A pain prediction and warning method and system based on medical pain scoring scale

Abstract

The invention discloses a pain prediction and warning method and system based on a medical pain scoring scale, and relates to the field of pain prediction and warning. The invention optimizes the classification center data of the collected pain characteristic data, thereby providing a classification tool for subsequent classification of the pain characteristic data of the patient to be predicted to have pain at a future moment; constructs multiple mapping models, thereby providing a mapping tool for subsequent classification and mapping of the psychological state data, environmental factor data and physiological index data of the patient to be predicted to have pain at a future moment; determines whether to give a primary warning and a secondary warning based on the pain type data at the predicted future moment, and when giving a primary warning, prompts the patient to take drug treatment to relieve the pain at the future moment; and when giving a secondary warning, indicates that the pain occurring at the future moment cannot be effectively relieved only by drug treatment, thereby prompting the patient to take further measures to relieve the pain.

Description

Technical Field

[0001] The present invention belongs to the field of pain prediction and warning, and in particular, relates to a pain prediction and warning method and system based on a medical pain scoring scale. Background Art

[0002] When patients develop symptoms, they are often accompanied by pain; these pains may be varied, some of which are inevitably more severe. If the pain that may occur in the future cannot be predicted and warned, it will not only increase the patient's pain level, but also cause medical staff to be passive, requiring them to take emergency measures to relieve the patient's pain when the pain occurs, increasing the burden on medical staff. Summary of the invention

[0003] In view of the problems in the related art, the present invention proposes a pain prediction and warning method and system based on a medical pain scoring scale to overcome the above-mentioned technical problems existing in the existing related art.

[0004] In order to solve the above technical problems, the present invention is achieved through the following technical solutions:

[0005] The present invention is a pain prediction and warning method based on a medical pain scoring scale, comprising the following steps:

[0006] S1. Collect pain characteristic data and corresponding pain type data, psychological state data, environmental factor data and physiological index data of multiple patients;

[0007] S2. Optimizing the classification center data of the pain characteristic data of multiple patients in conjunction with the collected pain type data to obtain a pain characteristic classification center data matrix; constructing multiple mapping models based on the collected psychological state data, environmental factor data, physiological indicator data and pain type data of multiple patients to obtain a mapping model set;

[0008] S3, predicting data at multiple future time points based on the psychological state data, environmental factor data, and physiological index data of the patient to be predicted for pain at multiple historical time points, to obtain a future psychological state data matrix, a future environmental factor data matrix, and a future physiological index data matrix;

[0009] S4, using a mapping model set to classify and map the future psychological state data matrix, the future environmental factor data matrix, and the future physiological indicator data matrix to obtain a future pain feature data matrix; using a pain feature classification center data matrix to classify the future pain feature data matrix to obtain a future pain category data set;

[0010] S5, determining whether to issue a warning once according to the future pain type data set, and when issuing a warning once, adjusting the physiological index data of the patient to be predicted to obtain an adjusted future pain type data set; and determining whether to issue a second warning according to the adjusted future pain type data set;

[0011] Since different types of pain have different characteristics, the current type of pain can be determined through a variety of characteristic data of pain. In addition, the occurrence of pain in pain patients is often closely related to their psychological state, physiological indicators and factors of the patient's environment. Based on the above, this solution collects pain characteristic data of multiple patients and corresponding pain type data, psychological state data, environmental factor data and physiological indicator data, which provides data support for the subsequent acquisition of classification center data of pain characteristic data and the construction of a mapping model between psychological state data, environmental factor data and physiological indicator data and pain characteristic data. By optimizing the classification center data of the collected pain characteristic data, a classification tool is provided for the subsequent classification of pain characteristic data of patients to be predicted at future moments. By constructing multiple mapping models, a mapping tool is provided for the subsequent classification and mapping of the psychological state data, environmental factor data and physiological indicator data of the patients with predicted pain in the future; by predicting the psychological state data, environmental factor data and physiological indicator data of the patients with predicted pain in the future, a data basis is provided for the subsequent analysis of the occurrence of pain in the patients with predicted pain in the future, and then timely warnings are given to pre-treat the patients, so that the patients will have less and milder pain in the future; when a warning is given once, the patient is prompted to take drug treatment to relieve the pain in the future; when a second warning is given, it indicates that drug treatment alone cannot effectively relieve the pain that occurs in the future, thereby prompting the patient to take further measures to relieve the pain.

[0012] Preferably, the S1 comprises the following steps:

[0013] S11, setting multiple pain patients, pain characteristic types and pain categories, obtaining a pain patient set, a pain category set and a pain characteristic type set; collecting pain characteristic data and corresponding pain category data of each patient in the pain patient set when pain occurs according to the pain characteristic type set and the pain category set in conjunction with a medical pain rating scale, to obtain a pain characteristic data matrix a And pain category dataset , Indicates the setting i Data on the types of pain experienced by the pain patients, represents the total number of pain patients set; ;

[0014] in, Indicates the setting i The first time pain occurred in patients with pain j Types of pain feature data, Indicates the total number of pain feature types set;

[0015] S12, setting a plurality of psychological state types, environmental factor types and physiological indicator types to obtain a psychological state type set, an environmental factor type set and a physiological indicator type set; collecting a pain type data set according to the psychological state type set, the environmental factor type set and the physiological indicator type set. The psychological state data, environmental factor data and physiological index data of the pain patient when each type of pain occurs are obtained to obtain the psychological state data matrix , Environmental Factors Data Matrix And the physiological index data matrix ; They are as follows,

[0016] ; ; ;

[0017] in, , , Respectively represent the set i The first time the pain occurred j types of psychological state data, environmental factor data, and physiological indicator data. , , Respectively represent the total number of set psychological state types, environmental factor types, and physiological indicator types;

[0018] Pain types include acute pain, chronic pain, severe pain, moderate pain and mild pain, among which, the medical pain rating scale can help determine the type of pain in a quantitative way; pain characteristic types include pain intensity and pain duration, through which the type of pain can be determined; psychological state types include anxiety levels, environmental factors types include temperature and humidity, and physiological indicators types include heart rate and blood pressure.

