Cloud computing-based intelligent ward patient nursing tracking and danger early warning system

Abstract

The present application relates to the technical field of healthcare informatics, in particular to a smart ward patient nursing tracking and danger early warning system based on cloud computing, which comprises a data acquisition device, a segmentation module, a importance analysis module, an attention weight analysis module and a data uploading module, by acquiring different types of physiological data and motion state data of a plurality of patients with the same disease condition in their recovery period, and different types of environmental data of the ward where the plurality of patients are located, determining the comprehensive importance of the patients to different types of physiological data, and further determining the attention weight of any patient to different types of physiological data; based on the attention weight, determining the uploading frequency of any patient to different types of physiological data and uploading data. The present application adaptively determines the uploading frequency of the patient to different types of physiological data and uploads data, effectively avoiding the delay of patient tracking and nursing, and ensuring the reliability of nursing tracking.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of healthcare informatics, in particular to a smart ward patient nursing tracking and danger early warning system based on cloud computing. BACKGROUND

[0002] With the deep integration of information technology and the medical and health field, medical informatization is gradually evolving towards intelligence. In view of the problems of low efficiency of nursing process and lagging response in the traditional ward management mode, the smart ward system realizes real-time monitoring of patient vital signs and behavior state by integrating big data and AI technology such as Internet of Things and cloud computing. Relying on intelligent analysis algorithm, the system can dynamically evaluate the patient's recovery status and give early warning to abnormal conditions, so as to transform passive response into active intervention, which can significantly improve the nursing quality and ensure patient safety.

[0003] In postoperative rehabilitation monitoring, the existing scheme collects multi-dimensional physiological data of patients through ward sensors and uploads them to the server at a fixed frequency for recovery status evaluation and danger warning. However, the network bandwidth of each node of smart hospital and smart ward is limited, and the basic system operation needs to be prioritized. In the face of massive monitoring data of patients, it is difficult to upload the multi-dimensional monitoring data of patients at a fixed frequency, which may lead to the failure of timely uploading of some important data of patients that need to be concerned, resulting in delay in tracking and nursing of patient behavior and recovery process, and further affecting the timeliness and accuracy of abnormal warning. SUMMARY

[0004] In order to solve the technical problem that uploading the monitoring data of patients at a fixed frequency may lead to the failure of timely uploading of some important data of patients that need to be concerned, the purpose of the present application is to provide a smart ward patient nursing tracking and danger early warning system based on cloud computing, and the technical solution adopted is as follows:

[0005] In the first aspect, the present application provides a smart ward patient nursing tracking and danger early warning system based on cloud computing, which comprises:

[0006] A data acquisition device is used to acquire different types of physiological data and motion state data of a plurality of patients with the same disease condition during their recovery period, and different types of environmental data of the ward where the patients are located;

[0007] A segmentation module is used to divide the recovery period of any patient into a plurality of sub-periods according to the changes of the motion state data and environmental data of the patient, and each sub-period corresponds to a type of rehabilitation activity;

[0008] a weight analysis module configured to determine a basic weight of each type of physiological data for each patient based on fluctuation of each type of physiological data in each sub-period of the patient;

[0009] a weight analysis module configured to determine a basic weight of each type of physiological data for each patient based on fluctuation of each type of physiological data in each sub-period of the patient;

[0010] a data uploading module configured to determine an uploading frequency of each type of physiological data for each patient based on the attention weight and perform data uploading.

[0011] In some possible implementation manners of the first aspect, the segmenting module comprises:

[0012] a motion state vector obtaining unit configured to construct a motion state vector of each time point based on all types of motion state data and environment data of the patient at the same time point;

[0013] a motion state sequence obtaining unit configured to construct a motion state sequence of each set time period based on all motion state vectors of the patient in the set time period;

[0014] a clustering unit configured to cluster all motion state sequences based on a metric distance of the motion state sequences of any two adjacent set time periods of the patient, to obtain a plurality of clustering clusters;

[0015] a sub-period dividing unit configured to determine a continuous time period formed by set time periods corresponding to all motion state sequences in each clustering cluster, and take the continuous time period as a sub-period of the rehabilitation period of the patient.

[0016] In some possible implementation manners of the first aspect, the clustering unit is configured to:

[0017] determine a cosine similarity between motion state vectors of two time points in the motion state sequences of any two adjacent set time periods of the patient;

[0018] determine a DTW distance of the motion state sequences of any two adjacent set time periods of the patient by using a DTW algorithm based on the cosine similarity;

[0019] take the DTW distance as a metric distance of the motion state sequences of any two adjacent set time periods of the patient.

[0020] In some possible implementation manners of the first aspect, the weight analysis module comprises:

[0021] a personal importance analysis unit configured to determine a personal importance of different types of physiological data to any patient based on fluctuation of the different types of physiological data in each of the sub-periods of different patients;

[0022] a basic importance analysis unit configured to determine a basic importance of different types of physiological data to patients based on distribution of the personal importance of different patients.

[0023] With reference to the first aspect, in some possible implementation manners, the personal importance analysis unit is configured to:

[0024] determine a difference value of the relative standard values of different types of physiological data in each of the sub-periods of any patient, and determine a recovery indicator of different types of physiological data in each of the sub-periods of any patient;

[0025] determine an importance component of different types of physiological data in each of the sub-periods of any patient based on dispersion of all the recovery indicators;

[0026] determine a personal importance of different types of physiological data to any patient based on distribution of the importance component in each of the sub-periods.

