A smart allocation method for hospital wards based on the Internet of Things

Abstract

The present invention provides a smart allocation method for hospital wards based on the Internet of Things, which belongs to the field of smart medical technology. The method includes analyzing the flow of wards based on ward types, and determining the monitoring needs and selecting corresponding sensing devices according to the ward flow, and then determining the layout of the sensing devices; analyzing the data collected by the sensing devices in real time, grading the patient's condition, and evaluating the ward resource utilization rate to obtain a resource evaluation result, and then determining the use of a hospital bed; judging the patient's recovery condition based on the patient's medical data and the corresponding medical time period sensing device data, and evaluating the patient's discharge standard based on the rehabilitation judgment result, thereby realizing smart allocation of wards, improving the utilization efficiency of medical resources, improving the efficiency of patient condition monitoring and treatment, reducing the waste of medical resources, and at the same time ensuring data privacy and security, and promoting technology integration and data sharing.

Description

Technical Field

[0001] The present invention relates to the field of smart medical technology, and in particular to a smart allocation method for hospital wards based on the Internet of Things. Background Art

[0002] In recent years, with the development of technologies such as the Internet of Things, artificial intelligence, and big data analysis, intelligent medical systems have been widely used and researched in medical management, diagnosis and treatment decision-making, and patient monitoring. At present, the technologies mainly include data analysis and artificial intelligence, sensor technology, and the Internet of Things. If a large amount of medical data is transmitted and stored, there may be risks of data leakage and privacy infringement. At the same time, there may be obstacles to the interoperability between different systems and devices, and there is a problem of waste of resources.

[0003] Therefore, the present invention provides a hospital ward intelligent allocation method based on the Internet of Things. Summary of the invention

[0004] The present invention provides an intelligent allocation method for hospital wards based on the Internet of Things. The method collects data in the ward through sensor equipment, analyzes the data in real time and evaluates the patient's condition, resource utilization and recovery status, makes intelligent decisions based on medical data, optimizes ward management and bed allocation, improves the efficiency of medical resource utilization, improves the efficiency of patient condition monitoring and treatment, reduces the waste of medical resources, ensures data privacy and security, and promotes technology integration and data sharing.

[0005] The present invention provides a hospital ward intelligent allocation method based on the Internet of Things, comprising:

[0006] Step 1: Analyze the ward flow situation based on the ward type, and select corresponding sensing devices according to the monitoring needs according to the ward flow situation, and then determine the layout of the sensing devices;

[0007] Step 2: Analyze the data collected by the sensing device in real time, classify the patient's condition, and evaluate the utilization rate of ward resources to obtain resource evaluation results, and then determine the use of beds;

[0008] Step 3: Determine the patient's recovery status based on the patient's medical data and the data from the corresponding medical time period sensing device. Evaluate the patient's discharge criteria based on the recovery judgment results, and then realize the intelligent allocation of wards.

[0009] The present invention provides a hospital ward intelligent allocation method based on the Internet of Things, which analyzes ward flow conditions based on ward types, including:

[0010] Classify different types of wards by function and collect patient flow data in each type of ward over a period of time;

[0011] Analyze the transfer paths of patients in various types of wards, mark them according to the number of times they flow between different wards, and sort them according to the number of marks to determine the first-stage transfer situation of various types of wards within a period of time;

[0012] Based on the distribution of patients’ stay time in each type of ward, additional time distribution information was added to obtain the second stage transfer situation;

[0013] Taking into account the patient admission situation and the frequency of ward use, determine the busyness of each type of ward;

[0014] Determine the patient flow pattern in each type of ward based on the second-stage transfer situation and busyness;

[0015] According to the distribution of the number of patients in each type of ward, the load of each type of ward is evaluated, and the ward flow situation is obtained by combining the load situation and flow pattern.

[0016] The present invention provides a hospital ward intelligent allocation method based on the Internet of Things, which is characterized in that the monitoring demand is determined according to the flow situation of the ward, and the corresponding sensing device is selected, and then the sensing device layout is determined, including:

[0017] According to the flow of wards and seasonal disease patterns, determine the monitoring needs of various types of wards, and then determine the necessary sensing equipment for different types of wards. At the same time, determine the commonly used sensing equipment according to different ward types;

[0018] Performing an AND-NOT operation on the data set corresponding to the necessary sensing device and the data set corresponding to the common sensing device, comparing the remaining parts of the data set belonging to the necessary sensing device with the remaining parts of the data set belonging to the common sensing device one by one according to the AND-NOT operation results, obtaining a device comparison result, and judging the matching degree and coverage of the necessary sensing device and the common sensing device in combination with the device comparison result, and obtaining a device judgment result;

[0019] According to the equipment judgment result and the equipment comparison result, whether the commonly used sensing equipment meets the monitoring requirements of the corresponding type of ward is evaluated, and the demand evaluation result is obtained, and the update requirements are obtained based on the demand evaluation result and the monitoring requirements, and then the equipment update requirements are obtained;

[0020] Determine the layout plan of the sensing devices based on the update equipment requirements.

