Intelligent analysis and adjustment method and system for hot perfusion liquid concentration

Abstract

The present disclosure proposes an intelligent analysis and adjustment method for the concentration of hot perfusion liquid medicine, which collects the patient's heart rate, body temperature and blood oxygen concentration vital sign data, monitors the patient's vital sign changes based on the vital sign data, obtains comparative reference features according to the change in the current patient's vital sign data during water perfusion and before perfusion, and rates the water perfusion adaptability of each patient according to the reference features, generates a matching cluster based on the sample library corresponding to it, and generates the initial liquid medicine concentration value and basic bias of the current patient during drug perfusion according to the liquid medicine concentration data of the patient in the matching cluster; based on the basic bias, analyzes the adaptability characteristics of the liquid medicine concentration according to the year-on-year change in the three groups of vital sign data of the patient during drug perfusion, obtains the acceptance value of the liquid medicine concentration in the next cycle during drug perfusion of the patient, and adjusts the concentration value of the perfusion liquid in real time. The implementation method can accurately find the appropriate perfusion liquid medicine concentration value for the patient, and make real-time adjustments based on the patient's treatment condition.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to a method and system for intelligent analysis and regulation of the concentration of a hot perfusion medicine solution. Background Art

[0002] During the process of thermal perfusion therapy, water perfusion is usually performed first, and drug circulation perfusion can be further used according to the patient's tolerance. Patients may experience discomfort from perfusion due to certain irritating drugs, and this discomfort cannot be directly obtained through sensors. During thermal perfusion, patients may experience excessive sweating, increased body temperature, increased heart rate, and respiratory depression. Therefore, medical staff need to monitor these signs of patients at all times to comprehensively judge whether the patients have abnormal conditions such as respiratory depression and decreased blood oxygen saturation;

[0003] However, we found that when using the above method to monitor patient vital signs, the following technical problems often occur:

[0004] First, when manually monitoring various vital signs data, it is impossible to accurately give the appropriate initial drug concentration based on the year-on-year changes in the patient's vital signs data before and during water perfusion;

[0005] Second, although some data collection equipment is used for manual monitoring, the adjustment of drug concentration during actual drug perfusion is prone to delays, resulting in patients receiving drug perfusion at inappropriate drug concentrations for a long time, which poses a risk of reducing the effect of thermal perfusion or causing iatrogenic damage to patients. Summary of the invention

[0006] The content of this disclosure is used to introduce concepts in a brief form, which will be described in detail in the detailed implementation section below. The content of this disclosure is not intended to identify the key features or essential features of the technical solution claimed for protection, nor is it intended to limit the scope of the technical solution claimed for protection.

[0007] Some embodiments of the present disclosure propose a method and system for intelligent analysis and adjustment of the concentration of hot perfusion liquid medicine to solve one or more of the technical problems mentioned in the above background technology section.

[0008] In a first aspect, some embodiments of the present disclosure provide a method for intelligent analysis and adjustment of hot perfusion liquid concentration, the method comprising: collecting three sets of vital sign data of the patient's heart rate H, body temperature T and blood oxygen concentration K, and monitoring changes in the patient's vital signs based on the three sets of vital sign data; obtaining comparative reference features according to the change in the three sets of vital sign data of the current patient during water perfusion and before perfusion; rating the water perfusion adaptability of each patient according to the reference features, and generating a matching cluster based on the corresponding one in the sample library; generating an initial liquid concentration value and a basic bias for the current patient during drug perfusion according to the liquid concentration data of the patient during drug perfusion in the matching clusters; based on the basic bias, analyzing the adaptability characteristics of the liquid concentration according to the year-on-year change in the three sets of vital sign data of the patient during drug perfusion, and obtaining the acceptance value of the liquid concentration in the next cycle during drug perfusion of the patient.

[0009] In a second aspect, some embodiments of the present disclosure provide a system for intelligent analysis and regulation of the concentration of hot perfusion liquid medicine, the method comprising: a memory, a processor, and a computer program stored in the above-mentioned memory and executable on the above-mentioned processor, wherein the above-mentioned processor implements any one of the steps of the above-mentioned method for intelligent analysis and regulation of the concentration of hot perfusion liquid medicine when executing the above-mentioned computer program.