[0019] Preferably, S2 comprises the following steps:

[0020] S21. According to the pain type dataset Pain feature data matrix a The pain feature data set in is classified to obtain the pain feature data classification matrix set , Represents the pain feature data matrix a The first i A pain feature data classification matrix, Represents the pain feature data matrix a The total number of pain feature data classification matrices obtained by classification; obtaining the optimal data center of each pain feature data classification matrix in the pain feature data classification matrix set, and obtaining the pain feature classification center data matrix ; , They are as follows: ; ;

[0021] in, express Middle j The first time pain occurred in patients with pain k Types of pain feature data, Indicates classification to The total number of pain patients in Represents the pain feature data matrix a The first i The classification center dataset of the pain feature data classification matrix j Types of pain characteristic data;

[0022] S22, according to the pain characteristic data matrix a Set multiple pain feature mapping model interfaces to obtain a mapping model interface set , It represents the mapping from the psychological state data, environmental factor data and physiological index data of the pain patient to the first i The interface corresponding to the mapping model of the types of pain characteristic data; according to the psychological state data matrix , Environmental Factors Data Matrix , Physiological index data matrix And the pain feature data matrix a In the mapping model interface set Build the corresponding mapping model under each mapping model interface to obtain the mapping model set , express The corresponding mapping model;

[0023] Since each group of pain feature data has corresponding pain type data, multiple groups of pain feature data can be classified according to the pain type data; by iteratively optimizing the classification center data of each pain feature data classification matrix in the pain feature data classification matrix set for multiple times, the classification center data finally obtained can more accurately represent the numerical characteristics of each data in the current pain feature data classification matrix; since the mapping relationship between different types of pain feature data and psychological state data, environmental factor data and physiological indicator data is different, a corresponding mapping model is constructed for each type of pain feature data in this scheme, thereby improving the accuracy of subsequent mapping of the psychological state data, environmental factor data and physiological indicator data of patients with predicted pain.

[0024] Preferably, in S21, the optimal data center of each pain feature data classification matrix in the pain feature data classification matrix set is obtained to obtain the pain feature classification center data matrix The following steps are involved:

[0025] S211. Constructing the first eagle population , Indicates the first eagle population i A hawk, represents the size of the first eagle population; sets the maximum number of iterations of the first eagle population to , the current number of iterations is , respectively recorded as the first maximum number of iterations and the first current number of iterations; the search space dimension of the first eagle population is ;

[0026] S212, multiple times from the pain feature data classification matrix set A row of data is randomly selected from each pain feature data classification matrix in the first population of eagles to form the initial position matrix of each eagle. The first initial position matrix set is obtained. , Indicates the first eagle population i The initial position matrix of the eagle is as follows: ;

[0027] in, express In the j The classification center dataset of the pain feature data classification matrix k The components of the pain feature data dimensions of the types;

[0028] S213, according to Construct the first eagle population i The fitness function of a hawk ;as follows,

[0029] ;

[0030] in, Represents the pain feature data matrix a The first The total number of pain patients in the pain feature data classification matrix, Represents the pain feature data matrix a The first In the pain feature data classification matrix j The first time pain occurred in patients with pain k Types of pain characteristic data;

[0031] S214, start iteration, before iteration, set the first current iteration number Set to 1; in the first round of iteration, the first eagle population is used. i The fitness function of a hawk Calculate the first initial position matrix set The fitness value of the initial position matrix of each eagle in the first fitness value set is obtained to obtain a first fitness value set; the maximum fitness value in the first fitness value set and the corresponding initial position matrix of the eagle are respectively used as the first global optimal fitness value and the first global optimal position; the first initial position matrix set is adjusted according to the first global optimal fitness value and the first global optimal position. The initial position matrix of each eagle in is updated; after the update is completed, the first current iteration number Add 1 and enter the next iteration;

[0032] In each other round of iteration, the first eagle population is used i The fitness function of a hawk Calculate the fitness value of the position matrix of each eagle updated in the previous round of iterations to obtain a second fitness value set; use the maximum fitness value in the second fitness value set and the corresponding initial position matrix of the eagle as the second global optimal fitness value and the second global optimal position respectively; update the position matrix of each eagle updated in the previous round of iterations according to the second global optimal fitness value and the second global optimal position; after the update is completed, the first current iteration number Add 1 and enter the next iteration;

[0033] S215, when When , the iteration is stopped and the first final global optimal position is output; the first final global optimal position is used as the pain feature classification center data matrix Otherwise, continue iterating until until the time;

[0034] The Eagle Optimization Algorithm simulates the various hunting behaviors of the Eagle. The algorithm can effectively explore in the search space to find the global optimal solution or the area close to the optimal solution. At the same time, through different search strategies, it can achieve a balance between exploring new search areas and developing known areas, and can better jump out of the local optimal solution and find the global optimal solution. Based on the above advantages of the Eagle Optimization Algorithm, this scheme adopts the Eagle Optimization Algorithm to iteratively optimize the classification center data of each pain feature data classification matrix in the pain feature data classification matrix set, wherein the overall discrete degree of the pain feature data classification matrix set is used as a measure of the quality of the solution found. As the iteration proceeds, the overall discrete degree of the pain feature data classification matrix set will become smaller and smaller, indicating that the classification center data found is more accurate.

