Anesthesia efficacy evaluation system and method based on artificial intelligence

Abstract

The invention discloses an artificial intelligence-based anesthesia efficacy evaluation system and method, and belongs to the field of efficacy evaluation, and the method comprises the steps: collecting all heart rate data of a patient in an anesthesia process, carrying out the time node division of the whole process of the patient from an anesthesia induction period to an anesthesia waking period, and constructing a time node heart rate data set; calculating an average value of the heart rate data of the time nodes, and calculating a heart rate change rate; calculating a comprehensive heart rate value and a comprehensive heart rate change degree; designing and analyzing an evaluation value based on the heart rate change rate and the comprehensive heart rate change degree; by collecting the heart rate data in the anesthesia process and combining the time node heart rate data set, comprehensive and staged monitoring and analysis of the heart rate are achieved; the comprehensive heart rate value and the change degree thereof are further calculated, and the influence of the anesthetic on the heart rate in different stages is accurately reflected; by designing the evaluation value and combining the preset threshold value, quantitative evaluation is performed on the anesthesia effect, so that anesthesia management is more accurate and safer.

Description

Technical Field

[0001] The present invention relates to the technical field of drug efficacy evaluation, and in particular to an anesthetic drug efficacy evaluation system and method based on artificial intelligence. Background Art

[0002] In modern surgery and medical procedures, anesthesia is one of the important means to ensure patient safety and comfort. With the development of medical technology, the types and methods of using anesthetic drugs are becoming increasingly diverse, and the evaluation of anesthetic effects has become more complex and important. Traditional anesthesia evaluation mainly relies on the experience of anesthesiologists and some physiological indicators of patients, such as heart rate, blood pressure, and respiratory rate. However, due to individual differences and environmental factors, relying solely on these indicators cannot fully and accurately reflect the actual effect of anesthetic drugs. With the rapid development of artificial intelligence (AI) technology, AI-based anesthetic efficacy evaluation methods have gradually become a research hotspot. These new methods can more accurately evaluate the effects of anesthetic drugs and improve the safety and effectiveness of anesthesia by analyzing a large amount of physiological data collected during the patient's surgery, combined with machine learning and deep learning algorithms.

[0003] However, the existing AI-based anesthetic efficacy evaluation methods still have some shortcomings. First, most methods only focus on a single or a few physiological indicators, ignoring the complexity and multidimensionality of the human physiological system. Secondly, most of the current models are based on static data analysis, lack the ability to dynamically process time series data, and it is difficult to accurately reflect the changes in the patient's physiological state during anesthesia. In addition, since the training of the model in the existing methods depends on a large amount of labeled data, the quality of the data and the accuracy of the annotations also have a significant impact on the evaluation results. In practical applications, how to analyze multidimensional data in real time and efficiently during anesthesia, and then provide accurate anesthetic efficacy evaluation results, is a major challenge facing existing technologies. Summary of the invention

[0004] The purpose of the present invention is to provide an artificial intelligence-based anesthetic efficacy evaluation system and method to solve the problems raised in the above-mentioned background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] An artificial intelligence-based anesthetic efficacy evaluation method comprises the following steps: constructing a heart rate sensor set, collecting all heart rate data detected and recorded by the heart rate sensors of patients during anesthesia, and constructing a heart rate data set; dividing the entire process from the anesthesia induction period to the anesthesia recovery period of the patient into time nodes, and constructing a time node heart rate data set based on the heart rate data set; calculating the average value of the time node heart rate data detected by the same heart rate sensor at the same time node, and calculating the heart rate change rate of the same heart rate sensor at different time nodes based on the average value; calculating the comprehensive heart rate value of the heart rate data collected by all heart rate sensors at the same time node, and calculating the comprehensive heart rate change degree at different time nodes; designing and analyzing the evaluation value based on the heart rate change rate and the comprehensive heart rate change degree.

[0007] As a preferred solution of the method for evaluating the anesthetic efficacy based on artificial intelligence described in the present invention, a heart rate sensor set BS = {BS a |a∈[1,A]}, where BS a represents the ath heart rate sensor, and A represents the total number of heart rate sensors; the heart rate sensor is used to detect and record the heart rate data of the human body.

[0008] Furthermore, the heart rate sensor BS of the patient during anesthesia is collected a All the heart rate data detected and recorded are used to construct the heart rate data set PD = {PD ab |a∈[1,A],b∈[1,B]}, where PD ab It represents the bth heart rate data detected by the ath heart rate sensor during the anesthesia process, and B represents the total number of heart rate data.

