Pediatric clinical data acquisition method and system

Abstract

The invention provides a pediatric clinical data acquisition method and system. The pediatric clinical data acquisition method comprises the following steps: acquiring a clinical parameter sample of each physiological parameter; based on the missing duration of each clinical observation value in the clinical parameter sample of each physiological parameter, performing segmentation reconstruction on the clinical parameter sample to obtain a clinical reconstruction value of each clinical observation value, and determining an information loss degree of each clinical reconstruction value; determining a plurality of missing observation points in all the clinical reconstruction values, and compensating each missing observation point through all the information loss degrees to obtain a plurality of filling samples; determining an abnormal characteristic value when the child sleeps, and dividing the filling sample into an abnormal sample and a normal sample through the abnormal characteristic value; and according to the abnormal sample and the normal sample of each physiological parameter, checking clinical data when the child sleeps. By adopting the scheme of the application, the missing value filling of the child clinical data can be realized under the condition of data fluctuation nonlinearity.

Description

Technical Field

[0001] This application relates to the technical field of healthcare information processing. In particular, this application relates to a pediatric clinical data collection method and system. Background Art

[0002] Healthcare information processing refers to the application of information technology to manage and process healthcare-related data, information, and knowledge to improve the efficiency, quality, and patient experience of medical services. Its application scope is extensive, covering multiple aspects such as electronic health records, telemedicine, medical data analysis, intelligent management of medical devices, patient monitoring systems, and hospital information systems. The aim is to digitally store patients' health information so that medical professionals can instantaneously access information such as patients' medical histories, diagnostic records, and drug prescriptions, thereby improving diagnostic accuracy, reducing duplicate tests, and accelerating decision-making speed.

[0003] Pediatric clinical data collection effectively solves the problems of complexity and individual differences in the data collection process through the application of innovative methods and system devices, improving the accuracy and work efficiency of the data. By utilizing big data and artificial intelligence technologies, it can provide strong decision-making support for the diagnosis, treatment, and research of pediatric diseases. The development of pediatric clinical data collection not only optimizes the data collection process but also provides new possibilities for the future development of pediatric medicine. During the pediatric clinical data collection process, data missing often occurs, which can lead to decision-making errors by medical professionals. In the existing technology, mathematical interpolation functions (such as linear interpolation) are often used to reduce the impact of data missing on medical professionals' decision-making. However, the linear interpolation method usually makes a straight-line estimation between missing values without considering the duration of the missing values. Even if the missing time is very long, it still assumes that the relationship between the missing values and adjacent values is linear. However, in physiological data, this assumption often does not hold. For example, a child's heart rate or respiratory rate may exhibit non-linear fluctuations due to various factors (such as sleep stage, activity status), making it impossible for conventional linear interpolation to accurately capture these dynamic characteristics. Therefore, how to fill in the missing values of children's clinical data in the case of non-linear data fluctuations has become a difficult problem faced by the industry. Summary of the Invention

[0004] This application provides a pediatric clinical data collection method and system that can fill in the missing values of children's clinical data in the case of non-linear data fluctuations.

[0005] In a first aspect, this application provides a pediatric clinical data collection method, including the following steps: Monitor various physiological parameters of a child during sleep to obtain clinical parameter samples of each physiological parameter; For each physiological parameter, the clinical parameter sample is segmented and reconstructed based on the missing duration of each clinical observation value in the clinical parameter sample corresponding to the physiological parameter, to obtain the clinical reconstruction value of each clinical observation value in the clinical parameter sample, and then the information loss degree of each clinical reconstruction value at the corresponding missing duration is determined; Determine multiple missing observation points among all clinical reconstruction values, and perform periodic compensation on the clinical data of each missing observation point through all information loss degrees, so as to obtain the filling sample of each physiological parameter; Based on all missing observation points, determine the abnormal characteristic values during children's sleep, and divide the filling sample into abnormal samples and normal samples of each physiological parameter through the abnormal characteristic values; Check the clinical data during children's sleep according to the abnormal samples and normal samples of each physiological parameter.

[0006] In some embodiments, segmenting and reconstructing the clinical parameter sample based on the missing duration of each clinical observation value in the clinical parameter sample corresponding to each physiological parameter to obtain the clinical reconstruction value of each clinical observation value in the clinical parameter sample specifically includes: Obtain the missing duration of each clinical observation value in the clinical parameter sample corresponding to each physiological parameter; Set the segmentation labels of each clinical observation value through all the missing durations; Reconstruct the clinical parameter sample based on the missing duration of each clinical observation value in the clinical parameter sample and the corresponding segmentation label, to obtain the clinical reconstruction value of each clinical observation value in the clinical parameter sample.

[0007] In some embodiments, determining the information loss degree of each clinical reconstruction value at the corresponding missing duration specifically includes: Select a clinical reconstruction value as the selected clinical reconstruction value; Obtain the missing duration corresponding to the selected clinical reconstruction value; Determine the information loss degree of the selected clinical reconstruction value affected by the missing duration based on the selected clinical reconstruction value and the missing duration; Continue to determine the information loss degree of the remaining clinical reconstruction values affected by the missing duration.