[0035] Preferably, in S22, according to the mental state data matrix , Environmental Factors Data Matrix , Physiological index data matrix And the pain feature data matrix a In the mapping model interface set Build the corresponding mapping model under each mapping model interface to obtain the mapping model set The following steps are involved:

[0036] S221, extracting the pain feature data matrix a Middle j Column data, get the current pain feature data set ; Reconstruct the j An initial multi-layer perceptron model;

[0037] S222: Using the current pain feature dataset , psychological state data matrix , Environmental Factors Data Matrix And the physiological index data matrix For j The initial multi-layer perceptron model is trained and tested; after the training and testing are completed, the first j A final multi-layer perceptron model;

[0038] The said j The final multilayer perceptron model is used as the mapping model interface set Middle j A mapping model under a mapping model interface;

[0039] S223, repeat S221 and S222 multiple times until the mapping model interface set The mapping models under each mapping model interface are trained and tested to obtain the mapping model set ;

[0040] Preferably, the current pain feature data set is used in S222 , psychological state data matrix , Environmental Factors Data Matrix And the physiological index data matrix For j The initial multi-layer perceptron model is trained and tested; after the training and testing are completed, the first j The final multilayer perceptron model includes the following steps:

[0041] S2221, setting a first training data ratio; and performing a training on the current pain feature data set according to the first training data ratio. , psychological state data matrix , Environmental Factors Data Matrix And the physiological index data matrix Perform data division to obtain a current pain feature training data set, a psychological state training data matrix, an environmental factor training data matrix, a physiological index training data matrix, a current pain feature test data set, a psychological state test data matrix, an environmental factor test data matrix, and a physiological index test data matrix;

[0042] S2222, setting a first training error threshold; inputting the current pain feature training data set, the psychological state training data matrix, the environmental factor training data matrix and the physiological index training data matrix into the first j The method comprises the following steps: performing training in an initial multilayer perceptron model; during the training process, when the training error is less than the first training error threshold, stopping the training to obtain the jth trained multilayer perceptron model; otherwise, continuing the training until the training error is less than the first training error threshold;

[0043] S2223, set a first test accuracy threshold; input the current pain feature test data set, psychological state test data matrix, environmental factor test data matrix and physiological index test data matrix into the first j After the test is completed, the first test accuracy is obtained; when the first test accuracy is greater than or equal to the first test accuracy threshold, the first j The trained multi-layer perceptron model is used as the j A final multilayer perceptron model; otherwise, return to S2222 to continue the jtraining a trained multi-layer perceptron model until the first test accuracy is greater than or equal to a first test accuracy threshold;

[0044] By using the current pain feature training data set, the psychological state training data matrix, the environmental factor training data matrix, the physiological index training data matrix, and the current pain feature test data set, the psychological state test data matrix, the environmental factor test data matrix, and the physiological index test data matrix respectively j The initial multi-layer perceptron model is trained and tested so that the obtained j The final multi-layer perceptron model has a good mapping ability for psychological state data, environmental factor data, physiological indicator data and pain characteristic data, thus providing an accurate mapping model for the subsequent mapping of psychological state data, environmental factor data and physiological indicator data of patients with predicted pain.

[0045] Preferably, S3 comprises the following steps:

[0046] S31, setting the pain patient to be predicted, multiple historical time points and future time points, and obtaining a historical time point set and future time points , , Respectively represent the set i historical time points and future time points, , Respectively represent the total number of set historical time points and future time points;

[0047] S32, matching the psychological state type set, the environmental factor type set and the physiological indicator type set and according to the historical time point set Collect the psychological state data, environmental factor data and physiological index data of the patient to be predicted at multiple historical time points to obtain the historical psychological state data matrix , Historical environmental factors data matrix And the historical physiological index data matrix ; They are as follows, ; ;

[0048] ;

[0049] in, , , They respectively represent the pain patients to be predicted in the i At a historical point in time j types of psychological state data, environmental factor data, and physiological indicator data;

[0050] S33, according to the historical psychological state data matrix , Historical environmental factors data matrix And the historical physiological index data matrix Construct the second final multi-layer perceptron model; cooperate with the future time point set The second final multi-layer perceptron model is used to predict the psychological state data, environmental factor data and physiological index data of the patient with pain to be predicted in the future, and the future psychological state data matrix is ​​obtained. , Future Environmental Factors Data Matrix And the future physiological index data matrix ; They are as follows, ; ; ;

[0051] in, , , They respectively represent the pain patients to be predicted in the i At a future time point j types of psychological state data, environmental factor data, and physiological indicator data;

[0052] Preferably, in S33, according to the historical psychological state data matrix , Historical environmental factors data matrix And the historical physiological index data matrix Construct the second final multilayer perceptron model, with the future time point set The second final multi-layer perceptron model is used to predict the psychological state data, environmental factor data and physiological index data of the patient with pain to be predicted in the future, and the future psychological state data matrix is ​​obtained. , Future Environmental Factors Data Matrix And the future physiological index data matrix The following steps are involved:

[0053] S331, construct a second initial multilayer perceptron model and set a second training data ratio and a test data ratio; according to the second training data ratio and the test data ratio, the historical psychological state data matrix , Historical environmental factors data matrix And the historical physiological index data matrix Perform data division to obtain a historical psychological state training data matrix, a historical environmental factor training data matrix, a historical physiological indicator training data matrix, a historical psychological state test data matrix, a historical environmental factor test data matrix, a historical physiological indicator test data matrix, a historical psychological state to-be-predicted data matrix, a historical environmental factor to-be-predicted data matrix, and a historical physiological indicator to-be-predicted data matrix;

[0054] S332, setting a second training error threshold; inputting the historical psychological state training data matrix, the historical environmental factor training data matrix, and the historical physiological index training data matrix into a second initial multilayer perceptron model for training; during the training process, when the training error is less than the second training error threshold, stopping the training to obtain a second trained multilayer perceptron model; otherwise, continuing the training until the training error is less than the second training error threshold;

[0055] S333, setting a second test accuracy threshold; inputting the historical psychological state test data matrix, the historical environmental factor test data matrix and the historical physiological index test data matrix into the second trained multilayer perceptron model for testing; after the test is completed, obtaining the second test accuracy; when the second test accuracy is greater than or equal to the second test accuracy threshold, using the second trained multilayer perceptron model as the second final multilayer perceptron model; otherwise, returning to S332 to continue training the second trained multilayer perceptron model until the second test accuracy is greater than or equal to the second test accuracy threshold;

[0056] S334, cooperate with future time point set The historical psychological state data matrix to be predicted, the historical environmental factor data matrix to be predicted, and the historical physiological index data matrix to be predicted are input into the second final multi-layer perceptron model for prediction to obtain the future psychological state data matrix , Future Environmental Factors Data Matrix And the future physiological index data matrix ;

[0057] By training and testing the second initial multilayer perceptron model using the historical psychological state training data matrix, the historical environmental factor training data matrix, the historical physiological indicator training data matrix, the historical psychological state test data matrix, the historical environmental factor test data matrix, and the historical physiological indicator test data matrix respectively, the obtained second final multilayer perceptron model has good prediction capabilities for psychological state data, environmental factor data, and physiological indicator data, providing an accurate prediction model for subsequent predictions of the historical psychological state data matrix to be predicted, the historical environmental factor data matrix to be predicted, and the historical physiological indicator data matrix to be predicted.

