Respiratory function training system and method for patient in cardiothoracic surgery

Abstract

The invention provides a respiratory function training system and method for a patient in the cardiothoracic surgery department, and the method comprises the steps: matching a similar patient group of a target patient from historical rehabilitation records based on obtained individual attribute data, and extracting the dynamic training data of the similar patient group; determining a tidal volume difference coefficient according to the average tidal volume of the target patient in the current training period and the tidal volume distribution of the similar patient group, and determining an endurance difference coefficient according to the sustainable training duration of the target patient in the current training period and the sustainable training duration of the similar patient group, and carrying out weighted fusion on the tidal volume difference coefficient and the endurance difference coefficient to obtain a training performance difference degree between the target patient and the similar patient group, and carrying out adaptive optimization on a training scheme of the next training period of the target patient according to the training performance difference degree and the pain score of the target patient. According to the scheme, the self-adaptive training scheme of the patient can be constructed based on the historical rehabilitation data.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of training optimization, in particular, the present application relates to a respiratory function training system and method for thoracic surgery patients. BACKGROUND

[0002] Traditional respiratory function training for thoracic surgery patients is mainly guided by artificial guidance, assisted by basic equipment, and relies on the active cooperation of patients. The operation is simple and does not require complex technical support. The traditional training methods include abdominal breathing training, pursed-lip breathing training, effective coughing training, and basic breathing training.

[0003] Training optimization refers to scientifically adjusting the target, method, process and evaluation of training, so as to reduce risks and make training more efficient and more in line with needs. Therefore, respiratory function training for thoracic surgery patients is a special improvement of respiratory training based on general training optimization and the special physiological state of thoracic surgery patients. Traditional rehabilitation programs are usually based on surgical types or general staging and belong to general programs for groups, which completely ignores the huge differences in age, baseline lung function, comorbidities, pain tolerance, etc. of individual patients, leading to over-risk for some patients and over-ease for others, resulting in poor effectiveness. Moreover, traditional rehabilitation programs only focus on objective physiological data, forcibly require patients to complete the target, ignore their pain, and lead to fear and resistance of patients. On the contrary, if only based on the subjective feelings of patients, training will be insufficient due to fear, affecting the rehabilitation progress. Therefore, how to construct an adaptive training program for patients based on historical rehabilitation data has become a difficult problem in the industry. SUMMARY

[0004] Based on this, the present application provides a respiratory function training system and method for thoracic surgery patients to construct an adaptive training program for patients based on historical rehabilitation data.

[0005] In a first aspect, the present application provides a respiratory training optimization method for a respiratory function training system for thoracic surgery patients, comprising the following steps: Obtaining individual attribute data of a target patient, the individual attribute data including baseline physiological characteristic parameters, clinical pathological characteristic parameters and dynamic rehabilitation state parameters; Based on the individual attribute data, a similar patient group of the target patient is matched from historical rehabilitation records, and dynamic training data of the similar patient group at the same postoperative rehabilitation day is extracted, the dynamic training data including tidal volume distribution and sustainable training duration; determine a tidal volume difference coefficient according to the average tidal volume of the target patient in the current training cycle and the tidal volume distribution of the similar patient group, and determine a tolerance difference coefficient according to the sustainable training duration of the target patient in the current training cycle and the sustainable training duration of the similar patient group, and then perform weighted fusion on the tidal volume difference coefficient and the tolerance difference coefficient to obtain the training performance difference degree between the target patient and the similar patient group; adaptively optimize the training scheme of the next training cycle of the target patient according to the training performance difference degree and the pain score of the target patient.

[0006] In some embodiments, matching the similar patient group of the target patient from the historical rehabilitation records based on the individual attribute data specifically includes: constructing a multi-dimensional feature vector containing the basic physiological feature parameters, the clinical pathological feature parameters and the dynamic rehabilitation state parameters; determining the comprehensive similarity of the multi-dimensional feature vector of the target patient and the multi-dimensional feature vector of each historical patient in the historical rehabilitation records; sorting each historical patient according to all the comprehensive similarities, and then selecting the top K historical patients as the similar patient group of the target patient.

[0007] In some embodiments, extracting the dynamic training data of the similar patient group at the same postoperative rehabilitation day specifically includes: screening the training records consistent with the postoperative rehabilitation days of the target patient from the historical rehabilitation records of the similar patient group; extracting all tidal volume measurement values from the screened training records to form a tidal volume sample set, and extracting all sustainable training duration data to form a duration sample set; determining the statistical characteristics of the tidal volume distribution in the tidal volume sample set, the statistical characteristics including the mean and the standard deviation; calculating the arithmetic mean of the duration sample set as the sustainable training duration of the similar patient group; integrating the statistical characteristics of the tidal volume distribution and the sustainable training duration of the similar patient group into the dynamic training data.