[0027] With reference to the first aspect, in some possible implementation manners, the importance component of different types of physiological data in each of the sub-periods of any patient is determined by:

[0028] determining a standard deviation of all the recovery indicators as the dispersion of all the recovery indicators;

[0029] determine a rehabilitation activity similarity in each of the sub-periods of any patient based on a metric distance between motion state sequences constituted by all types of the motion state data and the environmental data of any two adjacent setting time periods in each of the sub-periods of any patient;

[0030] determine the importance component of different types of physiological data in each of the sub-periods of any patient based on the rehabilitation activity similarity and the dispersion.

[0031] With reference to the first aspect, in some possible implementation manners, the attention weight analysis module comprises:

[0032] a recovery condition analysis unit configured to determine a difference value of the relative standard values of different types of physiological data in a rehabilitation period of any patient, and determine a comprehensive recovery indicator of different types of physiological data of any patient;

[0033] a difficulty recovery evaluation unit configured to determine a difficulty recovery degree of the patient to the different types of physiological data based on a difference of comprehensive recovery indicators of the different types of physiological data of the patient;

[0034] a focus weight determination unit configured to determine a focus weight of the patient to the different types of physiological data based on the difficulty recovery degree and the basic importance.

[0035] In some possible implementation manners of the first aspect, the determination of the difficulty recovery degree of the patient to the different types of physiological data comprises:

[0036] determining a maximum value in the comprehensive recovery indicators of the different types of physiological data of the patient, to obtain a maximum comprehensive recovery indicator;

[0037] determining a ratio of the comprehensive recovery indicators of the different types of physiological data of the patient to the maximum comprehensive recovery indicator, and performing negative correlation normalization on the ratio, to obtain the difficulty recovery degree of the patient to the different types of physiological data.

[0038] In some possible implementation manners of the first aspect, the data uploading module comprises:

[0039] a correction parameter acquisition unit configured to normalize the focus weight to a set range, to obtain an uploading frequency correction parameter of the patient to the different types of physiological data;

[0040] an uploading frequency determination unit configured to calculate a product of the uploading frequency correction parameter and a set uploading frequency, to obtain a final uploading frequency of the patient to the different types of physiological data;

[0041] a data uploading unit configured to upload the different types of physiological data of the patient in a rehabilitation period of the patient according to the final uploading frequency.

[0042] In some possible implementation manners of the first aspect, the system further comprises a danger early warning module, and the danger early warning module comprises:

[0043] an abnormality analysis unit configured to perform abnormality identification on the different types of physiological data of the patient uploaded, to obtain an abnormality indicator;

[0044] an abnormality early warning unit configured to perform recovery state abnormality early warning of the patient based on the abnormality indicator.

[0045] In the second aspect, the present application further provides a cloud computing-based intelligent ward patient nursing tracking and danger early warning method, which comprises:

[0046] acquire different types of physiological data and motion state data of a plurality of patients with the same disease during their recovery periods, and different types of environmental data of the wards where the plurality of patients are located;

[0047] divide the recovery period of any patient into a plurality of sub-periods based on changes in the motion state data and the environmental data of the patient, each of the sub-periods corresponding to a type of rehabilitation activity;

[0048] determine a basic emphasis of the patient on different types of physiological data based on fluctuations in the different types of physiological data in each of the sub-periods of different patients;

[0049] determine a concern weight of any patient on different types of physiological data based on the basic emphasis and fluctuations in the different types of physiological data of the patient during the recovery period;

[0050] determine an uploading frequency of any patient on different types of physiological data and perform data uploading based on the concern weight.

[0051] In a third aspect, the present application further provides a cloud computing-based intelligent ward patient care tracking and danger warning device, comprising a memory and a processor. The memory is used to store executable computer program codes, and the processor is used to call and run the executable computer program codes from the memory, so that the system performs the steps implemented by each module in the first aspect or any possible implementation manner of the first aspect.

[0052] In a fourth aspect, the present application further provides a computer program product, which comprises computer program codes, and when the computer program codes are run on a computer, the computer performs the steps implemented by each module in the first aspect or any possible implementation manner of the first aspect.

[0053] In a fifth aspect, the present application further provides a computer readable storage medium, which stores computer program codes, and when the computer program codes are run on a computer, the computer performs the steps implemented by each module in the first aspect or any possible implementation manner of the first aspect.

[0054] The present application has the following beneficial effects: the present application divides the rehabilitation period of any patient into several sub-periods corresponding to different rehabilitation activity types based on the change of the motion state data and the environmental data of the patient; then the basic attention degree of the patient to different types of physiological data is determined based on the fluctuation of the different types of physiological data in each sub-period of different patients, which reflects the universal recovery importance of patients with the same condition in each physiological aspect; then the attention weight of the patient to different types of physiological data is determined based on the basic attention degree and the fluctuation of different types of physiological data of the patient in the rehabilitation period, which reflects the actual recovery importance of the individual patient in each physiological aspect; finally, the uploading frequency of the patient to different types of physiological data is determined based on the attention weight, and the data is uploaded. The present application adaptively determines the attention weight of the patient to different types of physiological data, and uploads the different types of physiological data of the patient according to different uploading frequencies based on the attention weight, which effectively ensures that the important data that the patient needs to pay attention to can be uploaded in time, thereby avoiding the delay of tracking and nursing of the patient's behavior and recovery process, and accordingly ensuring the timeliness and accuracy of the patient's abnormal early warning. BRIEF DESCRIPTION OF DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, and the advantages thereof, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative labor based on these drawings.

