Children respiratory rehabilitation monitoring method and system

Abstract

The invention provides a children respiratory rehabilitation monitoring method and system, and relates to the technical field of respiratory rehabilitation monitoring. The method comprises the following steps: monitoring breathing index data of a child breathing patient through monitoring equipment; according to the medical record of the child respiratory patient, obtaining a respiratory disease type of the child respiratory patient; according to the respiratory index data and the respiratory disease type, determining a rehabilitation speed score of the child respiratory patient; inputting the breathing index data into a trained breathing frequency prediction model to obtain predicted breathing frequency data at multiple moments in the current monitoring period; acquiring respiratory rate data of the child respiratory patient; according to the predicted respiratory rate data and the respiratory rate data, determining a rehabilitation degree score of the child respiratory patient; and generating a respiratory rehabilitation monitoring report according to the rehabilitation speed score and the rehabilitation degree score. According to the breathing rehabilitation monitoring system and method, the breathing rehabilitation monitoring result can be comprehensively monitored according to the rehabilitation speed and the rehabilitation degree.

Description

Technical Field

[0001] The present invention relates to the technical field of respiratory rehabilitation monitoring, and in particular to a method and system for monitoring respiratory rehabilitation of children. Background Art

[0002] Although current related technologies can formulate respiratory assistance strategies for pediatric respiratory patients, they do not take into account the impact of recovery speed and degree on the respiratory rehabilitation monitoring results of pediatric respiratory patients. That is, it is impossible to comprehensively monitor the respiratory rehabilitation monitoring results based on the two aspects of recovery speed and degree.

[0003] The information disclosed in the background technology section of this application is only intended to deepen the understanding of the general background technology of this application, and should not be regarded as an admission or any form of suggestion that the information constitutes the prior art already known to those skilled in the art. Summary of the Invention

[0004] The present invention provides a child respiratory rehabilitation monitoring method and system, which can solve the technical problem that related technologies cannot comprehensively monitor respiratory rehabilitation monitoring results based on the two aspects of rehabilitation speed and rehabilitation degree.

[0005] According to a first aspect of the present invention, a method for monitoring respiratory rehabilitation in children is provided, comprising:

[0006] At multiple moments in the current monitoring cycle, respiratory index data of the pediatric respiratory patient is monitored by the monitoring device;

[0007] The respiratory disease types of pediatric respiratory patients were obtained from their medical records;

[0008] determining a recovery speed score for the pediatric respiratory patient based on the respiratory index data and the type of respiratory disease;

[0009] Inputting the respiratory index data into a trained respiratory rate prediction model to obtain predicted respiratory rate data at multiple moments in the current monitoring period;

[0010] Obtain respiratory rate data of a pediatric respiratory patient at multiple moments in a current monitoring cycle;

[0011] determining a recovery degree score for the pediatric respiratory patient based on the predicted respiratory rate data and the respiratory rate data;

[0012] A respiratory rehabilitation monitoring report is generated according to the rehabilitation speed score and the rehabilitation degree score.

[0013] Furthermore, the respiratory index data includes vital capacity and blood oxygen saturation.

[0014] Furthermore, determining a recovery speed score for a pediatric respiratory patient based on the respiratory index data and the respiratory disease type includes:

[0015] Divide the current monitoring period into two phases, namely the first phase and the second phase, wherein the duration of the first phase is equal to the duration of the second phase;

[0016] Fitting the respiratory index data of the first stage and the time in the first stage to obtain a first respiratory index function in the first stage;

[0017] determining a first breathing index derivative function according to the first breathing index function;

[0018] determining a first breathing index change rate at multiple moments in the first stage according to the first breathing index derivative function;

[0019] Fitting the breathing index data of the second stage and the moments in the second stage to obtain a second breathing index function in the second stage;

[0020] determining a second breathing index derivative function according to the second breathing index function;

[0021] determining a change rate of the second breathing index at multiple moments in the second stage according to the second breathing index derivative function;

[0022] Determine symptom weights based on the type of respiratory disease;

[0023] A recovery speed score for a pediatric respiratory patient is determined according to the first respiratory index change rate, the second respiratory index change rate, and the symptom weight.

[0024] Furthermore, according to the respiratory disease type, the symptom weight is determined, including:

[0025] According to the theoretical recovery time of the respiratory disease, the respiratory disease type is divided into mild disease, moderate disease and severe disease, wherein the theoretical recovery time of the mild disease is less than or equal to 7 days, the theoretical recovery time of the moderate disease is greater than 7 days and less than or equal to 14 days, and the theoretical recovery time of the severe disease is greater than 14 days;

[0026] If the respiratory disease type is a mild disease, the symptom weight is determined to be 1;

[0027] If the respiratory disease type is a moderate disease, the symptom weight is determined to be 0.5;

[0028] If the respiratory disease type is a severe disease, the symptom weight is determined to be 0.1.

[0029] Further, determining a recovery speed score of a pediatric respiratory patient according to the first respiratory index change rate, the second respiratory index change rate, and the symptom weight includes:

[0030] If the average value of the second respiratory index change rate at multiple moments in the second stage is greater than or equal to the average value of the first respiratory index change rate at multiple moments in the first stage, the conditional function takes a value of 1; otherwise, the conditional function takes a value of 0;

[0031] The conditional function values of the first respiratory index change rate of multiple respiratory index data are averaged and multiplied by the symptom weight to determine the recovery speed score of the pediatric respiratory patient.

