Intelligent management method and device for newborn incubator

Abstract

The invention provides an intelligent management method and device for a newborn incubator, and relates to the technical field of intelligent medical treatment. The lifting assembly can be controlled to adjust the angle of the bed body of the nursing bed relative to the horizontal plane. And in the later preset duration, receiving a plurality of vital sign parameters of the newborn lying on the nursing bed, which are collected by the parameter detection module at a plurality of sampling moments. According to the deep belief DBN network, the dependency relationship of the multiple vital sign parameters on the time sequence is determined, and according to the dependency relationship, the Nth parameter change feature is obtained. And according to a preset scoring strategy, scoring the Nth parameter change feature and the Nth vital sign parameter to obtain an Nth evaluation score. And under the condition that the Nth evaluation score is lower than a set evaluation score threshold value, updating the N angle based on the Nth angle and the Nth evaluation score by using a preset reinforcement learning model. And adding 1 to the value of N, and returning to execute the operation, so that the reliability of the adjusted Nth angle of the bed body relative to the horizontal plane is high.

Description

Technical Field

[0001] This application relates to the field of intelligent medical technology, and particularly to an intelligent management method and device for a neonatal incubator. Background Art

[0002] A neonatal incubator is a medical device used to maintain the body temperature of premature or weak newborns. A nursing bed is usually arranged inside the neonatal incubator, and a newborn usually lies on the nursing bed. Among them, the angle of the nursing bed can be adjusted according to the condition of the newborn. Moreover, the angle adjustment of the nursing bed in the neonatal incubator is crucial for the comfort and health of the newborn, especially for infants with special needs in terms of breathing, digestion, and body position.

[0003] Currently, the angle adjustment method of the nursing bed in a neonatal incubator usually requires medical staff to adjust it according to the health condition of the newborn, with low reliability, poor real-time performance, and waste of human resource costs. Summary of the Invention

[0004] This application provides an intelligent management method and device for a neonatal incubator, which is used to solve the problems of low reliability, poor real-time performance, and waste of human resource costs in the angle adjustment of the nursing bed in the existing neonatal incubator.

[0005] In a first aspect, this application provides an intelligent management method for a neonatal incubator, which is applied to an intelligent management system for a neonatal incubator. The system includes an incubator body, a base is arranged inside the incubator body, a main controller is embedded in the base, a nursing bed is arranged on the base, the end of the nursing bed is fixedly connected to one side of the base, the head of the nursing bed is connected to the base in a liftable manner, an angle sensor for detecting the angle of the head of the nursing bed relative to the horizontal plane is arranged at a position corresponding to the head of the bed on the base, and a parameter detection module for detecting the vital sign parameters of the newborn is arranged on one side of the incubator body close to the head of the bed. Among them, the vital sign parameters are associated with the angle of the head of the nursing bed relative to the horizontal plane. The method provided by this application includes:

[0006] The main controller controls the lifting component to perform a lifting operation to adjust the angle of the bed body of the nursing bed relative to the horizontal plane, and receives the angle of the bed body of the nursing bed relative to the horizontal plane collected by the angle sensor. When it is determined that the received angle reaches the Nth angle, the main controller stops controlling the lifting component to perform the lifting operation, where N is a positive integer and the initial value of N is 1;

[0007] The main controller receives, within a preset time period afterwards, a plurality of vital sign parameters of the newborn lying on the nursing bed collected by the parameter detection module at a plurality of sampling times, where the plurality of vital sign parameters include the Nth vital sign parameter collected at the last sampling time among the plurality of sampling times;

[0008] Determine the dependency relationships of multiple vital sign parameters in terms of time series according to the deep belief DBN network, and obtain the change characteristics of the Nth parameter according to the dependency relationships;

[0009] The main controller scores the change characteristics of the Nth parameter and the Nth vital sign parameter according to a preset scoring strategy to obtain the Nth evaluation score;

[0010] When the Nth evaluation score is lower than the set evaluation score threshold, the main controller updates the Nth angle based on the Nth angle and the Nth evaluation score by using a preset reinforcement learning model;

[0011] The main controller increments the value of N by 1, returns to execute the control of the lifting component to perform the lifting operation to adjust the angle of the nursing bed body relative to the horizontal plane, and receives the angle of the nursing bed body relative to the horizontal plane collected by the angle sensor. When it is determined that the received angle reaches the Nth angle, stop the step of controlling the lifting component to perform the lifting operation until the Nth evaluation score reaches the set evaluation score threshold.