[0021] The present invention provides a hospital ward intelligent allocation method based on the Internet of Things, which is characterized by analyzing the data collected by the sensing device in real time, grading the patient's condition, and evaluating the ward resource utilization rate to obtain the resource evaluation result, and then determining the use of the hospital bed, including:

[0022] According to the layout plan, sensing devices are set up to collect real-time data of corresponding wards, and the real-time data is classified to obtain classification results, wherein the classification results include patient-related data, family-related data, ward resource utilization data, and sensing device status data;

[0023] Based on the patient-related data, historical data is acquired, and the patient's condition and condition trend are predicted in combination with a deep learning model, and the patient's condition is graded based on the prediction result to obtain a condition grading result;

[0024] Taking the ward resource utilization data and the sensing device status data as additional conditions, the available beds are obtained according to the additional conditions, and the bed demand and the corresponding resource allocation are determined in combination with the disease classification result, and then the beds are first allocated to obtain a first allocation result;

[0025] Analyze family-related data, classify the family members' accompanying care, and make a second allocation of beds based on the accompanying care classification results to obtain a second allocation result;

[0026] The ward resource utilization rate is evaluated according to the intersection of the first allocation result and the second allocation result, and the occupied bed of the corresponding patient is determined according to the resource evaluation result.

[0027] The present invention provides a hospital ward intelligent allocation method based on the Internet of Things, which is characterized in that the ward resource utilization rate is evaluated according to the intersection of a first allocation result and a second allocation result, and the use bed of the corresponding patient is determined according to the resource evaluation result, including:

[0028] Perform intersection analysis on the first allocation result and the second allocation result of each time period, and determine the hospital beds to be confirmed for use according to the intersection analysis result;

[0029] Based on the beds to be confirmed for use, calculating the resource utilization rate of the ward, and evaluating the resource utilization of the ward according to the calculation result of the resource utilization rate to obtain a resource evaluation result;

[0030] The resource evaluation result and the resource utilization rate are judged to determine the bed status of the patient's pending confirmed bed, and then determine the patient's bed in use.

[0031] The present invention provides a hospital ward intelligent allocation method based on the Internet of Things, which is characterized by judging the resource evaluation result and the resource utilization rate, determining the bed status of the corresponding patient's bed to be confirmed for use, and then determining the patient's bed for use, including:

[0032] If the resource assessment results show that the resource utilization rate is fully utilized and all resources in the ward are in an efficient state, the beds to be confirmed for use will be determined as used beds;

[0033] If the resource assessment result shows that the resource utilization rate is partial utilization and some resources are idle, the actual allocation of various resources in the ward is verified, the reasons for the idleness of some resources are determined, and the cause determination result is obtained. Based on the cause determination result, the use determination of the beds to be confirmed is performed to obtain the use determination result, and the use of the beds is determined based on the use determination result;

[0034] If the resource assessment results show that resource utilization is inefficient and there are idle resources, the ward resource utilization will be re-evaluated for the corresponding patients' beds to be confirmed, and the beds to be confirmed with idle resources will be marked as unexpected. If the number of unexpected marks exceeds the preset number, the resource assessment results with unexpected marks will be corrected through on-site inspection.

[0035] The present invention provides a hospital ward intelligent allocation method based on the Internet of Things, which is characterized by judging the patient's recovery status according to the patient's medical data and the corresponding medical time period sensing device data, and evaluating the patient's discharge standard according to the recovery judgment result, thereby realizing the intelligent allocation of wards, including:

[0036] Collect the patient's medical data and determine the characteristics of the disease by combining the data of the corresponding medical period sensing equipment;

[0037] Conduct correlation analysis between disease characteristics to obtain correlations between different disease characteristics, set a preset threshold of correlation based on medical expertise, and screen out disease characteristics whose correlation with recovery and disease stability is higher than the preset threshold;

[0038] Use a machine learning model to evaluate the importance of the disease improvement features to obtain important evaluation results, and select the disease improvement features that have the greatest impact on the recovery and disease stabilization according to the important evaluation results to obtain multiple selected features;

[0039] The selected features are processed for dimensionality reduction, and a rehabilitation model is constructed based on the dimensionality reduction results combined with the support vector machine algorithm to output the predicted rehabilitation results. Furthermore, it is determined whether the patient has met the discharge criteria based on the predicted rehabilitation results, thereby realizing the intelligent allocation of wards.

[0040] The present invention provides a hospital ward intelligent allocation method based on the Internet of Things, which constructs a rehabilitation model based on the dimensionality reduction processing result combined with the support vector machine algorithm, and outputs the predicted rehabilitation results, including:

[0041] Obtaining standard features corresponding to selected features from the improvement-standard table, performing dimensionality reduction processing on the standard features and the selected features, and obtaining dimensionality reduction processing results;

[0042] A rehabilitation model is constructed according to the result of the dimensionality reduction process. The kernel function corresponding to the support vector machine algorithm corresponding to the rehabilitation model is:

[0043] Where K(x, x′) represents the kernel function for evaluating the similarity between the standard feature and the selected feature in the feature space; x represents the selected feature; x′ represents the standard feature; i represents the i-th selected feature; N represents a total of N selected features; α i represents the Lagrange multiplier corresponding to the i-th selected feature; γ i represents the decay rate corresponding to the i-th selected feature; (f(x) j -f(x′) j ) 2 represents the Euclidean distance between the selected feature and the standard feature in the jth mapping in the feature space; f(x) j represents the jth mapping of the selected feature in the feature space; f(x′) j represents the j-th mapping of the standard feature in the feature space; D represents a total of D mappings; j represents the j-th mapping in the feature space.