[0056] Figure 1 Structure diagram of the cloud computing-based intelligent ward patient nursing tracking and danger early warning system of the present application embodiment;

[0057] Figure 2 Step flow chart of the cloud computing-based intelligent ward patient nursing tracking and danger early warning method of the present application embodiment;

[0058] Figure 3 Structure diagram of the segmentation module of the present application embodiment;

[0059] Figure 4 Structure diagram of the attention degree analysis module of the present application embodiment;

[0060] Figure 5 Structure diagram of the attention weight analysis module of the present application embodiment;

[0061] Figure 6 Structure diagram of the data uploading module of the present application embodiment;

[0062] Figure 7 Structure diagram of a danger warning module according to an embodiment of the present application;

[0063] Figure 8 Structure diagram of a danger warning module according to an embodiment of the present application; DETAILED DESCRIPTION

[0064] To make the technical features of the present application clear, the present application will be described in detail below with reference to the specific embodiments and in conjunction with the drawings.

[0065] Embodiments of the present application will be described below in greater detail with reference to the accompanying drawings. While certain embodiments of the present application are shown in the drawings, it is understood that the present application can be embodied in various forms and should not be construed as being limited to the embodiments set forth herein; rather, these embodiments are provided so that the present application can be more thoroughly and completely understood. It is understood that the drawings and embodiments of the present application are for exemplary purposes only and should not be construed as limiting the scope of the present application.

[0066] It should be understood that the various steps of the method embodiments of the present application can be performed in different orders and / or in parallel. In addition, the method embodiments can include additional steps and / or omit performing the steps shown. The scope of the present application is not limited in this respect.

[0067] The term "comprising" and variations thereof as used herein are used inclusively, i.e., "comprising but not limited to." The term "based on" is "based at least in part on." The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments." Related definitions will be given in the description below.

[0068] It should be noted that the terms "first", "second", and so on used in the present application are only used to distinguish different devices, modules or units, and are not intended to limit the order or interdependence of the functions performed by these devices, modules or units.

[0069] In the embodiments of the present application, although the operations or steps are described in a specific order in the accompanying drawings, it should not be construed that the operations or steps must be performed in the specific order or in a serial order, or that all of the shown operations or steps must be performed to obtain a desired result. In the embodiments of the present application, the operations or steps can be performed in series; the operations or steps can be performed in parallel; or a part of the operations or steps can be performed.

[0070] Meanwhile, it can be understood that the data (including but not limited to the data itself, acquisition or use of the data) involved in the technical solutions of the present application should comply with the requirements of the corresponding laws, regulations and relevant provisions. Unless otherwise defined, all technical and scientific terms used in the present application have the same meaning as understood by those skilled in the art to which the present application belongs, and all parameters or indicators in the formulas involved in the present application are normalized values after eliminating the dimension influence.

[0071] In order to solve the problem that uploading the monitoring data of the patient according to the fixed frequency easily causes the important data of the patient that needs to be concerned not to be uploaded in time, the embodiment of the present application provides a smart ward patient nursing tracking and danger early warning system based on cloud computing. The system is essentially a software system, which is composed of various modules realizing corresponding functions, and the corresponding structural diagram is as shown in Figure 1 The core of the system is to realize a smart ward patient nursing tracking and danger early warning method based on cloud computing. Each module in the system corresponds to each step in the method, and the corresponding flow chart of the method is as shown in Figure 2 The modules of the system will be described in detail in combination with the specific steps in the method.

[0072] The data acquisition device 100 is used to acquire different types of physiological data and motion state data of a plurality of patients with the same disease condition during their rehabilitation period, and different types of environmental data of the wards where the plurality of patients are located.

[0073] Different types of sensors are arranged in the smart ward, and the sensors are used to monitor different types of physiological data and motion state data of the patients during their rehabilitation period, and different types of environmental data of the wards where the patients are located.

[0074] In the embodiment of the present application, different types of physiological index monitoring sensors are arranged in the smart ward, including heart rate sensors, blood pressure sensors, blood oxygen detectors, electrocardiogram ECG sensors, etc. on the patients, which are used to monitor the physiological data of the patients such as heart rate, blood pressure, blood oxygen saturation (SpO2), ECG (electrocardiogram), etc. Through the motion-related sensors such as wearable gyroscopes and accelerometers, the rotation rate of the head and each joint of the patient, the body motion acceleration and other motion state data are monitored. At the same time, environmental monitoring sensors such as temperature and humidity sensors and air sensors are installed indoors to monitor the environmental data such as temperature, humidity, particulate matter concentration, etc. in the ward. The detection frequency of each monitoring sensor is preset to 1 time / second.

[0075] In the case of obtaining authorization, the medical records of different patients are obtained through the medical management system, and the patients with the same disease are classified based on the medical records, and a plurality of patients with the same disease are obtained, and the physiological data and the motion state data of the plurality of patients with the same disease in the rehabilitation period are obtained, and the environmental data of the ward where the patients are located is obtained.

[0076] In the embodiment of the present application, the medical records of different patients in the same age group are obtained through the medical management system, and then the disease information in the medical records is extracted by using the jieba word segmentation tool, and the disease word segmentation groups of different patients are obtained, and the Jaccard correlation coefficient between the disease word segmentation groups of different patients is determined, which reflects the disease similarity between different patients, so that all patients in the same age group are classified based on the Jaccard correlation coefficient by using the existing clustering algorithm (such as DBSCAN (Density-Based Spatial Clustering of Applications with Noise)), and a plurality of categories are obtained, and the patients in each category are patients with the same disease in the same category. It should be understood that other ways in the prior art can also be used to determine the patients with the same disease in the same category, which is not limited here.