[0012] In some embodiments, the main controller scores the change characteristics of the Nth parameter and the Nth vital sign parameter according to a preset scoring strategy to obtain the Nth evaluation score, including:

[0013] The main controller inputs the change characteristics of the Nth parameter and the Nth vital sign parameter into a pre-trained parameter prediction model to predict the future vital sign parameters of the newborn on the nursing bed after a future time period. Among them, the parameter prediction model is trained by inputting multiple first training samples into a neural network, and each first training sample includes the historical parameter change characteristics and the actual future vital sign parameters of the newborn on the nursing bed after the corresponding historical future time period;

[0014] According to the future life characteristic parameters and a preset first mapping relationship table, determine the first sub-score corresponding to the future life characteristic parameters, where the first mapping relationship table records the mapping relationships between various different life characteristic parameters and the corresponding scores;

[0015] According to the Nth vital sign parameter and a preset first mapping relationship table, determine the second sub-score corresponding to the Nth vital sign parameter;

[0016] Perform weighted summation on the first sub-score and the second sub-score to obtain the Nth evaluation score.

[0017] In some embodiments, performing weighted summation on the first sub-score and the second sub-score to obtain the Nth evaluation score includes:

[0018] According to the formula P N = k1P1 + k2P2, determine the Nth evaluation score, where P NLet \(N\) be the \(N\)th evaluation score, \(P1\) be the first sub-score, \(P2\) be the second sub-score, \(k1\) be the first weighting coefficient, \(k2\) be the second weighting coefficient, and \(k1 + k2=1\).

[0019] In some embodiments, the \(N\)th vital sign parameter includes the \(N\)th respiratory rate, the \(N\)th intracranial pressure, and the \(N\)th heart rate. According to the \(N\)th vital sign parameter and a preset first mapping table, determining the second sub-score corresponding to the \(N\)th vital sign parameter includes:

[0020] According to the \(N\)th respiratory rate and the preset first mapping table, determining the third sub-score corresponding to the \(N\)th respiratory rate;

[0021] According to the \(N\)th intracranial pressure and the preset first mapping table, determining the fourth sub-score corresponding to the \(N\)th intracranial pressure;

[0022] According to the \(N\)th heart rate and the preset first mapping table, determining the fifth sub-score corresponding to the \(N\)th heart rate;

[0023] According to the formula \(P2 = k3P3 + k4P4 + k5P5\), determining the second sub-score, where \(P2\) is the second sub-score, \(P3\) is the third sub-score, \(P4\) is the fourth sub-score, \(P5\) is the fifth sub-score, \(k3\) is the third weighting coefficient, \(k4\) is the fourth weighting coefficient, \(k5\) is the fifth weighting coefficient, and \(k3 + k4 + k5 = 1\).

[0024] In some embodiments, the main controller scores the \(N\)th parameter change feature and the \(N\)th vital sign parameter according to a preset scoring strategy to obtain the \(N\)th evaluation score, including:

[0025] Fusing the \(N\)th parameter change feature and the \(N\)th vital sign parameter to obtain the \(N\)th vital sign feature;

[0026] Inputting the \(N\)th vital sign feature into a pre-trained second scoring model to obtain the \(N\)th evaluation score. The second scoring model is obtained by training a neural network with multiple second training samples, and each second training sample includes a historical vital sign feature and its corresponding historical actual score.

[0027] In some embodiments, the head of the nursing bed and the base are liftably connected by a lifting component. The lifting component includes a first telescopic rod and a second telescopic rod arranged oppositely, a first driving motor for driving the first telescopic rod to expand and contract, and a second driving motor for driving the second telescopic rod to expand and contract. The main controller controls the lifting component to perform a lifting operation, including:

[0028] The main controller controls the first driving motor to drive the first telescopic rod to expand and contract and controls the second driving motor to drive the second telescopic rod to expand and contract, so that the lifting component performs a lifting operation.

[0029] In some embodiments, the angle sensor is a distance sensor that receives the angle of the bed body of the nursing bed relative to the horizontal plane collected by the angle sensor, including:

[0030] Receiving the distance of the base relative to the head of the nursing bed collected by the distance sensor;

[0031] According to the distance, look up the angle of the bed body of the nursing bed relative to the horizontal plane from the preset second mapping relation table.

[0032] In a second aspect, the present application also provides a neonatal incubator intelligent management device configured in a main controller. The main controller belongs to a neonatal incubator intelligent management system. The system further includes an incubator body. A base is provided inside the incubator body. The main controller is embedded in the base. A nursing bed is provided on the base. The tail of the nursing bed is fixedly connected to one side of the base. The head of the nursing bed is liftably connected to the base. An angle sensor for detecting the angle of the head of the nursing bed relative to the horizontal plane is provided at a position corresponding to the head on the base. A parameter detection module for detecting the vital sign parameters of the neonate is provided on one side of the incubator body near the head. Among them, the vital sign parameters are associated with the angle of the head of the nursing bed relative to the horizontal plane. The device provided by the present application includes:

[0033] An angle adjustment unit for controlling the lifting component to perform a lifting operation to adjust the angle of the bed body of the nursing bed relative to the horizontal plane, and receiving the angle of the bed body of the nursing bed relative to the horizontal plane collected by the angle sensor. When it is determined that the received angle reaches the Nth angle, stop controlling the lifting component to perform the lifting operation, where N is a positive integer and the initial value of N is 1;