[0008] In some embodiments, performing periodic compensation on the clinical data of each missing observation point through all information loss degrees to obtain the filling sample of each physiological parameter specifically includes: Determine the periodic pattern characteristics of all clinical reconstruction values; Compensate the clinical data of each missing observation point through the periodic pattern characteristics and all information loss degrees to obtain the clinical filling value of each missing observation point; Construct an imputation sample for each physiological parameter based on all clinical imputation values.

[0009] In some embodiments, determining the abnormal characteristic values during children's sleep based on all missing observation points specifically includes: Determining the abnormal characteristic values during children's sleep based on all missing observation points specifically includes: Select a missing observation point as the selected missing observation point; Obtain the corresponding missing duration of the selected missing observation point; Determine the missing characteristic value of the selected missing observation point according to the selected missing observation point and the missing duration; Continue to determine the missing characteristic values of the remaining missing observation points; Determine the abnormal factors of the physiological parameters during children's sleep based on the imputation sample; Determine the abnormal characteristic values during children's sleep through all the missing characteristic values and the abnormal factors.

[0010] In some embodiments, dividing the imputation sample into abnormal samples and normal samples for each physiological parameter through the abnormal characteristic values specifically includes: Perform an abnormal judgment on the imputation sample through the abnormal characteristic values to obtain a plurality of clinical abnormal values; Use the set formed by sorting all the clinical abnormal values in chronological order of timestamps as the abnormal sample for each physiological parameter; Use the set formed by sorting the other clinical reconstruction value groups in the imputation sample in chronological order of timestamps as the normal sample for each physiological parameter.

[0011] In some embodiments, checking the clinical data during children's sleep based on the abnormal samples and normal samples of each physiological parameter specifically includes: Store the abnormal samples of each physiological parameter in the abnormal storage area of the children's sleep monitoring center; Store the normal samples of each physiological parameter in the normal storage area of the children's sleep monitoring center; Check the clinical data during children's sleep through the abnormal samples in the abnormal storage area and the normal samples in the normal storage area.

[0012] In a second aspect, the present application provides a pediatric clinical data acquisition system, including: A monitoring module, configured to monitor various physiological parameters during children's sleep to obtain a clinical parameter sample for each physiological parameter; A processing module, configured to, for each physiological parameter, segmentally reconstruct the clinical parameter sample based on the missing duration of each clinical observation value in the clinical parameter sample corresponding to the physiological parameter, obtain the clinical reconstruction value of each clinical observation value in the clinical parameter sample, and further determine the information loss degree of each clinical reconstruction value under the corresponding missing duration; The processing module is further configured to determine multiple missing observation points among all the clinical reconstruction values, perform periodic compensation on the clinical data of each missing observation point through all the information loss degrees, and further obtain the filled sample of each physiological parameter; The processing module is further configured to determine the abnormal characteristic value during children's sleep based on all the missing observation points, and divide the filled sample into the abnormal sample and the normal sample of each physiological parameter through the abnormal characteristic value; An execution module, configured to check the clinical data during children's sleep according to the abnormal sample and the normal sample of each physiological parameter.

[0013] In a third aspect, the present application provides a computer device, including a memory and a processor, where the memory stores a computer program, and when the processor executes the computer program, the steps of the above pediatric clinical data acquisition method are implemented.

[0014] In a fourth aspect, the present application provides a computer-readable storage medium, storing a computer program, and when the computer program is executed by a processor, the steps of the above pediatric clinical data acquisition method are implemented.

[0015] The technical solution provided by the disclosed embodiments of the present application has the following beneficial effects: In the pediatric clinical data acquisition method and system provided by the present application, various physiological parameters during children's sleep are monitored to obtain the clinical parameter sample of each physiological parameter; for each physiological parameter, the clinical parameter sample is segmentally reconstructed based on the missing duration of each clinical observation value in the clinical parameter sample corresponding to the physiological parameter, the clinical reconstruction value of each clinical observation value in the clinical parameter sample is obtained, and further the information loss degree of each clinical reconstruction value under the corresponding missing duration is determined; multiple missing observation points among all the clinical reconstruction values are determined, periodic compensation is performed on the clinical data of each missing observation point through all the information loss degrees, and further the filled sample of each physiological parameter is obtained; the abnormal characteristic value during children's sleep is determined based on all the missing observation points, and the filled sample is divided into the abnormal sample and the normal sample of each physiological parameter through the abnormal characteristic value; the clinical data during children's sleep is checked according to the abnormal sample and the normal sample of each physiological parameter.

