological parameter is within the range of the dynamic physiological parameter, then the initial dynamic physiological parameter is determined as the dynamic physiological parameter; If the initial dynamic physiological parameters are not within the range of dynamic physiological parameters, then the physiological parameters after adjusting the initial dynamic physiological parameters to the range of dynamic physiological parameters are determined as dynamic physiological parameters.

[0015] Furthermore, the digital twin is also pre-defined with external expressions corresponding to three emotional state types; When the dynamic physiological parameters and dynamic emotional states are rendered and output in a multimodal manner, Based on the comparison results between the dynamic emotional state and the external manifestations corresponding to the preset emotional state types, the output content of the dynamic emotional state is determined.

[0016] Furthermore, the preset emotional state types correspond to external manifestations including: facial expressions and body movements, and voice.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention constructs a digital twin of a virtual patient's emotional state, physiological parameters, and disease. It can receive real-time external stimuli from trainees' actions, speech, and environment. By dynamically determining physiological parameters and emotional states in conjunction with the underlying disease and the digital twin, it performs multimodal rendering output. This allows the virtual patient's response to break through the limitations of traditional pre-set scripts, more closely resembling the dynamic changes of patients under multiple stimuli in real clinical settings. This significantly enhances the immersion and realism of virtual nursing training. Simultaneously, trainees can make clinical responses based on the output results, effectively training their judgment in complex clinical scenarios and their comprehensive control over patients' physiological and emotional states. This avoids the limitations of traditional virtual training due to fixed responses, better meeting the clinical skills training needs of nursing staff and contributing to improving the teaching effectiveness and application value of virtual medical training.

[0018] On the other hand, the present invention also proposes an artificial intelligence-based virtual patient emotional and physiological dynamic response system, which is applicable to artificial intelligence-based virtual patient emotional and physiological dynamic response methods, including: an input module, a control module, an output rendering module, and a clinical response module; The input module is used to receive external stimulus information such as the trainee's operational stimuli, verbal stimuli, and environmental stimuli, as well as the virtual patient's basic disease information in real time, and transmit the information to the control module. The control module is used to construct a digital twin of the virtual patient's emotional state, physiological parameters, and disease; and to determine dynamic physiological parameters and dynamic emotional state based on the external stimulus information and the digital twin. The output rendering module is used to output the data of the dynamic physiological parameters and the external manifestation of the dynamic emotional state; The clinical response module is used to execute the clinical responses that trainees make based on the output results.

[0019] It is understood that the virtual patient's emotional and physiological dynamic response method and system based on artificial intelligence in the above embodiments of the present invention have the same beneficial effects, and will not be described again. Attached Figure Description

[0020] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 A flowchart illustrating the AI-based virtual patient's emotional and physiological dynamic response method provided in this embodiment of the invention; Figure 2 A functional block diagram of an AI-based virtual patient emotional and physiological dynamic response system provided in an embodiment of the present invention. Detailed Implementation

[0021] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, embodiments and features in the embodiments of the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0022] In traditional virtual medical training systems, the physiological parameter changes and emotional responses of virtual patients largely rely on preset scripts or simple state machines, exhibiting singular, linear, and predictable characteristics. This makes them unable to respond to complex clinical scenarios with multiple superimposed stimuli. For example, when students perform pain stimulation operations, the patient always makes the same painful expression and preset groans. This rigid interaction fails to simulate the complexity and uncertainty of patients in real clinical situations, resulting in poor immersion and hindering the effective training of nursing students' clinical adaptability and humanistic care abilities. While some highly realistic simulators can provide limited physiological feedback, their emotional expression is weak, and they are difficult to deploy in a distributed manner within a virtual environment, leading to insufficient immersion and hindering the effective training of trainees' judgment and emotional care abilities in dynamic clinical situations, thus limiting the teaching effectiveness of virtual training.

[0023] Moreover, its high cost makes it impossible to deploy on a large scale, in a distributed and personalized manner in the metaverse environment.