[0034] A parameter receiving unit for receiving, within a preset time period thereafter, multiple vital sign parameters of the neonate lying on the nursing bed collected by the parameter detection module at multiple sampling moments, where the multiple vital sign parameters include the Nth vital sign parameter collected at the last sampling moment among the multiple sampling moments;

[0035] A feature acquisition unit for determining the temporal dependence relationship of the multiple vital sign parameters according to the deep belief DBN network, and obtaining the Nth parameter change feature according to the dependence relationship;

[0036] A parameter scoring unit for scoring the Nth parameter change feature and the Nth vital sign parameter according to a preset scoring strategy to obtain the Nth evaluation score;

[0037] An angle update unit for updating the Nth angle based on the Nth angle and the Nth evaluation score using a preset reinforcement learning model when the Nth evaluation score is lower than a set evaluation score threshold;

[0038] A step loop execution unit is used to increment the value of N by 1, return to execute and control the lifting component to perform a lifting operation to adjust the angle of the bed body of the nursing bed relative to the horizontal plane, and receive the angle of the bed body of the nursing bed relative to the horizontal plane collected by the angle sensor. When it is determined that the received angle reaches the Nth angle, stop the step of controlling the lifting component to perform the lifting operation until the Nth evaluation score reaches the set evaluation score threshold.

[0039] In a third aspect, the present application provides a neonatal incubator intelligent management system, including an incubator body. A base is arranged inside the incubator body, and a main controller is embedded in the base. A nursing bed is arranged on the base. The tail of the nursing bed is fixedly connected to one side of the base, and the head of the nursing bed is liftably connected to the base. An angle sensor for detecting the angle of the head of the nursing bed relative to the horizontal plane is arranged at a position corresponding to the head of the bed on the base. A parameter detection module for detecting the vital sign parameters of the neonate is arranged on one side of the incubator body close to the head of the bed. Among them, the vital sign parameters are associated with the angle of the head of the nursing bed relative to the horizontal plane, and the main controller is used to execute the method provided in the first aspect of the present application.

[0040] In a fourth aspect, the present application further provides a storage medium storing a computer program. When the computer program is executed by a processor, the computer is enabled to execute the method provided in the first aspect of the present application.

[0041] The present application provides a neonatal incubator intelligent management method and device. The main controller can control the lifting component to perform a lifting operation to adjust the angle of the bed body of the nursing bed relative to the horizontal plane, and receive the angle of the bed body of the nursing bed relative to the horizontal plane collected by the angle sensor. When it is determined that the received angle reaches the Nth angle, stop controlling the lifting component to perform the lifting operation, where N is a positive integer and the initial value of N is 1.

[0042] The main controller receives, within a preset time period thereafter, multiple vital sign parameters of the neonate lying on the nursing bed collected by the parameter detection module at multiple sampling moments. Among them, the multiple vital sign parameters include the Nth vital sign parameter collected at the last sampling moment among the multiple sampling moments. Since the vital sign parameters are associated with the angle of the head of the nursing bed relative to the horizontal plane, and the vital sign parameters (such as including heart rate, intracranial pressure, and respiratory rate, etc.) can characterize the current health condition and comfort level of the neonate.

[0043] Determine the temporal dependence relationship of the multiple vital sign parameters according to the deep belief DBN network, and obtain the Nth parameter change feature according to the dependence relationship. It can be understood that the change trend of the vital sign parameters (i.e., the Nth parameter change feature) can characterize the future health condition and comfort level of the neonate.

[0044] The main controller scores the Nth parameter change feature and the Nth vital sign parameter according to a preset scoring strategy to obtain the Nth evaluation score. The Nth evaluation score is used to quantify the comprehensive health condition and comfort of the current and future newborns.

[0045] When the Nth evaluation score is lower than the set evaluation score threshold, it indicates that when the bed body is at the current Nth angle relative to the horizontal plane, the comprehensive health condition and comfort of the current and future newborns are low. Therefore, the preset reinforcement learning model is used to update the Nth angle based on the Nth angle and the Nth evaluation score.

[0046] The main controller increments the value of N by 1, returns to execute the lifting operation of the lifting component to adjust the angle of the bed body of the nursing bed relative to the horizontal plane, and receives the angle of the bed body of the nursing bed relative to the horizontal plane collected by the angle sensor. When it is determined that the received angle reaches the Nth angle, the step of controlling the lifting component to execute the lifting operation is stopped until the Nth evaluation score reaches the set evaluation score threshold. It can be understood that when the Nth evaluation score reaches the set evaluation score threshold, the comprehensive health condition and comfort of the current and future newborns are high, indicating that the reliability and real-time performance of the finally adjusted Nth angle of the bed body relative to the horizontal plane are high, and the labor cost is saved. Description of the Drawings

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0048] Figure 1 It is a schematic structural diagram of a neonatal incubator provided by an embodiment of the present application;

[0049] Figure 2 It is a flowchart of a neonatal incubator intelligent management method provided by an embodiment of the present application. Detailed Embodiments

[0050] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. Based on the embodiments of the present application, all other embodiments made by those of ordinary skill in the art under the inspiration of this embodiment belong to the protection scope of the present application.