[0032] Furthermore, the training step of the respiratory rate prediction model includes:

[0033] Acquire sample respiratory index data of multiple sample pediatric respiratory patients at multiple times in multiple historical monitoring periods, wherein the sample respiratory index data include sample vital capacity and sample blood oxygen saturation;

[0034] Inputting the sample respiratory index data into a trained respiratory rate prediction model to obtain sample predicted respiratory rate data of multiple sample pediatric respiratory patients at multiple times during multiple historical monitoring periods;

[0035] Obtain sample respiratory rate data of multiple sample pediatric respiratory patients at multiple times in multiple historical monitoring periods;

[0036] Obtain the age of the pediatric respiratory patient and the sample age of multiple samples of pediatric respiratory patients;

[0037] determining a loss function of the respiratory rate prediction model based on the sample predicted respiratory rate data, the sample respiratory rate data, the age, and the sample age;

[0038] The respiratory rate prediction model is trained according to the loss function of the respiratory rate prediction model to obtain a trained respiratory rate prediction model.

[0039] Furthermore, determining a loss function of the respiratory rate prediction model based on the sample predicted respiratory rate data, the sample respiratory rate data, the age, and the sample age includes:

[0040] According to the formula

[0041]

[0042] Determine the loss function Loss of the respiratory rate prediction model, where F i,h,kis the sample respiratory rate data of the i-th sample child respiratory patient at the k-th moment in the h-th historical monitoring cycle, F i,h,k,p Predict respiratory rate data for the i-th sample of a child respiratory patient at the k-th moment in the h-th historical monitoring period, Y e is the age of the child respiratory patient, Y i is the sample age of the i-th sample child respiratory patient, K is the number of monitoring cycle moments, H is the number of historical monitoring cycles, M j is the number of sample pediatric respiratory patients in the jth training batch, N is the number of training batches, k≤K, h≤H, i≤M j , j≤N, and k, h, i, j, K, H, M j and N are both positive integers.

[0043] Further, determining a recovery degree score of a pediatric respiratory patient based on the predicted respiratory rate data and the respiratory rate data includes:

[0044] Calculating the absolute value of the relative difference between the predicted respiratory rate data at multiple moments in the current monitoring period and the respiratory rate data;

[0045] The difference between 1 and the absolute value corresponding to each moment is averaged to determine the recovery degree score of the pediatric respiratory patient.

[0046] Furthermore, a respiratory rehabilitation monitoring report is generated based on the rehabilitation speed score and the rehabilitation degree score, including:

[0047] When the recovery speed score is greater than or equal to the preset recovery speed score, and the recovery degree score is greater than or equal to the preset recovery degree score, it is determined that the recovery effect of the current monitoring period is good and the respiratory symptoms are mild;

[0048] When the recovery speed score is less than the preset recovery speed score, and the recovery degree score is greater than or equal to the preset recovery degree score, it is determined that the recovery effect of the current monitoring period is poor, but the respiratory symptoms are mild;

[0049] When the recovery speed score is greater than or equal to the preset recovery speed score, and the recovery degree score is less than the preset recovery degree score, it is determined that the recovery effect of the current monitoring period is good, but the respiratory symptoms are serious;

[0050] When the recovery speed score is less than the preset recovery speed score, and the recovery degree score is less than the preset recovery degree score, it is determined that the recovery effect of the current monitoring period is poor and the respiratory disease is serious.

[0051] According to a second aspect of the present invention, there is provided a child respiratory rehabilitation monitoring system, comprising:

[0052] A respiratory index data module is used to monitor the respiratory index data of pediatric respiratory patients through monitoring equipment at multiple moments in the current monitoring cycle;

[0053] A respiratory disease type module is used to obtain the respiratory disease type of a pediatric respiratory patient based on the patient's medical record;

[0054] a recovery speed scoring module, configured to determine a recovery speed score for a pediatric respiratory patient based on the respiratory index data and the respiratory disease type;

[0055] A respiratory rate prediction data module is used to input the respiratory index data into a trained respiratory rate prediction model to obtain predicted respiratory rate data at multiple moments in the current monitoring period;

[0056] Respiratory rate data module, used to obtain respiratory rate data of pediatric respiratory patients at multiple moments in the current monitoring cycle;

[0057] a recovery degree scoring module, configured to determine a recovery degree score of a pediatric respiratory patient based on the predicted respiratory rate data and the respiratory rate data;

[0058] The respiratory rehabilitation monitoring report module is used to generate a respiratory rehabilitation monitoring report according to the rehabilitation speed score and the rehabilitation degree score.

[0059] Technical effect: According to the present invention, the recovery speed score can be determined based on the respiratory index data and the type of respiratory disease, which can intuitively reflect the speed of recovery of the child respiratory patient in the current monitoring period. The recovery degree score is obtained by comparing the predicted respiratory rate data and the actual respiratory rate data, which can accurately measure the degree of closeness between the respiratory function of the child respiratory patient in the current monitoring period and the normal level of the current monitoring period. The respiratory rehabilitation monitoring results can be comprehensively monitored based on the two aspects of recovery speed and recovery degree, thereby improving the treatment efficiency. When determining the recovery speed score of the child respiratory patient, the recovery speed score of the child respiratory patient can be determined based on the three factors of the first respiratory index change rate, the second respiratory index change rate and the symptom weight. By comparing the first respiratory index change rate and the second respiratory index change rate, the difference in recovery speed of the child respiratory patient at different rehabilitation stages can be reflected. The rehabilitation characteristics and needs of different types of respiratory diseases are different. The symptom weight can help to achieve individualized treatment, more accurately evaluate the recovery speed, and adjust the treatment plan in time. When determining the loss function of the respiratory rate prediction model, the relative error between the sample respiratory rate data and the sample predicted respiratory rate data can be used to set a weight based on the characteristic that the more similar the age of the child respiratory patient is to the sample age of the sample child respiratory patient, the greater the reference value. Furthermore, weights can be set based on the characteristic that the shorter the time interval from the start time, the higher the accuracy, and the shorter the time interval from the first training batch, the lower the accuracy. Thus, the error output of the respiratory rate prediction model for each sample child respiratory patient in the jth batch at multiple times across multiple historical monitoring cycles is weighted averaged to obtain a loss function. This improves the design accuracy and objectivity of the loss function, thereby improving training efficiency and the accuracy of the respiratory rate prediction model during training. When determining the recovery level score of a child respiratory patient, the recovery level score can be determined based on the predicted respiratory rate data and the respiratory rate data. By comparing the predicted respiratory rate data with the respiratory rate data, the recovery level of the child respiratory patient in the current monitoring cycle can be reflected, and problems in the rehabilitation process can be promptly identified, thereby improving treatment effectiveness.