[0024] Therefore, it is necessary to design a method for the dynamic emotional and physiological responses of virtual patients based on artificial intelligence, and to provide a method and system that enables virtual patients in the metaverse to generate realistic, dynamic, and unpredictable emotional and physiological responses, thereby improving the authenticity and effectiveness of nursing training.

[0025] Reference Figure 1 In some embodiments of this application, the artificial intelligence-based virtual patient emotional and physiological dynamic response method includes: S1. Construct a digital twin of the virtual patient's emotional state, physiological parameters, and disease.

[0026] S2. Receives external stimulus information from trainees in real time, including operational stimuli, verbal stimuli, and environmental stimuli.

[0027] S3. Obtain basic disease information of virtual patients, and determine dynamic physiological parameters by combining operational stimuli and digital twins.

[0028] S4. Determine dynamic emotional states based on external stimulus information, dynamic physiological parameters, and digital twins.

[0029] S5. Render and output dynamic physiological parameters and dynamic emotional states in a multimodal manner.

[0030] S6. Trainees make clinical responses based on the output results.

[0031] Specifically, in S1, the digital twin of the virtual patient is a digital model built on medical pathophysiology and psychology knowledge. It can map the relationship between the real patient's "emotional state, physiological parameters and disease characteristics". It can dynamically update its own state according to external stimuli and simulate the physiological and emotional changes of the real patient.

[0032] Specifically, when constructing the disease module, refer to clinical medical databases (such as the International Classification of Diseases ICD-11) and enter the characteristics of common underlying diseases (such as fractures, hypertension, and diabetes), including the natural evolution of the disease (such as pain sensitivity in the acute phase of fractures and pain relief in the chronic phase) and typical responses to different nursing procedures (such as stronger pain response in fracture patients during puncture than in healthy patients). When constructing the physiological parameter module, the normal range of each physiological parameter is determined based on human physiological standards, and the linkage between parameters is established. The linkage is based on the pathological logic of sympathetic nerve excitation. When constructing the emotional state module, common clinical emotional types are classified, and a correspondence between "physiological parameter range and emotional type" is established. Finally, the "disease characteristics" of the disease module are bound to the "basic range" of the physiological parameters module, and the "dynamic changes" of the physiological parameters module are bound to the "type switching" of the emotional state module, forming a complete digital twin. For example, when constructing a digital twin of a virtual patient with a "right femoral fracture", the disease module is entered with "increased pain response to puncture operation during the acute phase of the fracture (1-7 days)"; the initial values ​​of the physiological parameters module are set to a heart rate of 90 beats / min and a systolic blood pressure of 130 mmHg; and the emotional state module is set to "when the heart rate is >100 beats / min and the systolic blood pressure is >135 mmHg, the emotion switches to pain".

[0033] Specifically, in S2, external stimuli are various input signals applied to the virtual patient during the training process; operational stimuli are the nursing operations performed by the trainee on the virtual patient, including three types: injection, turning over, and back percussion; verbal stimuli are the voice interactions between the trainee and the virtual patient, including two types: comforting words and inquiries about the patient's condition; and environmental stimuli are environmental signals in the virtual training scene, including two types: noise and light intensity.

[0034] Specifically, the training identifies the trainees' operational stimuli, verbal stimuli, and environmental stimuli, and classifies and labels the occurrence time, type, and intensity of the stimuli according to the "operation-verbal-environment" category, providing input for subsequent calculation of dynamic parameters and emotional states.

[0035] Specifically, in S3, dynamic physiological parameters refer to the real-time changes in physiological indicators of virtual patients under the influence of external stimuli and disease, including cardiovascular system parameters (heart rate, blood pressure), respiratory system parameters (respiratory rate, blood oxygen saturation), and nervous system parameters (pain threshold), etc. The linkage between these parameters follows the logic of medical pathology and physiology.

[0036] It is understandable that the same operative stimulus may have different physiological effects on patients with different underlying diseases (e.g., puncture has a greater impact on the heart rate of patients with fractures than on healthy patients). Therefore, it is necessary to take the underlying disease as a premise and combine the "disease-operative-physiology" association rules preset in the digital twin to calculate dynamic physiological parameters.