[0058] Preferably, S4 comprises the following steps:

[0059] S41, the future mental state data matrix , Future Environmental Factors Data Matrix And the future physiological index data matrix Input to the mapping model interface set In each mapping model interface in the Each mapping model in the future psychological state data matrix , Future Environmental Factors Data Matrix And the future physiological index data matrix Perform classification mapping to obtain the future pain feature data matrix ;as follows, ;

[0060] in, It indicates that the predicted pain patient is in the i The pain occurs at a future time point j Types of pain characteristic data;

[0061] S42, classifying the central data matrix according to the pain characteristics Future pain feature data matrix Classify the pain feature data at each future time point in the future to obtain the future pain category dataset , Indicates that the patient with pain to be predicted is in i Data on pain type at future time points;

[0062] Preferably, the S42 comprises the following steps:

[0063] S421, calculating the future pain feature data matrix Each row of data in the pain feature classification center data matrix The Euclidean distance between each row of data in is obtained to obtain the Euclidean distance matrix;

[0064] S422, taking the pain feature classification center data corresponding to the minimum Euclidean distance in each row of the Euclidean distance matrix as the category of the corresponding future pain feature data, to obtain a future pain category data set ;

[0065] By calculating the Euclidean distance between each row of data in the future pain feature data matrix and each row of data in the pain feature classification center data matrix, we can find which row of data in the future pain feature data matrix has the shortest distance to the pain feature classification center data matrix, thereby achieving the classification of the future pain feature data matrix.

[0066] Preferably, the S5 comprises the following steps:

[0067] S51, setting a warning pain type set and based on the future pain type data set Set the therapeutic drug type set , Indicates the setting i Types of therapeutic drugs, Indicates the total number of set therapeutic drug types;

[0068] When the future pain type dataset When there is future pain type data in the warning pain type set, an alarm is issued and the future time point where the future pain type data is located is recorded as the warning future time point; in conjunction with the treatment drug type set and adjusting the physiological index data of the patient to be predicted to have pain at a warning point in the future; after the adjustment is completed, an adjusted data set of future pain categories is obtained;

[0069] Otherwise, no warning is given;

[0070] S52, when there is future pain type data in the adjusted future pain type data set that is in the warning pain type set, a second warning is issued; otherwise, no second warning is issued;

[0071] Warning pain types mainly include acute pain, severe pain and other more serious pain types; because the dosage of therapeutic drugs used for each type of pain is different, this plan optimizes the dosage of each therapeutic drug at each medication time point for different pain types to achieve the best pain relief effect.

[0072] Preferably, in S51, the therapeutic drug type set And the physiological index data of the patient to be predicted to have pain are adjusted at a warning time point in the future; after the adjustment is completed, obtaining the adjusted future pain type data set includes the following steps:

[0073] S511, setting multiple medication time points before the future warning time point to obtain a medication time point set , Indicates the setting i The time point of taking medicine, Indicates the total number of set medication time points; set the type of therapeutic drugs to be taken at each medication time point The lower and upper limits of the dosage of each therapeutic drug in the formula are used to obtain the dosage interval matrix of the therapeutic drugs. ;as follows,

[0074] ;

[0075] in, , Respectively expressed in i Take the medicine at the time of taking j The lower and upper limits of the dosage of the therapeutic drug;

[0076] S512, build the second eagle population , Indicates the second eagle population i A hawk, represents the size of the second eagle population; sets the maximum number of iterations of the second eagle population to , the current number of iterations is , respectively recorded as the second maximum number of iterations and the second current number of iterations; the search space dimension of the second eagle population is ;

[0077] S513, according to the dosage interval matrix of the therapeutic drug Set the initial position matrix of each eagle in the second eagle population to obtain the second initial position matrix set , Indicates the second eagle population i The initial position matrix of the eagle is as follows: ;

[0078] in, express In the i Take the medicine at the time of taking j The amount of the therapeutic drug used;

[0079] S514. Based on the future pain type dataset Construct the fitness function of the second eagle population ;as follows,

[0080] ;

[0081] In the formula, Represents the future pain category dataset The number of future pain category data in the warning pain category set, βA positive number indicates a correction parameter;

[0082] S515, start iteration, before iteration, set the second current iteration number Set to 1; the fitness function of the second eagle population is used in the first round of iteration And cooperate with the second final multi-layer perceptron model to calculate the second initial position matrix set The fitness value of the initial position matrix of each eagle in the third fitness value set is obtained to obtain a third fitness value set; the maximum fitness value in the third fitness value set and the corresponding initial position matrix of the eagle are respectively used as the third global optimal fitness value and the third global optimal position; the second initial position matrix set is adjusted according to the third global optimal fitness value and the third global optimal position The initial position matrix of each eagle in is updated; after the update is completed, the second current iteration number Add 1 and enter the next iteration;

[0083] In each other round of iteration, the fitness function of the second eagle population is used And cooperate with the second final multi-layer perceptron model to calculate the fitness value of the position matrix of each eagle updated in the previous round of iterations to obtain a fourth fitness value set; the maximum fitness value in the fourth fitness value set and the corresponding initial position matrix of the eagle are used as the fourth global optimal fitness value and the fourth global optimal position respectively; according to the fourth global optimal fitness value and the fourth global optimal position, the position matrix of each eagle updated in the previous round of iterations is updated again; after the update is completed, the second current iteration number Add 1 and enter the next iteration;

[0084] S516, when When , the iteration is stopped and the second final global optimal position is output; the second final global optimal position is used as the adjusted future pain category data set; otherwise, the iteration is continued until until the time;

[0085] In this scheme, the Eagle optimization algorithm is used to optimize the dosage of each therapeutic drug for different pain types at each medication time point, and the number of future pain type data in the future pain type data set that are in the warning pain type set is used as the fitness function. Therefore, with the progress of iterations, the number of future pain type data in the future pain type data set that are in the warning pain type set will become smaller and smaller; when the optimization is completed, if there are still future pain type data in the future pain type data set that are in the warning pain type set, a secondary warning will be issued to prompt the patient to take further measures to relieve the pain.