[0008] In some embodiments, determining the tidal volume difference coefficient according to the average tidal volume of the target patient in the current training cycle and the tidal volume distribution of the similar patient group specifically includes: obtaining the average tidal volume of the target patient in the current training cycle; calculating a standardized difference value according to the average tidal volume of the target patient in the current training cycle, the mean and the standard deviation in the tidal volume distribution, and taking the standardized difference value as the tidal volume difference coefficient.

[0009] In some embodiments, the determination of the endurance difference coefficient based on the sustainable training duration of the target patient in the current training cycle and the sustainable training duration of the similar patient group specifically comprises: obtaining the sustainable training duration actually reached by the target patient in the current training cycle; calculating the absolute difference between the sustainable training duration of the target patient and the sustainable training duration of the similar patient group; dividing the absolute difference by the standard deviation of the sustainable training duration of the similar patient group to obtain a standardized difference value as the endurance difference coefficient.

[0010] In some embodiments, the training performance difference between the target patient and the similar patient group is determined by weighted fusion of the tidal volume difference coefficient and the endurance difference coefficient specifically comprises: determining the fusion weight corresponding to the tidal volume difference coefficient and the endurance difference coefficient according to the type of surgery of the target patient; linearly weighting and summing the tidal volume difference coefficient and the endurance difference coefficient according to the corresponding fusion weight, and taking the weighted sum result as the training performance difference between the target patient and the similar patient group, and converting it into a level identifier.

[0011] In some embodiments, the training program for the next training cycle of the target patient is adaptively optimized based on the training performance difference and the pain score of the target patient specifically comprises: querying a pre-set optimization decision rule table based on the level of the training performance difference and the level of the pain score of the target patient; adjusting the training intensity, training duration and training frequency of the target patient in the next training cycle according to the query result; compliance verification of the adjusted training intensity, training duration and training frequency according to the clinical safety threshold, and then generating a final optimized training program.

[0012] In a second aspect, the present application provides a respiratory function training system for thoracic and cardiac surgery patients, which comprises a respiratory training optimization unit, wherein the respiratory training optimization unit comprises: an acquisition module configured to acquire individual attribute data of a target patient, wherein the individual attribute data comprises basic physiological characteristic parameters, clinical pathological characteristic parameters and dynamic rehabilitation state parameters; a processing module configured to match a similar patient group of the target patient from historical rehabilitation records based on the individual attribute data, and extract dynamic training data of the similar patient group at the same postoperative rehabilitation day, wherein the dynamic training data comprises tidal volume distribution and sustainable training duration; The processing module is further configured to determine a tidal volume difference coefficient according to an average tidal volume of a current training cycle of the target patient and a tidal volume distribution of the similar patient group, determine a tolerance difference coefficient by using a sustainable training duration of the current training cycle of the target patient and a sustainable training duration of the similar patient group, and then perform weighted fusion on the tidal volume difference coefficient and the tolerance difference coefficient to obtain a training performance difference degree between the target patient and the similar patient group. The execution module is configured to perform adaptive optimization on a training scheme of a next training cycle of the target patient according to the training performance difference degree and a pain score of the target patient.

[0013] In a third aspect, the present application provides a computer device, comprising a memory and a processor, the memory stores a computer program, and the processor implements the steps of the respiratory training optimization method of the thoracic and cardiac surgery patient respiratory function training system when executing the computer program.

[0014] In a fourth aspect, the present application provides a computer readable storage medium, which stores a computer program, and the computer program implements the steps of the respiratory training optimization method of the thoracic and cardiac surgery patient respiratory function training system when executed by a processor.