[0037] Specifically, in S4, dynamic emotional state refers to the type of emotion presented by the virtual patient under the combined effect of physiological state and external stimuli, including three types: calm, anxiety, and distress.

[0038] Understandably, a patient's emotional state is influenced by both their physiological state and various external stimuli. It is necessary to first determine the initial emotional tendency through physiological parameters, and then adjust it in conjunction with other stimuli (speech, environment) to ensure that the emotional state is consistent with the clinical reality.

[0039] Specifically, in S5, multimodal rendering is a technology that presents the dynamic physiological parameters and emotional state of virtual patients to trainees through multiple sensory output methods such as vision, hearing, and data visualization. This includes facial expression / body movement rendering (visual), emotionally colored speech generation (auditory), and virtual monitor data display (data).

[0040] Specifically, in S6, the trainee uses the output of the virtual patient's dynamic physiological parameters and dynamic emotional state; combined with the underlying disease (such as a fracture), the current procedure (such as a puncture), and the patient's condition, to determine the strategies that need adjustment (e.g., if the patient is in significant pain, the procedure needs to be paused and the patient comforted, or the puncture angle needs to be adjusted to alleviate pain). For example, if the trainee observes that the virtual patient's emotional state is "pain" and their heart rate is 110 beats per minute, they determine that the puncture needs to be paused to relieve the patient's discomfort. They then operate the virtual device to stop the puncture and say, "I'll stop for a moment; take a deep breath and relax." The system receives the "pause operation" (new operational stimulus) and the "new comforting words" (new verbal stimulus) and proceeds to the next round of dynamic response calculation.

[0041] Understandably, the trainee's new response, as a "new external stimulus," triggers the S2 steps again, initiating the next cycle of "stimulus reception → parameter calculation → emotion determination → rendering output," thus achieving dynamic interaction.

[0042] As can be seen, the above embodiments simulate the "emotion-physiology-disease" correlation logic through digital twins, making the virtual patient's response consistent with medical common sense and dynamically responding to multiple stimuli (such as puncture + comfort + noise), breaking through the singleness of traditional scripted responses and making the training scenario closer to real clinical practice; trainees need to comprehensively judge the patient's state by "visual expression, auditory speech, and data parameters" (such as judging excessive puncture pain by "painful expression + increased heart rate"), rather than operating according to fixed steps, effectively training their ability to analyze and make decisions in dynamic clinical situations; the emotional state output of the virtual patient (such as painful expression and speech) prompts trainees to pay attention to the patient's psychological feelings (such as pausing the operation to comfort), rather than just focusing on the operation itself, helping to cultivate their empathy and doctor-patient communication skills; the digital twin can quickly input the correlation rules of different underlying diseases (such as hypertension and diabetes) and different emotional types, supporting the construction of diverse training scenarios (such as emergency rescue and chronic disease care) to meet the needs of nursing training at different stages.

[0043] Reference Figure 1 In some embodiments of this application, the digital twin is also pre-defined with: intervention measures corresponding to the underlying disease and physiological parameter adjustment schemes corresponding to the intervention measures; Determining dynamic physiological parameters includes: Pre-defined intervention measures based on screening of underlying diseases; The physiological parameter adjustment plan is determined based on the comparison results between the manipulated stimulus and the preset intervention measures; Acquire initial physiological parameters, and determine initial dynamic physiological parameters based on the physiological parameter adjustment plan; If the manipulated stimulus matches the preset intervention, the initial dynamic physiological parameters are determined according to the physiological parameter adjustment plan and initial physiological parameters after the preset intervention. If the manipulation stimulus does not match the preset intervention measures for the underlying disease, then the physiological parameters of the virtual patient are collected after a preset duration, and the physiological parameters after the preset duration are determined as the initial dynamic physiological parameters.

[0044] Specifically, the intervention measures corresponding to the underlying diseases preset in the digital twin refer to the set of routine clinical nursing operations related to specific underlying diseases stored in the disease module of the virtual patient's digital twin.