[0060] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and not limiting of the present invention. Other features and aspects of the present invention will become more apparent from the following detailed description of exemplary embodiments with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can derive other embodiments based on these drawings without inventive efforts.

[0062] Figure 1 A schematic diagram exemplarily illustrates a flow chart of a child respiratory rehabilitation monitoring method according to an embodiment of the present invention;

[0063] Figure 2 A flowchart of calculating a recovery speed score according to an embodiment of the present invention is exemplarily shown;

[0064] Figure 3 A flowchart exemplarily illustrates the steps of training a respiratory rate prediction model according to an embodiment of the present invention;

[0065] Figure 4 A flowchart of calculating a rehabilitation degree score according to an embodiment of the present invention is exemplarily shown;

[0066] Figure 5 The following is an exemplary flowchart of generating a respiratory rehabilitation monitoring report according to an embodiment of the present invention;

[0067] Figure 6 A block diagram of a child respiratory rehabilitation monitoring system according to an embodiment of the present invention is exemplarily shown. DETAILED DESCRIPTION

[0068] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0069] The following specific embodiments are used to describe the technical solution of the present invention in detail. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0070] Figure 1 A flowchart of a child respiratory rehabilitation monitoring method according to an embodiment of the present invention is exemplarily shown. The method includes:

[0071] Step S1, monitoring respiratory index data of a child respiratory patient by a monitoring device at multiple moments in the current monitoring cycle;

[0072] Step S2, obtaining the respiratory disease type of the child respiratory patient based on the medical record of the child respiratory patient;

[0073] Step S3, determining a recovery speed score of the child respiratory patient based on the respiratory index data and the respiratory disease type;

[0074] Step S4, inputting the respiratory index data into the trained respiratory rate prediction model to obtain predicted respiratory rate data at multiple moments in the current monitoring period;

[0075] Step S5, obtaining respiratory rate data of the child respiratory patient at multiple moments in the current monitoring cycle;

[0076] Step S6, determining a recovery degree score of the child respiratory patient based on the predicted respiratory rate data and the respiratory rate data;

[0077] Step S7: generating a respiratory rehabilitation monitoring report according to the rehabilitation speed score and the rehabilitation degree score.

[0078] According to the pediatric respiratory rehabilitation monitoring method of an embodiment of the present invention, a recovery rate score can be determined based on respiratory index data and respiratory disease type, which can intuitively reflect the speed of recovery of the child respiratory patient during the current monitoring cycle. The recovery degree score is derived by comparing predicted respiratory rate data with actual respiratory rate data, which can accurately measure the degree to which the child respiratory function during the current monitoring cycle is close to the normal level during the current monitoring cycle. Comprehensive monitoring of respiratory rehabilitation monitoring results based on both recovery rate and recovery degree can improve treatment efficiency.

[0079] According to one embodiment of the present invention, in step S1, the interval between adjacent moments can be set to 12 hours, 24 hours, etc., and each monitoring period can be set to 3 days, 7 days, etc., which is not limited by the present invention. Respiratory index data includes vital capacity and blood oxygen saturation. Therefore, vital capacity can be measured by an electronic spirometer, and blood oxygen saturation can be measured by an oximeter.

[0080] According to one embodiment of the present invention, in step S2, medical records of pediatric respiratory patients are collected from a relevant medical record platform to obtain the respiratory disease type of the pediatric respiratory patient. For example, the respiratory disease for which the pediatric respiratory patient is undergoing rehabilitation treatment can be determined, such as the common cold, acute bronchitis, viral pneumonia, or acute asthma attack. Based on the theoretical recovery time of the disease, the respiratory disease type can be divided into mild, moderate, and severe diseases.

[0081] According to one embodiment of the present invention, in step S3, a recovery speed score of a child respiratory patient is determined based on the respiratory index data and the respiratory disease type.

[0082] Figure 2 The flowchart of calculating the recovery speed score according to an embodiment of the present invention is exemplarily shown.

[0083] According to one embodiment of the present invention, step S3 includes: step S31, dividing the current monitoring cycle into two stages, namely the first stage and the second stage, wherein the duration of the first stage is equal to the duration of the second stage; step S32, fitting the respiratory index data of the first stage and the moment in the first stage to obtain a first respiratory index function in the first stage; step S33, determining a first respiratory index derivative function based on the first respiratory index function; step S34, determining a first respiratory index change rate at multiple moments in the first stage based on the first respiratory index derivative function; step S35, fitting the respiratory index data of the second stage and the moment in the second stage to obtain a second respiratory index function in the second stage; step S36, determining a second respiratory index derivative function based on the second respiratory index function; step S37, determining a second respiratory index change rate at multiple moments in the second stage based on the second respiratory index derivative function; step S38, determining a symptom weight based on the type of respiratory disease; step S39, determining a recovery speed score for a pediatric respiratory patient based on the first respiratory index change rate, the second respiratory index change rate and the symptom weight.