[0015] The technical scheme provided by the embodiments of the present application has the following beneficial effects: The thoracic and cardiac surgery patient respiratory function training system and method provided by the application, first, based on the obtained individual attribute data, the similar patient group of the target patient is matched from the historical rehabilitation records. This step can accurately locate the physiological state, pathological condition and rehabilitation stage of the target patient, and thus a dynamic and reliable control benchmark can be established for subsequent personalized evaluation; secondly, the dynamic training data of the similar patient group at the same postoperative rehabilitation days is extracted, and the dynamic training data includes tidal volume distribution and sustainable training time length. This step can extract the statistical characteristics of key training indicators from the historical data of the similar group, so as to provide objective and quantitative group behavior standards for quantitative evaluation of the target patient's performance; then, the tidal volume difference coefficient is determined according to the average tidal volume of the target patient in the current training period and the tidal volume distribution of the similar patient group, and the endurance difference coefficient is determined by the sustainable training time length of the target patient in the current training period and the sustainable training time length of the similar patient group, and then the tidal volume difference coefficient and the endurance difference coefficient are weighted and fused to obtain the training performance difference between the target patient and the similar patient group. The training performance difference degree comprehensively represents the overall difference between the target patient and the similar patient group in the intensity and endurance of the two core dimensions of respiratory training. This step can quantify the performance gap between the target patient and the similar group from the two dimensions of respiratory depth and endurance, and obtain the comprehensive difference degree by weighted fusion, so as to comprehensively and accurately reflect the relative level of the patient in the rehabilitation queue; finally, the training scheme of the next training period of the target patient is adaptively optimized according to the training performance difference degree and the pain score of the target patient. This step can automatically adjust the training parameters based on the objective performance difference and subjective pain feeling through regularized decision and safety check, so as to dynamically generate a truly personalized training prescription on the premise of ensuring the safety and comfort of the patient. In summary, the scheme of the application can construct an adaptive training scheme for the patient based on historical rehabilitation data. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 is an exemplary flowchart of the respiratory training optimization method of the thoracic and cardiac surgery patient respiratory function training system according to some embodiments of the application; Figure 2 is a flowchart of matching a similar patient group according to some embodiments of the application; Figure 3 is an optimization decision rule table according to some embodiments of the application; Figure 4 is a structural schematic diagram of the respiratory training optimization unit according to some embodiments of the application; Figure 5is a structural schematic diagram of a computer device for implementing a respiratory training optimization method of a thoracic surgery patient respiratory function training system according to some embodiments of the present application. DETAILED DESCRIPTION

[0017] In order to better understand the above technical solutions, the above technical solutions will be described in detail below in combination with the drawings in the specification and specific embodiments.

[0018] Reference Figure 1 The figure is an exemplary flowchart of a respiratory training optimization method of a thoracic surgery patient respiratory function training system according to some embodiments of the present application, which mainly includes the following steps: In step 101, individual attribute data of a target patient is obtained, which includes basic physiological characteristic parameters, clinical pathological characteristic parameters and dynamic rehabilitation state parameters.

[0019] When specifically implemented, the individual attribute data of the target patient can be obtained in the following manner, that is, the basic physiological characteristic parameters of the target patient can be automatically extracted from an electronic medical record system, including the age (unit: years) of the target patient, the gender and the body mass index (unit: kg / m²) of the target patient; the clinical pathological characteristic parameters of the target patient are obtained from the operation record, including the operation mode code of the target patient, the FEV1 predicted value (unit: %) in the preoperative lung function detection value of the target patient and the diagnosed comorbidity code; at the same time, the dynamic rehabilitation state parameters of the target patient are collected, including the integer postoperative days (unit: days) of the target patient calculated based on the operation date and the current pain score (unit: points, range 0-10) collected by using the visual analogue scale method; in the respiratory training process, the tidal volume unit collected by the sensor is usually milliliter, and the continuous training duration unit is usually minute, and in other embodiments, other methods can also be used for implementation, which are not limited in the present application.

[0020] It should be noted that all the parameters in the present application are stored in the patient rehabilitation database after being standardized.

[0021] In step 102, a similar patient group of the target patient is matched from the historical rehabilitation records based on the individual attribute data, and dynamic training data of the similar patient group at the same postoperative rehabilitation days is extracted, which includes tidal volume distribution and continuous training duration.

[0022] In some embodiments, referring to Figure 2 The figure is a flowchart of matching a similar patient group in some embodiments of the present application, and in the present embodiment, the similar patient group of the target patient can be matched from the historical rehabilitation records based on the individual attribute data by using the following steps, that is: In step 1031, a multi-dimensional feature vector containing the basic physiological feature parameters, the clinical pathological feature parameters and the dynamic rehabilitation state parameters is constructed. In step 1032, a comprehensive similarity of the multi-dimensional feature vector of the target patient and the multi-dimensional feature vectors of each historical patient in the historical rehabilitation records is determined. In step 1033, each historical patient is sorted according to all the comprehensive similarities, and the first K historical patients are selected as the similar patient group of the target patient.

[0023] It should be noted that the comprehensive similarity in the present application represents the overall similarity of the target patient and a certain historical patient in the multiple key attributes affecting the respiratory rehabilitation outcome. The core role of the comprehensive similarity is to quickly and accurately screen out a set of individuals most comparable to the target patient in the historical database.

[0024] In a specific implementation, the multi-dimensional feature vector containing the basic physiological feature parameters, the clinical pathological feature parameters and the dynamic rehabilitation state parameters can be implemented in the following manner. Firstly, for the basic physiological feature parameters, the age, gender and body mass index of each patient are numerically coded, wherein the age is directly used as an integer value, the gender is binary coded, and the body mass index is kept to one decimal place. Secondly, for the clinical pathological feature parameters, the surgical method is one-hot coded according to the standard surgical classification table of thoracic and cardiovascular surgery, the preoperative lung function baseline uses the actual value of the FEV1 predicted value, and the comorbidities are converted into a feature vector through the international disease classification code. Then, for the dynamic rehabilitation state parameters, the postoperative rehabilitation days are used as an integer feature, and the pain score uses the actual score of the visual analogue scale method. Finally, all numerical features are subjected to minimum-maximum normalization processing to make the feature values in the range of 0-1. Finally, all the processed features are spliced into a fixed-dimensional numerical vector, thereby completing the construction of the multi-dimensional feature vector. In other embodiments, other methods can also be used, which are not limited in the present application.