[0016] It can be seen that in this application, first, various physiological parameters during a child's sleep are monitored. After obtaining the clinical parameter samples of each physiological parameter, for each physiological parameter, based on the missing duration of each clinical observation value in the clinical parameter sample corresponding to each physiological parameter, the clinical parameter sample is segmented and reconstructed to obtain the clinical reconstruction values of each clinical observation value in the clinical parameter sample, so as to more flexibly process the data characteristics in different time periods. Even if the data fluctuations show non-linearity, it can be more targeted for reconstruction to ensure that the data change pattern is appropriately retained. Subsequently, the degree of effectiveness decline (information loss degree) of each clinical reconstruction value due to missing data within a specific missing duration is determined, that is: by measuring the loss degree of missing information, the characteristics of non-linear fluctuations can be better captured, providing a reference direction for data recovery in subsequent filling. Then, through all the information loss degrees, cycle compensation is performed on the clinical data of each missing observation point, and then the filling sample of each physiological parameter is obtained, that is: the compensation frequency or cycle is adjusted according to the non-linear fluctuation characteristics to make the filling process more conform to the real data fluctuation trend. Again, after determining the abnormal conditions (abnormal characteristic values) within the entire sleep monitoring cycle of the child, based on the abnormal characteristic values, the filling sample is divided into abnormal samples and normal samples of each physiological parameter. Finally, the clinical data during the child's sleep is verified according to the abnormal samples and normal samples of each physiological parameter; In summary, the solution of this application can achieve the filling of missing values in children's clinical data in the case of non-linear data fluctuations. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a schematic flowchart of a pediatric clinical data collection method according to some embodiments of the present application; Figure 2 is a schematic flowchart of determining a filling sample according to some embodiments of the present application; Figure 3 Schematic flowchart of a storage method of a filling sample according to some embodiments of the present application; Figure 4 is a schematic structural diagram of a pediatric clinical data collection system according to some embodiments of the present application; Figure 5 is an internal structural diagram of a computer device for implementing the pediatric clinical data collection method according to some embodiments of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] In order to better understand the technical solutions in this application, the technical solutions in this application will be described in detail below in conjunction with the specification drawings and specific embodiments.

[0019] Reference Figure 1, The figure is a schematic flowchart of a pediatric clinical data collection method shown according to some embodiments of the present application. The pediatric clinical data collection method 100 mainly includes the following steps: In step 101, various physiological parameters of a child during sleep are monitored to obtain clinical parameter samples of each physiological parameter.

[0020] Specifically, during implementation, a sleep monitoring device can be used to collect clinical observation values of various physiological parameters of a child during sleep at a preset sampling period. Then, a set of all clinical observation values corresponding to each physiological parameter arranged in the order of time stamps is used as the clinical parameter sample of the corresponding physiological parameter, thereby obtaining the clinical parameter sample of each physiological parameter.

[0021] It should be noted that the physiological parameters in the present application include heart rate, respiratory rate, blood oxygen saturation, body temperature, electroencephalogram, and body movement data. Preferably, according to the comfort level and monitoring requirements of the child, multiple devices can be used in combination, such as a mattress sensor, a wearable device, and a non-contact device, as the sleep monitoring device to collect comprehensive physiological parameters of the child during sleep. For example: precise heart rate data can be obtained by detecting the electrical activity of the heart through an electrocardiogram patch or a chest strap electrocardiogram monitor; the respiratory movement of the chest and abdomen can be detected by a chest strap with a gas pressure sensor to monitor the respiratory rate; real-time blood oxygen saturation data can be provided by detecting the blood oxygen level of the fingertip or wrist through a finger clip pulse oximeter. In addition, the preset sampling period can be, for example, 5 minutes.

[0022] In step 102, for each physiological parameter, based on the missing duration of each clinical observation value in the clinical parameter sample corresponding to the physiological parameter, the clinical parameter sample is segmented and reconstructed to obtain the clinical reconstruction value of each clinical observation value in the clinical parameter sample, and then the information loss degree of each clinical reconstruction value under the corresponding missing duration is determined.

[0023] In some embodiments, segmenting and reconstructing the clinical parameter sample based on the missing duration of each clinical observation value in the clinical parameter sample corresponding to each physiological parameter to obtain the clinical reconstruction value of each clinical observation value in the clinical parameter sample can be implemented by the following steps: Obtain the missing duration of each clinical observation value in the clinical parameter sample corresponding to the physiological parameter; Set segment labels for each clinical observation value through all the missing durations; Based on the missing duration of each clinical observation value in the clinical parameter sample and the corresponding segment label, the clinical parameter sample is reconstructed to obtain the clinical reconstruction value of each clinical observation value in the clinical parameter sample.

[0024] In specific implementation, the missing duration of each clinical observation value in the clinical parameter sample corresponding to each physiological parameter can be obtained in the following manner: First, obtain all the missing observation points and corresponding timestamps in each clinical parameter sample. Then, select a clinical observation value. When the observation point corresponding to this clinical observation value is not a missing observation point, set the missing duration of this clinical observation value to 1. When the observation point corresponding to this clinical observation value is a missing observation point, obtain the sorting number of the timestamp of this clinical observation value among all the missing observation points, and set the missing duration of this clinical observation value to the product of 1 and the sorting number. Repeat the above steps to determine the missing durations of the remaining clinical observation values, so as to obtain the missing durations of each clinical observation value in the clinical parameter sample corresponding to each physiological parameter. In addition, it should be noted that when the observation point corresponding to the clinical observation value is a missing observation point, set the segmentation label of the clinical observation value to 1. When the observation point corresponding to the clinical observation value is not a missing observation point, set the segmentation label of the clinical observation value to 0. In other embodiments, other methods can also be used to determine this, which is not limited here.