[0010] The above-mentioned embodiments of the present disclosure have the following beneficial effects:

[0011] 1. It can first select a matching cluster from the sample library based on the current patient's vital sign data during water perfusion, and then use the drug concentration of all previous patients in the matching cluster during drug perfusion as a reference to accurately calculate an initial drug concentration suitable for the current patient during drug perfusion. In this way, while improving the effect of hot perfusion treatment, it also ensures that the patient's vital signs tend to be stable.

[0012] 2. In the case of untimely human monitoring, this method can automatically calculate the appropriate acceptance value of the drug solution concentration according to the real-time changes of the patient's vital signs during drug infusion, realize the monitoring of the entire drug infusion process, and avoid harm to the patient's body due to untimely adjustment of the drug solution concentration. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a flow chart of the method for intelligent analysis and adjustment of the concentration of hot perfusion liquid medicine disclosed in the present invention. DETAILED DESCRIPTION

[0014] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0015] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present disclosure can be combined with each other.

[0016] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0017] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0018] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0019] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0020] like Figure 1 As shown, in some embodiments, a method for intelligent analysis and adjustment of the concentration of hot perfusion liquid is proposed, which may specifically include the following steps.

[0021] Step 101 , collecting three sets of vital sign data of the patient, namely, heart rate H, body temperature T and blood oxygen concentration K, and monitoring changes in the patient's vital signs based on the three sets of vital sign data.

[0022] In some embodiments, in order to obtain the patient's current heart rate change H during chemotherapy, the patient needs to wear a heart rate monitoring device to record the patient's heart rate data during the water circulation and drug circulation process; a body temperature sensor is used to obtain the patient's body temperature data T, and the monitoring position of the body temperature sensor is the armpit and the abdominal surface; a smart wearable device is used to detect the patient's blood oxygen concentration data K.

[0023] Step 102, obtaining a comparative reference feature according to the change amount of the three groups of vital sign data of the patient before and during water perfusion.

[0024] In some embodiments, based on the physical condition information sequence of the current patient before perfusion (i.e., a sequence consisting of three sets of vital sign data) and the physical condition information sequence of the current patient during water perfusion, a dynamic time warping algorithm is used to obtain the reference feature R_a:

[0025] R_A=tanh{DTW(H_0,H_A )*DTW(T_0,T_A )*DTW(K_0,K_A )}

[0026] Among them, DTW is dynamic time warping, 0 is the subscript index of the patient's physical condition data information when no perfusion is performed, and A is the subscript index of the physical condition data information of the perfusion water cycle.

[0027] The reference feature R_a is used as the patient's stress level for perfusion chemotherapy. For the year-on-year changes in the three sets of vital sign data of the current patient before and after water perfusion, the greater the degree of change, the greater the stress level of the current patient for perfusion chemotherapy.

[0028] Step 103 , rating the water perfusion adaptability of each patient according to the reference features, and generating a corresponding matching cluster based on the sample library.

[0029] In some embodiments, the sample library contains data collected from patients undergoing thermal perfusion each time, and the collected data include the three groups of vital sign data. Among them, the vital sign data of patients currently undergoing thermal perfusion treatment will also be collected into the sample library, that is, the matching cluster mentioned below actually already includes the real-time data of the current patient.

[0030] Based on the reference feature R_a and three sets of vital sign data of all patients in the sample library during water perfusion, the difference distance D is calculated:

[0031] D(X,Y)=abs(R_A (X)-R_A (Y)) / (1+similarity(U ̅_X,U ̅_Y ) )

[0032] Among them, X and Y represent different patients in the sample library (including patients currently undergoing hot perfusion treatment), abs is the absolute value function, and U ̅ is the average value of the physical condition information collected per unit time before perfusion for each patient. Specifically, U ̅={(H_0 ) ̅,(T_0 ) ̅,(K_0 ) ̅}, which is the physical condition data of each patient before perfusion. The unit time can be 1min, 2min, 3min, 4min, or 5min.

[0033] When the physical sign data of different patients before perfusion are close, and the difference distance D between them is small during water perfusion, it means that the basic perfusion adaptability of the two patients is similar, otherwise it means that the basic perfusion adaptability of the two patients is quite different.

[0034] During each treatment, medical staff need to set the appropriate perfusion solution concentration for the patient first. Since the adjustment parameters of each perfusion chemotherapy machine are different, this solution is based on the unsupervised cluster analysis method to analyze the typicality of the patients and group them based on the difference distance D(X,Y) between the physical conditions of two types of patients using the k-means mean clustering algorithm.