[0051] The terms "first", "second", "third", "fourth", etc. (if any) in the description, claims and above-mentioned drawings of this application are used to distinguish similar objects, and do not necessarily describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of this application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0052] An embodiment of this application provides a method for intelligent management of a neonatal incubator, which is applied to a neonatal incubator intelligent management system. As Figure 1 shown, the system includes an incubator body 101. A base 103 is provided inside the incubator body 101. A main controller (not shown in the drawings) is embedded in the base 103. A nursing bed 102 is provided on the base 103. The end of the nursing bed 102 is fixedly connected to one side of the base 103, and the head of the nursing bed 102 is liftably connected to the base 103. An angle sensor for detecting the angle of the head of the nursing bed 102 relative to the horizontal plane is provided at a position corresponding to the head of the bed on the base 103. A parameter detection module 106 for detecting the vital sign parameters of the neonate is provided on one side of the incubator body 101 near the head of the bed. Among them, the vital sign parameters are associated with the angle of the head of the nursing bed 102 relative to the horizontal plane. As Figure 2 shown, the method provided by the embodiment of this application includes:

[0053] S101: The main controller controls the lifting assembly to perform a lifting operation to adjust the angle of the bed body of the nursing bed 102 relative to the horizontal plane, and receives the angle of the bed body of the nursing bed 102 relative to the horizontal plane collected by the angle sensor. When it is determined that the received angle reaches the Nth angle, the main controller stops controlling the lifting assembly to perform the lifting operation.

[0054] Wherein, N is a positive integer, and the initial value of N is 1.

[0055] The angle sensor is a distance sensor 104. The specific implementation of receiving the angle of the bed body of the nursing bed 102 relative to the horizontal plane collected by the angle sensor can be: receiving the distance from the base 103 to the head of the nursing bed 102 collected by the distance sensor 104; and looking up the angle of the bed body of the nursing bed 102 relative to the horizontal plane from a preset second mapping table according to the distance.

[0056] S102: The main controller receives multiple vital sign parameters of the newborn lying on the nursing bed 102 collected by the parameter detection module 106 at multiple sampling times within a preset duration thereafter.

[0057] Exemplarily, each vital sign parameter may include respiratory rate, intracranial pressure, and heart rate. The parameter detection module 106 may include a respiratory sensor for detecting the respiratory rate, a transcranial Doppler ultrasound module for measuring the cerebral blood flow velocity, and thus the intracranial pressure can be determined based on the cerebral blood flow velocity, and a heart rate sensor for collecting the heart rate.

[0058] Exemplarily, the preset duration may be, but is not limited to, 20 min, and the interval duration between every two sampling times may be, but is not limited to, 2 min. Among them, the multiple vital sign parameters include the Nth vital sign parameter collected at the last sampling time among the multiple sampling times.

[0059] S103: Determine the temporal dependence relationship of the multiple vital sign parameters according to the deep belief DBN network, and obtain the Nth parameter change feature according to the dependence relationship.

[0060] The deep belief network DBN (Deep Belief Networks, DBN) is a neural network model in deep learning methods. The DBN network has multiple hidden layers. The hidden layer of the DBN is connected to the next hidden layer, and the last hidden layer is connected to the visible layer, but the units in the same layer are not connected. The multiple hidden layers can perform higher-stage correlation operations on the data, which also makes the hidden layers before and after training have more sequence. After the previous hidden layer is fully trained, the next hidden layer can use the training data obtained by the previous hidden layer to perform the same type of training again, and this process is repeated continuously until the trained data is finally obtained. In this way, the temporal dependence relationship of the multiple vital sign parameters can be determined according to the deep belief DBN network.

[0061] S104: The main controller scores the Nth parameter change feature and the Nth vital sign parameter according to a preset scoring strategy to obtain the Nth evaluation score.

[0062] Exemplarily, the specific implementation of S104 includes, but is not limited to, the following two methods:

[0063] The first one:

[0064] Step 1: The main controller inputs the Nth parameter change feature and the Nth vital sign parameter into a pre-trained parameter prediction model to predict the future vital sign parameters of the newborn on the nursing bed 102 after a future duration.

[0065] Among them, the parameter prediction model is obtained by training multiple first training samples input into a neural network. Each first training sample includes historical parameter change characteristics and the actual future vital sign parameters of the newborn on the nursing bed 102 after a corresponding historical future duration.

[0066] Step 2: Determine the first sub-score corresponding to the future life feature parameter according to the future life feature parameter and the preset first mapping relationship table.

[0067] Among them, the first mapping relationship table records the mapping relationships between various different life feature parameters and the corresponding scores.

[0068] Step 3: Determine the second sub-score corresponding to the Nth vital sign parameter according to the Nth vital sign parameter and the preset first mapping relationship table.