[0025] It should be noted that the missing duration in this application is the mapping value of the time interval during which the observation point is missing. The larger the missing duration, the longer the time interval during which the observation point is missing. The smaller the missing duration, the shorter the time interval during which the observation point is missing. In addition, the segmentation label is a binary flag used to mark whether a clinical observation value is missing. When the segmentation label is 1, it indicates that the observation point is a missing observation point. When the segmentation label is 0, it indicates that the observation point is a valid observation point.

[0026] In specific implementation, the clinical parameter sample can be reconstructed based on the missing durations and corresponding segmentation labels of each clinical observation value in the clinical parameter sample to obtain the clinical reconstruction value of each clinical observation value in the clinical parameter sample in the following manner: First, select a clinical observation value. Then, obtain the left adjacent clinical observation value and the right adjacent clinical observation value of this clinical observation value. Subsequently, add the result of multiplying the left adjacent clinical observation value by its missing duration to the result of multiplying the right adjacent clinical observation value by its missing duration. Finally, divide the obtained result by the result of adding the missing durations of the left adjacent clinical observation value and the right adjacent clinical observation value as the clinical reconstruction value of this clinical observation value. Repeat the above steps to determine the clinical reconstruction values of the remaining clinical observation values in the clinical parameter sample.

[0027] It should be noted that the clinical reconstruction value is the value calculated for the physiological parameter at the missing observation points and valid observation points; when the physiological parameter is abnormal, not only the missing observation points are affected, but the data at the valid observation points will also be affected by fluctuations. In this application, by combining the missing duration, the clinical reconstruction value can better capture the changes of the physiological parameter in the time series, and then reconstruct all the observation points through the adjacent observed values of each observation point in the clinical parameter sample, so as to better identify the abnormal patterns in the overall data set and smooth the sudden changes to ensure more stable subsequent data analysis.

[0028] In some embodiments, the information loss degree of each clinical reconstruction value under the corresponding missing duration can be determined by the following steps: Select a clinical reconstruction value as the selected clinical reconstruction value; Obtain the missing duration corresponding to the selected clinical reconstruction value; Determine the information loss degree of the selected clinical reconstruction value affected by the missing duration according to the selected clinical reconstruction value and the missing duration; Continue to determine the information loss degree of the remaining clinical reconstruction values affected by the missing duration.

[0029] Specifically, when implementing, the information loss degree of the selected clinical reconstruction value affected by the missing duration can be determined according to the selected clinical reconstruction value and the missing duration in the following way, that is: first, determine that the missing duration is the negative exponent of the natural constant e, then, determine the result obtained by dividing the selected clinical reconstruction value, and finally, use the obtained result as the information loss degree of the selected clinical reconstruction value affected by the missing duration.

[0030] It should be noted that the information loss degree in this application reflects the degree of decrease in the effectiveness of the clinical reconstruction value due to missing data within a specific missing duration. The greater the information loss degree, the higher the decrease in the effectiveness of the clinical reconstruction value due to missing data within a specific missing duration, and the smaller the information loss degree, the lower the decrease in the effectiveness of the clinical reconstruction value due to missing data within a specific missing duration.

[0031] In step 103, determine multiple missing observation points among all the clinical reconstruction values, and perform periodic compensation on the clinical data of each missing observation point through all the information loss degrees, so as to obtain the filled sample of each physiological parameter.

[0032] As a preferred embodiment, when the segmentation label corresponding to the clinical reconstruction value is 1, the time stamp corresponding to the clinical reconstruction value is used as a missing observation point. When the segmentation label corresponding to the clinical reconstruction value is 0, no processing is performed. In other embodiments, other methods can also be used to determine, which are not limited here.

[0033] In some embodiments, with reference to Figure 2 As shown, the figure is a schematic flowchart of determining the filling samples shown in some embodiments of the present application. By using all the information loss degrees to perform cyclic compensation on the clinical data of each missing observation point, and then obtaining the filling samples of each physiological parameter can be achieved by the following steps: First, in 1031, determine the cyclic pattern characteristics of all clinical reconstruction values; Then, in 1032, compensate the clinical data of each missing observation point by using the cyclic pattern characteristics and all the information loss degrees to obtain the clinical filling values of each missing observation point; Finally, in 1033, construct the filling samples of each physiological parameter based on all the clinical filling values.

[0034] When specifically implemented, determining the cyclic pattern characteristics of all clinical reconstruction values can be achieved in the following manner, that is: First, take all clinical reconstruction values as inputs, and use the discrete Fourier transform in the prior art to convert all clinical reconstruction values from the time domain to the frequency domain, and then obtain the spectrogram (amplitude information at different frequencies) of all clinical reconstruction values. Then, obtain the frequency with the largest amplitude from the spectrogram as the cyclic pattern characteristics of all clinical reconstruction values. In other embodiments, other methods can also be used to determine, which are not limited here.