[0084] According to one embodiment of the present invention, in step S31 , the duration of the first stage is equal to the duration of the second stage, so that the two stages have the same metric in the time dimension.

[0085] According to one embodiment of the present invention, in step S32, a suitable mathematical fitting method (for example, linear fitting, polynomial fitting, etc.) is used for the respiratory index data collected in the first stage and the corresponding time points in the first stage to construct a function that can reflect the change pattern of the respiratory index in the first stage over time, that is, a first respiratory index function, wherein the first respiratory index function includes the first respiratory index function of vital capacity in the first stage and the first respiratory index function of blood oxygen saturation in the first stage.

[0086] According to one embodiment of the present invention, in step S33, based on the first breathing index function, the first-order derivative function of the first breathing index function, that is, the first breathing index derivative function, is calculated using the derivation rule in calculus.

[0087] According to one embodiment of the present invention, in step S34, the breathing index change rates corresponding to multiple moments in the first stage, ie, the first breathing index change rates, are calculated based on the first breathing index derivative function.

[0088] According to one embodiment of the present invention, in step S35, similar to the first stage, the breathing index data in the second stage and the corresponding time are fitted to obtain a second breathing index function in the second stage, which describes the change pattern of the breathing index in the second stage over time, wherein the second breathing index function includes the second breathing index function of vital capacity in the second stage and the second breathing index function of blood oxygen saturation in the second stage.

[0089] According to one embodiment of the present invention, in step S36, a second breathing index derivative function is obtained by performing a derivative operation based on the second breathing index function.

[0090] According to one embodiment of the present invention, in step S37, the breathing index change rate corresponding to multiple moments in the second stage, ie, the second breathing index change rate, is calculated based on the second breathing index derivative function.

[0091] According to one embodiment of the present invention, in step S38, different types of respiratory diseases may have different symptoms and different degrees of impact on recovery speed. Therefore, it is necessary to determine symptom weights to improve the accuracy and reliability of the recovery speed score.

[0092] According to one embodiment of the present invention, step S38 includes: step S381, dividing the respiratory disease type into mild disease, moderate disease and severe disease according to the theoretical recovery time of the respiratory disease, wherein the theoretical recovery time of the mild disease is less than or equal to 7 days, the theoretical recovery time of the moderate disease is greater than 7 days and less than or equal to 14 days, and the theoretical recovery time of the severe disease is greater than 14 days; step S382, if the respiratory disease type is a mild disease, determining the symptom weight to be 1; step S383, if the respiratory disease type is a moderate disease, determining the symptom weight to be 0.5; step S384, if the respiratory disease type is a severe disease, determining the symptom weight to be 0.1.

[0093] According to one embodiment of the present invention, the theoretical recovery time of respiratory diseases is used as a classification standard to divide respiratory disease types into three types: mild diseases, moderate diseases and severe diseases. The theoretical recovery time of mild diseases is within one week, for example, the common cold (theoretical recovery time: 3 to 7 days), the theoretical recovery time of moderate diseases is within two weeks, for example, acute bronchitis (theoretical recovery time: 7 to 14 days), and the theoretical recovery time of severe diseases is more than two weeks, for example, viral pneumonia (theoretical recovery time: 14 to 21 days) and acute asthma attack (theoretical recovery time: 14 to 28 days). The longer the theoretical recovery time of the respiratory disease type, the slower the recovery speed. Therefore, the symptom weight of mild diseases is determined to be 1, the symptom weight of moderate diseases is determined to be 0.5, and the symptom weight of severe diseases is determined to be 0.1.

[0094] According to one embodiment of the present invention, in step S39, the recovery speed score of the pediatric respiratory patient is determined by combining the three factors of the first respiratory index change rate, the second respiratory index change rate and the symptom weight, which can comprehensively reflect the recovery progress of the pediatric respiratory patient in the two monitoring stages.

[0095] According to one embodiment of the present invention, step S39 includes: step S391, if the average value of the second respiratory index change rate at multiple moments in the second stage is greater than or equal to the average value of the first respiratory index change rate at multiple moments in the first stage, the conditional function takes a value of 1, otherwise, the conditional function takes a value of 0; step S392, averaging the conditional function values of the first respiratory index change rate of multiple respiratory index data, and multiplying them by the symptom weight to determine the recovery speed score of the child respiratory patient.

[0096] According to one embodiment of the present invention, the calculation formula for the recovery speed score is: Among them, V R I′ is the recovery speed score. 1,t,s I′ is the first respiratory index change rate of the sth respiratory index data at the tth moment in the first stage, 2,t,s is the rate of change of the second respiratory index of the sth respiratory index data at the tth moment in the second stage, δ is the symptom weight, K is the number of monitoring cycle moments, t≤K, s≤2, and t, s, and K are all positive integers. If is a conditional function. The greater the symptom weight of the child respiratory patient, the milder the child respiratory disease and the faster the child respiratory patient's recovery. If the average of the second respiratory index change rate at multiple moments in the second stage is greater than or equal to the average of the first respiratory index change rate at multiple moments in the first stage, it means that the recovery rate in the second stage is greater than or equal to the recovery rate in the first stage. That is, a faster increase in vital capacity indicates improved respiratory function, and a faster increase in blood oxygen saturation indicates enhanced oxygen uptake capacity. Therefore, a higher recovery rate score indicates a faster recovery rate for the child respiratory patient in the current monitoring cycle.