[0044] Compared with the existing technology, the beneficial effects of this application are as follows: collecting data in the ward through sensor equipment, analyzing data in real time and evaluating the patient's condition, resource utilization and recovery status, making intelligent decisions based on medical data, optimizing ward management and bed allocation, improving the efficiency of medical resource utilization, improving the efficiency of patient condition monitoring and treatment, reducing the waste of medical resources, while ensuring data privacy and security, and promoting technology integration and data sharing.

[0045] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0046] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0048] Figure 1 It is a flow chart of a hospital ward intelligent allocation method based on the Internet of Things provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0049] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0050] Embodiment 1:

[0051] The embodiment of the present invention provides a hospital ward intelligent allocation method based on the Internet of Things, such as Figure 1 As shown, including:

[0052] Step 1: Analyze the ward flow situation based on the ward type, and select corresponding sensing devices according to the monitoring needs according to the ward flow situation, and then determine the layout of the sensing devices;

[0053] Step 2: Analyze the data collected by the sensing device in real time, classify the patient's condition, and evaluate the utilization rate of ward resources to obtain resource evaluation results, and then determine the use of beds;

[0054] Step 3: Determine the patient's recovery status based on the patient's medical data and the data from the corresponding medical time period sensing device. Evaluate the patient's discharge criteria based on the recovery judgment results, and then realize the intelligent allocation of wards.

[0055] In this embodiment, the ward flow situation refers to the overall situation and trend of patient flow in various types of wards after comprehensively considering the patient flow pattern, busyness and load conditions, so that hospital managers can make reasonable resource allocation and management.

[0056] In this embodiment, the monitoring needs are constantly changing according to the flow of wards and seasonal disease patterns. For example, according to the pattern of seasonal influenza, it is necessary to monitor data such as the respiratory rate and temperature changes of patients in different types of wards, as well as monitoring needs such as air quality.

[0057] In this embodiment, the sensing device layout process is to determine the monitoring needs of the ward according to the ward flow and seasonal disease patterns, select the necessary sensing device types, and evaluate the coverage and matching degree of commonly used sensing devices, thereby updating the equipment requirements and determining the layout plan of the sensing devices.

[0058] In this embodiment, the specific process of disease classification is to collect patient-related data, including physiological indicators, disease manifestations, etc., and classify and organize them; based on the patient-related data, historical data is obtained to provide a basis for training and prediction of the deep learning model, wherein the deep learning model prediction uses the deep learning model to predict the patient's current condition and disease development trend; the patient's condition is classified and rated according to the prediction results in order to determine appropriate treatment and resource allocation, etc.

[0059] In this embodiment, the process of determining the use of a hospital bed is to collect real-time data by setting up sensing devices, classifying and analyzing the data, using a deep learning model to predict the patient's condition, and combining resource utilization data and equipment status data to perform bed allocation and family care classification, and ultimately determine bed allocation and resource utilization.

[0060] In this embodiment, the resource assessment result is a conclusion drawn based on conditions such as patient condition classification, resource utilization data, and sensor device status, and is used to determine available beds and resource allocation.

[0061] In this embodiment, using a hospital bed refers to determining the required medical beds, resources, monitoring equipment, etc. according to the patient's condition and classification results.

[0062] In this embodiment, the patient's medical data and medical period sensing device data are collected to determine the characteristics of the disease, and correlation analysis is performed to screen out the characteristics of the disease improvement. The machine learning model is used to evaluate the importance of the features, and the features that have the greatest impact on the recovery are selected. Then, the support vector machine algorithm is combined to build a rehabilitation model to predict the rehabilitation results.

[0063] In this embodiment, the medical data includes the patient's physiological indicators (such as blood pressure, heart rate), clinical manifestations (such as symptoms, signs), diagnosis results (such as disease type, condition assessment), medical history, etc.

[0064] The working principle and beneficial effects of the above technical solution are: collecting data in the ward through sensor equipment, analyzing the data in real time and evaluating the patient's condition, resource utilization and recovery status, making intelligent decisions based on medical data, optimizing ward management and bed allocation, improving the efficiency of medical resource utilization, improving the efficiency of patient condition monitoring and treatment, reducing the waste of medical resources, while ensuring data privacy and security, and promoting technology integration and data sharing.