[0035] First, assume that the number of cluster subsets of the physical condition set K=1, and calculate the Euclidean distance between the difference distances of each physical condition in the feature space, and use the largest Euclidean distance distMAX as the evaluation index of the global clustering characteristics of feature differences. If the maximum Euclidean distance is less than the set cluster verification threshold, it is considered that all elements of the above physical condition features belong to the same cluster adaptability type, otherwise let K=k+1.

[0036] Then, K-means is used to perform feature clustering on the above-mentioned multiple types of physical conditions to obtain the final clustering result {Csub(i), i=1,2...K}; then, a clustering verification characteristic analysis is performed on the subsets in the clustering result {Csub(i), i=1,2...K} to find the maximum value of the Euclidean distance between the physical condition features in each cluster subset in the feature space; the global maximum Euclidean distance distMAX(Cglobal)' in the cluster subset is selected, and if distMAX(Cglobal)' is less than the set clustering verification threshold, it is considered that the obtained cluster subset {Csub(i), i=1,2...K} meets the clustering requirements, otherwise K-means clustering is performed again.

[0037] Finally, the physical condition type of each patient included in the sample library is classified into a specific cluster among the K clusters, thereby obtaining the matching cluster group corresponding to the current patient in the sample library.

[0038] Step 104 : Based on the drug concentration data of all previous patients in the matching cluster during drug infusion, an initial drug concentration value and a basic offset during drug infusion of the current patient are generated.

[0039] Since the distribution of each patient in the matching cluster is not completely appropriate, the calculation can only be done for the current patient in the current matching cluster, and the specific perfusion concentration of the current patient cannot be calculated. When calculating the initial drug concentration value of the current patient's drug perfusion, the program will first give a virtual drug concentration of the current patient's last drug perfusion (if the sample library already has various signs data of the current patient during hot perfusion treatment, the actual drug concentration of the last drug perfusion will be called).

[0040] In some embodiments, the average N_G of the drug concentration of each patient in the matching cluster is calculated:

[0041] N_G=mean(N_x) , where N_x is the initial drug concentration of the patients in the matching cluster (i.e., the drug concentration initially set by the medical staff).

[0042] First, the typicality δ_X is calculated based on the sum of the differences between the drug concentrations of each patient in the matching cluster and the rest of the patients in the cluster:

[0043] δ_X=Σabs(N_(Q_X )-N_(G_i ) )

[0044] Here, N_(Q_X) is the concentration of the drug solution during the last drug infusion of the current patient (which may be an assumed value), and N_(G_i) is the concentration of the drug solution during drug infusion of the patients in the matching cluster, thereby obtaining the typical degree δ of each patient. The larger the δ value at this time, the more accurately the patient is considered to represent the typical patient at the matching cluster level.

[0045] Then, the typical degree δ_X of all patients in the matching cluster is standardized by the range to ensure that the typical degree δ of patients in the matching cluster does not exceed 1. Then the δ of all patients is processed to make δ_X=1-δ_X, and the typical degree δ_x of the drug concentration during drug infusion of the current patient relative to the drug concentration that the patients in the matching cluster can accept is obtained.

[0046] Finally, based on the average value N_G and typical degree δ_X of the drug solution concentration during drug infusion of patients in the above matching clusters, the initial drug solution concentration Ni_i=min(δ_X*N_G+(1-δ_X ) N_(Q_X ),N_(Q_X ) ) during drug infusion of the patient is determined.

[0047] Here, the initial drug concentration Ni is based on not exceeding N_(Q_X), so that the drug concentration of the current patient during hot perfusion is adjusted based on the average drug concentration of all patients in the matching cluster, thereby alleviating the problem of severe discomfort to patients caused by the high initial drug concentration given by manual judgment.

[0048] Based on the calculated initial drug concentration Ni during drug infusion of the current patient and the drug concentration of previous patients in the matching cluster, the basic deviation of the current patient relative to the patients in the matching cluster can be known; the larger the typicality δ value of the current patient, the closer the initial drug concentration Ni is to the average drug concentration N_G of the patients in the matching cluster, and vice versa, the closer the infusion drug concentration is to the patient's last infusion drug concentration N_(Q_X).