[0097] In this way, the recovery speed score of pediatric respiratory patients can be determined based on the three factors of the first respiratory index change rate, the second respiratory index change rate and the symptom weight. Comparing the change rate of the first respiratory index and the second respiratory index change rate can reflect the difference in recovery speed of pediatric respiratory patients at different rehabilitation stages. Different types of respiratory diseases have different rehabilitation characteristics and needs. Through symptom weight, it is helpful to achieve individualized treatment, more accurately evaluate the recovery speed, and adjust the treatment plan in time.

[0098] According to one embodiment of the present invention, in step S4, the respiratory index data is input into a trained respiratory rate prediction model. The respiratory rate prediction model can be a neural network model. Through machine learning or deep learning technology, it is trained based on a large amount of sample data to obtain predicted respiratory rate data at multiple moments in the current monitoring cycle.

[0099] Figure 3 The flowchart of the respiratory rate prediction model training steps according to an embodiment of the present invention is exemplarily shown.

[0100] According to one embodiment of the present invention, the training step of the respiratory rate prediction model includes: step S41, obtaining sample respiratory index data of multiple sample child respiratory patients at multiple moments in multiple historical monitoring cycles, wherein the sample respiratory index data include sample vital capacity and sample blood oxygen saturation; step S42, inputting the sample respiratory index data into the trained respiratory rate prediction model to obtain sample predicted respiratory rate data of multiple sample child respiratory patients at multiple moments in multiple historical monitoring cycles; step S43, obtaining sample respiratory rate data of multiple sample child respiratory patients at multiple moments in multiple historical monitoring cycles; step S44, obtaining the age of the child respiratory patient and the sample age of multiple sample child respiratory patients; step S45, determining the loss function of the respiratory rate prediction model based on the sample predicted respiratory rate data, the sample respiratory rate data, the age and the sample age; step S46, training the respiratory rate prediction model based on the loss function of the respiratory rate prediction model to obtain a trained respiratory rate prediction model.

[0101] According to one embodiment of the present invention, the historical monitoring period is an actual period, and data at each moment can be actually collected. The medical system collects sample respiratory index data, including sample vital capacity and sample blood oxygen saturation, from multiple sample child respiratory patients. The multiple sample child respiratory patients are divided into different training batches of sample child respiratory patients, and the number of sample child respiratory patients in each training batch is the same, so as to perform respiratory rate prediction model training for different training batches, for example, M1+M2+…+M N =M, where M is the number of sampled pediatric respiratory patients, M1, M2, ..., M Nwhere _{\theta} ...

[0102] According to one embodiment of the present invention, determining the loss function of the respiratory rate prediction model based on the sample predicted respiratory rate data, the sample respiratory rate data, the age, and the sample age includes: determining the loss function Loss of the respiratory rate prediction model according to formula (1),

[0103]

[0104] Among them, F i,h,k is the sample respiratory rate data of the i-th sample child respiratory patient at the k-th moment in the h-th historical monitoring cycle, F i,h,k,p Predict respiratory rate data for the i-th sample of a child respiratory patient at the k-th moment in the h-th historical monitoring period, Y e is the age of the child respiratory patient, Y i is the sample age of the i-th sample child respiratory patient, K is the number of monitoring cycle moments, H is the number of historical monitoring cycles, M j is the number of sample pediatric respiratory patients in the jth training batch, N is the number of training batches, k≤K, h≤H, i≤M j , j≤N, and k, h, i, j, K, H, M j and N are both positive integers.

[0105] According to one embodiment of the present invention, in formula (1), |F i,h,k -F i,h,k,p | is the difference between the sample respiratory rate data of the i-th sample child respiratory patient at the k-th moment in the h-th historical monitoring cycle and the sample predicted respiratory rate data of the i-th sample child respiratory patient at the k-th moment in the h-th historical monitoring cycle. The difference represents the error between the sample respiratory rate data and the sample predicted respiratory rate data. is the similarity between the age of the child respiratory patient and the sample age of the i-th sample child respiratory patient. In order to achieve a similar respiratory rate detection effect, if the age of the child respiratory patient is closer to the sample age of the sample child respiratory patient, that is, The larger the value of , the more similar the age of the pediatric respiratory patient is to the sample age of the sample pediatric respiratory patient, and the greater its reference value, therefore, the higher its weight. is the weight at the kth moment, which is used to reasonably weight the errors at different moments in the loss function. For the hth historical monitoring period, the accuracy of the sample predicted respiratory frequency data at the k+1th moment output by the respiratory frequency prediction model is usually higher than the accuracy of the sample predicted respiratory frequency data at the kth moment. That is, the shorter the time interval between a certain moment in a historical monitoring period and the first moment, the less accurate the prediction result. In order to improve the training efficiency, the higher the weight is set, and vice versa, the more accurate the prediction result, the lower the weight is. is the weight of the j-th training batch, which is used to reasonably weight the errors of different training batches in the loss function. For the sample pediatric respiratory patients in the j+1-th training batch, the accuracy of the respiratory rate prediction model outputting the sample predicted respiratory rate data of the j+1-th training batch is usually higher than the accuracy of the sample predicted respiratory rate data of the j-th training batch. That is, the shorter the time interval between a training batch and the first training batch, the less accurate the prediction result. In order to improve the training efficiency, the higher the weight is set. Conversely, the more accurate the prediction result, the lower the weight is. Therefore, a higher weight can be given to items with lower accuracy, thereby improving the training intensity and training efficiency.