[0065] Embodiment 2:

[0066] The embodiment of the present invention provides a hospital ward intelligent allocation method based on the Internet of Things, which analyzes ward flow conditions based on ward types, including:

[0067] Classify different types of wards by function and collect patient flow data in each type of ward over a period of time;

[0068] Analyze the transfer paths of patients in various types of wards, mark them according to the number of times they flow between different wards, and sort them according to the number of marks to determine the first-stage transfer situation of various types of wards within a period of time;

[0069] Based on the distribution of patients’ stay time in each type of ward, additional time distribution information was added to obtain the second stage transfer situation;

[0070] Taking into account the patient admission situation and the frequency of ward use, determine the busyness of each type of ward;

[0071] Determine the patient flow pattern in each type of ward based on the second-stage transfer situation and busyness;

[0072] According to the distribution of the number of patients in each type of ward, the load of each type of ward is evaluated, and the ward flow situation is obtained by combining the load situation and flow pattern.

[0073] In this embodiment, classifying wards by function refers to classifying different types of wards in the hospital according to their functions or service objects, such as ICU, general ward, operating room, etc.

[0074] In this embodiment, the flow data is a record of the transfer of patients between various types of wards within a period of time, including the flow record and information of patients from one ward to another.

[0075] In this embodiment, the first stage transfer situation refers to analyzing and sorting the transfer paths and the number of marks of the patients in each type of ward to determine the first transfer situation of the patients between the various types of wards within a period of time.

[0076] In this embodiment, the transfer path refers to the path for the patient to move between different types of wards, including the travel route and sequence of the patient from one ward to another.

[0077] In this embodiment, the second stage transfer situation refers to the transfer situation of the patient between the various types of wards within a period of time, which is obtained based on the distribution and time information of the patient's stay time in the various types of wards, and further adding the time distribution information.

[0078] In this embodiment, the busyness indicates the usage frequency and patient admission status of each type of ward, and is used to evaluate the busyness or workload of the ward.

[0079] In this embodiment, the flow pattern refers to the flow pattern or pattern of patients between wards of various types determined after comprehensive consideration of the transfer situation and busyness in the second stage.

[0080] In this embodiment, the load condition refers to evaluating the load condition of the ward according to the distribution of the number of patients in each type of ward, that is, the workload or number of patients currently borne by the ward.

[0081] The working principle and beneficial effects of the above technical solution are: by collecting and analyzing patient flow data in different types of wards, determining the patient's transfer path and residence time distribution, comprehensively considering the ward busyness and flow pattern, evaluating the ward load and optimizing the flow, optimizing the patient's medical experience, improving the quality of medical services, reducing waiting time for medical treatment, and promoting intelligent management of ward flow, it helps to improve the overall operational efficiency and medical effects of medical institutions.

[0082] Embodiment 3:

[0083] The embodiment of the present invention provides a hospital ward intelligent allocation method based on the Internet of Things, which determines the monitoring needs and selects corresponding sensing devices according to the flow conditions of the wards, and then determines the layout of the sensing devices, including:

[0084] According to the flow of wards and seasonal disease patterns, determine the monitoring needs of various types of wards, and then determine the necessary sensing equipment for different types of wards. At the same time, determine the commonly used sensing equipment according to different ward types;

[0085] Performing an AND-NOT operation on the data set corresponding to the necessary sensing device and the data set corresponding to the common sensing device, comparing the remaining parts of the data set belonging to the necessary sensing device with the remaining parts of the data set belonging to the common sensing device one by one according to the AND-NOT operation results, obtaining a device comparison result, and judging the matching degree and coverage of the necessary sensing device and the common sensing device in combination with the device comparison result, and obtaining a device judgment result;

[0086] According to the equipment judgment result and the equipment comparison result, whether the commonly used sensing equipment meets the monitoring requirements of the corresponding type of ward is evaluated, and the demand evaluation result is obtained, and the update requirements are obtained based on the demand evaluation result and the monitoring requirements, and then the equipment update requirements are obtained;

[0087] Determine the layout plan of the sensing devices based on the update equipment requirements.

[0088] In this embodiment, ward flow refers to the movement and interaction of patients between different types of wards, including aspects such as patient transfer, resource utilization, and equipment monitoring requirements; seasonal disease patterns refer to changes in the incidence and epidemic patterns of different diseases in different seasons, such as the higher incidence of influenza in winter and diarrhea in summer.

[0089] In this embodiment, a distinction is made between commonly used sensing device types and necessary sensing device types. Commonly used sensing device types refer to devices that are routinely used for monitoring, while necessary sensing device types are essential devices determined based on specific monitoring requirements and may include more accurate or professional monitoring functions.

[0090] In this embodiment, the matching degree refers to the matching degree between the necessary sensing device type and the commonly used sensing device type, and the coverage range refers to the applicable scope of the device when monitoring different types of wards.

[0091] In this embodiment, the type judgment result is a conclusion drawn after comparing and evaluating the necessary sensing device type with the commonly used sensing device type, and is the result of judging whether the device meets the monitoring requirements; the demand assessment result is a conclusion drawn after comparing the device types and updating the monitoring requirements, and is an assessment of whether the currently commonly used sensing devices meet the monitoring requirements of the ward to determine whether the equipment needs to be updated or the layout plan needs to be adjusted.