[0025] In a specific implementation, the comprehensive similarity of the multi-dimensional feature vector of the target patient and the multi-dimensional feature vectors of each historical patient in the historical rehabilitation records can be implemented in the following manner. Firstly, the weighted Euclidean distance d of the multi-dimensional feature vector of the target patient and the multi-dimensional feature vector of each historical patient in each dimension is calculated. The calculation formula of the weighted Euclidean distance is as follows: wherein, is the normalized numerical value of the target patient in the i-th feature dimension, is the normalized numerical value of the historical patient in the i-th feature dimension, is the preset weight coefficient of the i-th feature dimension, is the total dimension of the features; then, the comprehensive distance value is converted into a comprehensive similarity in the range of 0-1 through distance reciprocal transformation, and the calculation formula can be: comprehensive similarity = 1 / (1+d). In other embodiments, other methods can also be used to achieve this, which is not limited in the present application.

[0026] In a specific implementation, the following method can be used to sort all historical patients according to all comprehensive similarities, and then select the top K historical patients as the similar patient group of the target patient, that is: first, the comprehensive similarity scores of the target patient and all historical patients can be arranged in descending order, and then a suitable K value can be determined according to the size and data density of the historical rehabilitation database. In a preferred embodiment, K is 50, that is, the top 50 historical patients with the highest comprehensive similarity scores are selected to form the similar patient group; in another embodiment, the K value can be dynamically adjusted according to the data distribution. When the similarity difference in the high-score segment is small, the K value is appropriately increased, and when the difference is large, the K value is correspondingly reduced, so as to finally ensure that the selected group can represent the characteristics of the target patient and has sufficient statistical significance. In other embodiments, other methods can also be used to achieve this, which is not limited in the present application.

[0027] It should be noted that the above steps can accurately locate the historical reference group with highly comparable physiological state, pathological condition and rehabilitation stage for the target patient, so as to establish a dynamic and reliable control benchmark for subsequent personalized assessment.

[0028] In some embodiments, the following steps can be used to extract the dynamic training data of the similar patient group at the same postoperative rehabilitation day, that is: screening training records consistent with the postoperative rehabilitation day of the target patient from the historical rehabilitation records of the similar patient group; extracting all tidal volume measurement values from the screened training records to form a tidal volume sample set, and extracting all sustainable training duration data to form a duration sample set; determining the statistical characteristics of the tidal volume distribution in the tidal volume sample set, including the mean and the standard deviation; calculating the arithmetic mean of the duration sample set as the sustainable training duration of the similar patient group; integrating the statistical characteristics of the tidal volume distribution and the sustainable training duration of the similar patient group into the dynamic training data.

[0029] In a specific implementation, determining the statistical characteristics of the tidal volume distribution in the tidal volume sample set can be implemented in the following manner: the statistical characteristics of the tidal volume distribution in the tidal volume sample set can be calculated using standard statistical methods. Specifically, the mean value of all tidal volume measurements in the sample set is calculated as a measure of the central position of the tidal volume distribution, and the standard deviation of all tidal volume measurements is calculated as a measure of the dispersion of the tidal volume distribution. The standard deviation is calculated using an unbiased estimation method, i.e., the denominator uses the sample size minus one. The final mean value and standard deviation together constitute the statistical characteristics of the tidal volume distribution. In other embodiments, other methods can be used, which are not limited in the present application.

[0030] In a specific implementation, integrating the statistical characteristics of the tidal volume distribution and the sustainable training duration of the similar patient population into dynamic training data can be implemented in the following manner: the mean value of the calculated tidal volume, the standard deviation of the tidal volume, and the mean value and standard deviation of the sustainable training duration are combined into dynamic training data. In other embodiments, other methods can be used, which are not limited in the present application.

[0031] It should be noted that the above steps can extract the statistical characteristics of the key training indicators from the historical data of the similar population, thereby providing objective and quantitative population behavior standards for quantitative evaluation of the target patient performance.

[0032] In step 103, the tidal volume difference coefficient is determined according to the mean tidal volume of the target patient in the current training period and the tidal volume distribution of the similar patient population, and the endurance difference coefficient is determined by the sustainable training duration of the target patient in the current training period and the sustainable training duration of the similar patient population. Then, the tidal volume difference coefficient and the endurance difference coefficient are weighted and fused to obtain the training performance difference between the target patient and the similar patient population.