[0049] The above steps can solve the problem in the background technology that "when monitoring various vital signs data manually, it is impossible to accurately give a suitable initial drug concentration based on the year-on-year changes in the vital signs data of the patient before and after water infusion". First, based on the concentration of the drug solution at the last drug infusion of the current patient (which may be an assumed value), according to the sum of the differences between the drug solution concentrations of each patient in the matching cluster and the rest of the patients in the cluster, the typicality of each patient in the sample library is calculated to determine the matching degree between the patient and the matching cluster. Then, the typicality of all patients in the matching cluster is standardized by the extreme difference. A typical degree value not greater than 1 is obtained. Finally, based on the average value and typical degree of the drug solution concentration during drug infusion of patients in the above matching clusters, the initial drug solution concentration during infusion of the current patient can be comprehensively calculated. In this way, first, according to the vital sign data of the current patient during water infusion, a matching cluster that matches it is selected from the sample library, and then the drug solution concentration of all previous patients in the matching cluster during drug infusion is used as a reference to calculate in turn an initial drug solution concentration that accurately matches the current patient, which can improve the effect of hot perfusion treatment while ensuring that the patient's vital signs tend to be stable.

[0050] One point that needs to be emphasized is that the initial drug solution concentration Ni obtained here will not necessarily be directly used as the setting value of the patient's actual initial drug solution concentration. It is only a conservative value calculated. According to other vital sign data not detected in this method, the medical staff can also actively modify the calculated initial drug solution concentration after subjective judgment.

[0051] Step 105 , based on the basic bias, according to the year-on-year changes of the three groups of vital sign data of the patient during drug infusion, analyze the adaptability characteristics of the drug solution concentration to obtain the acceptance value of the drug solution concentration in the next cycle during drug infusion of the patient.

[0052] In some embodiments, first, the difference feature R_B of the current patient's vital sign data during drug infusion and water infusion is calculated. The calculation method of the difference feature R_B is consistent with the calculation method of the reference feature R_A, except that the corresponding data before infusion and during water infusion are changed to data during water infusion and drug infusion.

[0053] Then, the patient's adaptability S to the concentration of the drug solution is calculated based on the difference feature R_B:

[0054] S=ReLU(mean(〖R_B〗_last )-mean(〖R_B〗_now ))

[0055] Among them, mean(〖R_B〗_last ) is the average value of the difference characteristics of the drug solution concentration in the previous cycle when the current patient is infused with medicine; mean(〖R_B〗_now )) is the average value of the difference characteristics of the drug solution concentration in the current cycle when the current patient is infused with medicine; one cycle period is a sampling window, that is, the drug solution concentration is changed once for one drug cycle. After each cycle is determined for a drug solution concentration, mean(〖R_B〗_last ) and mean(〖R_B〗_now ) will be updated.

[0056] When S>0, S is the original value, when S<0, S=0;

[0057] After the patient has completed a drug cycle, the range of S is standardized to ensure that the S range is [0,1]. The closer S is to 1, the stronger the patient's adaptability to the drug is, indicating that the concentration of the drug solution can be appropriately increased in the next drug cycle. Otherwise, the concentration of the drug solution in this cycle remains unchanged.

[0058] In this link, an alarm prompt program can also be added. If the value of S is always less than 0 within a unit time under the cycle of the same drug concentration, an alarm signal will be issued. The unit time can be set to 1min, 2min, 3min, 4min, or 5min.

[0059] Finally, based on the average value N_G of the drug solution concentration during drug infusion of patients in the matching cluster, the offline dynamic mixing estimate N_D=S*N_G+(1-S)*N_i is calculated;

[0060] The offline dynamic mixing estimation value N_D is used as the drug solution concentration value in the next drug cycle when the current patient is infused with drugs. The above steps are repeated to judge the drug solution concentration value in the next drug cycle, thereby making adjustments again.

[0061] By adopting the above steps, the problem in the background technology that "although some acquisition equipment is used for manual monitoring, the adjustment of the concentration of the drug solution during actual drug perfusion is prone to delay, resulting in the patient being perfused for a long time at an inappropriate drug solution concentration, which has the risk of reducing the effect of thermal perfusion or causing iatrogenic damage to the patient" can be solved. First, a difference feature is obtained by monitoring the changes in the patient's vital signs data during drug perfusion. Then, the stress intensity of the current patient is calculated according to the average value of the difference feature of the drug solution concentration in the previous cycle of drug perfusion and the average value of the difference feature of the drug solution concentration in the current cycle. Finally, the stress intensity of the current patient is combined with the average value of the drug solution concentration of the patient in the matching cluster during drug perfusion to perform offline dynamic mixed estimation, thereby obtaining an estimated value, which is used as the drug solution concentration value in the next cycle. In this way, in the case of untimely manual monitoring, the system in the thermal perfusion can automatically calculate the appropriate drug solution concentration acceptance value according to the real-time changes of the patient's vital signs during drug perfusion by running this method, thereby realizing automatic monitoring of the entire drug perfusion process and avoiding damage to the patient's body due to untimely adjustment of the drug solution concentration.