[0035] It should be noted that the cyclic pattern characteristics described in the present application reflect the degree of cyclic change of all clinical reconstruction values. The larger the cyclic pattern characteristics, the more significant the cyclic change of all clinical reconstruction values, that is: the clinical reconstruction values of physiological parameters during the observation period repeat or fluctuate regularly at a higher frequency. The smaller the cyclic pattern characteristics, the less significant the cyclic change of all clinical reconstruction values, that is: it indicates that the changes of the clinical reconstruction values of physiological parameters during the observation period are irregular or the fluctuation frequency is low.

[0036] Among them, in some embodiments, compensating the clinical data of each missing observation point by using the cyclic pattern characteristics and all the information loss degrees to obtain the clinical filling values of each missing observation point can be achieved by the following steps: Select a missing observation point as the selected missing observation point; Determine the fading ratio of the selected missing observation point by using all the information loss degrees; Determine the clinical filling value of the selected missing observation point based on the fading ratio, the cyclic pattern characteristics, and the clinical reconstruction value corresponding to the selected missing observation point; Continue to determine the clinical filling values of the remaining missing observation points.

[0037] It should be noted that the fading ratio described in this application reflects the proportion of the selected missing observation point in the overall information loss. The larger the fading ratio, the greater the proportion of the selected missing observation point in the overall information loss; the smaller the fading ratio, the smaller the proportion of the selected missing observation point in the overall information loss. Specifically, when implemented, the fading ratio of the selected missing observation point can be determined through all information loss degrees in the following manner: that is, the result of dividing the information loss degree of the clinical reconstruction value corresponding to the selected missing observation point by the sum of all information loss degrees is used as the fading ratio of the selected missing observation point. In other embodiments, other methods can also be used to determine it, which is not limited here.

[0038] It should be noted that in this application, based on the periodic pattern features, the final clinical filling value is adjusted by combining the fading degree of the missing observation point, so as to ensure that the compensation result not only conforms to the periodic change of the data, but also can reflect the influence degree of the missing observation point by the missing duration. As a preferred embodiment, the clinical filling value of the selected missing observation point can be determined based on the fading ratio and the clinical reconstruction value corresponding to the selected missing observation point in the following manner: that is, multiply the result of subtracting the clinical reconstruction value from the periodic pattern feature by the fading ratio. Finally, the value obtained by adding the result to the clinical reconstruction value is used as the clinical filling value of the selected missing observation point. In addition, when specifically implemented, the filling sample of each physiological parameter can be constructed based on all clinical filling values in the following manner: that is, first, the clinical filling value of each missing observation point is used as a new clinical reconstruction value, and then, the set of all clinical reconstruction values sorted in the order of time stamps is used as the filling sample of each physiological parameter. In other embodiments, other methods can also be used to determine it, which will not be elaborated here.

[0039] In step 104, based on all missing observation points, the abnormal characteristic values during children's sleep are determined, and the filling sample is divided into abnormal samples and normal samples of each physiological parameter through the abnormal characteristic values.

[0040] In some embodiments, the abnormal characteristic values during children's sleep can be determined based on all missing observation points by the following steps: Select a missing observation point as the selected missing observation point; Obtain the corresponding missing duration of the selected missing observation point; Determine the missing characteristic value of the selected missing observation point according to the selected missing observation point and the missing duration; Continue to determine the missing characteristic values of the remaining missing observation points; Determine the abnormal factor of the physiological parameter during children's sleep according to the filling sample; Determine the abnormal characteristic values during children's sleep through all missing characteristic values and the abnormal factor.

[0041] In specific implementation, the missing feature value of the selected missing observation point can be determined according to the selected missing observation point and the missing duration by the following method: First, obtain the clinical filling value of the selected missing observation point. Then, determine the natural logarithm of the missing duration. Finally, use the result of multiplying the clinical filling value of the selected missing observation point by the natural logarithm as the missing feature value of the selected missing observation point. Additionally, as a preferred embodiment, the abnormal factor of the physiological parameters during children's sleep can be determined according to the filling sample by the following method, that is: use the variance of all clinical reconstruction values in the filling sample as the abnormal factor of the physiological parameters during children's sleep. In other embodiments, other methods can also be used to determine it, which is not limited here.

[0042] It should be noted that in this application, the missing feature value reflects the influence degree of the missing observation point in the missing time dimension. The larger the missing feature value, the higher the influence degree of the missing observation point in the missing time dimension; the smaller the missing feature value, the lower the influence degree of the missing observation point in the missing time dimension. In this application, the missing feature value is determined by combining the time span of the missing observation point (the natural logarithm of the missing duration) and the value obtained after data compensation (the clinical filling value), so as to analyze the possible influence of the missing of the missing observation point on the overall monitoring. Additionally, the abnormal factor is a measure of the abnormal situation of the physiological parameters during children's sleep, used to characterize the fluctuation degree of the physiological parameters in the filling sample. The larger the abnormal factor, the higher the fluctuation degree of the physiological parameters in the filling sample; the smaller the abnormal factor, the lower the fluctuation degree of the physiological parameters in the filling sample.