[0086] A pain prediction and warning system based on a medical pain scoring scale comprises a patient pain data collection module, a patient pain influencing factor data collection module, a pain feature classification center data optimization module, a mapping model construction module, a future pain influencing factor data prediction module, a primary classification module, a secondary classification module, a primary judgment and warning module, and a secondary judgment and warning module.

[0087] The present invention has the following beneficial effects:

[0088] 1. The present invention optimizes the classification center data of the collected pain feature data, thereby providing a classification tool for subsequent classification of the pain feature data of the patient to be predicted to have pain in the future; by constructing multiple mapping models, a mapping tool is provided for subsequent classification and mapping of the psychological state data, environmental factor data and physiological indicator data of the patient to be predicted to have pain in the future; and by using the predicted pain type data at the future moment, it is determined whether to issue a primary warning and a secondary warning. When a primary warning is issued, the patient is prompted to take drug treatment to relieve the pain at the future moment; when a secondary warning is issued, it indicates that drug treatment alone cannot effectively relieve the pain occurring at the future moment, thereby prompting the patient to take further measures to relieve the pain.

[0089] 2. The present invention uses the Eagle optimization algorithm to optimize the dosage of each therapeutic drug for different pain types at each medication time point. As iterations proceed, the number of future pain type data in the future pain type data set that is in the warning pain type set will become smaller and smaller. When the optimization is completed, if there is still future pain type data in the future pain type data set that is in the warning pain type set, a secondary warning will be issued to prompt the patient to take further measures to relieve the pain.

[0090] 3. In the present invention, a corresponding mapping model is constructed for each type of pain characteristic data, thereby improving the accuracy of subsequent mapping of the psychological state data, environmental factor data and physiological indicator data of the patient to be predicted to have pain.

[0091] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] In order to more clearly illustrate the technical solutions of the embodiments of the invention, the drawings required for describing the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the invention. For ordinary technicians in this field, they can also obtain drawings based on these drawings without paying creative work.

[0093] Figure 1The present invention is a schematic diagram of a process of predicting and warning pain by a pain prediction and warning system based on a medical pain scoring scale. DETAILED DESCRIPTION

[0094] The following will be combined with the drawings in the embodiments of the invention to clearly and completely describe the technical solutions in the embodiments of the invention. Obviously, the described embodiments are only part of the embodiments of the invention, not all of the embodiments. Based on the embodiments in the invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the invention.

[0095] In the description of the present invention, it is necessary to understand that the terms "opening", "upper", "lower", "top", "middle", "inside" and the like indicating orientation or positional relationship are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the components or elements referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as limiting the invention.

[0096] Embodiment 1

[0097] This embodiment is a pain prediction and warning method based on a medical pain rating scale, comprising the following steps:

[0098] S1. Collect pain characteristic data and corresponding pain type data, psychological state data, environmental factor data and physiological index data of multiple patients;

[0099] The S1 comprises the following steps:

[0100] S11, setting multiple pain patients, pain characteristic types and pain categories, obtaining a pain patient set, a pain category set and a pain characteristic type set; collecting pain characteristic data and corresponding pain category data of each patient in the pain patient set when pain occurs according to the pain characteristic type set and the pain category set in conjunction with a medical pain rating scale, to obtain a pain characteristic data matrix a And pain category dataset , Indicates the setting i Data on the types of pain experienced by the pain patients, represents the total number of pain patients set; ;

[0101] in, Indicates the setting i The first time pain occurred in patients with pain j Types of pain feature data, Indicates the total number of pain feature types set;

[0102] S12, setting a plurality of psychological state types, environmental factor types and physiological indicator types to obtain a psychological state type set, an environmental factor type set and a physiological indicator type set; collecting a pain type data set according to the psychological state type set, the environmental factor type set and the physiological indicator type set. The psychological state data, environmental factor data and physiological index data of the pain patient when each type of pain occurs are obtained to obtain the psychological state data matrix , Environmental Factors Data Matrix And the physiological index data matrix ; They are as follows, ; ; ;

[0103] in, , , Respectively represent the set i The first time the pain occurred j types of psychological state data, environmental factor data, and physiological indicator data. , , Respectively represent the total number of set psychological state types, environmental factor types, and physiological indicator types;

[0104] S2. Optimizing the classification center data of the pain characteristic data of multiple patients in conjunction with the collected pain type data to obtain a pain characteristic classification center data matrix; constructing multiple mapping models based on the collected psychological state data, environmental factor data, physiological indicator data and pain type data of multiple patients to obtain a mapping model set;

[0105] The S2 comprises the following steps:

[0106] S21. According to the pain type dataset Pain feature data matrix a The pain feature data set in is classified to obtain the pain feature data classification matrix set , Represents the pain feature data matrix a The first i A pain feature data classification matrix, Represents the pain feature data matrix a The total number of pain feature data classification matrices obtained by classification; obtaining the optimal data center of each pain feature data classification matrix in the pain feature data classification matrix set, and obtaining the pain feature classification center data matrix ; , They are as follows: ; ;

[0107] in, express Middle j The first time pain occurred in patients with pain k Types of pain feature data, Indicates classification to The total number of pain patients in Represents the pain feature data matrix a The first i The classification center dataset of the pain feature data classification matrix j Types of pain characteristic data;