[0045] The physiological parameter adjustment plan is a rule for the change of physiological parameters, which clarifies the direction (increase / decrease / no change), amplitude (such as heart rate change ±X beats / min), and calculation method of parameter change. It is based on the clinical law of "operational intervention-physiological response" in medical pathophysiology, such as "venipuncture (pain intervention) → heart rate increases by 15-25 beats / min, systolic blood pressure increases by 10-20 mmHg".

[0046] The preset duration is a time interval pre-set in the digital twin, used to capture the time interval after the implementation of "non-interventional stimuli".

[0047] Specifically, after obtaining the basic disease information of the virtual patient, the routine intervention measures for the disease are selected based on the preset intervention measures corresponding to the basic disease. The operation stimulus is compared with the selected routine intervention measures for the disease. If the comparison is consistent, the physiological parameter adjustment plan corresponding to the intervention measure is retrieved. If the comparison is inconsistent, the operation stimulus is marked as a non-preset intervention operation, and the adjustment plan is not retrieved for the time being.

[0048] Specifically, after determining the physiological parameter adjustment plan, the initial physiological parameters of the virtual patient are obtained. If the physiological parameter adjustment plan (operational stimulus matching preset intervention measures) has been retrieved, the initial physiological parameters are combined with the change range in the adjustment plan, and the initial dynamic physiological parameters are calculated by "initial value + change range".

[0049] It is understandable that when the manipulation stimulus is a pre-set intervention for the underlying disease, its effect on physiological parameters has clear clinical patterns. Directly using the pre-set adjustment scheme for calculation can ensure the accuracy and repeatability of parameter changes and avoid the distortion of the training scenario caused by random calculation.

[0050] Specifically, when a mismatch is confirmed between the manipulation stimulus and the preset intervention, a timer is started after the manipulation stimulus ends for a preset duration; after the timer ends, the physiological parameters of the virtual patient at this time are collected; the collected parameters are marked as "non-intervention-related dynamic physiological parameters" and passed to the subsequent "determine dynamic emotional state" step.

[0051] As can be seen, this embodiment ensures that changes in physiological parameters strictly follow the clinical logic of "disease-operation-physiology" by "basic disease screening and intervention measures" and "matching operations to determine adjustment plans," avoiding the randomness of parameter changes in traditional virtual patients and making the training scenario closer to real clinical practice.

[0052] Reference Figure 1 In some embodiments of this application, the preset intervention measures include a first preset operation intervention, a second preset operation intervention, and a third preset operation intervention; The pre-set intervention measures include a first pre-set operational intervention, a second pre-set operational intervention, and a third pre-set operational intervention; Initial dynamic physiological parameters are determined based on the physiological parameter adjustment plan and initial physiological parameters following the pre-set intervention measures, including: If the external stimulus matches the first preset intervention, then the dynamic physiological parameter is determined as the first dynamic physiological parameter; If the external stimulus matches the second preset intervention, then the dynamic physiological parameter is determined as the second dynamic physiological parameter; If the external stimulus matches the third pre-set intervention, then the dynamic physiological parameter is determined as the third dynamic physiological parameter.

[0053] Specifically, the first, second, and third preset interventions are three types of core nursing operations preset in the digital twin for the underlying disease, covering the main scenarios of disease care: the first preset intervention is an invasive treatment, such as intravenous puncture, used for drug administration and causing pain stimulation; the second preset intervention is a fixation and immobilization, such as limb traction fixation, used to relieve pain from fracture displacement; and the third preset intervention is a physical relief, such as local cold compress, used to reduce swelling and mildly relieve pain.

[0054] The first / second / third dynamic physiological parameters are the physiological indicators calculated by "initial physiological parameters + adjustment plan" after the virtual patient receives the corresponding preset operation intervention. The clinical impact of each type of parameter is consistent with the corresponding operation.

[0055] It is understandable that under the same underlying disease, different types of procedures have fundamentally different physiological effects: invasive procedures (puncture) cause sympathetic nerve excitation due to pain, leading to increased heart rate and blood pressure; immobilization procedures (traction) cause decreased heart rate and blood pressure due to pain relief; physical procedures (cold compresses) have an effect in between. Calculating parameters using a single approach would lead to the clinical distortion that "both puncture and immobilization increase heart rate."