[0077] The segmentation module 200 is used for dividing the rehabilitation period of any patient into a plurality of sub-sections based on the changes of the motion state data and the environmental data of the patient, and each sub-section corresponds to a rehabilitation activity type.

[0078] During the rehabilitation period, the patient will perform certain rehabilitation activities according to the arrangement of the medical order. Since there is a clear correlation between the similarity of the patient's rehabilitation activities and the consistency of the recovery process, for the same patient, when the activities in the ward are basically consistent during the rehabilitation period, the recovery of the patient will also have certain similarity. Therefore, it is necessary to compare the rehabilitation activities of the patient every day, and if the motion parameters of the patient on two days are basically similar, it means that the rehabilitation activities of the patient on the two days are basically consistent, and the recovery of the patient is basically consistent, so the time period in which the patient's recovery activities are basically consistent can be determined, and each time period corresponds to a rehabilitation activity type.

[0079] Further, as shown in Figure 3 The segmentation module 200 includes a motion state vector acquisition unit 201, a motion state sequence acquisition unit 202, a clustering unit 203 and a sub-section division unit 204, and specifically includes:

[0080] The motion state vector acquisition unit 201 is configured to construct a motion state vector at each time point based on all types of motion state data and environmental data of an arbitrary patient at the same time point.

[0081] In the embodiment, for an arbitrary patient , all types of motion state data and environmental data (the environmental data can also reflect the recovery activity of the patient) of the patient at each time point are arranged in a set order to obtain a column vector, and the column vector is taken as the motion state vector of the patient at each time point.

[0082] The motion state sequence acquisition unit 202 is configured to construct a motion state sequence of each set time period based on all motion state vectors of an arbitrary patient in each set time period during the recovery period of the patient.

[0083] In the embodiment, for an arbitrary patient , the motion state vectors of the patient at all time points in each set time period (for example, every 1 day) are arranged in time sequence to construct a motion state sequence of each set time period.

[0084] The clustering unit 203 is configured to cluster all motion state sequences based on a metric distance of the motion state sequences of an arbitrary patient in any two adjacent set time periods to obtain a plurality of clustering clusters.

[0085] A metric distance between motion state sequences of an arbitrary patient in any two adjacent set time periods (for example, adjacent 2 days) is determined, the metric distance reflects the similarity between the motion state sequences, and based on the metric distance, an existing clustering algorithm is used to cluster all motion state sequences, and similar motion state sequences are clustered into a class, so that a plurality of clustering clusters can be obtained.

[0086] The clustering unit 203 is configured to: determine a cosine similarity between the motion state vectors of two time points in the motion state sequences of an arbitrary patient in any two adjacent set time periods; determine a DTW distance of the motion state sequences of an arbitrary patient in any two adjacent set time periods by using a DTW algorithm based on the cosine similarity; and take the DTW distance as the metric distance of the motion state sequences of an arbitrary patient in any two adjacent set time periods.

[0087] In the embodiment, for an arbitrary patient , a cosine similarity between two motion state vectors is used to evaluate the similarity of the motion state of the patient at any two time points in any two adjacent set time periods (for example, the first day and the second day). In the formula,​ and They represent the first At some point in the day and the first The motion state vector at a certain moment in a day. Let represent the cosine similarity function. Therefore, based on the similarity of this motion state... Using the DTW (Dynamic Time Warping) algorithm, the patient's... In the Heaven and the Di The motion state sequence corresponding to the day and The DTW distance between them is recorded as the patient's DTW distance. In the Heaven and the Di The motion state sequence corresponding to the day and The distance metric.

[0088] Based on any patient The distance between any two adjacent set time periods is used to measure the distance between motion state sequences. Existing clustering algorithms (such as DBSCAN algorithm) are used to cluster all motion state sequences, thereby obtaining multiple clusters.

[0089] The sub-segment division unit 204 is used to determine a continuous time period consisting of a set time period corresponding to all the motion state sequences in each cluster, and to use the continuous time period as a sub-segment of the rehabilitation period of any patient.

[0090] In this embodiment of the invention, for any patient The continuous time period (daily) corresponding to all motion state sequences in the cluster is taken as a sub-segment, thereby allowing the patient to... The rehabilitation period is divided into multiple sub-segments. Thus, the numerous time periods (days) during the patient's rehabilitation period are divided into numerous sub-segments based on the similarity of the patient's daily rehabilitation activities, assuming a total of... Each sub-segment is essentially the same as the rehabilitation activities performed by the patient within that sub-segment; that is, each sub-segment corresponds to a type of rehabilitation activity.

[0091] The importance analysis module 300 is used to determine the patient's basic importance to different types of physiological data based on the fluctuation of different types of physiological data in each of the sub-segments of different patients.

[0092] Patient recovery is a relatively complex process, not only affected by patient rehabilitation activities, but also affected by internal factors such as individual physical fitness, psychological state, and external factors such as dietary nutrition structure and other multiple factors. Changes in these factors often change the patient's physical condition, thereby interfering with the established rehabilitation plan, making it difficult for the patient to recover. Therefore, according to the fluctuation of physiological data of different patients in the similar rehabilitation activities, a basic importance of physiological data during patient recovery is constructed, which reflects the universal recovery importance of patients with the same disease in each physiological aspect.

[0093] Further, as shown in Figure 4 The above-mentioned importance analysis module 300 includes a personal importance analysis unit 301 and a basic importance analysis unit 302, specifically:

[0094] The personal importance analysis unit 301 determines the personal importance of different types of physiological data of any patient based on the fluctuation of different types of physiological data in each of the sub-sections of different patients.