[0009] As a preferred solution of the artificial intelligence-based anesthetic efficacy evaluation method of the present invention, the entire process from the anesthesia induction period to the anesthesia recovery period of the patient is divided into x time nodes, and the time node set TS is defined as {TS t |t∈[1,x]}, where TS t represents the tth time node; the anesthesia induction period refers to the initial stage of the anesthesia process in which the patient enters an anesthesia state; the anesthesia recovery period refers to the end stage in which the patient recovers consciousness after the anesthesia is stopped.

[0010] Further, based on the heart rate data set and the time node set, a time node heart rate data set TPD is constructed. ab (TS t )|a∈[1,A],b∈[1,B],t∈[1,x]},where TPD ab (TS t ) indicates that at time node TS t Heart rate data at the time node.

[0011] As a preferred solution of the method for evaluating the efficacy of anesthetic drugs based on artificial intelligence described in the present invention, the heart rate sensor BS is calculated. a At time node TS t Heart rate data TPD at the time point of detection ab (TS t ), calculated as follows:

[0012]

[0013] Among them, μ ab (TS t ) represents the time node heart rate data TPD ab (TS t )’s average value.

[0014] Further, based on the heart rate data TPD at the time node ab (TS t ) to calculate the heart rate sensor BS a At time node TS t To time node TS t+1 Heart rate data TPD at the time point of detection ab (TS t ) is calculated by the following formula:

[0015]

[0016] Among them, μ ab (TS t+1 ) represents the time node TS t+1 Time node heart rate data TPD ab (TS t+1 ) average value, Δμ ab (TS t →TS t+1 ) indicates the heart rate variability.

[0017] In the present invention, if Δμ ab (TS t →TS t+1 )>0, indicating that the heart rate is at time node TS t+1 Relative to time node TS t Increased, if Δμ ab (TS t →TS t+1 )<0, indicating that the heart rate is at time node TS t+1 Relative to time node TS tBy monitoring this change rate, we can further analyze the effect of anesthetic drugs on heart rate and thus evaluate the efficacy of anesthetic drugs.

[0018] Further, the calculation is done at time node TS t When the heart rate data collected by all heart rate sensors is calculated, the comprehensive heart rate value is calculated as follows:

[0019]

[0020] Among them, μ com (TS t ) represents the comprehensive heart rate value, ω a Indicates the ath heart rate sensor BS a Preset coefficients.

[0021] In the present invention, the heart rate sensors are usually placed at different parts of the body (such as the chest, wrist, ankle, etc.), and each sensor will monitor the heart rate data of different parts, which may vary depending on the location of the sensor; the weight ω a It reflects the importance of each sensor, which is determined by the sensor location. For example, sensors close to the heart may get higher weights because these locations more directly reflect heart activity. The data of sensors with higher weights have a greater impact on the final comprehensive heart rate value.

[0022] Further, based on the comprehensive heart rate value μ com (TS t ), calculated at time node TS t To time node TS t+1 The comprehensive heart rate variability at 1000 Hz is calculated as follows:

[0023] Δμ com (TSS t →TS t+1 )=μ com (TS t+1 )-μ com (TS t );

[0024] Among them, μ com (TS t+1 ) represents the time node TS t+1 The comprehensive heart rate value at the time, Δμ com (TS t →TS t+1 ) represents the comprehensive heart rate variability.

[0025] In the present invention, if the comprehensive heart rate variation is positive, it indicates that the patient's heart rate is rising, which means that the depth of anesthesia is weakening, and the patient may be gradually shifting from a deep anesthesia state to a superficial anesthesia or awakening state. If the heart rate increase continues, it may indicate that the efficacy of the anesthetic drug is gradually weakening, indicating the need for drug supplementation or adjustment. If the comprehensive heart rate variation is negative, it indicates that the patient's heart rate is decreasing, which means that the depth of anesthesia is deepening and the patient has entered a deeper state of anesthesia. If the heart rate drops too quickly or too much, it may indicate that the patient has entered an over-anesthesia state, which may have adverse effects on the patient's physiological functions, such as inhibiting breathing or lowering blood pressure. At this time, it may be necessary to reduce the dose of the anesthetic drug or take other measures to adjust the depth of anesthesia.

[0026] As a preferred embodiment of the method for evaluating the anesthetic efficacy based on artificial intelligence described in the present invention, an evaluation value is designed based on the heart rate change rate and the comprehensive heart rate change degree, and the calculation formula is as follows:

[0027] E(TS t →TS t+1 )=α·Δμ ab (TS t →TS t+1 )+β·Δμ com (TS t →TS t+1 );

[0028] Among them, α and β are the preset coefficients of variation, E(TS t →TS t+1 ) represents the evaluation value.