[0056] As can be seen, this embodiment uses "three types of operational interventions + corresponding dynamic parameters" to precisely bind operations and parameters, ensuring that parameter changes reflect differences in operational mechanisms and improving the realism of virtual responses.

[0057] See Figure 1 As shown, in some embodiments of this application, the digital twin is also pre-defined with three emotional state types and their corresponding physiological parameter ranges; Determining dynamic emotional states includes: The initial dynamic emotional state is determined by comparing the dynamic physiological parameters after the operation stimulus with the physiological parameter ranges corresponding to the three preset emotional state types. Determine whether it is a single operant stimulus based on external stimulus information; If so, then the initial dynamic emotional state is determined to be a dynamic emotional state; If not, the dynamic emotional state is determined by adjusting the order of the preset stimulus types. The preset stimulus type confirmation order is as follows: operation stimulus is the first confirmation order, language stimulus is the second confirmation order, and environmental stimulus is the third confirmation order.

[0058] Specifically, the three emotional states include: the first emotional state, such as calmness; the second emotional state, such as anxiety; and the third emotional state, such as distress; each state corresponds to a specific range of physiological parameters.

[0059] Specifically, the initial dynamic emotional state is determined by comparing the dynamic physiological parameters after the operation stimulus with the range of physiological parameters of the three emotional states; the received external stimulus information is retrieved, and when the judgment result involves verbal and environmental stimuli, rather than a single operation stimulus, the emotional state needs to be adjusted in a preset order.

[0060] Understandably, this embodiment first establishes a physiological basis and then adjusts stimuli in sequence, so that the emotional state not only conforms to the physiological logic of "increased heart rate → pain" but also reflects the verbal influence of "comfort → pain relief," thus avoiding emotional responses that are out of touch with reality. The preset priority order makes emotional adjustment systematic, avoids chaotic responses when multiple stimuli are superimposed, and improves the stability of the training scenario.

[0061] See Figure 1 As shown, in some embodiments of this application, the three preset emotional state types correspond to physiological parameter ranges including: a first preset physiological parameter range, a second preset physiological parameter range, and a third preset physiological parameter range; Determine the initial dynamic emotional state, including: When the dynamic physiological parameters after the operation stimulus are within the first preset physiological parameter range, the initial dynamic emotional state is determined to be the first emotional state type. When the dynamic physiological parameters after the operation stimulus are within the range of the second preset physiological parameters, the initial dynamic emotional state is determined to be the second emotional state type. When the dynamic physiological parameters after the operation stimulus are within the range of the third preset physiological parameters, the initial dynamic emotional state is determined to be the third emotional state type.

[0062] Specifically, the first / second / third preset physiological parameter ranges are preset physiological parameter intervals in the digital twin that correspond one-to-one with the three emotional state types, used to determine the initial dynamic emotional state. The first preset physiological parameter range corresponds to the first emotional state (calm); the second preset physiological parameter range corresponds to the second emotional state (anxiety); and the third preset physiological parameter range corresponds to the third emotional state (pain).

[0063] Understandably, physiological parameters are an important objective basis for emotional state, and relying solely on subjective judgment of emotions can lead to distorted responses. This embodiment transforms the determination of the initial dynamic emotional state from "subjective" to "objective" by establishing a one-to-one correspondence between "physiological parameter range and emotional state," ensuring a high degree of matching between emotional state and physiological basis, and avoiding the clinical contradiction of "physiological calm but emotional distress."

[0064] See Figure 1As shown, in some embodiments of this application, the digital twin is also pre-defined with: two types of verbal stimuli and their corresponding emotional state change types, and two types of environmental stimuli and their corresponding emotional state change types. When determining the dynamic emotional state by adjusting the order of preset stimulus types, Based on the comparison results between the verbal stimulus and the preset verbal stimulus type, the first level of emotional state change type is determined; Based on the comparison results between environmental stimuli and preset environmental stimulus types, the type of change in the second level of emotional state is determined. Based on the integrated analysis of the initial dynamic emotional state, the types of changes in the first-level emotional state, and the types of changes in the second-level emotional state, the dynamic emotional state is determined.