[0043] It should be noted that in this application, the abnormal feature value reflects the overall degree of the abnormal situation during the entire sleep monitoring period of children. The larger the abnormal feature value, the higher the overall degree of the abnormal situation during the entire sleep monitoring period of children; the smaller the abnormal feature value, the lower the overall degree of the abnormal situation during the entire sleep monitoring period of children. For example: when the abnormal feature value is relatively large, it indicates that there is more missing data and a longer time during the monitoring process, and at the same time, the abnormal fluctuation of the data is also larger, indicating a higher abnormal risk of the physiological parameters. On the contrary, when the abnormal feature value is relatively small, it indicates that the monitoring data is relatively complete and stable, the fluctuation of the physiological parameters is small, suggesting fewer abnormal situations during the sleep process; as a preferred embodiment, the abnormal feature value during children's sleep monitoring can be determined by all the missing feature values and the abnormal factor by the following method, that is: use the result of adding the mean value of all the missing feature values and the abnormal factor as the abnormal feature value during children's sleep monitoring. In other embodiments, other methods can also be used to determine it, which is not limited here.

[0044] In some embodiments, dividing the filled samples into abnormal samples and normal samples of each physiological parameter by the abnormal eigenvalue can be achieved by the following steps: Perform abnormal judgment on the filled samples by the abnormal eigenvalue to obtain a plurality of clinical abnormal values; Take the set formed by sorting all clinical abnormal values in chronological order of timestamps as the abnormal samples of each physiological parameter; Take the set formed by sorting the other clinical reconstruction value groups in the filled samples in chronological order of timestamps as the normal samples of each physiological parameter.

[0045] In addition, in some embodiments, performing abnormal judgment on the filled samples by the abnormal eigenvalue to obtain a plurality of clinical abnormal values can be achieved by the following steps: Set an abnormal judgment threshold; Determine the relative abnormality degree of each clinical reconstruction value in the filled samples according to the abnormal eigenvalue; Perform judgment through the abnormal judgment threshold and all relative abnormality degrees to obtain a plurality of clinical abnormal values.

[0046] As a preferred embodiment, the abnormal judgment threshold can be set according to the different normal fluctuation ranges of the physiological parameter during children's sleep. For example, if the physiological parameter changes greatly during sleep (such as the short-term fluctuation of heart rate is relatively intense), a larger abnormal judgment threshold (such as 15%) is set. On the contrary, when the physiological parameter is relatively stable during children's sleep (such as blood oxygen saturation), a smaller abnormal judgment threshold (such as 10%) can be set. Specifically, when implementing, determining the relative abnormality degree of each clinical reconstruction value in the filled samples according to the abnormal eigenvalue can be achieved in the following way, that is: First, select a clinical reconstruction value in the filled samples as the selected clinical reconstruction value, determine the absolute value of the difference between the selected clinical reconstruction value and the abnormal eigenvalue, and then divide the obtained result by the abnormal eigenvalue as the relative abnormality degree of the selected clinical reconstruction value, and continue to determine the relative abnormality degrees of the remaining clinical reconstruction values. In addition, when implementing specifically, performing judgment through the abnormal judgment threshold and all relative abnormality degrees to obtain a plurality of clinical abnormal values can be achieved in the following way, that is: Select a relative abnormality degree. When this relative abnormality degree is greater than or equal to the abnormal judgment threshold, the clinical reconstruction value corresponding to this relative abnormality degree is used as a clinical abnormal value. When this relative abnormality degree is less than the abnormal judgment threshold, no processing is performed. In other embodiments, other methods can also be used to determine, which are not limited here.

[0047] In step 105, check the clinical data during children's sleep based on the abnormal samples and normal samples of each physiological parameter.

[0048] In some embodiments, the verification of clinical data during children's sleep against abnormal samples and normal samples of each physiological parameter can be achieved through the following steps: Store the abnormal samples of each physiological parameter in the abnormal storage area of the children's sleep monitoring center; Store the normal samples of each physiological parameter in the normal storage area of the children's sleep monitoring center; Verify the clinical data during children's sleep through the abnormal samples in the abnormal storage area and the normal samples in the normal storage area.

[0049] It should be noted that after obtaining the abnormal samples and normal samples of each physiological parameter, in this embodiment, in order to facilitate doctors to analyze only the abnormal samples or use the normal samples for comparison to determine whether the clinical samples are abnormal when analyzing the clinical parameter samples of children's sleep, the present invention constructs two memory systems: one is an abnormal storage area for storing abnormal samples, and the other is a normal storage area for storing normal samples. Then, all clinical reconstruction values in the abnormal samples are stored in the abnormal storage area, and all clinical reconstruction values in the normal samples are stored in the normal storage area. Refer to Figure 3 shown in the figure, which is a schematic flowchart of the storage method of filling samples shown in some embodiments of the present application; when the big data analysis system needs to perform sleep data analysis, it is necessary to verify the clinical data during children's sleep. The abnormal samples in the abnormal storage area or the normal samples in the normal storage area can be called separately to compare with the clinical data during children's sleep to verify whether the children's clinical data is abnormal or normal, thereby helping doctors more accurately judge the situation of children's sleep disorders.