[0092] The working principle and beneficial effects of the above technical solution are: according to the ward flow and seasonal disease patterns, the monitoring needs of the ward are determined, the necessary types of sensing equipment are selected, and the coverage and matching degree of commonly used sensing equipment are evaluated, so as to update the equipment requirements and determine the layout plan of the sensing equipment, improve the monitoring accuracy and efficiency, enhance the scientific nature of ward management and diagnosis and treatment decisions, reduce data leakage and privacy risks, promote interoperability between different systems, enhance data integration and sharing capabilities, and promote the application and development of intelligent medical systems in medical management and patient monitoring.

[0093] Embodiment 4:

[0094] The embodiment of the present invention provides a hospital ward intelligent allocation method based on the Internet of Things, which analyzes the data collected by the sensing device in real time, grades the patient's condition, and evaluates the ward resource utilization rate to obtain a resource evaluation result, and then determines the use of the hospital bed, including:

[0095] According to the layout plan, sensing devices are set up to collect real-time data of corresponding wards, and the real-time data is classified to obtain classification results, wherein the classification results include patient-related data, family-related data, ward resource utilization data, and sensing device status data;

[0096] Based on the patient-related data, historical data is acquired, and the patient's condition and condition trend are predicted in combination with a deep learning model, and the patient's condition is graded based on the prediction result to obtain a condition grading result;

[0097] Taking the ward resource utilization data and the sensing device status data as additional conditions, the available beds are obtained according to the additional conditions, and the bed demand and the corresponding resource allocation are determined in combination with the disease classification result, and then the beds are first allocated to obtain a first allocation result;

[0098] Analyze family-related data, classify the family members' accompanying care, and make a second allocation of beds based on the accompanying care classification results to obtain a second allocation result;

[0099] The ward resource utilization rate is evaluated according to the intersection of the first allocation result and the second allocation result, and the occupied bed of the corresponding patient is determined according to the resource evaluation result.

[0100] In this embodiment, the specific process of accompanying classification collects family-related data, including accompanying needs, accompanying time and other information; analyzes and evaluates the accompanying situation of family members, including the division and classification of accompanying levels; arranges or allocates beds according to the accompanying classification results of family members to meet the accompanying needs of family members and the needs of caring for patients.

[0101] The working principle and beneficial effects of the above technical solution are: by setting up sensing devices to collect real-time data, classifying and analyzing the data, using deep learning models to predict the patient's condition, and combining resource utilization data and equipment status data to perform bed allocation and family care grading, ultimately determining bed allocation and resource utilization, improving patient monitoring accuracy and efficiency, optimizing resource allocation and bed utilization, improving the scientific nature of diagnosis and treatment decisions, and reducing the risk of data leakage and privacy infringement.

[0102] Embodiment 5:

[0103] The embodiment of the present invention provides a hospital ward intelligent allocation method based on the Internet of Things, which evaluates the ward resource utilization rate according to the intersection of a first allocation result and a second allocation result, and determines the use bed of the corresponding patient according to the resource evaluation result, including:

[0104] Perform intersection analysis on the first allocation result and the second allocation result of each time period, and determine the hospital beds to be confirmed for use according to the intersection analysis result;

[0105] Based on the beds to be confirmed for use, calculating the resource utilization rate of the ward, and evaluating the resource utilization of the ward according to the calculation result of the resource utilization rate to obtain a resource evaluation result;

[0106] The resource evaluation result and the resource utilization rate are judged to determine the bed status of the patient's pending confirmed bed, and then determine the patient's bed in use.

[0107] In this embodiment, the specific process of intersection analysis is to intersect the first allocation result and the second allocation result: perform intersection analysis on the first allocation result and the second allocation result of each time period to determine the beds to be confirmed for use that are both first allocated and second allocated.

[0108] In this embodiment, the resource utilization calculation process is to determine the number of beds to be confirmed for use based on the intersection analysis results, and then calculate the resource utilization, compare the number of beds to be confirmed for use with the total number of beds, and obtain the resource utilization of the ward.

[0109] In this embodiment, the resource evaluation result is a conclusion obtained by evaluating and judging the resource utilization of the ward based on the calculation result of the resource utilization rate.

[0110] In this embodiment, the resource utilization rate status refers to the evaluation status of the ward resource utilization status, and the resource utilization efficiency and the ward operation status can be judged according to the resource utilization rate.

[0111] In this embodiment, the difference between a bed to be confirmed for use and an in-use bed is that a bed to be confirmed for use refers to a bed that has been confirmed as available by the first allocation and the second allocation, but has not yet been finally determined to be allocated to a specific patient; while an in-use bed is a bed that has been determined to be allocated to a specific patient, and a bed to be confirmed for use requires further final allocation decision based on resource utilization and other conditions.

[0112] The working principle and beneficial effects of the above technical solution are: by performing an intersection analysis on the first allocation result and the second allocation result, the beds that can be confirmed for use at the same time are determined, the resource utilization rate is calculated to evaluate the resource utilization, and the patient's bed is determined based on the resource evaluation results, thereby improving the reasonable allocation and utilization efficiency of ward resources, optimizing the bed allocation process, reducing resource waste, and improving medical resource management capabilities.