[0029] Furthermore, an evaluation threshold τ is preset, where τ is 0.

[0030] If E(TS t →TS t+1 )>τ, it is judged that the anesthetic effect is weakened and the drug efficacy is poor.

[0031] If E(TS t →TS t+1 )≤τ, the anesthetic effect is judged to be effective and the drug efficacy is good.

[0032] In the present invention, if the evaluation value is a positive value, it usually indicates that the patient's physiological response has become more active, which may mean that the anesthetic effect is weakened, the patient may be gradually waking up from the anesthesia state or the anesthesia depth is insufficient, then the drug efficacy is poor, and the dosage or concentration of the anesthetic drug may need to be adjusted; if the evaluation value is a negative value, it usually indicates that the patient's physiological response has become more stable and the heart rate has decreased, which may mean that the anesthetic drug is working effectively and the patient is in a controlled anesthesia state, then the drug efficacy is good and the anesthetic dose is appropriate; if the evaluation value is zero, it indicates that the effect of the anesthetic drug on the heart rate is in a stable state without significant fluctuations, which means that the drug efficacy is stable during this time period and the patient may be in a stable anesthesia state, but continued monitoring is still required to ensure that the patient's state remains under control.

[0033] An artificial intelligence-based anesthetic efficacy evaluation system comprises a heart rate data acquisition module, a time node data processing module, a heart rate data analysis module and an anesthetic efficacy evaluation module.

[0034] The heart rate data collection module constructs and manages a set of heart rate sensors and collects the data of each sensor into a unified heart rate data set.

[0035] The time node data processing module divides the anesthesia process into multiple time nodes and associates the collected heart rate data with these time nodes.

[0036] The heart rate data analysis module calculates the average heart rate, heart rate change rate and comprehensive heart rate value based on the time node data, and identifies the relationship between the heart rate change and the depth of anesthesia.

[0037] The anesthetic efficacy evaluation module determines whether the anesthetic efficacy is good by calculating the evaluation value and comparing it with a preset threshold value.

[0038] Furthermore, the heart rate data acquisition module also includes a heart rate sensor set construction unit and a heart rate data acquisition unit.

[0039] The heart rate sensor set building unit builds a heart rate sensor set BS={BS a |a∈[1,A]}, where BS a represents the ath heart rate sensor, and A represents the total number of heart rate sensors; the heart rate sensor is used to detect and record the heart rate data of the human body.

[0040] The heart rate data acquisition unit collects the heart rate data of the patient during anesthesia. a All the heart rate data detected and recorded are used to construct the heart rate data set PD = {PD ab |a∈[1,A],b∈[1,B]}, where PD abIt represents the bth heart rate data detected by the ath heart rate sensor during the anesthesia process, and B represents the total number of heart rate data.

[0041] Furthermore, the time node data processing module also includes a time node division unit and a time node heart rate data set construction unit.

[0042] The time node division unit divides the entire process from the anesthesia induction period to the anesthesia awakening period of the patient into x time nodes, and defines the time node set TS = {TS t |t∈[1,x]}, where TS t represents the tth time node; the anesthesia induction period refers to the initial stage of the anesthesia process in which the patient enters an anesthesia state; the anesthesia recovery period refers to the end stage in which the patient recovers consciousness after the anesthesia is stopped.

[0043] The time node heart rate data set construction unit constructs a time node heart rate data set TPD based on the heart rate data set and the time node set. ab (TS t )|a∈[1,A],b∈[1,B],t∈[1,x]}, where TPD ab (TS t ) indicates that at time node TS t Heart rate data at the time node.

[0044] Furthermore, the heart rate data analysis module also includes a heart rate average value calculation unit and a comprehensive heart rate value calculation unit.

[0045] The heart rate average value calculation unit calculates the heart rate sensor BS a At time node TS t Heart rate data TPD at the time point of detection ab (TS t ) average value; based on the heart rate data TPD at the time node ab (TS t ) to calculate the heart rate sensor BS a At time node TS t To time node TS t+1 Heart rate data TPD at the time point of detection ab (TS t )’s heart rate variability.

[0046] The comprehensive heart rate value calculation unit calculates the time node TS t When the heart rate data collected by all heart rate sensors is a comprehensive heart rate value, based on the comprehensive heart rate value μ com (TS t ), calculated at time node TS tTo time node TS t+1 The comprehensive heart rate variability over time.

[0047] Furthermore, the anesthetic efficacy evaluation module also includes an evaluation value calculation unit and an anesthetic effect determination unit.