[0050] In addition, on the other hand of the present application, in some embodiments, the present application provides a pediatric clinical data acquisition system. Refer to Figure 4 , which is a schematic structural diagram of the pediatric clinical data acquisition system shown in some embodiments of the present application. The pediatric clinical data acquisition system 200 includes: The monitoring module 201, the processing module 202, and the execution module 203 are described as follows: Monitoring module 201. In the present application, the monitoring module 201 is mainly used to monitor various physiological parameters during children's sleep to obtain clinical parameter samples of each physiological parameter; Processing module 202. In the present application, the processing module 202 is mainly used for each physiological parameter to segment and reconstruct the clinical parameter sample based on the missing duration of each clinical observation value in the clinical parameter sample corresponding to each physiological parameter, to obtain the clinical reconstruction value of each clinical observation value in the clinical parameter sample, and further determine the information loss degree of each clinical reconstruction value under the corresponding missing duration; In addition, the processing module 202 in the present application is further configured to determine multiple missing observation points among all clinical reconstruction values, perform periodic compensation on the clinical data of each missing observation point through all information loss degrees, and then obtain the filled samples of each physiological parameter; In addition, the processing module 202 in the present application is further configured to determine abnormal characteristic values during children's sleep based on all missing observation points, and divide the filled samples into abnormal samples and normal samples of each physiological parameter through the abnormal characteristic values; The execution module 203. In the present application, the execution module 203 is mainly configured to check the clinical data during children's sleep according to the abnormal samples and normal samples of each physiological parameter.

[0051] In addition, the present application further provides a computer device, which includes a memory and a processor. The memory stores code, and the processor is configured to obtain the code and execute the above-mentioned pediatric clinical data acquisition method.

[0052] In some embodiments, refer to Figure 5 , this figure is an internal structural diagram of a computer device for implementing the pediatric clinical data acquisition method according to some embodiments of the present application. The pediatric clinical data acquisition method in the above embodiments can be implemented by Figure 5 The computer device shown. The computer device 300 includes at least one processor 301, a communication bus 302, a memory 303, and at least one communication interface 304.

[0053] The processor 301 can be a general-purpose central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more for controlling the execution of the pediatric clinical data acquisition method in the present application.

[0054] The communication bus 302 is used to transmit information between the above components.

[0055] The memory 303 may be a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM), or other type of dynamic storage device that can store information and instructions. It may also be an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM), or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), a magnetic disk, or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 303 may exist independently and be connected to the processor 301 through the communication bus 302. The memory 303 may also be integrated with the processor 301.

[0056] Among them, the memory 303 is used to store the program code for executing the solution of this application and is controlled by the processor 301 for execution. The processor 301 is used to execute the program code stored in the memory 303. The program code may include one or more software modules. The pediatric clinical data collection method in the above embodiments may be implemented through one or more software modules in the program code in the processor 301 and the memory 303.

[0057] The communication interface 304 uses any device such as a transceiver to communicate with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0058] In a specific implementation, as an embodiment, the computer device may include multiple processors, and each of these processors may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. Here, the processor may refer to one or more devices, circuits, and / or processing cores for processing data (such as computer program instructions).

[0059] The above computer device can be a general-purpose computer device or a special-purpose computer device. In specific implementations, the computer device can be a desktop computer, a laptop computer, a network server, a personal digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. The embodiments of the present application do not limit the type of the computer device.

[0060] In addition, the present application also provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the above pediatric clinical data acquisition method is implemented.

[0061] In summary, in the pediatric clinical data acquisition method and system disclosed in the embodiments of the present application, by monitoring various physiological parameters during children's sleep, clinical parameter samples of each physiological parameter are obtained; for each physiological parameter, the clinical parameter sample is segmented and reconstructed based on the missing duration of each clinical observation value in the clinical parameter sample corresponding to the physiological parameter, so as to obtain the clinical reconstruction value of each clinical observation value in the clinical parameter sample, and then the information loss degree of each clinical reconstruction value under the corresponding missing duration is determined; multiple missing observation points among all clinical reconstruction values are determined, and the clinical data of each missing observation point is cyclically compensated through all information loss degrees, so as to obtain the filled sample of each physiological parameter; based on all missing observation points, abnormal characteristic values during children's sleep are determined, and the filled sample is divided into abnormal samples and normal samples of each physiological parameter through the abnormal characteristic values; according to the abnormal samples and normal samples of each physiological parameter, the clinical data during children's sleep is checked, and the missing value filling of children's clinical data can be realized under the condition that the data fluctuation is non-linear.

[0062] Although the preferred embodiments of the present application have been described, those skilled in the art can make additional changes and modifications once they know the basic creative concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0063] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application also intends to include these changes and modifications.