[0069] Step 4: Perform weighted summation on the first sub-score and the second sub-score to obtain the Nth evaluation score.

[0070] Exemplarily, according to the formula P N = k1P1 + k2P2, determine the Nth evaluation score, where P N is the Nth evaluation score, P1 is the first sub-score, P2 is the second sub-score, k1 is the first weighting coefficient, k2 is the second weighting coefficient, and k1 + k2 = 1. For example, k1 = 0.4; k2 = 0.6.

[0071] It can be understood that based on the above Steps 1 - 4, the Nth evaluation score can be accurately obtained.

[0072] Further, the Nth vital sign parameter includes the Nth respiratory rate, the Nth intracranial pressure, and the Nth heart rate. The acquisition method of the above second score can be:

[0073] Step A1: Determine the third sub-score corresponding to the Nth respiratory rate according to the Nth respiratory rate and the preset first mapping relationship table.

[0074] Step A2: Determine the fourth sub-score corresponding to the Nth intracranial pressure according to the Nth intracranial pressure and the preset first mapping relationship table.

[0075] Step A3: Determine the fifth sub-score corresponding to the Nth heart rate according to the Nth heart rate and the preset first mapping relationship table.

[0076] Step A4: Determine the second sub-score according to the formula P2 = k3P3 + k4P4 + k5P5, where P2 is the second sub-score, P3 is the third sub-score, P4 is the fourth sub-score, P5 is the fifth sub-score, k3 is the third weighting coefficient, k4 is the fourth weighting coefficient, k5 is the fifth weighting coefficient, and k3 + k4 + k5 = 1. For example, k3 = 0.3; k3 = 0.3, k4 = 0.4.

[0077] It can be understood that based on the above Step A1 - Step A4, the second sub-score can be accurately obtained.

[0078] The second method: Integrate the Nth parameter change feature and the Nth vital sign parameter to obtain the Nth vital sign feature; input the Nth vital sign feature into the pre-trained second scoring model to obtain the Nth evaluation score. The second scoring model is obtained by training a neural network with multiple second training samples, and each second training sample includes historical vital sign features and their corresponding historical actual scores. In this way, the Nth evaluation score can be accurately obtained.

[0079] S105: The main controller determines whether the Nth evaluation score is lower than the set evaluation score threshold. If so, execute S106; if not, execute S108.

[0080] S106: Update the Nth angle based on the Nth angle and the Nth evaluation score using a preset reinforcement learning model.

[0081] It can be understood that the Nth angle is the state of the reinforcement learning model; updating the Nth angle is the action of the reinforcement learning model, and the updated Nth evaluation score is the reward of the reinforcement learning model.

[0082] S107: The main controller increments the value of N by 1 and returns to execute S101.

[0083] S108: The main controller controls the lifting component to stop working so as to maintain the angle of the bed body of the nursing bed 102 relative to the horizontal plane.

[0084] In some embodiments, the head of the nursing bed 102 is telescopically connected to the base 103 through a lifting component. The lifting component includes a first telescopic rod 105 and a second telescopic rod (obscured in the drawing) arranged opposite to each other, as well as a first drive motor (not shown in the drawing) for driving the first telescopic rod 105 to extend and retract and a second drive motor (not shown in the drawing) for driving the second telescopic rod to extend and retract. S108 can be specifically implemented as: The main controller controls the first drive motor to drive the first telescopic rod 105 to extend and retract and controls the second drive motor to drive the second telescopic rod to extend and retract, so that the lifting component performs a lifting operation.

[0085] In summary, the embodiment of the present application provides an intelligent management method for a neonatal incubator. The main controller can control the lifting component to perform a lifting operation to adjust the angle of the bed body of the nursing bed 102 relative to the horizontal plane, and receive the angle of the bed body of the nursing bed 102 relative to the horizontal plane collected by the angle sensor. When it is determined that the received angle reaches the Nth angle, the control of the lifting component to perform the lifting operation is stopped, where N is a positive integer and the initial value of N is 1.

[0086] Within a preset duration thereafter, the main controller receives multiple vital sign parameters of the neonate lying on the nursing bed 102 collected by the parameter detection module 106 at multiple sampling times, where the multiple vital sign parameters include the Nth vital sign parameter collected at the last sampling time among the multiple sampling times. Since the vital sign parameters are related to the angle of the head of the nursing bed 102 relative to the horizontal plane, and the vital sign parameters (such as including heart rate, intracranial pressure, and respiratory rate, etc.) can characterize the health condition and comfort level of the current neonate.

[0087] Determine the temporal dependence relationship of the multiple vital sign parameters according to the deep belief DBN network, and obtain the Nth parameter change feature according to the dependence relationship. It can be understood that the change trend of the vital sign parameters (i.e., the Nth parameter change feature) can characterize the health condition and comfort level of the neonate in the future.