[0048] The evaluation value calculation unit designs an evaluation value E(TS) based on the heart rate change rate and the comprehensive heart rate change degree. t →TS t+1 ).

[0049] The anesthetic effect determination unit presets an evaluation threshold τ, if E(TS t →TS t+1 )>τ, it is judged that the anesthetic effect is weakened and the drug efficacy is poor; if E(TS t →TS t+1 )≤τ, the anesthetic effect is judged to be effective and the drug efficacy is good.

[0050] Compared with the prior art, the beneficial effects achieved by the present invention are: by constructing a heart rate sensor set and collecting heart rate data during anesthesia, and combining the division of the anesthesia process into multiple time nodes to construct a time node heart rate data set, a comprehensive and staged monitoring and analysis of the heart rate is achieved. Based on the average value and change rate of the heart rate data, the comprehensive heart rate value and its degree of change are further calculated, which can accurately reflect the effect of anesthetic drugs on the heart rate at different stages. Finally, by designing an evaluation value and combining it with a preset threshold, the anesthetic effect is quantitatively evaluated, making anesthesia management more accurate and safe, and effectively reducing the risk during anesthesia. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0052] in:

[0053] Figure 1 Schematic diagram of a flow chart of an anesthetic efficacy evaluation method based on artificial intelligence according to an embodiment of the present invention;

[0054] Figure 2 It is a structural schematic diagram of an anesthetic efficacy evaluation system based on artificial intelligence of the present invention. DETAILED DESCRIPTION

[0055] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] See also Figure 1-Figure 2 , the present invention provides a technical solution:

[0057] See also Figure 1 In the first embodiment, a method for evaluating the efficacy of anesthetic drugs based on artificial intelligence is provided, and the method comprises the following steps:

[0058] Step S1: construct a heart rate sensor set, and collect all heart rate data detected and recorded by the heart rate sensor during the anesthesia process of the patient to construct a heart rate data set.

[0059] Specifically, construct a heart rate sensor set BS = {BS a |a∈[1,a]}, where BS a represents the ath heart rate sensor, and A represents the total number of heart rate sensors; the heart rate sensor is used to detect and record the heart rate data of the human body.

[0060] Furthermore, the heart rate sensor BS of the patient during anesthesia is collected a All the heart rate data detected and recorded are used to construct the heart rate data set PD = {PD ab |a∈[1,A],b∈[1,B]}, where PD ab It represents the bth heart rate data detected by the ath heart rate sensor during the anesthesia process, and B represents the total number of heart rate data.

[0061] Step S2: Divide the entire process from the anesthesia induction period to the anesthesia recovery period of the patient into time nodes, and construct a time node heart rate dataset based on the heart rate dataset.

[0062] Specifically, the entire process from the anesthesia induction period to the anesthesia awakening period is divided into x time nodes, and the time node set TS is defined as {TS t |t∈[1,x]}, where TS t represents the tth time node; the anesthesia induction period refers to the initial stage of the anesthesia process in which the patient enters an anesthesia state; the anesthesia recovery period refers to the end stage in which the patient recovers consciousness after the anesthesia is stopped.

[0063] Further, based on the heart rate data set and the time node set, a time node heart rate data set TPD is constructed. ab (TSt )|a∈[1,A],b∈[1,B],t∈[1,x]}, where TPD ab (TS t ) indicates that at time node TS t Heart rate data at the time node.

[0064] Step S3: Calculate the average value of the time node heart rate data detected by the same heart rate sensor at the same time node, and based on the average value, calculate the heart rate change rate of the same heart rate sensor at different time nodes; calculate the comprehensive heart rate value of the heart rate data collected by all heart rate sensors at the same time node, and calculate the comprehensive heart rate change degree at different time nodes.

[0065] Specifically, calculate the heart rate sensor BS a At time node TS t Heart rate data TPD at the time point of detection ab (TS t ), calculated as follows:

[0066]

[0067] Among them, μ ab (TS t ) represents the time node heart rate data TPD ab (TS t )’s average value.

[0068] For example, assuming that the total number of heart rate sensors is 2, that is, A = 2, and the total number of heart rate data is 2, that is, B = 2, then the heart rate data TPD of the first time node collected by the first heart rate sensor is ab (TS t ) is 3, the heart rate data TPD of the second time node ab (TS t ) is 5, and we can get μ by substituting it into the formula. ab (TS t ) is 0.5*(3+5)=4.