[0113] Embodiment 6:

[0114] The embodiment of the present invention provides a hospital ward intelligent allocation method based on the Internet of Things, which judges the resource evaluation result and the resource utilization status, determines the bed status of the patient's to-be-confirmed bed, and further determines the patient's bed usage, including:

[0115] If the resource assessment results show that the resource utilization rate is fully utilized and all resources in the ward are in an efficient state, the beds to be confirmed for use will be determined as used beds;

[0116] If the resource assessment result shows that the resource utilization rate is partial utilization and some resources are idle, the actual allocation of various resources in the ward is verified, the reasons for the idleness of some resources are determined, and the cause determination result is obtained. Based on the cause determination result, the use determination of the beds to be confirmed is performed to obtain the use determination result, and the use of the beds is determined based on the use determination result;

[0117] If the resource assessment results show that resource utilization is inefficient and there are idle resources, the ward resource utilization will be re-evaluated for the corresponding patients' beds to be confirmed, and the beds to be confirmed with idle resources will be marked as unexpected. If the number of unexpected marks exceeds the preset number, the resource assessment results with unexpected marks will be corrected through on-site inspection.

[0118] In this embodiment, it is assumed that in a hospital's ward resource assessment, the resource utilization rate is assessed as partially utilized, and some resources are idle. After verifying the allocation of various resources in the ward, the hospital management team found that some beds were idle during the day, which may be because some patients have recovered and been discharged, but the beds have not been rearranged. According to this situation, the management team decided to make a use judgment on the beds to be confirmed for use: Cause judgment: After investigation and verification, it was found that the idle beds were because some patients had recovered, but new patients were not arranged in time; Use judgment result: According to the cause judgment result of some idle resources, the management team decided to rearrange patients to these beds to improve resource utilization and ward efficiency; Determine the use of beds: According to the use judgment result, the hospital management team rearranges patients to stay in idle beds to ensure full utilization of resources and improve ward efficiency.

[0119] In this embodiment, unexpected marking means that when the resource assessment results show inefficient resource utilization and idle resources, some beds are temporarily marked or labeled when the beds to be confirmed for use by the corresponding patients are re-evaluated to indicate that these beds may have certain abnormal conditions or problems.

[0120] The working principle and beneficial effects of the above technical solution are: by processing different situations of ward resource evaluation results, including full utilization, partial utilization and inefficient utilization, the resource status in the ward is verified and the cause is judged according to the resource utilization rate, and then the use of the beds to be confirmed is judged and determined, thereby improving the full utilization and efficient status maintenance of ward resources and reducing resource idleness and waste.

[0121] Embodiment 7:

[0122] The embodiment of the present invention provides a hospital ward intelligent allocation method based on the Internet of Things, which determines the patient's recovery status according to the patient's medical data and the corresponding medical period sensing device data, and evaluates the patient's discharge criteria according to the recovery judgment result, thereby realizing intelligent allocation of wards, including:

[0123] Collect the patient's medical data and determine the characteristics of the disease by combining the data of the corresponding medical period sensing equipment;

[0124] Conduct correlation analysis between disease characteristics to obtain correlations between different disease characteristics, set a preset threshold of correlation based on medical expertise, and screen out disease characteristics whose correlation with recovery and disease stability is higher than the preset threshold;

[0125] Use a machine learning model to evaluate the importance of the disease improvement features to obtain important evaluation results, and select the disease improvement features that have the greatest impact on the recovery and disease stabilization according to the important evaluation results to obtain multiple selected features;

[0126] The selected features are processed for dimensionality reduction, and a rehabilitation model is constructed based on the dimensionality reduction results combined with the support vector machine algorithm to output the predicted rehabilitation results. Furthermore, it is determined whether the patient has met the discharge criteria based on the predicted rehabilitation results, thereby realizing the intelligent allocation of wards.

[0127] In this embodiment, the specific process of correlation analysis is correlation analysis between features, statistical analysis of the correlation between disease features, calculation of correlation coefficients or other indicators to measure the correlation between features; screening of correlation, and screening out disease improvement features that are higher than a preset threshold based on the correlation.

[0128] In this embodiment, the disease characteristics are determined based on the collected medical data and sensing device data to determine the patient's disease manifestations and characteristics, which may include disease status, symptom manifestations, physiological indicators, etc.; the disease improvement characteristics are the disease characteristics related to recovery status and disease stability that are screened out in the correlation analysis and are higher than a preset correlation threshold.

[0129] In this embodiment, the selected features are selected after important evaluation to select the features with the greatest influence for building a rehabilitation model and predicting rehabilitation results. There are multiple selected features.

[0130] In this embodiment, the important evaluation results are the importance of the features of disease improvement assessed by the machine learning model, and the important evaluation results are output for selecting the most influential features.

[0131] In this embodiment, the input of the rehabilitation model includes medical data, sensory device data, disease characteristics and selection characteristics.

[0132] The working principle and beneficial effects of the above technical solution are: by collecting patient medical data and medical time sensing device data, determining the characteristics of the disease, and performing correlation analysis to screen out the characteristics of disease improvement, using machine learning models to evaluate the importance of features, selecting the features that have the greatest impact on rehabilitation, and then combining support vector machine algorithms to build a rehabilitation model and predict rehabilitation results. According to the prediction results, it is judged whether the patient meets the discharge criteria, realizing smart ward allocation, optimizing rehabilitation models and resource utilization, reducing the waste of medical resources, and improving the quality of medical services.