[0088] The main controller scores the Nth parameter change feature and the Nth vital sign parameter according to a preset scoring strategy to obtain the Nth evaluation score. The Nth evaluation score is used to quantify the comprehensive health condition and comfort level of the current and future neonates.

[0089] When the Nth evaluation score is lower than the set evaluation score threshold, it indicates that when the bed body is at the current Nth angle relative to the horizontal plane, the comprehensive health condition and comfort level of the current and future neonates are low. Furthermore, a preset reinforcement learning model is used to update the Nth angle based on the Nth angle and the Nth evaluation score.

[0090] The main controller increments the value of N by 1, returns to execute the step of controlling the lifting component to perform a lifting operation to adjust the angle of the bed body of the nursing bed 102 relative to the horizontal plane, and receives the angle of the bed body of the nursing bed 102 relative to the horizontal plane collected by the angle sensor. When it is determined that the received angle reaches the Nth angle, the step of stopping the control of the lifting component to perform the lifting operation is executed until the Nth evaluation score reaches the set evaluation score threshold. It can be understood that when the Nth evaluation score reaches the set evaluation score threshold, the comprehensive health condition and comfort level of the current and future neonates are high, indicating that the reliability and real-time performance of the finally adjusted Nth angle of the bed body relative to the horizontal plane are high, saving labor costs.

[0091] In addition, an intelligent management device for a neonatal incubator is provided in an embodiment of the present application. The device is configured in a main controller, and the main controller belongs to an intelligent management system for a neonatal incubator. It should be noted that the basic principle and technical effects of the intelligent management system for a neonatal incubator provided in the embodiment of the present application are the same as those of the above embodiment. For the sake of brief description, for the parts not mentioned in the embodiment of the present application, reference can be made to the corresponding content in the above embodiment.

[0092] The system further includes an incubator body 101. A base 103 is arranged inside the incubator body 101. A main controller is embedded in the base 103. A nursing bed 102 is arranged on the base 103. The tail of the nursing bed 102 is fixedly connected to one side of the base 103, and the head of the nursing bed 102 is connected to the base 103 in a liftable manner. An angle sensor for detecting the angle of the head of the nursing bed 102 relative to the horizontal plane is arranged at a position corresponding to the head of the bed on the base 103. A parameter detection module 106 for detecting the vital sign parameters of a neonate is arranged on one side close to the head of the bed inside the incubator body 101. Among them, the vital sign parameters are associated with the angle of the head of the nursing bed 102 relative to the horizontal plane. The device provided in the embodiment of the present application includes an angle adjustment unit, a parameter receiving unit, a feature obtaining unit, a parameter scoring unit, an angle updating unit, and a step loop execution unit.

[0093] The angle adjustment unit is used to control the lifting assembly to perform a lifting operation to adjust the angle of the bed body of the nursing bed 102 relative to the horizontal plane, and receive the angle of the bed body of the nursing bed 102 relative to the horizontal plane collected by the angle sensor. When it is determined that the received angle reaches the Nth angle, the control of the lifting assembly to perform the lifting operation is stopped, where N is a positive integer and the initial value of N is 1.

[0094] The parameter receiving unit is used to receive, within a preset time period afterwards, multiple vital sign parameters of a neonate lying on the nursing bed 102 collected by the parameter detection module 106 at multiple sampling times, where the multiple vital sign parameters include the Nth vital sign parameter collected at the last sampling time among the multiple sampling times.

[0095] The feature obtaining unit is used to determine the temporal dependence relationship of the multiple vital sign parameters according to a deep belief DBN network, and obtain the Nth parameter change feature according to the dependence relationship.

[0096] The parameter scoring unit is used to score the Nth parameter change feature and the Nth vital sign parameter according to a preset scoring strategy to obtain the Nth evaluation score.

[0097] The angle updating unit is used to update the Nth angle based on the Nth angle and the Nth evaluation score by using a preset reinforcement learning model when the Nth evaluation score is lower than a set evaluation score threshold.

[0098] A step loop execution unit is configured to increment the value of N by 1, return to execute and control the lifting component to perform a lifting operation to adjust the angle of the bed body of the nursing bed 102 relative to the horizontal plane, and receive the angle of the bed body of the nursing bed 102 relative to the horizontal plane collected by the angle sensor. When it is determined that the received angle reaches the Nth angle, the step of controlling the lifting component to perform the lifting operation is stopped until the Nth evaluation score reaches the set evaluation score threshold.

[0099] In addition, still as Figure 2 shown, the present application provides a neonatal incubator intelligent management system, including an incubator body 101. A base 103 is arranged inside the incubator body 101. A main controller is embedded in the base 103. A nursing bed 102 is arranged on the base 103. The tail of the nursing bed 102 is fixedly connected to one side of the base 103. The head of the nursing bed 102 is liftably connected to the base 103. An angle sensor for detecting the angle of the head of the nursing bed 102 relative to the horizontal plane is arranged at a position corresponding to the head of the base 103. A parameter detection module 106 for detecting the vital sign parameters of the neonate is arranged on one side of the incubator body 101 close to the head of the bed. Among them, the vital sign parameters are associated with the angle of the head of the nursing bed 102 relative to the horizontal plane. The main controller is configured to execute the method provided in the above embodiments of the present application.