[0065] Specifically, the first type of verbal stimulus is comforting language, such as "Don't be afraid, it will be better soon"; the second type of verbal stimulus is inquiring language, such as "How are you feeling?", and each type of verbal stimulus corresponds to a specific type of emotional state change. The first type of environmental stimulus is light intensity, and the second type of environmental stimulus is noise; each type of environmental stimulus corresponds to a specific type of emotional state change.

[0066] The emotional adjustment direction is bound to the types of changes in the first / secondary emotional states and the types of verbal / environmental stimuli. The first-level emotional state change type changes step by step from the third emotional state to the second emotional state and then back to the first emotional state; the second-level emotional state change type changes step by step from the first emotional state to the second emotional state and then back to the third emotional state.

[0067] Understandably, in clinical practice, language and environment are important factors in adjusting patients' emotions (e.g., comfort can alleviate pain, while noise can exacerbate anxiety), and their impact on emotions has a priority (language has a greater impact than environment). This step, through a graded adjustment of "language first, then environment," combined with a pre-set "stimulus type - emotional change" rule, makes the emotional adjustment when multiple stimuli are superimposed delicate and consistent with clinical logic, avoiding the problem of "multiple stimuli causing confusion and affecting emotions," and improving the authenticity of emotional responses.

[0068] See Figure 1 As shown, in some embodiments of this application, determining dynamic physiological parameters further includes: After determining the dynamic emotional state Based on the comparison results between the dynamic emotional state and the three preset emotional state types corresponding to the dynamic physiological parameter range, the dynamic physiological parameter range is determined. If the initial dynamic physiological parameters are within the range of dynamic physiological parameters, then the initial dynamic physiological parameters are determined as dynamic physiological parameters; If the initial dynamic physiological parameters are not within the dynamic physiological parameter range, then the physiological parameters after adjusting the initial dynamic physiological parameters to the dynamic physiological parameter range shall be determined as the dynamic physiological parameters.

[0069] Specifically, the dynamic physiological parameter range corresponding to the emotional state is a range of physiological parameters that corresponds one-to-one with the three emotional state types. It is a reasonable range that the physiological parameters need to meet after the emotion reacts to the physiology. For example, the first emotion "pain" corresponds to a heart rate of 110-130 beats / min, which is higher than the initial range of 105-120 beats / min after the operation stimulus.

[0070] Understandably, in real clinical settings, emotion and physiology are "bidirectionally coupled": physiological changes trigger emotional fluctuations (e.g., increased heart rate → pain), and emotional fluctuations, in turn, affect physiology (e.g., pain → further increase in heart rate). Traditional virtual patients only achieve a unidirectional effect of "physiology → emotion," ignoring the counter-effect of "emotion → physiology," resulting in incomplete responses. This embodiment achieves a closed-loop coupling of "physiology-emotion" through "emotion → determining the physiological range → adjusting initial parameters," making the virtual patient's response closer to the physiological and emotional interaction patterns of the real human body.

[0071] See Figure 1 As shown, in some embodiments of this application, the digital twin also has pre-defined external expressions corresponding to three emotional state types; When rendering and outputting dynamic physiological parameters and dynamic emotional states in a multimodal manner, Based on the comparison results between the dynamic emotional state and the external manifestations corresponding to the preset emotional state type, the output content of the dynamic emotional state is determined.

[0072] Specifically, the preset emotional state types correspond to external manifestations including: facial expressions and body movements, and voice. Specifically, the external manifestations corresponding to the emotional state types are the sensory output contents that correspond one-to-one with the three emotional state types: facial expressions and body movements are visual manifestations; speech is an auditory manifestation; based on clinical observation data of "emotion-external manifestation", such as pain corresponding to frowning and groaning.

[0073] Specifically, it retrieves the external manifestations and final dynamic physiological parameters corresponding to the dynamic emotional state: the output of the external manifestations, namely "visual manifestations + auditory manifestations", is transmitted to the "output rendering module" and output synchronously with the physiological parameter data (monitor display).