[0062] The present invention also proposes a hot perfusion liquid concentration intelligent analysis and adjustment system, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any step of a hot perfusion liquid concentration intelligent analysis and adjustment method.

[0063] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0064] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, sample library or other media used in the embodiments provided in the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0065] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0066] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

[0067] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for intelligent analysis and adjustment of the concentration of hot perfusion liquid medicine, characterized in that: The method comprises: Collect three sets of vital sign data of the patient's heart rate H, body temperature T and blood oxygen concentration K, and monitor the changes of the patient's vital signs based on the three sets of vital sign data; According to the changes in the three sets of vital sign data of the current patient during water perfusion and before perfusion, comparative reference features are obtained, including: The three sets of physical sign data collected during and before water perfusion are combined into two sets of physical condition information sequences; Based on the physical condition information sequence, the reference feature R is obtained according to the dynamic time warping formula A ; R A =tanh{DTW(H0,H A )*DTW(T0,T A )*DTW(K0,K A )} Among them, DTW is dynamic time warping, 0 is the subscript index of the patient's physical condition data information when no perfusion is performed, and A is the subscript index of the physical condition data information of the perfusion water cycle; The reference feature R A As the patient's stress level for infusion chemotherapy; Each patient's water perfusion adaptability was rated based on reference characteristics, generating matching clusters based on the sample library, including: The sample library is used to collect various data of patients undergoing thermal perfusion each time, and the collected data includes the three groups of vital sign data, wherein the data of the current patient will also be collected in the sample library; Calculate the difference distance D based on the stress level and three sets of physical sign data of the current patient and all previous patients in the sample library; Among them, X and Y represent different patients in the sample library, abs is the absolute value function, The average value of the physical condition information collected per unit time for each patient before perfusion, specifically: As the physical condition data of each patient before perfusion; Based on the difference distance D between patients, the k-means unsupervised clustering method is used to divide the patients into matching clusters of corresponding levels; According to the drug concentration data of the patients in the matching cluster during drug infusion, the initial drug concentration value and basic bias of the current patient during drug infusion are generated, including: According to the sum of the differences between the drug concentrations of each patient in the matching cluster and the rest of the patients in the cluster, the typical degree δ of each patient is obtained. X ; is the concentration of the drug solution when the current patient was last infused. To match the concentration of drug solution during drug infusion in the cluster, the typical level of each patient is obtained. X ; Based on the concentration of the drug solution of the patients in the matching cluster and the typical degree of the current patient in the matching cluster δ X , get the initial perfusion concentration N of the current patient in this perfusion chemotherapy i ; in, is the concentration of the drug solution at the last drug infusion of the current patient, N G is the average value of the drug solution concentration during drug infusion of patients in the matching cluster; Based on the basic bias, according to the year-on-year changes in the three groups of vital signs data of the patient during drug infusion, the adaptability characteristics of the drug solution concentration are analyzed to obtain the acceptance value of the drug solution concentration in the next cycle during drug infusion, including: Calculate the difference feature R between the current patient's vital sign data during drug infusion and water infusion B , wherein the difference feature R B The calculation method is the same as the reference feature R A The calculation method is consistent with R B =tanh{DTW(H A ,H B )*DTW(T A ,T B )*DTW(K A ,K B )} Wherein, DTW is dynamic time warping, A is the subscript index of the patient's physical condition data information during water infusion, and B is the subscript index of the patient's physical condition data information during drug infusion; Calculate the patient's adaptability to drug concentration during drug infusion in, It is the average value of the difference characteristics of the drug solution concentration in the previous cycle when the current patient is infused with drugs; It is the average value of the difference characteristics of the drug solution concentration under this cycle during the current patient's drug infusion; According to the patient's adaptability to the drug solution S, the initial drug solution concentration N i and the average N of the drug concentration of the patients in the matching cluster during drug infusion G , calculate the offline dynamic hybrid estimate N D ; N D =S*N G +(1-S)*N i The estimated value N D Used as the drug solution concentration value in the next drug cycle when the patient is infused with drugs.

2. A hot perfusion liquid concentration intelligent analysis and regulation system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to claim 1 are implemented.

Images