[0069] Further, based on the heart rate data TPD at the time node ab (TS t ) to calculate the heart rate sensor BS a At time node TS t To time node TS t+1 Heart rate data TPDah (TS t ) is calculated by the following formula:

[0070]

[0071] Among them, μ ab (TS t+1 ) represents the time node TS t+1 Time node heart rate data TPD ab (TS t+1 ) average value, Δμ ab (TS t →TS t+1 ) indicates the heart rate variability.

[0072] For example, suppose the first heart rate sensor collects data at time node TS t+1 The heart rate data of the first time node is 2, and the heart rate data of the second time node is 4, then the time node TS t+1 When the heart rate data at the time node is 0.5*(2+4)=3, substituting it into the formula, we get Δμ ab (TS t →TS t+1 ) is -0.25.

[0073] In the present invention, if Δμ ab (TS t →TS t+1 )>0, indicating that the heart rate is at time node TS t+1 Relative to time node TS t Increased, if Δμ ab (TS t →TS t+1 )<0, indicating that the heart rate is at time node TS t+1 Relative to time node TS t By monitoring this change rate, we can further analyze the effect of anesthetic drugs on heart rate and thus evaluate the efficacy of anesthetic drugs.

[0074] Calculated at time node TS t When the heart rate data collected by all heart rate sensors is calculated, the comprehensive heart rate value is calculated as follows:

[0075]

[0076] Among them, μ com (TS t ) represents the comprehensive heart rate value, ω a Indicates the preset coefficient of the a-th heart rate sensor a.

[0077] For example, suppose the second heart rate sensor collects the heart rate data TPD at the time node ab (TS t ) has an average value of 2, and the preset coefficient ω a is 1, and the comprehensive heart rate value μ is obtained by substituting it into the formula com (TS t) is 0.5*(4+2)=3.

[0078] In the present invention, the heart rate sensors are usually placed at different parts of the body (such as the chest, wrist, ankle, etc.), and each sensor will monitor the heart rate data of different parts, which may vary depending on the location of the sensor; the weight ω a It reflects the importance of each sensor, which is determined by the sensor location. For example, sensors close to the heart may get higher weights because these locations more directly reflect heart activity. The data of sensors with higher weights have a greater impact on the final comprehensive heart rate value.

[0079] Furthermore, based on the comprehensive heart rate value μ com (TS t ), calculated at time node TS t To time node TS t+1 The comprehensive heart rate variability at 1000 Hz is calculated as follows:

[0080] Δμ com ((TS t →TS t+1 )=μ com (TS t+1 )-μ com ((TS t );

[0081] Among them, μ com (TS t+1 ) represents the time node TS t+1 The comprehensive heart rate value at

[0082] Δμ com (TS t →TS t+1 ) represents the comprehensive heart rate variability.

[0083] For example, suppose the time node TS t+1 The comprehensive heart rate value μ com (TS t+1 ) is 2, and the comprehensive heart rate variation Δμ is obtained by substituting it into the formula com (TS t →TS t+1 ) is -1.

[0084] In the present invention, if the comprehensive heart rate variation is positive, it indicates that the patient's heart rate is rising, which means that the depth of anesthesia is weakening, and the patient may be gradually shifting from a deep anesthesia state to a superficial anesthesia or awakening state. If the heart rate increase continues, it may indicate that the efficacy of the anesthetic drug is gradually weakening, indicating the need for drug supplementation or adjustment. If the comprehensive heart rate variation is negative, it indicates that the patient's heart rate is decreasing, which means that the depth of anesthesia is deepening and the patient has entered a deeper state of anesthesia. If the heart rate drops too quickly or too much, it may indicate that the patient has entered an over-anesthesia state, which may have adverse effects on the patient's physiological functions, such as inhibiting breathing or lowering blood pressure. At this time, it may be necessary to reduce the dose of the anesthetic drug or take other measures to adjust the depth of anesthesia.

[0085] Step S4: Design and analyze evaluation values ​​based on the heart rate variability and the comprehensive heart rate variability.

[0086] Based on the heart rate change rate and the comprehensive heart rate change degree, an evaluation value is designed, and the calculation formula is as follows:

[0087] E(TS t →TS t+1 )=α·Δμ ab (TS t →TS t+1 )+β·Δμ com (TS t →TS t+1 );

[0088] Among them, α and β are the preset coefficients of variation, E(TS t →TS t+1 ) represents the evaluation value.

[0089] An evaluation threshold τ is preset, where τ is 0.

[0090] If E(TS t →TS t+1 )>τ, it is judged that the anesthetic effect is weakened and the drug efficacy is poor.

[0091] If E(TS t →TS t+1 )≤τ, the anesthetic effect is judged to be effective and the drug efficacy is good.