[0100] In addition, an embodiment of the present application further provides a storage medium storing a computer program, which when executed by a processor causes the computer to execute the method provided in the above embodiments of the present application.

[0101] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all 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.

Claims

1. An intelligent management method for neonatal incubators, characterized in that, Applied to the intelligent management system of neonatal incubators. The system includes an incubator body, a base is arranged inside the incubator body, a main controller is embedded in the base, a nursing bed is arranged on the base, the tail of the nursing bed is fixedly connected to one side of the base, the head of the nursing bed is liftably connected to the base, and an angle sensor for detecting the angle of the head of the nursing bed relative to the horizontal plane is arranged at a position corresponding to the head of the nursing bed on the base. A parameter detection module for detecting the vital sign parameters of the neonate is arranged on one side of the incubator body close to the head of the bed. Among them, the vital sign parameters are associated with the angle of the head of the nursing bed relative to the horizontal plane. The method includes: The main controller controls the lifting component to perform a lifting operation to adjust the angle of the nursing bed body relative to the horizontal plane, and receives the angle of the nursing bed body relative to the horizontal plane collected by the angle sensor. When it is determined that the received angle reaches the Nth angle, the main controller stops controlling the lifting component to perform the lifting operation, where N is a positive integer and the initial value of N is 1; Within a preset time period afterwards, the main controller receives multiple vital sign parameters of the neonate lying on the nursing bed collected by the parameter detection module at multiple sampling moments, where the multiple vital sign parameters include the Nth vital sign parameter collected at the last sampling moment among the multiple sampling moments; Determine the temporal dependence relationship of the multiple vital sign parameters according to the deep belief DBN network, and obtain the Nth parameter change feature according to the dependence relationship; The main controller scores the Nth parameter change feature and the Nth vital sign parameter according to a preset scoring strategy to obtain the Nth evaluation score; When the Nth evaluation score is lower than the set evaluation score threshold, the main controller updates the Nth angle based on the Nth angle and the Nth evaluation score by using a preset reinforcement learning model; The main controller increments the value of N by 1, returns to execute the step of controlling the lifting component to perform a lifting operation to adjust the angle of the nursing bed body relative to the horizontal plane, and receives the angle of the nursing bed body relative to the horizontal plane collected by the angle sensor. When it is determined that the received angle reaches the Nth angle, the main controller stops controlling the lifting component to perform the lifting operation, until the Nth evaluation score reaches the set evaluation score threshold.

2. The method according to claim 1, wherein The main controller scores the Nth parameter change feature and the Nth vital sign parameter according to a preset scoring strategy to obtain the Nth evaluation score, including: The main controller inputs the Nth parameter change feature and the Nth vital sign parameter into a pre-trained parameter prediction model to predict the future vital sign parameters of the neonate on the nursing bed after a future time period. The parameter prediction model is trained by inputting multiple first training samples into a neural network. Each first training sample includes a historical parameter change feature and the actual future vital sign parameters of the neonate on the nursing bed after the corresponding historical future time period; Determine a first sub-score corresponding to the future life characteristic parameter according to the future life characteristic parameter and a preset first mapping relation table, wherein the first mapping relation table records the mapping relations between various different life characteristic parameters and corresponding scores; Determine a second sub-score corresponding to the Nth vital sign parameter according to the Nth vital sign parameter and the preset first mapping relation table; Perform weighted summation on the first sub-score and the second sub-score to obtain an Nth evaluation score.

3. The method according to claim 2, wherein The performing weighted summation on the first sub-score and the second sub-score to obtain an Nth evaluation score includes: According to the formula P N = k1P1 + k2P2, determine the Nth evaluation score, where P N is the Nth evaluation score, P1 is the first sub-score, P2 is the second sub-score, k1 is the first weighting coefficient, k2 is the second weighting coefficient, and k1 + k2 = 1.

4. The method according to claim 2, wherein The Nth vital sign parameter includes an Nth respiration rate, an Nth intracranial pressure, and an Nth heart rate. The determining a second sub-score corresponding to the Nth vital sign parameter according to the Nth vital sign parameter and the preset first mapping relation table includes: Determine a third sub-score corresponding to the Nth respiration rate according to the Nth respiration rate and the preset first mapping relation table; Determine a fourth sub-score corresponding to the Nth intracranial pressure according to the Nth intracranial pressure and the preset first mapping relation table; Determine a fifth sub-score corresponding to the Nth heart rate according to the Nth heart rate and the preset first mapping relation table; Determine the second sub-score according to the formula P2 = k3P3 + k4P4 + k5P5, where P2 is the second sub-score, P3 is the third sub-score, P4 is the fourth sub-score, P5 is the fifth sub-score, k3 is a third weighting coefficient, k4 is a fourth weighting coefficient, k5 is a fifth weighting coefficient, and k3 + k4 + k5 = 1.