[0106] According to one embodiment of the present invention, using and The training loss function is obtained by taking a weighted average of the errors in the respiratory rate data of multiple samples of pediatric respiratory patients from multiple historical monitoring periods in the jth training batch. During the training of the respiratory rate prediction model, the loss function is back-propagated and some internal parameters of the model are adjusted to reduce the loss value of the respiratory rate prediction model, thereby improving the accuracy of the respiratory rate prediction model and obtaining the trained respiratory rate prediction model.

[0107] In this way, the relative error between the sample respiratory rate data and the sample predicted respiratory rate data can be used to set a weight based on the characteristic that the more similar the age of the child respiratory patient is to the sample age of the sample child respiratory patient, the greater the reference value, and the weight can be set based on the characteristic that the shorter the time interval from the start moment, the higher the accuracy, and the shorter the time interval from the first training batch, the lower the accuracy. In this way, the errors output by the respiratory rate prediction model of each sample child respiratory patient in the jth batch at multiple moments in multiple historical monitoring cycles are weighted averaged to obtain a loss function, so as to improve the design accuracy and objectivity of the loss function, thereby improving the training efficiency during the training process and improving the accuracy of the respiratory rate prediction model.

[0108] According to one embodiment of the present invention, in step S5, respiratory rate data of the child respiratory patient at multiple moments in the current monitoring cycle is obtained through a respiratory monitor.

[0109] According to one embodiment of the present invention, in step S6, a recovery degree score of the child respiratory patient is determined based on the predicted respiratory rate data and the respiratory rate data.

[0110] Figure 4 The following is an exemplary flowchart of calculating a rehabilitation degree score according to an embodiment of the present invention.

[0111] According to one embodiment of the present invention, step S6 includes: step S61, calculating the absolute value of the relative difference between the predicted respiratory rate data at multiple moments in the current monitoring cycle and the respiratory rate data; step S62, averaging the difference between 1 corresponding to each moment and the absolute value to determine the recovery degree score of the child respiratory patient.

[0112] According to one embodiment of the present invention, the calculation formula for the rehabilitation degree score is: Among them, D R The degree of recovery is scored, f e,k,p is the predicted respiratory rate data at the kth moment of the current monitoring period, f e,k is the respiratory rate data at the kth moment of the current monitoring cycle, K is the number of moments in the monitoring cycle, k ≤ K, and both k and K are positive integers. The closer the actual respiratory rate data of the pediatric respiratory patient is to the predicted respiratory rate data, the better the pediatric respiratory patient's recovery in the current monitoring cycle. Therefore, the higher the recovery level score, the better the pediatric respiratory patient's recovery in the current monitoring cycle.

[0113] In this way, the recovery degree score of pediatric respiratory patients can be determined based on the predicted respiratory rate data and respiratory rate data. By comparing the predicted respiratory rate data and respiratory rate data, the recovery degree of pediatric respiratory patients in the current monitoring cycle can be reflected, and problems in the rehabilitation process can be discovered in time, thereby improving the treatment effect.

[0114] According to one embodiment of the present invention, in step S7, a respiratory rehabilitation monitoring report is generated based on the recovery speed score and the recovery degree score. The respiratory rehabilitation monitoring report can intuitively understand the rehabilitation progress of the pediatric respiratory patient, help formulate a reasonable rehabilitation plan, and improve rehabilitation results.

[0115] Figure 5 A flowchart for generating a respiratory rehabilitation monitoring report according to an embodiment of the present invention is exemplarily shown.

[0116] According to one embodiment of the present invention, step S7 includes: step S71, when the recovery speed score is greater than or equal to a preset recovery speed score, and the recovery degree score is greater than or equal to the preset recovery degree score, it is determined that the recovery effect of the current monitoring period is good and the respiratory symptoms are mild; step S72, when the recovery speed score is less than the preset recovery speed score, and the recovery degree score is greater than or equal to the preset recovery degree score, it is determined that the recovery effect of the current monitoring period is poor, but the respiratory symptoms are mild; step S73, when the recovery speed score is greater than or equal to the preset recovery speed score, and the recovery degree score is less than the preset recovery degree score, it is determined that the recovery effect of the current monitoring period is good, but the respiratory symptoms are severe; step S74, when the recovery speed score is less than the preset recovery speed score, and the recovery degree score is less than the preset recovery degree score, it is determined that the recovery effect of the current monitoring period is poor, and the respiratory symptoms are severe.

[0117] According to one embodiment of the present invention, when the recovery speed score and the recovery degree score are respectively greater than or equal to the preset recovery speed score (e.g., 0.5) and the preset recovery degree score (e.g., 0.9), it indicates that the child respiratory patient has a better recovery effect (faster recovery speed) and a milder respiratory disease (better recovery degree) in the current monitoring period.