[0074] As can be seen, this embodiment transforms abstract emotional states into intuitive visual and auditory signals by binding "emotional state → external expression", thereby enhancing the immersion of the training and allowing trainees to experience judgment scenarios "as if facing a real patient".

[0075] As can be seen from the above embodiments.

[0076] See Figure 2 As shown in some embodiments of this application, the AI-based virtual patient emotional and physiological dynamic response system is applicable to the AI-based virtual patient emotional and physiological dynamic response method, and includes: an input module, a control module, an output rendering module, and a clinical response module.

[0077] Specifically, the input module is used to receive external stimulus information such as the trainee's operational stimuli, verbal stimuli, and environmental stimuli, as well as the virtual patient's basic disease information in real time, and transmit the information to the control module. The control module is used to construct a digital twin of the virtual patient's emotional state, physiological parameters, and disease; and to determine dynamic physiological parameters and dynamic emotional state based on external stimulus information and the digital twin. The output rendering module is used to output data on dynamic physiological parameters and external manifestations of dynamic emotional states; The clinical response module is used to execute the clinical responses that trainees make based on the output results.

[0078] It is understood that the artificial intelligence-based virtual patient emotional and physiological dynamic response method and system in the above embodiments of the present invention have the same beneficial effects, and will not be described again.

[0079] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program goods. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program goods embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0080] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program goods according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0081] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0082] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. An artificial intelligence-based virtual patient emotional and physiological dynamic response method, characterized in that: a digital twin of a virtual patient's emotional state, physiological parameters, and disease is constructed; real-time external stimulus information of an operator's operation stimulus, verbal stimulus, and environmental stimulus is received; basic disease information of the virtual patient is obtained, and dynamic physiological parameters are determined in combination with the operation stimulus and the digital twin; dynamic emotional states are determined based on the external stimulus information, the dynamic physiological parameters, and the digital twin; the dynamic physiological parameters and the dynamic emotional states are multi-modal rendered and output; and the operator makes a clinical response based on the output result.

2. The artificial intelligence-based virtual patient emotional and physiological dynamic response method according to claim 1, characterized in that: the digital twin is further provided with preset intervention measures corresponding to the basic disease and physiological parameter adjustment schemes corresponding to the intervention measures; the determination of the dynamic physiological parameters comprises: screening the corresponding preset intervention measures based on the basic disease; determining the physiological parameter adjustment scheme based on the comparison result of the operation stimulus and the preset intervention measures; obtaining initial physiological parameters and determining the initial dynamic physiological parameters based on the physiological parameter adjustment scheme; if the operation stimulus matches the preset intervention measures, determining the initial dynamic physiological parameters based on the physiological parameter adjustment scheme corresponding to the preset intervention measures and the initial physiological parameters; and if the operation stimulus does not match the preset intervention measures, collecting the physiological parameters of the virtual patient after a preset time period and determining the physiological parameters after the preset time period as the initial dynamic physiological parameters; the preset intervention measures include a first preset operation intervention, a second preset operation intervention, and a third preset operation intervention; determining the initial dynamic physiological parameters based on the physiological parameter adjustment scheme corresponding to the preset intervention measures and the initial physiological parameters comprises: if the external stimulus matches the first preset operation intervention, determining the dynamic physiological parameters as the first dynamic physiological parameters; if the external stimulus matches the second preset operation intervention, determining the dynamic physiological parameters as the second dynamic physiological parameters; and if the external stimulus matches the third preset operation intervention, determining the dynamic physiological parameters as the third dynamic physiological parameters; the digital twin is further provided with three emotional state types and corresponding physiological parameter ranges; the determination of the dynamic emotional states comprises: determining the initial dynamic emotional states based on the comparison result of the dynamic physiological parameters after the operation stimulus and the preset three emotional state type corresponding physiological parameter ranges; judging whether it is a single operation stimulus based on the external stimulus information; if yes, determining the initial dynamic emotional states as the dynamic emotional states; and if no, determining the dynamic emotional states according to a preset stimulus type adjustment sequence; the preset stimulus type adjustment sequence is: operation stimulus as the first adjustment sequence, verbal stimulus as the second adjustment sequence, and environmental stimulus as the third adjustment sequence; the preset three emotional state type corresponding physiological parameter ranges include a first preset physiological parameter range, a second preset physiological parameter range, and a third preset physiological parameter range; and the determination of the initial dynamic emotional states comprises: ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ 3. The artificial intelligence-based virtual patient emotional and physiological dynamic response method of claim 2, wherein, ​ ​ ​ ​ ​ 4. The method of claim 3, wherein, ​ ​ ​ ​ ​ ​ ​ 5. The artificial intelligence-based virtual patient emotional and physiological dynamic response method of claim 4, wherein, ​ ​ When the dynamic physiological parameter after the operation stimulation is in a first preset physiological parameter range, the initial dynamic emotional state is determined as a first emotional state type; When the dynamic physiological parameter after the operation stimulation is in a second preset physiological parameter range, the initial dynamic emotional state is determined as a second emotional state type; When the dynamic physiological parameter after the operation stimulation is in a third preset physiological parameter range, the initial dynamic emotional state is determined as a third emotional state type.