5. The method according to claim 2, wherein The main controller scores the Nth parameter change characteristic and the Nth vital sign parameter according to a preset scoring strategy to obtain an Nth evaluation score, including: Fuse the Nth parameter change characteristic and the Nth vital sign parameter to obtain an Nth vital sign characteristic; Input the Nth vital sign characteristic into a pre-trained second scoring model to obtain an Nth evaluation score. The second scoring model is obtained by training a neural network with a plurality of second training samples, and each second training sample includes a historical vital sign characteristic and its corresponding historical actual score.

6. The method according to any one of claims 1-5, characterized in that The head of the nursing bed is liftably connected to the base through a lifting assembly. The lifting assembly includes a first telescopic rod and a second telescopic rod arranged oppositely, a first driving motor for driving the first telescopic rod to extend and retract, and a second driving motor for driving the second telescopic rod to extend and retract. The main controller controls the lifting assembly to perform a lifting operation, including: The main controller controls the first driving motor to drive the first telescopic rod to extend and retract and controls the second driving motor to drive the second telescopic rod to extend and retract so that the lifting assembly performs a lifting operation.

7. According to the method as claimed in any one of claims 1 to 5, characterized in that The angle sensor is a distance sensor. The receiving the angle of the bed body of the nursing bed relative to the horizontal plane collected by the angle sensor includes: Receive the distance between the base and the head of the nursing bed collected by the distance sensor; Look up the angle of the bed body of the nursing bed relative to the horizontal plane according to the distance from a preset second mapping table.

8. An intelligent management device for a neonatal incubator, characterized in that, Configured in a main controller, the main controller belongs to a neonatal incubator intelligent management system. The system further includes an incubator body. A base is provided inside the incubator body. The main controller is embedded in the base. A nursing bed is provided on the base. The tail of the nursing bed is fixedly connected to one side of the base. The head of the nursing bed is connected to the base in a liftable manner. An angle sensor for detecting the angle of the head of the nursing bed relative to the horizontal plane is provided at a position corresponding to the head of the bed on the base. A parameter detection module for detecting the vital sign parameters of a neonate is provided on one side of the incubator body near the head of the bed. Among them, the vital sign parameters are associated with the angle of the head of the nursing bed relative to the horizontal plane. The device includes: An angle adjustment unit, configured to control the lifting assembly to perform a lifting operation to adjust the angle of the bed body of the nursing bed relative to the horizontal plane, and receive the angle of the bed body of the nursing bed relative to the horizontal plane collected by the angle sensor. When it is determined that the received angle reaches the Nth angle, stop controlling the lifting assembly to perform the lifting operation, where N is a positive integer and the initial value of N is 1; A parameter receiving unit, configured to receive, within a preset time period afterwards, multiple vital sign parameters of a neonate lying on the nursing bed collected by the parameter detection module at multiple sampling moments, where the multiple vital sign parameters include the Nth vital sign parameter collected at the last sampling moment among the multiple sampling moments; A feature acquisition unit, configured to determine the temporal dependence relationship of the multiple vital sign parameters according to a deep belief DBN network, and obtain the Nth parameter change feature according to the dependence relationship; A parameter scoring unit, configured to score the Nth parameter change feature and the Nth vital sign parameter according to a preset scoring strategy to obtain the Nth evaluation score; An angle update unit, configured to, when the Nth evaluation score is lower than a set evaluation score threshold, update the Nth angle based on the Nth angle and the Nth evaluation score by using a preset reinforcement learning model; A step loop execution unit, configured to increment the value of N by 1, return to execute the step of controlling the lifting assembly to perform a lifting operation to adjust the angle of the bed body of the nursing bed relative to the horizontal plane, and receive the angle of the bed body of the nursing bed relative to the horizontal plane collected by the angle sensor. When it is determined that the received angle reaches the Nth angle, stop controlling the lifting assembly to perform the lifting operation, until the Nth evaluation score reaches the set evaluation score threshold.

9. A neonatal incubator intelligent management system, characterized in that, It includes an incubator body, a base is arranged inside the incubator body, a main controller is embedded in the base, a nursing bed is arranged on the base, the tail of the nursing bed is fixedly connected to one side of the base, the head of the nursing bed is connected to the base in a liftable manner, an angle sensor for detecting the angle of the head of the nursing bed relative to the horizontal plane is arranged at a position corresponding to the head of the bed on the base, and a parameter detection module for detecting the vital sign parameters of a newborn is arranged on one side of the incubator body close to the head of the bed. Among them, the vital sign parameters are associated with the angle of the head of the nursing bed relative to the horizontal plane, and the main controller is used to execute the method according to any one of claims 1-7.

10. A storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it causes the computer to execute the method according to any one of claims 1 to 7.

Images