[0108] S22, according to the pain characteristic data matrix a Set multiple pain feature mapping model interfaces to obtain a mapping model interface set , It represents the mapping from the psychological state data, environmental factor data and physiological index data of the pain patient to the first i The interface corresponding to the mapping model of the types of pain characteristic data; according to the psychological state data matrix , Environmental Factors Data Matrix , Physiological index data matrix And the pain feature data matrix a In the mapping model interface set Build the corresponding mapping model under each mapping model interface to obtain the mapping model set , express The corresponding mapping model;

[0109] S22 is based on the mental state data matrix , Environmental Factors Data Matrix , Physiological index data matrix And the pain feature data matrix a In the mapping model interface set Build the corresponding mapping model under each mapping model interface to obtain the mapping model set The following steps are involved:

[0110] S221, extracting the pain feature data matrix a Middle j Column data, get the current pain feature data set ; Reconstruct the j An initial multi-layer perceptron model;

[0111] S222: Using the current pain feature dataset , psychological state data matrix , Environmental Factors Data Matrix And the physiological index data matrix For j The initial multi-layer perceptron model is trained and tested; after the training and testing are completed, the first j A final multi-layer perceptron model;

[0112] The said j The final multilayer perceptron model is used as the mapping model interface set Middle j A mapping model under a mapping model interface;

[0113] S223, repeat S221 and S222 multiple times until the mapping model interface set The mapping models under each mapping model interface are trained and tested to obtain the mapping model set ;

[0114] S3, predicting data at multiple future time points based on the psychological state data, environmental factor data, and physiological index data of the patient to be predicted for pain at multiple historical time points, to obtain a future psychological state data matrix, a future environmental factor data matrix, and a future physiological index data matrix;

[0115] The S3 comprises the following steps:

[0116] S31, setting the pain patient to be predicted, multiple historical time points and future time points, and obtaining a historical time point set and future time points , , Respectively represent the set i historical time points and future time points, , Respectively represent the total number of set historical time points and future time points;

[0117] S32, matching the psychological state type set, the environmental factor type set and the physiological indicator type set and according to the historical time point set Collect the psychological state data, environmental factor data and physiological index data of the patient to be predicted at multiple historical time points to obtain the historical psychological state data matrix , Historical environmental factors data matrix And the historical physiological index data matrix ; They are as follows, ; ; ;

[0118] in, , , They respectively represent the pain patients to be predicted in the i At a historical point in time j types of psychological state data, environmental factor data, and physiological indicator data;

[0119] S33, according to the historical psychological state data matrix , Historical environmental factors data matrix And the historical physiological index data matrix Construct the second final multi-layer perceptron model; cooperate with the future time point set The second final multi-layer perceptron model is used to predict the psychological state data, environmental factor data and physiological index data of the patient with pain to be predicted in the future, and the future psychological state data matrix is ​​obtained. , Future Environmental Factors Data Matrix And the future physiological index data matrix ; They are as follows, ; ; ;

[0120] in, , , They respectively represent the pain patients to be predicted in the i At a future time point j types of psychological state data, environmental factor data, and physiological indicator data;

[0121] S4, using a mapping model set to classify and map the future psychological state data matrix, the future environmental factor data matrix, and the future physiological indicator data matrix to obtain a future pain feature data matrix; using a pain feature classification center data matrix to classify the future pain feature data matrix to obtain a future pain category data set;

[0122] The S4 comprises the following steps:

[0123] S41, the future mental state data matrix , Future Environmental Factors Data Matrix And the future physiological index data matrix Input to the mapping model interface set In each mapping model interface in the Each mapping model in the future psychological state data matrix , Future Environmental Factors Data Matrix And the future physiological index data matrix Perform classification mapping to obtain the future pain feature data matrix ;as follows, ;

[0124] in, It indicates that the predicted pain patient is in the i The pain occurs at a future time point j Types of pain characteristic data;

[0125] S42, classifying the central data matrix according to the pain characteristics Future pain feature data matrix Classify the pain feature data at each future time point in the future to obtain the future pain category dataset , Indicates that the patient with pain to be predicted is in i Data on pain type at future time points;

[0126] S5, determining whether to issue a warning once according to the future pain type data set, and when issuing a warning once, adjusting the physiological index data of the patient to be predicted to obtain an adjusted future pain type data set; and determining whether to issue a second warning according to the adjusted future pain type data set;

[0127] The S5 comprises the following steps:

[0128] S51, setting a warning pain type set and based on the future pain type data set Set the therapeutic drug type set , Indicates the setting i Types of therapeutic drugs, Indicates the total number of set therapeutic drug types;

[0129] When the future pain type dataset When there is future pain type data in the warning pain type set, an alarm is issued and the future time point where the future pain type data is located is recorded as the warning future time point; in conjunction with the treatment drug type set and adjusting the physiological index data of the patient to be predicted to have pain at a warning point in the future; after the adjustment is completed, an adjusted data set of future pain categories is obtained;

[0130] Otherwise, no warning is given;

[0131] S51 is combined with the therapeutic drug type set And the physiological index data of the patient to be predicted to have pain are adjusted at a warning time point in the future; after the adjustment is completed, obtaining the adjusted future pain type data set includes the following steps:

[0132] S511, setting multiple medication time points before the future warning time point to obtain a medication time point set , Indicates the setting i The time point of taking medicine, Indicates the total number of set medication time points; set the type of therapeutic drugs to be taken at each medication time point The lower and upper limits of the dosage of each therapeutic drug in the formula are used to obtain the dosage interval matrix of the therapeutic drugs. ;as follows,

[0133] ;

[0134] in, , Respectively expressed in i Take the medicine at the time of taking j The lower and upper limits of the dosage of the therapeutic drug;

[0135] S512, build the second eagle population , Indicates the second eagle population i A hawk, represents the size of the second eagle population; sets the maximum number of iterations of the second eagle population to , the current number of iterations is , respectively recorded as the second maximum number of iterations and the second current number of iterations; the search space dimension of the second eagle population is ;