[0095] The personal importance analysis unit 301 is used to:

[0096] First, determine the difference value of the relative standard value of different types of physiological data in each of the sub-sections of any patient, and determine the recovery index of different types of physiological data in each of the sub-sections of any patient.

[0097] For patients with the same disease, there are similar trends in recovery speed, recovery level, etc. in the recovery stage after treatment, but due to the influence of different patient's physical differences and other factors, which leads to the patient's poor recovery ability in some aspects, which needs to be focused on, then the difference of the standard value determined by the physiological data of the patient compared with the physiological data of the majority of patients with the same disease is analyzed and compared, to evaluate the patient's recovery in each aspect, thereby determining the personal importance of different types of physiological data of any patient.

[0098] In the embodiment of the present application, for any patient In each sub-section of any type of physiological data , first use the Symbolic Aggregate Approximation (SAX) algorithm to process it to obtain the corresponding processing data change sequence of any type of physiological data of the patient . Wherein, the Symbolic Aggregate Approximation (SAX) algorithm is used to obtain the physiological data In the process of corresponding processing data change sequence, the average value of the data in one hour is preset as the fitting value of the data in the corresponding time period, and all fitting values are arranged in time sequence, so as to obtain the corresponding processing data change sequence.

[0099] Then, the DTW algorithm is used to align the physiological data of all patients with the same disease in the same sub-period The corresponding processing data change sequence is aligned, and the average value of each set of aligned values (each value corresponds to a time period) in all processing data change sequences is taken as a standard value, and all standard values are arranged in time sequence, so as to obtain the physiological data of the patient in each sub-period Corresponding ideal recovery sequence , wherein, The standard value in the ideal recovery sequence at time period t is represented.

[0100] Further, for any patient In any type of physiological data in each sub-period , the value of each time period (1 hour) in the corresponding processing data change sequence of the patient Corresponding ideal recovery sequence The absolute value of the difference between the value in the same time period and the value in the same time period in the ideal recovery sequence is calculated, and the ratio of the absolute value of the difference to the value in the same time period in the ideal recovery sequence The smaller the ratio, the closer the physiological data of the patient to the standard value, and the better the recovery. Therefore, the ratio is negatively correlated (such as calculating the reciprocal of the ratio), and the negatively correlated mapping value is taken as a recovery indicator, so that multiple recovery indicators of any type of physiological data of the patient in each sub-period can be obtained. The larger the recovery indicator, the better the recovery of the patient in this physiological aspect.

[0101] Secondly, based on the dispersion degree of all said recovery indicators, the importance component of different types of physiological data in each said sub-period of any patient is determined.

[0102] The importance degree component of different types of physiological data in each sub-section of any patient is determined, including: determining the standard deviation of all the recovery indicators as the dispersion degree of all the recovery indicators in each sub-section of any patient; determining the rehabilitation activity similarity in each sub-section of any patient based on the metric distance between the motion state sequences composed of all types of the motion state data and the environmental data of any two adjacent setting time periods in each sub-section of any patient; and determining the importance degree component of different types of physiological data in each sub-section of any patient based on the rehabilitation activity similarity and the dispersion degree.

[0103] In the embodiments of the present application, for any patient In each sub-section of any type of physiological data of any patient , the standard deviation of all the recovery indicators corresponding thereto is calculated, and the standard deviation is taken as the dispersion degree of the physiological data in the corresponding sub-section; meanwhile, based on the metric distance between the motion state sequences composed of all types of the motion state data and the environmental data of any two adjacent setting time periods (1 day) in each sub-section of any patient determined in the clustering unit 203, the metric distance is negatively correlated mapped (such as calculating the reciprocal of the metric distance), and the average value of the negatively correlated mapped values of all the metric distances is taken as the rehabilitation activity similarity in each sub-section of any patient.

[0104] When the dispersion degree of any type of physiological data in a certain sub-section of any patient is higher, and the rehabilitation activity similarity is higher, it indicates that the physiological data in the sub-section is greatly affected by factors other than rehabilitation training, is more likely to be not recovered well due to external influences, and should be paid more attention to, and the corresponding importance degree component is larger. Therefore, the product of the rehabilitation activity similarity and the dispersion degree of any type of physiological data in each sub-section of any patient is calculated, the product is normalized by using the existing norm normalization function, and the normalized value of the product is taken as the importance degree component of any type of physiological data in each sub-section of any patient. Finally, the personal importance degree of any patient to different types of physiological data is determined based on the distribution of the importance degree components in each sub-section.

[0105] The distribution of the importance degree components of any type of physiological data in all sub-sections of any patient is counted, so that the personal importance degree of any patient to different types of physiological data is obtained.

[0106] The distribution of the importance degree components of any type of physiological data in all sub-sections of any patient is counted, so that the personal importance degree of any patient to different types of physiological data is obtained.​​​ personal importance of the arbitrary patient.

[0107] In the embodiment of the present application, the average value of the importance components of the arbitrary type of physiological data in all sub-sections is determined, and the average value is taken as the personal importance of the arbitrary patient. personal importance of the arbitrary patient. . Thus, the personal importance of each patient to different types of physiological data can be determined.

[0108] The basic importance analysis unit 302 is configured to determine the basic importance of the patient to different types of physiological data based on the distribution of the personal importance of different patients.

[0109] When the personal importance of all patients with the same condition to a certain type of physiological data is greater, it means that the recovery of the patient's condition on the physiological index during the recovery period is more susceptible to other factors outside the rehabilitation training, and more attention should be paid. Thus, the basic importance of the patients with the same condition to various physiological conditions can be calculated.