[0033] In some embodiments, determining the tidal volume difference coefficient according to the mean tidal volume of the target patient in the current training period and the tidal volume distribution of the similar patient population can be implemented in the following steps: Obtaining the mean tidal volume of the target patient in the current training period; Calculating a standardized difference value according to the mean tidal volume of the target patient in the current training period, the mean value and the standard deviation in the tidal volume distribution, and taking the standardized difference value as the tidal volume difference coefficient.

[0034] It should be noted that the tidal volume difference coefficient in the present application is used to quantify the deviation of the target patient from the average level of the similar patient group in a single breath depth, and its core function is to eliminate the dimension effect, so that the tidal volume performance of different patients can be compared with each other. A higher positive value indicates that the patient's tidal volume is significantly better than the similar group, and a higher negative value indicates that it is significantly behind. It provides an objective and comparable index for evaluating the "strength" dimension of the patient's respiratory function.

[0035] In specific implementation, the average tidal volume of the target patient in the current training period can be obtained in the following manner: the tidal volume raw data of each breath of the target patient in the current training period can be collected by the sensor of the respiratory training device, the collected tidal volume raw data is subjected to validity check to exclude abnormal values caused by coughing, speaking or device interference, and then the arithmetic mean of all tidal volume measurement values obtained after the check is taken as the average tidal volume of the current training period. In other embodiments, other methods can also be used, which are not limited in the present application.

[0036] In specific implementation, the standardized difference value is calculated according to the average tidal volume of the target patient in the current training period, the average value and the standard deviation in the tidal volume distribution, and the standardized difference value is taken as the tidal volume difference coefficient, which can be realized in the following manner: first, the pre-calculated average value and standard deviation of the tidal volume are read from the dynamic training data of the similar patient group; then, the difference between the average tidal volume of the target patient in the current training period and the average value in the tidal volume distribution is calculated; finally, the difference value is divided by the standard deviation of the tidal volume distribution, and the quotient value is the tidal volume difference coefficient; wherein the calculated tidal volume difference coefficient is a signed value, a positive number indicates that the tidal volume of the target patient is better than the average level of the similar group, and a negative number indicates that the tidal volume of the target patient lags behind the average level of the similar group. In other embodiments, other methods can also be used, which are not limited in the present application.

[0037] In some embodiments, the determination of the endurance difference coefficient by the sustainable training duration of the target patient in the current training period and the sustainable training duration of the similar patient group can be realized in the following steps: The actual sustainable training duration of the target patient in the current training period is obtained; The difference between the sustainable training duration of the target patient and the sustainable training duration of the similar patient group is calculated; The difference value is divided by the standard deviation of the sustainable training duration of the similar patient group to obtain a standardized difference value as the endurance difference coefficient.

[0038] It should be noted that the endurance difference coefficient in the present application is a signed value, and a positive number indicates that the target patient's training endurance is better than the average level of the similar group, and a negative number indicates that the target patient's training endurance lags behind the average level of the similar group.

[0039] In addition, it should be noted that the endurance difference coefficient in the present application is used to quantify the deviation of the target patient from the average level of the similar patient group in maintaining training endurance, and its core function is to objectively measure the performance of the patient's respiratory muscle endurance and cardiopulmonary function endurance relative to the reference group. The endurance difference coefficient focuses on evaluating the "endurance" dimension of the patient's respiratory function.

[0040] In specific implementation, the actual sustainable training duration of the target patient in the current training period can be achieved in the following manner, that is, the complete duration from the start of training to the end of training can be automatically recorded by the timer built-in the respiratory training device, the start of training is marked by the first breath of the patient reaching the effective tidal volume threshold, and the end of training is judged by the patient actively pressing the stop button or continuously exceeding the preset time without reaching the effective tidal volume threshold. The time interval is recorded and automatically stored in the patient training database, wherein data validity verification is required before storage to exclude abnormal records caused by device failure or accidental interruption. Finally, the time value that passes the verification is taken as the sustainable training duration of the current training period. In other embodiments, other methods can also be used to achieve this, which are not limited in the present application.

[0041] In some embodiments, the tidal volume difference coefficient and the endurance difference coefficient are weighted and fused to obtain the training performance difference degree between the target patient and the similar patient group, which can be achieved in the following steps, that is: According to the type of surgery of the target patient, determine the fusion weight corresponding to the tidal volume difference coefficient and the endurance difference coefficient; According to the corresponding fusion weight, linearly weight and sum the tidal volume difference coefficient and the endurance difference coefficient, and take the weighted sum result as the training performance difference degree between the target patient and the similar patient group, and convert it into a level identifier.