[0092] For example, assuming that αβ and β are 0.2 and 0.4 respectively, substituting them into the formula gives the evaluation value E(TS t →TS t+1 ) is 0.2*(-0.25)+0.4*(-1)=(-0.05)+(-0.4)=-0.45, and the evaluation value is less than the evaluation threshold, which means that the anesthetic effect is effective and the drug efficacy is good.

[0093] In the present invention, if the evaluation value is a positive value, it usually indicates that the patient's physiological response has become more active, which may mean that the anesthetic effect is weakened, the patient may be gradually waking up from the anesthesia state or the anesthesia depth is insufficient, then the drug efficacy is poor, and the dosage or concentration of the anesthetic drug may need to be adjusted; if the evaluation value is a negative value, it usually indicates that the patient's physiological response has become more stable and the heart rate has decreased, which may mean that the anesthetic drug is working effectively and the patient is in a controlled anesthesia state, then the drug efficacy is good and the anesthetic dose is appropriate; if the evaluation value is zero, it indicates that the effect of the anesthetic drug on the heart rate is in a stable state without significant fluctuations, which means that the drug efficacy is stable during this time period and the patient may be in a stable anesthesia state, but continued monitoring is still required to ensure that the patient's state remains under control.

[0094] See also Figure 2 In the second embodiment of the present invention, an artificial intelligence-based anesthetic efficacy evaluation system is provided, the system comprising: a heart rate data acquisition module, a time node data processing module, a heart rate data analysis module and an anesthetic efficacy evaluation module.

[0095] The heart rate data collection module constructs and manages a set of heart rate sensors and collects the data of each sensor into a unified heart rate data set.

[0096] The time node data processing module divides the anesthesia process into multiple time nodes and associates the collected heart rate data with these time nodes.

[0097] The heart rate data analysis module calculates the average heart rate, heart rate change rate and comprehensive heart rate value based on the time node data, and identifies the relationship between the heart rate change and the depth of anesthesia.

[0098] The anesthetic efficacy evaluation module determines whether the anesthetic efficacy is good by calculating the evaluation value and comparing it with a preset threshold value.

[0099] Furthermore, the heart rate data acquisition module also includes a heart rate sensor set construction unit and a heart rate data acquisition unit.

[0100] The heart rate sensor set building unit builds a heart rate sensor set BS={BS a |a∈[1,A]}, where BS a represents the ath heart rate sensor, and A represents the total number of heart rate sensors; the heart rate sensor is used to detect and record the heart rate data of the human body.

[0101] The heart rate data acquisition unit collects the heart rate data of the patient during anesthesia. a All the heart rate data detected and recorded are used to construct the heart rate data set PD = {PD ab|a∈[1,A],b∈[1,B]}, where PD ab It represents the bth heart rate data detected by the ath heart rate sensor during the anesthesia process, and B represents the total number of heart rate data.

[0102] Furthermore, the time node data processing module also includes a time node division unit and a time node heart rate data set construction unit.

[0103] The time node division unit divides the entire process from the anesthesia induction period to the anesthesia awakening period of the patient into x time nodes, and defines the time node set TS = {TS t |t∈[1,x]}, where TS t represents the tth time node; the anesthesia induction period refers to the initial stage of the anesthesia process in which the patient enters an anesthesia state; the anesthesia recovery period refers to the end stage in which the patient recovers consciousness after the anesthesia is stopped.

[0104] The time node heart rate data set construction unit constructs a time node heart rate data set TPD based on the heart rate data set and the time node set. ab (TS t )|a∈[1,A],b∈[1,B],t∈[1,x]}, where TPD ab (TS t ) indicates that at time node TS t Heart rate data at the time node.

[0105] Furthermore, the heart rate data analysis module also includes a heart rate average value calculation unit and a comprehensive heart rate value calculation unit.

[0106] The heart rate average value calculation unit calculates the heart rate sensor BS a At time node TS t Heart rate data TPD at the time point of detection ab (TS t ) average value; based on the heart rate data TPD at the time node ab (TS t ) to calculate the heart rate sensor BS a At time node TS t To time node TS t+1 Heart rate data TPD at the time point of detection ab (TS t )’s heart rate variability.

[0107] The comprehensive heart rate value calculation unit calculates the time node TS t When the heart rate data collected by all heart rate sensors is a comprehensive heart rate value, based on the comprehensive heart rate value μ com(TS t ), calculated at time node TS t To time node TS t+1 The comprehensive heart rate variability over time.

[0108] Furthermore, the anesthetic efficacy evaluation module also includes an evaluation value calculation unit and an anesthetic effect determination unit.