[0110] In the embodiment of the present application, the average value of the personal importance of the arbitrary type of physiological data of all patients with the same condition is calculated, and the average value is taken as the basic importance of the patient to the arbitrary type of physiological data . Thus, the basic importance of the patient to different types of physiological data can be obtained.

[0111] The attention weight analysis module 400 is configured to determine the attention weight of the arbitrary patient to different types of physiological data based on the basic importance and in combination with the fluctuation of the different types of physiological data of the arbitrary patient during the rehabilitation period.

[0112] ​​​​​​​While patients with similar conditions may exhibit some similarities in postoperative recovery, the actual recovery of each patient is constrained by various factors affecting their individual health. When a patient struggles to recover in a particular area, it can hinder the return of numerous related physiological parameters to normal, thus impeding a full recovery. Therefore, based on patients' prioritization of different types of physiological data, this study analyzes the fluctuations of various physiological data points during the recovery period for any given patient. It quantifies the impact of different types of physiological data on the patient's recovery, thereby determining the weight of each patient's attention to different types of physiological data. Data with a greater impact requires more focused attention and should therefore receive a higher weight.

[0113] Furthermore, such as Figure 5 As shown, the aforementioned attention weight analysis module 400 includes a recovery status analysis unit 401, a difficult-to-recover assessment unit 402, and an attention weight determination unit 403, specifically including:

[0114] The recovery analysis unit 401 is used to determine the difference values ​​of different types of physiological data of any patient relative to the standard values ​​during their recovery period, and to determine the comprehensive recovery index of different types of physiological data of any patient.

[0115] In this embodiment of the invention, for any patient Any type of physiological data Following the method described in the individual attention analysis unit 301 above, which determines that the recovery indicators of different types of physiological data in each of the aforementioned sub-segments of any patient are the same, it is possible to determine the patient's... Physiological data throughout the recovery period Multiple recovery indicators were used to determine the average value of all recovery indicators, and this average value was used as the patient's... physiological data The comprehensive recovery indicators. Therefore, comprehensive recovery indicators for different types of physiological data in any patient can be determined.

[0116] The recovery difficulty assessment unit 402 is used to determine the degree of recovery difficulty of any patient for different types of physiological data based on the differences in comprehensive recovery indicators of different types of physiological data.

[0117] The determination of the degree of difficulty in recovery of different types of physiological data for any patient includes: determining the maximum value among the comprehensive recovery indicators of different types of physiological data for any patient to obtain the maximum comprehensive recovery indicator; determining the ratio of the comprehensive recovery indicator of different types of physiological data for any patient to the maximum comprehensive recovery indicator, and performing negative correlation normalization on the ratio to obtain the degree of difficulty in recovery of different types of physiological data for any patient.

[0118] In this embodiment of the invention, for any patient Identify the patient The maximum comprehensive recovery index is obtained by finding the maximum value among the comprehensive recovery indices of different types of physiological data. Then, the maximum comprehensive recovery index is calculated for any patient. Any type of physiological data The ratio of the comprehensive recovery index to the maximum comprehensive recovery index; the smaller the ratio, the better the physiological data. The worse the recovery, the more difficult it is to recover this type of physiological data compared to other physiological data. Therefore, this ratio is negatively correlated and normalized (e.g., by calculating the reciprocal), and the result of this negative correlation normalization is used as the physiological data. The degree of difficulty in recovery, and recorded as Therefore, the degree of difficulty in recovering different types of physiological data for any patient can be obtained.

[0119] The attention weight determination unit 403 is used to determine the attention weight of any patient for different types of physiological data based on the degree of difficulty in recovery and the basic importance.

[0120] In this embodiment of the invention, for any patient Any type of physiological data Calculate this physiological data The degree of difficulty in recovery With this type of physiological data Basic importance The product of, and use that product as the patient's... physiological data The weight of attention given to recovery is denoted as follows: Therefore, we can obtain the attention weights that any patient gives to different types of physiological data.

[0121] The data upload module 500 is used to determine the upload frequency of different types of physiological data for any patient based on the attention weight and to upload the data.

[0122] Based on the patient's weight in different types of physiological data, the upload frequency for each type of physiological data is determined. When a certain type of physiological data has a higher weight, it is uploaded to the cloud server at a higher upload frequency; otherwise, it is uploaded to the cloud server at a lower upload frequency. This ensures reliable monitoring of patient care while guaranteeing that important data that patients are likely to need to monitor is uploaded in a timely manner, avoiding delays in tracking and caring for patient behavior and recovery.

[0123] Furthermore, such as Figure 6As shown, the data upload module 500 includes a correction parameter acquisition unit 501, an upload frequency determination unit 502, and a data upload unit 503, specifically comprising:

[0124] The parameter acquisition unit 501 is used to normalize the attention weight to a set range to obtain the upload frequency correction parameter for any patient's physiological data of different types.

[0125] Because patients require a huge amount of physiological data to be monitored during recovery, while the network bandwidth in the smart ward is relatively limited, it is necessary to prioritize the transmission of more important data. That is, the update frequency of each type of physiological data should be set according to the patient's attention weight for each type of physiological data.

[0126] In this embodiment of the invention, for any patient Any type of physiological data The focus weights are normalized to a set range (e.g., [0.5, 1]), and the normalized values ​​are used as physiological data. The upload frequency correction parameter is denoted as . Since normalizing the values ​​to a set range is existing technology, it will not be elaborated here. Therefore, the upload frequency correction parameters for different types of physiological data from any patient can be obtained.