[0042] It should be noted that the training performance difference degree in the present application is a signed value, which comprehensively reflects the overall difference degree and direction of the target patient in the intensity and endurance of respiratory training in two core dimensions from the overall performance of the similar patient group. A positive value indicates that the comprehensive performance is better than the similar group, and a negative value indicates that the comprehensive performance lags behind the similar group.

[0043] In a specific implementation, the fusion weight corresponding to the tidal volume difference coefficient and the endurance difference coefficient can be determined according to the type of surgery of the target patient, which can be realized in the following manner: based on the clinical pathway consensus of postoperative respiratory rehabilitation of thoracic and cardiovascular surgery, for patients undergoing lung surgery, the focus of rehabilitation is to restore lung parenchymal function and gas exchange efficiency, so the fusion weight of the tidal volume difference coefficient is set to 0.6, and the fusion weight of the endurance difference coefficient is set to 0.4; for patients undergoing cardiac surgery, the focus of rehabilitation is to improve the endurance and functional capacity of the heart-lung circulation, so the fusion weight of the tidal volume difference coefficient is set to 0.4, and the fusion weight of the endurance difference coefficient is set to 0.6. In other embodiments, other methods can also be used, which are not limited in the present application.

[0044] In a specific implementation, the linear weighted sum calculation of the tidal volume difference coefficient and the endurance difference coefficient according to the corresponding fusion weight can be realized in the following manner: the tidal volume difference coefficient is multiplied by the corresponding fusion weight value, and the endurance difference coefficient is multiplied by the corresponding fusion weight value, then the two product results are added to obtain a weighted sum, which is the specific value of the training performance difference degree. To further improve the operability of the scheme, the value can also be converted into a grade identifier, for example: training performance difference degree less than or equal to -1.0 is defined as'significant lag', greater than -1.0 and less than 0 is defined as'slight lag', greater than or equal to 0 and less than 1.0 is defined as'slight lead', and greater than or equal to 1.0 is defined as'significant lead'. In other embodiments, other methods can also be used, which are not limited in the present application.

[0045] It should be noted that the above steps can quantify the performance gap of the target patient and the similar group from the depth of respiration and endurance, and obtain a comprehensive difference degree by weighted fusion, so as to comprehensively and accurately reflect the relative level of the patient in the rehabilitation queue.

[0046] In step 104, the training scheme of the next training cycle of the target patient is adaptively optimized according to the training performance difference degree and the pain score of the target patient.

[0047] In some embodiments, the training scheme of the next training cycle of the target patient can be adaptively optimized according to the training performance difference degree and the pain score of the target patient, which can be realized in the following steps: querying a preset optimization decision rule table based on the grade of the training performance difference degree and the grade of the pain score of the current training cycle of the target patient; adjusting the training intensity, training duration and training frequency of the next training cycle of the target patient according to the query result; performing compliance verification on the adjusted training intensity, training duration and training frequency according to a clinical safety threshold, and then generating a final optimized training scheme.

[0048] In specific implementation, querying the preset optimization decision rule table based on the level of training performance difference and the level of the target patient's pain score in the current training cycle can be achieved in the following way: First, a preset optimization decision rule table containing all possible level combinations is maintained, as shown in the following figure. Figure 3 As shown, the training intensity adjustment refers to the percentage adjustment of the target tidal volume, and the training duration adjustment refers to the number of minutes to adjust the duration of a single training session. The optimization decision rule table uses the level of training performance difference as the row and the level of the pain score of the target patient in the current training cycle as the column. When a query is needed, the currently calculated level of training performance difference and the level of pain score are used as input parameters, and the corresponding matrix cell is located through the table lookup algorithm to obtain the preset training scheme adjustment strategy under this combination. Other methods can also be used in other embodiments, and this application does not limit them.

[0049] In specific implementation, adjusting the training intensity, training duration, and training frequency for the next training cycle of the target patient based on the query results can be achieved in the following way: After obtaining the corresponding adjustment strategy from the optimization decision rule table, the existing training plan can be adjusted according to the parameter modification rules defined in the adjustment strategy. Specifically, for training intensity, the current target tidal volume can be adjusted according to the percentage indicated by the adjustment strategy; for training duration, the single training time can be increased or decreased according to the number of minutes specified by the adjustment strategy; for training frequency, the number of daily training sessions can be adjusted according to the requirements of the adjustment strategy. All adjustment operations need to be modified relative to the parameters of the current training plan to ensure the continuity and gradualness of the adjustment. Other methods can also be used in other embodiments, which are not limited in this application.

[0050] In specific implementation, the adjusted training intensity, training duration, and training frequency are verified for compliance based on clinical safety thresholds to generate the final optimized training plan. This can be achieved in the following way: After parameter adjustment, the adjusted training intensity is compared with the preset safety threshold to ensure that the target tidal volume is not less than 60% of the patient's preoperative predicted value, the duration of a single training session does not exceed the maximum safe duration of 120 minutes, and the training frequency does not exceed 4 times per day. If any parameter exceeds the safe range, it must be automatically corrected to the nearest safe threshold range. Finally, all verified parameters can be combined to generate a structured training plan document. This training plan document contains specific training parameter settings and execution instructions and is output to the respiratory training device control system. Other methods can also be used in other embodiments, and this application does not limit them.