[0109] The evaluation value calculation unit designs an evaluation value E(TS) based on the heart rate change rate and the comprehensive heart rate change degree. t →TS t+1 ).

[0110] The anesthetic effect determination unit presets an evaluation threshold τ, if E(TS t →TS t+1 )>τ, it is judged that the anesthetic effect is weakened and the drug efficacy is poor; if E(TS t →TS t+1 )≤τ, the anesthetic effect is judged to be effective and the drug efficacy is good.

[0111] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0112] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An artificial intelligence-based anesthetic efficacy evaluation method, characterized in that: The method comprises the following steps: Step S1: construct a heart rate sensor set, and collect all heart rate data detected and recorded by the heart rate sensor during the anesthesia process of the patient to construct a heart rate data set; Step S2: dividing the entire process from the anesthesia induction period to the anesthesia recovery period of the patient into time nodes, and constructing a time node heart rate data set based on the heart rate data set; Step S3: Calculate the average value of the time node heart rate data detected by the same heart rate sensor at the same time node, and calculate the heart rate change rate of the same heart rate sensor at different time nodes based on the average value; calculate the comprehensive heart rate value of the heart rate data collected by all heart rate sensors at the same time node, and calculate the comprehensive heart rate change degree at different time nodes; Step S4: Design and analyze evaluation values ​​based on the heart rate variability and the comprehensive heart rate variability.

2. The method for evaluating anesthetic efficacy based on artificial intelligence according to claim 1, characterized in that: The step S1 comprises the following steps: Construct heart rate sensor set BS = {BS a |a∈[1,A]}, where BS a represents the ath heart rate sensor, and A represents the total number of heart rate sensors; the heart rate sensor is used to detect and record the heart rate data of a human body; Collect the patient's heart rate sensor during anesthesia BS a All the heart rate data detected and recorded are used to construct the heart rate data set PD = {PD ab |a∈[1,A],b∈[1,B]}, where PD ab It represents the bth heart rate data detected by the ath heart rate sensor during the anesthesia process, and B represents the total number of heart rate data.

3. The method for evaluating anesthetic efficacy based on artificial intelligence according to claim 2, characterized in that: The step S2 comprises the following steps: The whole process from anesthesia induction to anesthesia recovery is divided into x time nodes, and the time node set TS = {TS t |t∈[1,x]}, where TS t represents the tth time node; the anesthesia induction period refers to the initial stage of the anesthesia process in which the patient enters the anesthesia state; the anesthesia recovery period refers to the end stage in which the patient recovers consciousness after the anesthesia is stopped; Based on the heart rate data set and the time node set, a time node heart rate data set TPD is constructed. ab (TSt)|a∈[1,A],b∈[1,B],t∈[1,x]},where TPD ab (TSt) means at time node TS t Heart rate data at the time node.

4. The method for evaluating anesthetic efficacy based on artificial intelligence according to claim 3, characterized in that: The step S3 comprises the following steps: Calculate heart rate sensor BS a At time node TS t Heart rate data TPD at the time point of detection ab (TS t ), calculated as follows: Among them, μ ab (TS t ) represents the time node heart rate data TPD ab (TS t )’s average value; Based on the heart rate data TPD at the time node ab (TS t ) to calculate the heart rate sensor BS a At time node TS t To time node TS t+1 Heart rate data TPD at the time point of detection ab (TS t ) is calculated by the following formula: Among them, μ ab (TS t+1 ) represents the time node TS t+1 Time node heart rate data TPD ab (TS t+1 ) average value, Δμ ab (TS t →TS t+1 ) represents the heart rate change rate; Calculated at time node TS t When the heart rate data collected by all heart rate sensors is calculated, the comprehensive heart rate value is calculated as follows: Among them, μ com (TS t ) represents the comprehensive heart rate value, ω a Indicates the ath heart rate sensor BS a Preset coefficients; Based on the comprehensive heart rate value μ com (TS t ), calculated at time node TS t To time node TS t+1 The comprehensive heart rate variability at 1000 Hz is calculated as follows: Dm com (TS t →TS t+1 )=μ com (TS t+1 )-m com (TS t ); Among them, μ com (TS t+1 ) represents the time node TS t+1 The comprehensive heart rate value at the time, Δμ com (TS t →TS t+1 ) represents the comprehensive heart rate variability.