[0118] According to the pediatric respiratory rehabilitation monitoring method of an embodiment of the present invention, the recovery speed score can be determined based on the respiratory index data and the type of respiratory disease, which can intuitively reflect the speed of recovery of the pediatric respiratory patient in the current monitoring period. The recovery degree score is obtained by comparing the predicted respiratory rate data with the actual respiratory rate data, which can accurately measure the degree of closeness between the respiratory function of the pediatric respiratory patient in the current monitoring period and the normal level of the current monitoring period. The respiratory rehabilitation monitoring results can be comprehensively monitored based on the two aspects of recovery speed and recovery degree, thereby improving the treatment efficiency. When determining the recovery speed score of the pediatric respiratory patient, the recovery speed score of the pediatric respiratory patient can be determined based on three factors: the first respiratory index change rate, the second respiratory index change rate, and the symptom weight. Comparing the first respiratory index change rate and the second respiratory index change rate can reflect the difference in recovery speed of the pediatric respiratory patient at different rehabilitation stages. The rehabilitation characteristics and needs of different respiratory disease types are different. The symptom weight can help to achieve individualized treatment, more accurately evaluate the recovery speed, and adjust the treatment plan in time. When determining the loss function of the respiratory rate prediction model, the relative error between the sample respiratory rate data and the sample predicted respiratory rate data can be used to set a weight based on the characteristic that the more similar the age of the child respiratory patient is to the sample age of the sample child respiratory patient, the greater the reference value. Furthermore, weights can be set based on the characteristic that the shorter the time interval from the start time, the higher the accuracy, and the shorter the time interval from the first training batch, the lower the accuracy. Thus, the error output of the respiratory rate prediction model for each sample child respiratory patient in the jth batch at multiple times across multiple historical monitoring cycles is weighted averaged to obtain a loss function. This improves the design accuracy and objectivity of the loss function, thereby improving training efficiency and the accuracy of the respiratory rate prediction model during training. When determining the recovery level score of a child respiratory patient, the recovery level score can be determined based on the predicted respiratory rate data and the respiratory rate data. By comparing the predicted respiratory rate data with the respiratory rate data, the recovery level of the child respiratory patient in the current monitoring cycle can be reflected, and problems in the rehabilitation process can be promptly identified, thereby improving treatment effectiveness.

[0119] Figure 6 A block diagram of a child respiratory rehabilitation monitoring system according to an embodiment of the present invention is exemplarily shown, wherein the system includes:

[0120] A respiratory index data module is used to monitor the respiratory index data of pediatric respiratory patients through monitoring equipment at multiple moments in the current monitoring cycle;

[0121] A respiratory disease type module is used to obtain the respiratory disease type of a pediatric respiratory patient based on the patient's medical record;

[0122] a recovery speed scoring module, configured to determine a recovery speed score for a pediatric respiratory patient based on the respiratory index data and the respiratory disease type;

[0123] A respiratory rate prediction data module is used to input the respiratory index data into a trained respiratory rate prediction model to obtain predicted respiratory rate data at multiple moments in the current monitoring period;

[0124] Respiratory rate data module, used to obtain respiratory rate data of pediatric respiratory patients at multiple moments in the current monitoring cycle;

[0125] a recovery degree scoring module, configured to determine a recovery degree score of a pediatric respiratory patient based on the predicted respiratory rate data and the respiratory rate data;

[0126] The respiratory rehabilitation monitoring report module is used to generate a respiratory rehabilitation monitoring report according to the rehabilitation speed score and the rehabilitation degree score.

[0127] The present invention may be a method, an apparatus, a system and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present invention.

[0128] Those skilled in the art will appreciate that the embodiments of the present invention described above and shown in the accompanying drawings are intended to be illustrative only and are not intended to limit the present invention. The objectives of the present invention have been fully and effectively achieved. The functional and structural principles of the present invention have been demonstrated and illustrated in the embodiments. Any variations or modifications may be made to the embodiments of the present invention without departing from the principles described.

Claims

1. A method for monitoring respiratory rehabilitation in children, characterized in that: include: At multiple moments in the current monitoring cycle, respiratory index data of the pediatric respiratory patient is monitored by the monitoring device; The respiratory disease types of pediatric respiratory patients were obtained from their medical records; determining a recovery speed score for the pediatric respiratory patient based on the respiratory index data and the type of respiratory disease; Inputting the respiratory index data into a trained respiratory rate prediction model to obtain predicted respiratory rate data at multiple moments in the current monitoring period; Obtain respiratory rate data of a pediatric respiratory patient at multiple moments in a current monitoring cycle; determining a recovery degree score for the pediatric respiratory patient based on the predicted respiratory rate data and the respiratory rate data; A respiratory rehabilitation monitoring report is generated according to the rehabilitation speed score and the rehabilitation degree score.

2. The child respiratory rehabilitation monitoring method according to claim 1, characterized in that: The respiratory index data includes vital capacity and blood oxygen saturation.

3. The child respiratory rehabilitation monitoring method according to claim 1, characterized in that: Determine a recovery speed score for a pediatric respiratory patient based on the respiratory index data and the type of respiratory disease, including: Divide the current monitoring period into two phases, namely the first phase and the second phase, wherein the duration of the first phase is equal to the duration of the second phase; Fitting the respiratory index data of the first stage and the time in the first stage to obtain a first respiratory index function in the first stage; determining a first breathing index derivative function according to the first breathing index function; determining a first breathing index change rate at multiple moments in the first stage according to the first breathing index derivative function; Fitting the breathing index data of the second stage and the moments in the second stage to obtain a second breathing index function in the second stage; determining a second breathing index derivative function according to the second breathing index function; determining a change rate of the second breathing index at multiple moments in the second stage according to the second breathing index derivative function; Determine symptom weights based on the type of respiratory disease; A recovery speed score for a pediatric respiratory patient is determined according to the first respiratory index change rate, the second respiratory index change rate, and the symptom weight.