[0051] It should be noted that the above steps can automatically adjust the training parameters based on the objective performance difference and subjective pain perception through rule-based decision and safety check, so as to dynamically generate a truly personalized training prescription under the premise of ensuring patient safety and comfort.

[0052] In addition, another aspect of the present application, in some embodiments, the present application provides a thoracic and cardiac surgery patient respiratory function training system, which comprises a respiratory training optimization unit, which is used for Figure 4 The figure is a structural schematic diagram of a respiratory training optimization unit according to some embodiments of the present application, which comprises an acquisition module 401, a processing module 402 and an execution module 403, which are described as follows: The acquisition module 401 is mainly used for acquiring individual attribute data of a target patient in the present application, and the individual attribute data includes basic physiological characteristic parameters, clinical pathological characteristic parameters and dynamic rehabilitation state parameters. The processing module 402 is mainly used for matching a similar patient group of the target patient from historical rehabilitation records based on the individual attribute data, and extracting dynamic training data of the similar patient group at the same postoperative rehabilitation day, which includes tidal volume distribution and sustainable training duration. The processing module 402 is also used for determining a tidal volume difference coefficient according to the average tidal volume of the current training cycle of the target patient and the tidal volume distribution of the similar patient group, and determining a tolerance difference coefficient by the sustainable training duration of the current training cycle of the target patient and the sustainable training duration of the similar patient group, and then weighting and fusing the tidal volume difference coefficient and the tolerance difference coefficient to obtain a training performance difference degree between the target patient and the similar patient group. The execution module 403 is mainly used for adaptively optimizing the training scheme of the next training cycle of the target patient according to the training performance difference degree and the pain score of the target patient.

[0053] The above-mentioned modules in the respiratory training optimization unit can be realized by software, hardware and their combinations in whole or in part. The above-mentioned modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so as to be called and executed by the processor to perform the operations corresponding to the above-mentioned modules.

[0054] In addition, in one embodiment, the present application provides a computer device, which can be a server, and its internal structure diagram can be as shown in Figure 5As shown in the figure. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium, an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the respiratory training optimization data of the thoracic surgery patient respiratory function training system. The network interface of the computer device is used to communicate with the external terminal through the network connection. The computer program is executed by the processor to implement a respiratory training optimization method of a thoracic surgery patient respiratory function training system.

[0055] Those skilled in the art can understand that, Figure 5 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0056] In one embodiment, a computer device is also provided, including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps in the above-mentioned thoracic surgery patient respiratory function training system respiratory training optimization method embodiments.

[0057] In one embodiment, a computer readable storage medium is provided, storing a computer program, which is executed by a processor to implement the steps in the above-mentioned thoracic surgery patient respiratory function training system respiratory training optimization method embodiments.

[0058] In one embodiment, a computer program product or computer program is provided, which includes computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions, so that the computer device executes the steps in the above-mentioned thoracic surgery patient respiratory function training system respiratory training optimization method embodiments.

[0059] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiment methods. Any reference to memory, storage, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0060] Any combination of the technical features of the above embodiments can be made. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, but as long as the combination of the technical features does not exist, it should be considered as the scope of the present application.

[0061] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent. It should be pointed out that for those skilled in the art, without departing from the concept of the present application, some modifications and improvements can be made, which are all within the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A method for optimizing respiratory training in a respiratory function training system for cardiothoracic surgery patients, characterized in that, The method comprises the following steps: Obtaining individual attribute data of a target patient, the individual attribute data comprising basic physiological characteristic parameters, clinical pathological characteristic parameters and dynamic rehabilitation state parameters; Matching a similar patient group of the target patient from historical rehabilitation records based on the individual attribute data, and extracting dynamic training data of the similar patient group at the same postoperative rehabilitation day, the dynamic training data comprising tidal volume distribution and sustainable training duration; Determining a tidal volume difference coefficient according to the average tidal volume of the current training cycle of the target patient and the tidal volume distribution of the similar patient group, and determining a tolerance difference coefficient by the sustainable training duration of the current training cycle of the target patient and the sustainable training duration of the similar patient group, and then weighting and fusing the tidal volume difference coefficient and the tolerance difference coefficient to obtain a training performance difference degree between the target patient and the similar patient group; Adaptively optimizing a training scheme of the next training cycle of the target patient according to the training performance difference degree and the pain score of the target patient.