5. The method for evaluating anesthetic efficacy based on artificial intelligence according to claim 4, characterized in that: The step S4 comprises the following steps: Based on the heart rate change rate and the comprehensive heart rate change degree, an evaluation value is designed, and the calculation formula is as follows: E(TS t →TS t+1 )=a·Dm ab (TS t →TS t+1 )+β·Dm com (TS t →TS t+1 ); Among them, α and β are the preset coefficients of variation, E(TS t →TS t+1 ) represents the evaluation value; Preset evaluation threshold τ; If E(TS t →TS t+1 )>τ, it is judged that the anesthetic effect is weakened and the drug efficacy is poor; If E(TS t →TS t+1 )≤τ, the anesthetic effect is judged to be effective and the drug efficacy is good.

6. An artificial intelligence-based anesthetic efficacy evaluation system, which executes an artificial intelligence-based anesthetic efficacy evaluation method as described in any one of claims 1 to 5, characterized in that: The system includes: a heart rate data acquisition module, a time node data processing module, a heart rate data analysis module and an anesthetic efficacy evaluation module; The heart rate data collection module constructs and manages a set of heart rate sensors and collects data from each sensor into a unified heart rate data set; The time node data processing module divides the anesthesia process into multiple time nodes and associates the collected heart rate data with these time nodes; The heart rate data analysis module calculates the average heart rate, heart rate change rate and comprehensive heart rate value based on the time node data, and identifies the relationship between the heart rate change and the depth of anesthesia; The anesthetic efficacy evaluation module determines whether the anesthetic efficacy is good by calculating the evaluation value and comparing it with a preset threshold value.

7. The artificial intelligence-based anesthetic efficacy evaluation system according to claim 6, characterized in that: The heart rate data acquisition module also includes a heart rate sensor set construction unit and a heart rate data acquisition unit; The heart rate sensor set building unit builds a heart rate sensor set BS={BS a |a∈[1,A]}, where BS a represents the ath heart rate sensor, and A represents the total number of heart rate sensors; the heart rate sensor is used to detect and record the heart rate data of a human body; The heart rate data acquisition unit collects the heart rate data of the patient during anesthesia. a All the heart rate data detected and recorded are used to construct the heart rate data set PD = {PD ab |a∈[1,A],b∈[1,B]}, where PD ab It represents the bth heart rate data detected by the ath heart rate sensor during the anesthesia process, and B represents the total number of heart rate data.

8. The artificial intelligence-based anesthetic efficacy evaluation system according to claim 7, characterized in that: The time node data processing module also includes a time node division unit and a time node heart rate data set construction unit; The time node division unit divides the entire process from the anesthesia induction period to the anesthesia awakening period of the patient into x time nodes, and defines the time node set TS = {TS t |t∈[1,x]}, where TS t represents the tth time node; the anesthesia induction period refers to the initial stage of the anesthesia process in which the patient enters the anesthesia state; the anesthesia recovery period refers to the end stage in which the patient recovers consciousness after the anesthesia is stopped; The time node heart rate data set construction unit constructs a time node heart rate data set TPD based on the heart rate data set and the time node set. ab (TS t )|a∈[1,A],b∈[1,B],t∈[1,x]},where TPD ab (TS t ) indicates that at time node TS t Heart rate data at the time node.

9. The artificial intelligence-based anesthetic efficacy evaluation system according to claim 8, characterized in that: The heart rate data analysis module also includes a heart rate average value calculation unit and a comprehensive heart rate value calculation unit; The heart rate average value calculation unit calculates the heart rate sensor BS a At time node TS t Heart rate data TPD at the time point of detection ab (TS t ) average value; based on the heart rate data TPD at the time node ab (TS t ) to calculate the heart rate sensor BS a At time node TS t To time node TS t+1 Heart rate data TPD at the time point of detection ab (TS t )’s heart rate variability; The comprehensive heart rate value calculation unit calculates the time node TS t When the heart rate data collected by all heart rate sensors is a comprehensive heart rate value, based on the comprehensive heart rate value μ com (TS t ), calculated at time node TS t To time node TS t+1 The comprehensive heart rate variability over time.

10. The artificial intelligence-based anesthetic efficacy evaluation system according to claim 9, characterized in that: The anesthetic efficacy evaluation module also includes an evaluation value calculation unit and an anesthetic effect determination unit; The evaluation value calculation unit designs an evaluation value E(TS) based on the heart rate change rate and the comprehensive heart rate change degree. t →TS t+1 ); The anesthetic effect determination unit presets an evaluation threshold τ, if E(TS t →TS t+1 )>τ, it is judged that the anesthetic effect is weakened and the drug efficacy is poor; if E(TS t →TS t+1 )≤τ, the anesthetic effect is judged to be effective and the drug efficacy is good.

Images