[0136] S513, according to the dosage interval matrix of the therapeutic drug Set the initial position matrix of each eagle in the second eagle population to obtain the second initial position matrix set , Indicates the second eagle population i The initial position matrix of the eagle is as follows: ;

[0137] in, express In the i Take the medicine at the time of taking j The amount of the therapeutic drug used;

[0138] S514. Based on the future pain type dataset Construct the fitness function of the second eagle population ;as follows,

[0139] ;

[0140] In the formula, Represents the future pain category dataset The number of future pain category data in the warning pain category set, β A positive number indicates a correction parameter;

[0141] S515, start iteration, before iteration, set the second current iteration number Set to 1; the fitness function of the second eagle population is used in the first round of iteration And cooperate with the second final multi-layer perceptron model to calculate the second initial position matrix set The fitness value of the initial position matrix of each eagle in the third fitness value set is obtained to obtain a third fitness value set; the maximum fitness value in the third fitness value set and the corresponding initial position matrix of the eagle are respectively used as the third global optimal fitness value and the third global optimal position; the second initial position matrix set is adjusted according to the third global optimal fitness value and the third global optimal position The initial position matrix of each eagle in is updated; after the update is completed, the second current iteration number Add 1 and enter the next iteration;

[0142] In each other round of iteration, the fitness function of the second eagle population is used And cooperate with the second final multi-layer perceptron model to calculate the fitness value of the position matrix of each eagle updated in the previous round of iterations to obtain a fourth fitness value set; the maximum fitness value in the fourth fitness value set and the corresponding initial position matrix of the eagle are used as the fourth global optimal fitness value and the fourth global optimal position respectively; according to the fourth global optimal fitness value and the fourth global optimal position, the position matrix of each eagle updated in the previous round of iterations is updated again; after the update is completed, the second current iteration number Add 1 and enter the next iteration;

[0143] S516, when When , the iteration is stopped and the second final global optimal position is output; the second final global optimal position is used as the adjusted future pain category data set; otherwise, the iteration is continued until until the time;

[0144] S52: When there is future pain type data in the adjusted future pain type data set that is located in the warning pain type set, a secondary warning is issued; otherwise, no secondary warning is issued.

[0145] Embodiment 2

[0146] This embodiment discloses a pain prediction and warning system based on a medical pain scoring scale, the system can implement the method of the above embodiment, including a patient pain data collection module, a patient pain influencing factor data collection module, a pain feature classification center data optimization module, a mapping model construction module, a future pain influencing factor data prediction module, a primary classification module, a secondary classification module, a primary judgment and warning module, and a secondary judgment and warning module;

[0147] The patient pain data collection module is used to collect pain characteristic data and corresponding pain type data of multiple patients to obtain a pain characteristic data matrix and a pain type data set;

[0148] The patient pain influencing factor data collection module is used to collect psychological state data, environmental factor data and physiological index data of multiple patients when they experience pain, and obtain a psychological state data matrix, an environmental factor data matrix and a physiological index data matrix;

[0149] The pain feature classification center data optimization module is used to optimize the classification center data of the pain feature data matrix in cooperation with the pain type data set to obtain the pain feature classification center data matrix;

[0150] The mapping model construction module is used to construct multiple mapping models according to the psychological state data matrix, the environmental factor data matrix, the physiological index data matrix and the pain feature data matrix to obtain a mapping model set;

[0151] The future pain influencing factor data prediction module is used to predict the psychological state data, environmental factor data and physiological index data at multiple future time points based on the psychological state data, environmental factor data and physiological index data of the patient to be predicted at multiple historical time points, and obtain a future psychological state data matrix, a future environmental factor data matrix and a future physiological index data matrix;

[0152] The primary classification module is used to classify and map the future psychological state data matrix, the future environmental factor data matrix and the future physiological index data matrix using a mapping model set to obtain a future pain feature data matrix;

[0153] The secondary classification module is used to classify the future pain feature data matrix using the pain feature classification center data matrix to obtain a future pain category data set;

[0154] The one-time determination and warning module is used to determine whether to issue a warning based on the future pain type data set, and when issuing a warning, the physiological indicator data of the patient to be predicted to have pain is adjusted to obtain an adjusted future pain type data set;

[0155] The secondary determination warning module is used to determine whether to issue a secondary warning based on the adjusted future pain category data set.

[0156] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0157] The preferred embodiments of the invention disclosed above are only used to help explain the invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the invention, so that those skilled in the art can understand and use the invention well.

Claims

1. A pain prediction and warning method based on a medical pain rating scale, characterized in that: The following steps are involved: S1. Collect pain characteristic data and corresponding pain type data, psychological state data, environmental factor data and physiological index data of multiple patients; S2. Optimizing the classification center data of the pain characteristic data of multiple patients in conjunction with the collected pain type data to obtain a pain characteristic classification center data matrix; constructing multiple mapping models based on the collected psychological state data, environmental factor data, physiological indicator data and pain type data of multiple patients to obtain a mapping model set; S3, predicting data at multiple future time points based on the psychological state data, environmental factor data, and physiological index data of the patient to be predicted for pain at multiple historical time points, to obtain a future psychological state data matrix, a future environmental factor data matrix, and a future physiological index data matrix; S4, using a mapping model set to classify and map the future psychological state data matrix, the future environmental factor data matrix, and the future physiological index data matrix to obtain a future pain feature data matrix; The future pain feature data matrix is ​​classified using the pain feature classification center data matrix to obtain the future pain category data set; S5, determining whether to issue an alarm based on the future pain category data set, and when an alarm is issued, adjusting the physiological indicator data of the patient to be predicted to obtain an adjusted future pain category data set; Then, a decision is made whether to issue a second warning based on the adjusted future pain type data set.