2. The method of claim 1, wherein, The matching of the similar patient group of the target patient from the historical rehabilitation records based on the individual attribute data specifically comprises: Constructing a multi-dimensional feature vector comprising the basic physiological characteristic parameters, the clinical pathological characteristic parameters and the dynamic rehabilitation state parameters; Determining the comprehensive similarity of the multi-dimensional feature vector of the target patient and the multi-dimensional feature vector of each historical patient in the historical rehabilitation records; Ranking each historical patient according to all the comprehensive similarities, and then selecting the top K historical patients as the similar patient group of the target patient.

3. The method of claim 1, wherein, The extraction of the dynamic training data of the similar patient group at the same postoperative rehabilitation day specifically comprises: Screening training records consistent with the postoperative rehabilitation day of the target patient from the historical rehabilitation records of the similar patient group; Extracting all tidal volume measurement values from the screened training records to form a tidal volume sample set, and extracting all sustainable training duration data to form a duration sample set; Determining the statistical characteristics of the tidal volume distribution in the tidal volume sample set, the statistical characteristics comprising mean value and standard deviation; Calculating the arithmetic mean value of the duration sample set as the sustainable training duration of the similar patient group; Integrating the statistical characteristics of the tidal volume distribution and the sustainable training duration of the similar patient group into the dynamic training data.

4. The method of claim 1, wherein, The determination of the tidal volume difference coefficient according to the average tidal volume of the current training cycle of the target patient and the tidal volume distribution of the similar patient group specifically comprises: Obtaining the average tidal volume of the current training cycle of the target patient; Calculating a standardized difference value according to the average tidal volume of the current training cycle of the target patient, the mean value and the standard deviation in the tidal volume distribution, and taking the standardized difference value as the tidal volume difference coefficient.

5. The method of claim 1, wherein, The determination of the tolerance difference coefficient by the sustainable training duration of the current training cycle of the target patient and the sustainable training duration of the similar patient group specifically comprises: Obtaining the sustainable training duration actually reached by the target patient in the current training cycle; Calculating the absolute difference value between the sustainable training duration of the target patient and the sustainable training duration of the similar patient group; The absolute difference value is divided by the standard deviation of sustainable training duration of the similar patient population to obtain a standardized difference value as a tolerance difference coefficient.

6. The method of claim 1, wherein, The tidal volume difference coefficient and the tolerance difference coefficient are weighted and fused to obtain a training performance difference degree between the target patient and the similar patient population, and the training performance difference degree is converted into a grade identifier. The fusion weight corresponding to the tidal volume difference coefficient and the tolerance difference coefficient is determined according to the type of surgery of the target patient. The tidal volume difference coefficient and the tolerance difference coefficient are linearly weighted and summed according to the corresponding fusion weight, and the weighted sum result is taken as the training performance difference degree between the target patient and the similar patient population, and is converted into a grade identifier.

7. The method of claim 1, wherein, The training scheme of the next training cycle of the target patient is adaptively optimized according to the training performance difference degree and the pain score of the target patient, and the adaptive optimization specifically includes: The preset optimization decision rule table is queried based on the grade of the training performance difference degree and the grade of the pain score of the target patient. The training intensity, training duration and training frequency of the next training cycle of the target patient are adjusted according to the query result. The adjusted training intensity, training duration and training frequency are checked for compliance according to a clinical safety threshold, and a finally optimized training scheme is generated.

8. A cardiothoracic surgery patient respiratory function training system comprising a respiratory training optimization unit, characterized in that, The respiratory training optimization unit includes: An acquisition module is configured to obtain individual attribute data of a target patient, the individual attribute data including basic physiological characteristic parameters, clinical pathological characteristic parameters and dynamic rehabilitation state parameters. A processing module is configured to match a similar patient population of the target patient from historical rehabilitation records based on the individual attribute data, and extract dynamic training data of the similar patient population at the same postoperative rehabilitation day, the dynamic training data including tidal volume distribution and sustainable training duration. The processing module is further configured to determine a tidal volume difference coefficient according to the average tidal volume of the current training cycle of the target patient and the tidal volume distribution of the similar patient population, and determine a tolerance difference coefficient by the sustainable training duration of the current training cycle of the target patient and the sustainable training duration of the similar patient population, and then weight and fuse the tidal volume difference coefficient and the tolerance difference coefficient to obtain a training performance difference degree between the target patient and the similar patient population. An execution module is configured to adaptively optimize the training scheme of the next training cycle of the target patient according to the training performance difference degree and the pain score of the target patient. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-8 when the computer program is executed by the processor. The processor executes the computer program to implement the steps of the respiratory training optimization method of the thoracic and cardiovascular surgery patient respiratory function training system in any one of claims 1 to 7.

10. A computer readable storage medium storing a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the respiratory training optimization method of the thoracic and cardiovascular surgery patient respiratory function training system in any one of claims 1 to 7.

Images