6. The artificial intelligence-based virtual patient emotional and physiological dynamic response method of claim 5, wherein, The digital twin is further preset with: two types of verbal stimulation and corresponding emotional state change types, two types of environmental stimulation and corresponding emotional state change types; When determining the dynamic emotional state according to the preset stimulation type adjustment sequence, Based on the comparison result of the verbal stimulation and the preset verbal stimulation type, the first-level emotional state change type is determined; Based on the comparison result of the environmental stimulation and the preset environmental stimulation type, the second-level emotional state change type is determined; Based on the integration analysis of the initial dynamic emotional state, the first-level emotional state change type and the second-level emotional state change type, the dynamic emotional state is determined.

7. The artificial intelligence-based virtual patient emotional and physiological dynamic response method of claim 6, wherein, The determination of the dynamic physiological parameter further comprises: After determining the dynamic emotional state, Based on the comparison result of the dynamic emotional state and the preset dynamic physiological parameter range corresponding to the three emotional state types, the dynamic physiological parameter range is determined; If the initial dynamic physiological parameter is in the dynamic physiological parameter range, the initial dynamic physiological parameter is determined as the dynamic physiological parameter; If the initial dynamic physiological parameter is not in the dynamic physiological parameter range, the physiological parameter after the initial dynamic physiological parameter is adjusted to the dynamic physiological parameter range is determined as the dynamic physiological parameter.

8. The artificial intelligence-based virtual patient emotional and physiological dynamic response method of claim 7, wherein, The digital twin is further preset with: three types of emotional state types corresponding to external manifestations; When the dynamic physiological parameter and the dynamic emotional state are rendered and output in multiple modes, Based on the comparison result of the dynamic emotional state and the preset external manifestations corresponding to the emotional state types, the dynamic emotional state output content is determined.

9. The artificial intelligence-based virtual patient emotional and physiological dynamic response method of claim 8, wherein, The preset external manifestations corresponding to the emotional state types include: facial expressions and body movements, speech.

10. The system for dynamically responding the virtual patient's emotion and physiology based on artificial intelligence, applied to the method for dynamically responding the virtual patient's emotion and physiology based on artificial intelligence as claimed in any one of claims 1-9, characterized in that, Comprise: An input module, a control module, an output rendering module and a clinical response module; The input module is used to receive external stimulation information of operation stimulation, verbal stimulation and environmental stimulation of the real training person and basic disease information of the virtual patient in real time, and transmit the information to the control module; The control module is used to construct the digital twin of the emotional state, physiological parameter and disease of the virtual patient; and determine the dynamic physiological parameter and the dynamic emotional state according to the external stimulation information and the digital twin; The output rendering module is used to output the data of the dynamic physiological parameter and the external manifestations of the dynamic emotional state; The clinical response module is used to execute the clinical response made by the real training person according to the output result.

Images