2. The method for predicting and warning pain based on a medical pain scale according to claim 1, characterized in that: The S1 comprises the following steps: S11, setting a plurality of pain patients, pain characteristic types and pain categories, obtaining a pain patient set, a pain category set and a pain characteristic type set; using a medical pain rating scale and according to the pain characteristic type set and the pain category set, collecting pain characteristic data and corresponding pain category data of each patient in the pain patient set when pain occurs, to obtain a pain characteristic data matrix and a pain category data set; S12. Set a plurality of psychological state types, environmental factor types, and physiological indicator types to obtain a psychological state type set, an environmental factor type set, and a physiological indicator type set; collect the psychological state data, environmental factor data, and physiological indicator data of the pain patient when the pain of each type of pain occurs in the pain type data set according to the psychological state type set, environmental factor type set, and physiological indicator type set to obtain a psychological state data matrix, an environmental factor data matrix, and a physiological indicator data matrix.

3. The method for predicting and warning pain based on a medical pain scale according to claim 2, characterized in that: The S2 comprises the following steps: S21, classifying the pain feature data set in the pain feature data matrix according to the pain type data set to obtain a pain feature data classification matrix set; obtaining the optimal data center of each pain feature data classification matrix in the pain feature data classification matrix set to obtain a pain feature classification center data matrix; S22. Set multiple pain feature mapping model interfaces according to the pain feature data matrix to obtain a mapping model interface set; construct a corresponding mapping model under each mapping model interface in the mapping model interface set according to the psychological state data matrix, environmental factor data matrix, physiological indicator data matrix and pain feature data matrix to obtain a mapping model set.

4. The method for predicting and warning pain based on a medical pain scale according to claim 3, characterized in that: In S22, a corresponding mapping model is constructed under each mapping model interface in the mapping model interface set according to the psychological state data matrix, the environmental factor data matrix, the physiological indicator data matrix, and the pain feature data matrix, and obtaining the mapping model set includes the following steps: S221, extract the pain feature data matrix j Column data, get the current pain feature data set; Reconstruct j An initial multi-layer perceptron model; S222, using the current pain feature data set, the psychological state data matrix, the environmental factor data matrix and the physiological index data matrix to j An initial multi-layer perceptron model is trained and tested; After training and testing are completed, we get j A final multi-layer perceptron model; The said j The final multilayer perceptron model is used as the mapping model interface set j A mapping model under a mapping model interface; S223, repeat S221 and S222 multiple times until the mapping models under each mapping model interface in the mapping model interface set are trained and tested to obtain a mapping model set.

5. The method for predicting and warning pain based on a medical pain scale according to claim 4, characterized in that: The S3 comprises the following steps: S31, setting a patient with pain to be predicted, multiple historical time points and future time points, and obtaining a historical time point set and a future time point set; S32, collecting psychological state data, environmental factor data and physiological indicator data of the patient to be predicted at multiple historical time points according to the historical time point set in conjunction with the psychological state type set, the environmental factor type set and the physiological indicator type set, to obtain a historical psychological state data matrix, a historical environmental factor data matrix and a historical physiological indicator data matrix; S33. Construct a second final multi-layer perceptron model based on the historical psychological state data matrix, the historical environmental factor data matrix, and the historical physiological indicator data matrix; cooperate with the future time point set and use the second final multi-layer perceptron model to predict the psychological state data, environmental factor data, and physiological indicator data of the pain patient to be predicted at the future moment, and obtain the future psychological state data matrix, the future environmental factor data matrix, and the future physiological indicator data matrix.

6. The method for predicting and warning pain based on a medical pain scale according to claim 5, characterized in that: The S4 comprises the following steps: S41, inputting the future psychological state data matrix, the future environmental factor data matrix and the future physiological index data matrix into each mapping model interface in the mapping model interface set respectively, and each mapping model in the mapping model set classifies and maps the future psychological state data matrix, the future environmental factor data matrix and the future physiological index data matrix to obtain the future pain feature data matrix; S42. Classify the pain feature data at each future time point in the future pain feature data matrix according to the pain feature classification center data matrix to obtain a future pain category data set.

7. The method for predicting and warning pain based on a medical pain scale according to claim 6, characterized in that: The S5 comprises the following steps: S51, setting a warning pain category set and setting a therapeutic drug category set according to the future pain category data set; when there is future pain category data in the future pain category data set that is located in the warning pain category set, issuing an alarm and recording the future time point where the future pain category data is located, which is recorded as the alarm future time point; adjusting the physiological indicator data of the patient to be predicted to have pain in accordance with the therapeutic drug category set and the alarm future time point; after the adjustment is completed, obtaining the adjusted future pain category data set; otherwise, not issuing an alarm; S52: When there is future pain type data in the adjusted future pain type data set that is located in the warning pain type set, a secondary warning is issued; otherwise, no secondary warning is issued.

8. The method for predicting and warning pain based on a medical pain scale according to claim 7, characterized in that: In S51, the physiological index data of the patient to be predicted to have pain are adjusted in accordance with the therapeutic drug type set and the future warning time point; after the adjustment is completed, obtaining the adjusted future pain type data set includes the following steps: S511, setting multiple medication time points before the future warning time point to obtain a medication time point set; setting a lower limit and an upper limit on the dosage of each therapeutic drug in the therapeutic drug type set at each medication time point to obtain a therapeutic drug dosage interval matrix; S512, construct the second eagle population; set the maximum number of iterations of the second eagle population to , the current number of iterations is , respectively recorded as the second maximum number of iterations and the second current number of iterations; S513, setting the initial position matrix of each eagle in the second eagle population according to the therapeutic drug dosage interval matrix to obtain a second initial position matrix set; S514, constructing a fitness function of the second eagle population according to the future pain type data set; S515, start iteration; in each round of iteration, use the fitness function of the second eagle population and cooperate with the second final multi-layer perceptron model to calculate the fitness value of the position matrix of each eagle updated in the previous round of iteration, and update the position matrix of each eagle updated in the previous round of iteration again; S516, when When , the iteration is stopped and the second final global optimal position is output; the second final global optimal position is used as the adjusted future pain category data set; otherwise, the iteration is continued until Until then.

9. A system for implementing the pain prediction and warning method based on a medical pain score scale as described in any one of claims 1 to 8.

Images