[0133] Embodiment 8:

[0134] The embodiment of the present invention provides a hospital ward intelligent allocation method based on the Internet of Things, which constructs a rehabilitation model based on the dimensionality reduction processing result combined with the support vector machine algorithm, and outputs the predicted rehabilitation result, including:

[0135] Obtaining standard features corresponding to selected features from the improvement-standard table, performing dimensionality reduction processing on the standard features and the selected features, and obtaining dimensionality reduction processing results;

[0136] A rehabilitation model is constructed according to the result of the dimensionality reduction process. The kernel function corresponding to the support vector machine algorithm corresponding to the rehabilitation model is:

[0137]

[0138] The kernel function of the similarity between the standard feature and the selected feature in the feature space; x represents the selected feature; x ′ represents the standard feature; i represents the i-th selected feature; N represents a total of N selected features; α i represents the Lagrange multiplier corresponding to the i-th selected feature; γ i represents the decay rate corresponding to the i-th selected feature; (f(x) j -f(x′) j ) 2 represents the Euclidean distance between the selected feature and the standard feature in the jth mapping in the feature space; f(x) j represents the jth mapping of the selected feature in the feature space; f(x′) j represents the j-th mapping of the standard feature in the feature space; D represents a total of D mappings; j represents the j-th mapping in the feature space.

[0139] In this embodiment, the improvement-standard table is a table or data set that describes the relationship between the patient's improvement status and standard features, and is used to determine which features meet the improvement standards.

[0140] In this embodiment, the standard feature is a feature or attribute that meets the improvement standard corresponding to the selected feature, which can help distinguish which features are more important for the improvement of the disease condition.

[0141] In this embodiment, the dimensionality reduction process is to reduce the dimension of data features by some method, remove redundant information and noise, and retain the most representative and influential features to improve the efficiency and accuracy of model training and prediction.

[0142] In this embodiment, mapping refers to the process of mapping the selected features into a new feature space so as to perform dimensionality reduction processing or perform similarity calculation in different feature spaces.

[0143] The working principle and beneficial effects of the above technical solution are: by collecting patient medical data and medical time sensing device data, determining the characteristics of the disease, and performing correlation analysis to screen out the characteristics of disease improvement, using machine learning models to evaluate the importance of features, selecting the features that have the greatest impact on rehabilitation, and then combining support vector machine algorithms to build a rehabilitation model and predict rehabilitation results. According to the prediction results, it is judged whether the patient meets the discharge criteria, realizing smart ward allocation, optimizing rehabilitation models and resource utilization, reducing the waste of medical resources, and improving the quality of medical services.