[0127] It should be understood that, in the initial case, since patients have just been admitted to the smart ward and there is a lack of relevant recovery data, the basic importance that patients with similar conditions attach to different types of physiological data is used as their attention weight, and this attention weight is normalized to a set range (e.g., [0.5,1]), thereby obtaining the correction parameter for the upload frequency of different types of physiological data by patients.

[0128] The upload frequency determination unit 502 is used to calculate the product of the upload frequency correction parameter and the set upload frequency to obtain the final upload frequency of any patient for different types of physiological data.

[0129] In this embodiment of the invention, a preset upload frequency is set. (as preset) (1 time / second), for any patient Any type of physiological data Calculate the corresponding transmission frequency correction parameters. With setting upload frequency The product of these factors, and use this product as physiological data. The final upload frequency, and recorded as Therefore, the final upload frequency of different types of physiological data for any patient can be obtained.

[0130] The data uploading unit 503 is used to upload different types of physiological data of any patient during their recovery period according to the final upload frequency.

[0131] In this embodiment of the invention, the different types of physiological data of each patient are uploaded to the cloud server according to the final upload frequency of each patient's different types of physiological data.

[0132] Once the cloud server receives the data, it identifies and issues warnings for different types of physiological data from the patient.

[0133] Furthermore, such as Figure 7 As shown, the cloud-based smart ward patient care tracking and risk warning system also includes a risk warning module 600, which comprises an anomaly analysis unit 601 and an anomaly warning unit 602, specifically:

[0134] The anomaly analysis unit 601 is used to identify anomalies in any type of physiological data from any patient uploaded to the system, and to obtain anomaly indicators.

[0135] In this embodiment of the invention, for any patient The latest physiological data of different types are arranged in a set order to obtain a real-time physiological index vector, which is denoted as... Simultaneously, the DTW algorithm was used to align the same type of physiological data from all patients with the same condition, and the average value of each aligned group of physiological data was used as a reference value. These reference values ​​were then arranged chronologically to obtain a sequence of reference values ​​for different types of physiological data. For patients... For any physiological data point in the real-time physiological indicator vector, its aligned reference value is determined within a reference value sequence of similar physiological data. These reference values ​​are then arranged in a predetermined order to obtain the reference value vector of the real-time physiological indicator vector, denoted as [reference value]. Calculate real-time physiological indicator vectors. Its reference value vector cosine similarity Furthermore, based on this cosine similarity Abnormal indicators were obtained. Among them, when the cosine similarity A larger value indicates a higher real-time physiological indicator vector. Its reference value vector The higher the similarity, the more likely the patients are to be diagnosed with COVID-19. The smaller the difference between various types of physiological data and ideal physiological data, the lower the corresponding abnormal indicators. The smaller the value, the better.

[0136] The abnormality early warning unit 602 is configured to perform recovery state abnormality early warning for any patient based on the abnormality index.

[0137] In the embodiment of the present application, for any patient , if the recovery state of the patient continuously appears abnormal in a continuous period of time (preset as 1 minute), that is, the abnormality index at the corresponding time is greater than or equal to a preset threshold (preset as 0.3), or the abnormality index of the patient continuously increases and exceeds a preset upper limit (preset as 0.5) in a short period of time (preset as 0.5 minutes), it is considered that the recovery state of the patient appears abnormal, and recovery state abnormality early warning information is sent to inform nurses and doctors to perform specific inspection on the patient.

[0138] Based on the same inventive concept, the embodiment of the present application further provides a cloud computing-based intelligent ward patient nursing tracking and danger early warning method, as shown in Figure 2 , the method comprises the following steps:

[0139] Based on the change of the motion state data and the environmental data of any patient, the rehabilitation period of the patient is divided into a plurality of sub-periods, and each sub-period corresponds to a rehabilitation activity type.

[0140] Based on the fluctuation of the physiological data of different types in each sub-period of different patients, the basic attention degree of the patient to the physiological data of different types is determined.

[0141] Based on the basic attention degree and in combination with the fluctuation of the physiological data of different types of any patient in the rehabilitation period, the attention weight of any patient to the physiological data of different types is determined.

[0142] Based on the attention weight, the uploading frequency of any patient to the physiological data of different types is determined and data uploading is performed.

[0143] Based on the same inventive concept, the embodiment of the present application further provides a cloud computing-based intelligent ward patient nursing tracking and danger early warning device, as shown in Figure 8 , the device comprises a memory 801, a processor 802, and computer program code 803 stored in the memory 801 and running on the processor 802, wherein when the processor 802 executes the computer program code 803, the system can execute the steps implemented by each module of any one of the cloud computing-based intelligent ward patient nursing tracking and danger early warning systems introduced above.

[0144] The embodiment of the present application can divide the device into functional modules according to the examples of the steps implemented by each module in the system, for example, each functional module can be obtained, or two or more functions can be integrated into one processing module, and the integrated module can be realized in the form of hardware. It should be noted that the division of the modules in the embodiment is illustrative, and is only a logical functional division, and another division mode can be used in actual implementation.

[0145] Based on the same inventive concept, the embodiment of the present application also provides a computer program product, which comprises computer program code, when the computer program code runs on a computer, so that the computer executes the steps implemented by each module of any one of the foregoing cloud computing-based smart ward patient care tracking and danger early warning systems.

[0146] Based on the same inventive concept, the embodiment of the present application also provides a computer readable storage medium, which stores computer program code, when the computer program code runs on a computer, so that the computer executes the steps implemented by each module of any one of the foregoing cloud computing-based smart ward patient care tracking and danger early warning systems.