Claims

1. A pediatric clinical data collection method, characterized in that: The steps include: Monitor various physiological parameters of children during sleep and obtain clinical parameter samples of each physiological parameter; For each physiological parameter, based on the missing duration of each clinical observation value in the clinical parameter sample corresponding to each physiological parameter, the clinical parameter sample is segmentedly reconstructed to obtain a clinical reconstructed value of each clinical observation value in the clinical parameter sample, and then the information loss degree of each clinical reconstructed value under the corresponding missing duration is determined; Determine multiple missing observation points in all clinical reconstruction values, and periodically compensate clinical data of each missing observation point through all information loss degrees, thereby obtaining a filling sample for each physiological parameter; Determine the abnormal characteristic values ​​of the child's sleep based on all missing observation points, and divide the filled samples into abnormal samples and normal samples of each physiological parameter according to the abnormal characteristic values; The clinical data of children during sleep were collated based on abnormal and normal samples of each physiological parameter.

2. The method according to claim 1, characterized in that The clinical parameter samples are segmented and reconstructed based on the missing duration of each clinical observation value in the clinical parameter samples corresponding to each physiological parameter, and the clinical reconstructed values ​​of each clinical observation value in the clinical parameter samples are obtained, specifically including: Obtaining the missing duration of each clinical observation value in the clinical parameter sample corresponding to each physiological parameter; Set segment labels for each clinical observation by all missing durations; The clinical parameter sample is reconstructed based on the missing duration of each clinical observation value in the clinical parameter sample and the corresponding segmentation label to obtain a clinically reconstructed value of each clinical observation value in the clinical parameter sample.

3. The method according to claim 1, characterized in that Determining the information loss of each clinical reconstruction value under the corresponding missing time specifically includes: selecting a clinical reconstruction value as a selected clinical reconstruction value; Obtain the missing duration corresponding to the selected clinical reconstruction value; Determining, according to the selected clinical reconstruction value and the missing duration, the information loss degree of the selected clinical reconstruction value after being affected by the missing duration; The information loss degree of the remaining clinical reconstruction values ​​affected by the missing time is then determined.

4. The method according to claim 1, characterized in that The clinical data of each missing observation point is periodically compensated by all information loss degrees, and the filling samples of each physiological parameter are obtained, which specifically include: Determine the periodic pattern characteristics of all clinical reconstruction values; Compensating the clinical data of each missing observation point by using the periodic pattern characteristics and all information loss degrees to obtain a clinical filling value for each missing observation point; An imputed sample for each physiological parameter was constructed based on all clinical imputed values.

5. The method according to claim 1, characterized in that The abnormal characteristic values ​​of children's sleep determined based on all missing observation points include: Select a missing observation point as the selected missing observation point; Get the corresponding missing duration of the selected missing observation point; Determine the missing characteristic value of the selected missing observation point according to the selected missing observation point and the missing duration; Continue to determine the missing feature values ​​of the remaining missing observations; Determining abnormal factors of physiological parameters of children during sleep based on the imputed samples; The abnormal feature values ​​of children's sleep are determined by all the missing feature values ​​and the abnormal factors.

6. The method according to claim 1, characterized in that Classifying the filled samples into abnormal samples and normal samples of each physiological parameter according to the abnormal feature value specifically includes: Performing abnormality judgment on the filled sample according to the abnormal characteristic value to obtain multiple clinical abnormal values; A set of all clinical abnormal values ​​sorted in the order of timestamps is used as an abnormal sample of each physiological parameter; The set formed by sorting the other clinical reconstruction value groups in the filling sample in the order of the timestamps is used as the normal sample of each physiological parameter.

7. The method according to claim 1, characterized in that The clinical data of children's sleep are checked based on abnormal samples and normal samples of each physiological parameter, including: The abnormal samples of each physiological parameter are stored in the abnormal storage area of ​​the children's sleep monitoring center; Store the normal samples of each physiological parameter in the normal storage area of ​​the Children's Sleep Monitoring Center; The clinical data of the child during sleep is checked by using the abnormal samples in the abnormal storage area and the normal samples in the normal storage area.

8. A pediatric clinical data acquisition system, characterized in that: include: A monitoring module is used to monitor various physiological parameters of children during sleep and obtain clinical parameter samples of each physiological parameter; A processing module, for reconstructing the clinical parameter samples in segments based on the missing duration of each clinical observation value in the clinical parameter samples corresponding to each physiological parameter, to obtain a clinically reconstructed value of each clinical observation value in the clinical parameter samples, and further determining the information loss degree of each clinically reconstructed value under the corresponding missing duration; The processing module is further used to determine multiple missing observation points in all clinical reconstruction values, and periodically compensate clinical data of each missing observation point through all information loss degrees, thereby obtaining a filling sample for each physiological parameter; The processing module is further used to determine abnormal characteristic values ​​of the child's sleep based on all missing observation points, and divide the filled samples into abnormal samples and normal samples of each physiological parameter according to the abnormal characteristic values; The execution module is used to check the clinical data of children during sleep based on the abnormal samples and normal samples of each physiological parameter.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the pediatric clinical data collection method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the pediatric clinical data collection method according to any one of claims 1 to 7 are implemented.

Images