4. The child respiratory rehabilitation monitoring method according to claim 3, characterized in that: Determine the symptom weight based on the type of respiratory disease, including: According to the theoretical recovery time of the respiratory disease, the respiratory disease type is divided into mild disease, moderate disease and severe disease, wherein the theoretical recovery time of the mild disease is less than or equal to 7 days, the theoretical recovery time of the moderate disease is greater than 7 days and less than or equal to 14 days, and the theoretical recovery time of the severe disease is greater than 14 days; If the respiratory disease type is a mild disease, the symptom weight is determined to be 1; If the respiratory disease type is a moderate disease, the symptom weight is determined to be 0.5; If the respiratory disease type is a severe disease, the symptom weight is determined to be 0.

1.

5. The child respiratory rehabilitation monitoring method according to claim 3, characterized in that: Determining a recovery speed score for a pediatric respiratory patient according to the first respiratory index change rate, the second respiratory index change rate, and the symptom weight includes: If the average value of the second respiratory index change rate at multiple moments in the second stage is greater than or equal to the average value of the first respiratory index change rate at multiple moments in the first stage, the conditional function takes a value of 1; otherwise, the conditional function takes a value of 0; The conditional function values of the first respiratory index change rate of multiple respiratory index data are averaged and multiplied by the symptom weight to determine the recovery speed score of the pediatric respiratory patient.

6. The child respiratory rehabilitation monitoring method according to claim 1, characterized in that: The training steps of the respiratory rate prediction model include: Acquire sample respiratory index data of multiple sample pediatric respiratory patients at multiple times in multiple historical monitoring periods, wherein the sample respiratory index data include sample vital capacity and sample blood oxygen saturation; Inputting the sample respiratory index data into a trained respiratory rate prediction model to obtain sample predicted respiratory rate data of multiple sample pediatric respiratory patients at multiple times during multiple historical monitoring periods; Obtain sample respiratory rate data of multiple sample pediatric respiratory patients at multiple times in multiple historical monitoring periods; Obtain the age of the pediatric respiratory patient and the sample age of multiple samples of pediatric respiratory patients; determining a loss function of the respiratory rate prediction model based on the sample predicted respiratory rate data, the sample respiratory rate data, the age, and the sample age; The respiratory rate prediction model is trained according to the loss function of the respiratory rate prediction model to obtain a trained respiratory rate prediction model.

7. The child respiratory rehabilitation monitoring method according to claim 6, characterized in that: Determining a loss function of the respiratory rate prediction model according to the sample predicted respiratory rate data, the sample respiratory rate data, the age, and the sample age includes: According to the formula Determine the loss function Loss of the respiratory rate prediction model, where F i,h,k is the sample respiratory rate data of the i-th sample child respiratory patient at the k-th moment in the h-th historical monitoring cycle, F i,h,k,p Predict the respiratory rate data for the i-th sample of a child respiratory patient at the k-th moment in the h-th historical monitoring period, Y e is the age of the child respiratory patient, Y i is the sample age of the i-th sample child respiratory patient, K is the number of monitoring cycle moments, H is the number of historical monitoring cycles, M j is the number of sample pediatric respiratory patients in the jth training batch, N is the number of training batches, k≤K, h≤H, i≤M j , j≤N, and k, h, i, j, K, H, M j and N are both positive integers.

8. The child respiratory rehabilitation monitoring method according to claim 1, characterized in that: Determining a recovery degree score of a pediatric respiratory patient based on the predicted respiratory rate data and the respiratory rate data includes: Calculating the absolute value of the relative difference between the predicted respiratory rate data at multiple moments in the current monitoring period and the respiratory rate data; The difference between 1 and the absolute value corresponding to each moment is averaged to determine the recovery degree score of the pediatric respiratory patient.

9. The child respiratory rehabilitation monitoring method according to claim 1, characterized in that: Generate a respiratory rehabilitation monitoring report based on the rehabilitation speed score and the rehabilitation degree score, including: When the recovery speed score is greater than or equal to the preset recovery speed score, and the recovery degree score is greater than or equal to the preset recovery degree score, it is determined that the recovery effect of the current monitoring period is good and the respiratory symptoms are mild; When the recovery speed score is less than the preset recovery speed score, and the recovery degree score is greater than or equal to the preset recovery degree score, it is determined that the recovery effect of the current monitoring period is poor, but the respiratory symptoms are mild; When the recovery speed score is greater than or equal to the preset recovery speed score, and the recovery degree score is less than the preset recovery degree score, it is determined that the recovery effect of the current monitoring period is good, but the respiratory symptoms are serious; When the recovery speed score is less than the preset recovery speed score, and the recovery degree score is less than the preset recovery degree score, it is determined that the recovery effect of the current monitoring period is poor and the respiratory disease is serious.

10. A child respiratory rehabilitation monitoring system, configured to execute the child respiratory rehabilitation monitoring method according to any one of claims 1 to 9, characterized in that: include: A respiratory index data module is used to monitor the respiratory index data of pediatric respiratory patients through monitoring equipment at multiple moments in the current monitoring cycle; A respiratory disease type module is used to obtain the respiratory disease type of a pediatric respiratory patient based on the patient's medical record; a recovery speed scoring module, configured to determine a recovery speed score for a pediatric respiratory patient based on the respiratory index data and the respiratory disease type; A respiratory rate prediction data module is used to input the respiratory index data into a trained respiratory rate prediction model to obtain predicted respiratory rate data at multiple moments in the current monitoring period; Respiratory rate data module, used to obtain respiratory rate data of pediatric respiratory patients at multiple moments in the current monitoring cycle; a recovery degree scoring module, configured to determine a recovery degree score of a pediatric respiratory patient based on the predicted respiratory rate data and the respiratory rate data; The respiratory rehabilitation monitoring report module is used to generate a respiratory rehabilitation monitoring report according to the rehabilitation speed score and the rehabilitation degree score.

Images