[0147] It should be noted that: the above-mentioned embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A cloud computing-based intelligent ward patient care tracking and danger warning system, characterized in that, The system comprises: a data acquisition device configured to acquire different types of physiological data and motion state data of a plurality of patients with the same condition during their rehabilitation period, and different types of environmental data of the wards where the plurality of patients are located; a segmentation module configured to divide the rehabilitation period of any patient into a plurality of sub-periods based on changes in the motion state data and environmental data of the patient, each of the sub-periods corresponding to a type of rehabilitation activity; an importance analysis module configured to determine a basic importance of different types of physiological data for a patient based on fluctuations in the different types of physiological data in each of the sub-periods of the patient; an attention weight analysis module configured to determine an attention weight of different types of physiological data for any patient based on the basic importance and in combination with fluctuations in the different types of physiological data of the patient during the rehabilitation period; a data uploading module configured to determine an uploading frequency of different types of physiological data for any patient based on the attention weight and perform data uploading; the segmentation module comprises: a motion state vector acquisition unit configured to construct a motion state vector at each time point based on all types of motion state data and environmental data of any patient at the same time point; a motion state sequence acquisition unit configured to construct a motion state sequence of each set period based on all motion state vectors of any patient in each set period during the rehabilitation period of the patient; a clustering unit configured to cluster all motion state sequences based on a metric distance of the motion state sequences of any patient in any two adjacent set periods to obtain a plurality of clustering clusters; a sub-period division unit configured to determine a continuous time period formed by set periods corresponding to all motion state sequences in each of the clustering clusters, and take the continuous time period as a sub-period of the rehabilitation period of any patient; the attention weight analysis module comprises: a recovery condition analysis unit configured to determine a difference value of different types of physiological data relative to a standard value for any patient in the rehabilitation period of the patient, and determine a comprehensive recovery indicator of different types of physiological data for the patient; a difficult recovery evaluation unit configured to determine a difficult recovery degree of different types of physiological data for any patient based on a difference in the comprehensive recovery indicator of different types of physiological data for the patient; an attention weight determination unit configured to determine an attention weight of different types of physiological data for any patient based on the difficult recovery degree and the basic importance; the importance analysis module comprises: a personal importance degree analysis unit configured to determine a personal importance degree of different types of physiological data for any patient based on fluctuations in the different types of physiological data in each of the sub-periods of the patient; a basic importance analysis unit configured to determine a basic importance of different types of physiological data for a patient based on a distribution of the personal importance degree of the patient; 2.The cloud-computing-based smart ward patient care tracking and danger warning system according to claim 1, wherein, the clustering unit is configured to: determine a cosine similarity between the motion state vectors at two time points in the motion state sequences of any patient in any two adjacent set periods; Based on the cosine similarity, a DTW algorithm is used to determine the DTW distance of the motion state sequence of any patient in any two adjacent setting periods; The DTW distance is used as the metric distance of the motion state sequence of any patient in any two adjacent setting periods. 3.The cloud-computing-based smart ward patient care tracking and danger warning system according to claim 1, wherein, The personal importance analysis unit is used to: Determine the difference value of the relative standard value of different types of physiological data in each sub-period of any patient, and determine the recovery index of different types of physiological data in each sub-period of any patient; Based on the dispersion degree of all the recovery indexes, determine the importance component of different types of physiological data in each sub-period of any patient; Based on the distribution of the importance component in each sub-period, determine the personal importance of different types of physiological data of any patient. 4.The cloud-computing-based smart ward patient care tracking and danger warning system according to claim 3, wherein, Determine the importance component of different types of physiological data in each sub-period of any patient, including: Determine the standard deviation of all the recovery indexes as the dispersion degree of all the recovery indexes; Based on the metric distance between the motion state sequence composed of all types of motion state data and environmental data in any two adjacent setting periods in each sub-period of any patient, determine the rehabilitation activity similarity in each sub-period of any patient; Based on the rehabilitation activity similarity and the dispersion degree, determine the importance component of different types of physiological data in each sub-period of any patient. 5.The cloud-computing-based smart ward patient care tracking and danger warning system according to claim 1, wherein, Determine the difficult recovery degree of different types of physiological data of any patient, including: Determine the maximum value of the comprehensive recovery index of different types of physiological data of any patient to obtain the maximum comprehensive recovery index; Determine the ratio of the comprehensive recovery index of different types of physiological data of any patient to the maximum comprehensive recovery index, and perform negative correlation normalization on the ratio to obtain the difficult recovery degree of different types of physiological data of any patient. 6.The cloud-computing-based smart ward patient care tracking and danger pre-warning system according to claim 1, wherein, The data uploading module includes: A correction parameter acquisition unit is used to normalize the attention weight to a set range to obtain an uploading frequency correction parameter of different types of physiological data of any patient; An uploading frequency determination unit is used to calculate the product of the uploading frequency correction parameter and a set uploading frequency to obtain the final uploading frequency of different types of physiological data of any patient; A data uploading unit is used to upload different types of physiological data of any patient in its rehabilitation period according to the final uploading frequency. 7.The cloud-computing-based smart ward patient care tracking and danger pre-warning system according to claim 1, wherein, The system further includes a danger warning module, which includes: An anomaly analysis unit is used to identify the abnormality of different types of physiological data of any patient obtained by uploading to obtain an abnormality index; An abnormality warning unit is used to perform recovery state abnormality warning of any patient based on the abnormality index.

Images