[0144] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A hospital ward intelligent allocation method based on the Internet of Things, characterized in that: include: Step 1: Analyze the ward flow situation based on the ward type, and select corresponding sensing devices according to the monitoring needs according to the ward flow situation, and then determine the layout of the sensing devices; Step 2: Analyze the data collected by the sensing device in real time, classify the patient's condition, and evaluate the utilization rate of ward resources to obtain resource evaluation results, and then determine the use of beds; Step 3: Determine the patient's recovery status based on the patient's medical data and the corresponding medical time period sensing device data, and evaluate the patient's discharge criteria based on the recovery judgment results, thereby realizing intelligent allocation of wards; Among them, the ward flow situation is analyzed based on the ward type, including: Classify different types of wards by function and collect patient flow data in each type of ward; Analyze the transfer paths of patients in each type of ward, mark them according to the number of times they flow between different wards, and sort them according to the number of marks to determine the first stage of transfer of each type of ward; Based on the distribution of patients’ stay time in each type of ward, additional time distribution information was added to obtain the second stage transfer situation; Taking into account the patient admission situation and the frequency of ward use, determine the busyness of each type of ward; Determine the patient flow pattern in each type of ward based on the second-stage transfer situation and busyness; According to the distribution of the number of patients in each type of ward, the load of each type of ward is evaluated, and the flow of wards is obtained by combining the load and flow patterns; Among them, the monitoring needs are determined according to the flow conditions of the ward, and the corresponding sensing devices are selected, and then the layout of the sensing devices is determined, including: According to the flow of wards and seasonal disease patterns, determine the monitoring needs of various types of wards, and then determine the necessary sensing equipment for different types of wards. At the same time, determine the commonly used sensing equipment according to different ward types; Performing an AND-NOT operation on the data set corresponding to the necessary sensing device and the data set corresponding to the common sensing device, comparing the remaining parts of the data set belonging to the necessary sensing device and the remaining parts of the data set belonging to the common sensing device one by one according to the AND-NOT operation results, obtaining a device comparison result, and judging the matching degree and coverage of the necessary sensing device and the common sensing device in combination with the device comparison result, and obtaining a device judgment result; According to the equipment judgment result and the equipment comparison result, whether the commonly used sensing equipment meets the monitoring requirements of the corresponding type of ward is evaluated, and the demand evaluation result is obtained, and the update requirements are obtained based on the demand evaluation result and the monitoring requirements, and then the equipment update requirements are obtained; Determine the layout plan of the sensing equipment according to the requirements of the updated equipment; The data collected by the sensing equipment is analyzed in real time to grade the patient's condition and evaluate the utilization of ward resources. The resource evaluation results are obtained to determine the use of beds, including: According to the layout plan, sensing devices are set up to collect real-time data of corresponding wards, and the real-time data is classified to obtain classification results, wherein the classification results include patient-related data, family-related data, ward resource utilization data, and sensing device status data; Based on the patient-related data, historical data is acquired, and the patient's condition and condition trend are predicted in combination with a deep learning model, and the patient's condition is graded based on the prediction result to obtain a condition grading result; Taking the ward resource utilization data and the sensing device status data as additional conditions, the available beds are obtained according to the additional conditions, and the bed demand and the corresponding resource allocation are determined in combination with the disease classification result, and then the beds are first allocated to obtain a first allocation result; Analyze family-related data, classify the family members' accompanying care, and make a second allocation of beds based on the accompanying care classification results to obtain a second allocation result; Evaluate the ward resource utilization rate according to the intersection of the first allocation result and the second allocation result, and determine the hospital bed for the corresponding patient according to the resource evaluation result; Among them, the patient's recovery status is judged based on the patient's medical data and the corresponding medical period sensing device data, and the patient's discharge criteria are evaluated based on the recovery judgment results, thereby realizing the intelligent allocation of wards, including: Collect the patient's medical data and determine the characteristics of the disease by combining the data of the corresponding medical period sensing equipment; Conduct correlation analysis between disease characteristics to obtain correlations between different disease characteristics, set a preset threshold of correlation based on medical expertise, and screen out disease characteristics whose correlation with recovery and disease stability is higher than the preset threshold; Using a machine learning model to evaluate the importance of the disease improvement features, obtaining important evaluation results, and selecting disease improvement features of multiple dimensions that have the greatest impact on the recovery and disease stability according to the important evaluation results, to obtain multiple selected features; Perform dimensionality reduction processing on the selected features, and build a rehabilitation model based on the dimensionality reduction processing results combined with the support vector machine algorithm to output the predicted rehabilitation results, and then judge whether the patient meets the discharge criteria based on the predicted rehabilitation results to achieve intelligent allocation of wards; Among them, the rehabilitation model is constructed based on the dimensionality reduction processing results combined with the support vector machine algorithm, and the predicted rehabilitation results are output, including: Obtaining standard features corresponding to selected features from the improvement-standard table, performing dimensionality reduction processing on the standard features and the selected features, and obtaining dimensionality reduction processing results; A rehabilitation model is constructed according to the result of dimensionality reduction processing. The kernel function corresponding to the support vector machine algorithm corresponding to the rehabilitation model is: ,in, The kernel function that represents the similarity between the evaluation standard feature and the selected feature in the feature space; Indicates the selection feature; Indicates standard features; represents the i-th selected feature; Indicates that there are N selected features; Represents the Lagrange multiplier corresponding to the i-th selected feature; Indicates the decay rate corresponding to the i-th selected feature; Represents the Euclidean distance between the selected feature and the standard feature in the jth mapping in the feature space; Represents the jth mapping of the selected feature in the feature space; represents the jth mapping of the standard feature in the feature space; Indicates that there are D mappings in total; represents the jth mapping in the feature space.

2. According to the method of intelligent allocation of hospital wards based on the Internet of Things in claim 1, it is characterized in that: The ward resource utilization rate is evaluated according to the intersection of the first allocation result and the second allocation result, and the use bed of the corresponding patient is determined according to the resource evaluation result, including: Perform intersection analysis on the first allocation result and the second allocation result of each time period, and determine the beds to be confirmed for use according to the intersection analysis result; Calculate the resource utilization rate of the ward based on the beds to be confirmed for use, and evaluate the resource utilization of the ward according to the calculation result of the resource utilization rate to obtain a resource evaluation result; The resource evaluation result and the resource utilization rate are judged to determine the bed status of the corresponding patient's pending confirmed bed, and then determine the patient's bed in use.

3. The method for intelligent allocation of hospital wards based on the Internet of Things according to claim 2 is characterized in that: Determining the resource evaluation result and the resource utilization status, determining the bed status of the corresponding patient's pending confirmed bed, and then determining the patient's bed, including: If the resource assessment results show that the resource utilization rate is fully utilized and all resources in the ward are in an efficient state, the beds to be confirmed for use will be determined as used beds; If the resource assessment result shows that the resource utilization rate is partial utilization and some resources are idle, the actual allocation of various resources in the ward is verified, the reasons for the idleness of some resources are determined, and the cause determination result is obtained. Based on the cause determination result, the use determination of the beds to be confirmed is performed to obtain the use determination result, and the use of the beds is determined based on the use determination result; If the resource assessment results show that resource utilization is inefficient and there are idle resources, the ward resource utilization will be re-evaluated for the corresponding patients' beds to be confirmed, and the beds to be confirmed with idle resources will be marked as unexpected. If the number of unexpected marks exceeds the preset number, the resource assessment results with unexpected marks will be corrected through on-site inspection.

Images