Electric nursing bed, urine wetting detection method and device thereof and storage medium

Abstract

The invention provides an electric nursing bed, a urine detection method and device thereof and a storage medium. The method comprises the steps that electric signals collected at all detection points in a urine sensing pad are received; detecting the surface state of the urine sensing pad according to the distribution information of the electric signal in time and space; if the surface state is a soaking state, calling a head ring to collect electroencephalogram signals and head posture data of the user; identifying the body state of the user according to the electroencephalogram signal and the head posture data; and executing a wetting alarm operation according to the body state. According to the embodiment of the invention, the urination detection is carried out according to the distribution information of the electrical signals on the urination sensing pad in time and space, the information amount of the urination detection is increased, the accuracy of the urination detection can be improved, the urination detection is corrected according to the electroencephalogram signal and the head posture data of the user, the probability of false alarm is effectively reduced by the dual mechanism, and the user experience is improved. And the interference to the user and nursing personnel is reduced.

Description

Technical Field

[0001] This invention belongs to the technical field of brain-computer interfaces, and in particular relates to an electric nursing bed and its urine detection method, device and storage medium. Background Technology

[0002] In settings such as convalescence and surgical rehabilitation, electric nursing beds have become one of the main tools to help caregivers. Electric nursing beds can be equipped with brain-computer interfaces, allowing users to perform various operations independently on the electric nursing bed.

[0003] Some electric nursing beds are equipped with urine-sensing pads, which rely on changes in electrical signals such as resistance and capacitance when urine wets the bed to detect the problem.

[0004] Currently, wetness-sensing pads typically use a threshold method for alarms, meaning that an alarm is triggered when the value of the electrical signal exceeds a threshold.

[0005] However, since the threshold is relatively fixed, false alarms are easily triggered when liquids such as sweat, medication, or food from the user's skin come into contact with the wetness sensor. Frequent false alarms can be disruptive to both users and caregivers. Summary of the Invention

[0006] In view of this, the present invention provides an electric nursing bed and a method, device and storage medium for detecting urine wetness, so as to improve the alarm accuracy of the urine wetness sensing pad.

[0007] A first aspect of the present invention provides a method for detecting urine wetness in an electric nursing bed, wherein the electric nursing bed is covered with a urine wetness sensing pad and is equipped with a headband, the method comprising:

[0008] Receive electrical signals collected at various detection points in the urine-sensing pad;

[0009] The surface state of the urine-sensing pad is detected based on the distribution information of the electrical signal in time and space;

[0010] If the surface is wet, the headband is invoked to collect EEG signals and head posture data from the user.

[0011] The user's physical state is identified based on the electroencephalogram (EEG) signals and the head posture data;

[0012] The urine-wet alarm will be activated based on the described physical condition.

[0013] Optionally, detecting the surface state of the urine-sensing pad based on the temporal and spatial distribution information of the electrical signal includes:

[0014] For the same moment, the electrical signal is converted into a two-dimensional electrical distribution map based on the spatial distribution information of the detection points;

[0015] The electrical distribution maps of multiple times are stacked in time to obtain a three-dimensional electrical distribution volume;

[0016] The surface condition of the urine-sensing pad is detected based on the electrical distributor.

[0017] Optionally, converting the electrical signal into a two-dimensional electrical distribution map based on the spatial distribution information of the detection points at the same time includes:

[0018] Query the coordinates of the detection points distributed in the space of the urine-sensing pad;

[0019] A two-dimensional electrical distribution map is constructed based on the dimensions of the urine-sensing pad;

[0020] At any given moment, if the coordinates of a pixel in the electrical distribution map match the coordinates of the detection point, then the intensity value of the electrical signal is assigned as the pixel value of that pixel.

[0021] If the coordinates of the pixel in the electrical distribution map are not the same as the coordinates of the detection point, then the specified value is assigned as the pixel value of the pixel.

[0022] Optionally, detecting the surface state of the urine-sensing pad based on the electrical distributor includes:

[0023] Load the urine moisture sensing detection model;

[0024] The electrical distributor is input into the urine wetness detection model to detect the probability that the urine wetness sensing pad is in a wet state;

[0025] The surface state of the urine-sensing pad is determined to be wet based on the changing trends of multiple probabilities.

[0026] Optionally, determining whether the surface state of the urine-sensing pad is wet based on the changing trends of the plurality of probabilities includes:

[0027] Add windows to multiple of the probabilities;

[0028] During the sliding of the window, the average value of the probabilities within the window is calculated;

[0029] If the difference between the current average value and the previous average value is greater than a preset change threshold, then the surface state of the urine wetness sensor pad is determined to be wet.

[0030] Optionally, the body state includes a state of consciousness and a state of movement; the step of identifying the user's body state based on the electroencephalogram (EEG) signal and the head posture data includes:

[0031] Load a state detection model; the state detection model includes a first convolutional module, a second convolutional module, a third convolutional module, a first head structure, and a second head structure;

[0032] The EEG signal was converted by performing continuous wavelet transformation to obtain the original EEG features;

[0033] The original EEG features are input into the first convolutional module to extract the first candidate EEG features;

[0034] The first candidate EEG features are input into the first head structure for binary classification to obtain the user's state of consciousness.

[0035] If the state of consciousness is conscious, then the head posture data is converted into a two-dimensional posture matrix;

[0036] The first candidate EEG features are input into the second convolution module to extract the second candidate EEG features;

[0037] The pose matrix is ​​input into the third convolutional module to extract pose features;

[0038] The second candidate EEG features are fused with the posture features to form a multimodal feature;

[0039] The multimodal features are input into the second head structure for binary classification to obtain the user's movement state.

[0040] Optionally, the step of performing the wetness alarm operation based on the physical condition includes:

[0041] If the state of consciousness is impaired or the movement state is non-turning, an alarm signal indicating that urine is wet is generated.

[0042] The alarm signal is sent to the designated nursing management device.

[0043] A second aspect of the present invention provides a urine detection device for an electric nursing bed, wherein the electric nursing bed is covered with a urine-sensing pad and is equipped with a headband, and the device includes:

[0044] An electrical signal receiving module is used to receive electrical signals collected at various detection points in the urine-sensing pad.

[0045] A surface condition detection module is used to detect the surface condition of the urine wetness sensor pad based on the distribution information of the electrical signal in time and space.

[0046] The user data acquisition module is used to call the headband to collect EEG signals and head posture data from the user if the surface is wet.

[0047] A body state recognition module is used to identify the user's body state based on the electroencephalogram (EEG) signals and the head posture data.

[0048] The urine wetness alarm module is used to perform urine wetness alarm operations based on the said physical condition.

[0049] Optionally, the surface condition detection module is further configured to:

[0050] For the same moment, the electrical signal is converted into a two-dimensional electrical distribution map based on the spatial distribution information of the detection points;

[0051] The electrical distribution maps of multiple times are stacked in time to obtain a three-dimensional electrical distribution volume;

[0052] The surface condition of the urine-sensing pad is detected based on the electrical distributor.

[0053] Optionally, the surface condition detection module is further configured to:

[0054] Query the coordinates of the detection points distributed in the space of the urine-sensing pad;

[0055] A two-dimensional electrical distribution map is constructed based on the dimensions of the urine-sensing pad;

[0056] At any given moment, if the coordinates of a pixel in the electrical distribution map match the coordinates of the detection point, then the intensity value of the electrical signal is assigned as the pixel value of that pixel.

[0057] If the coordinates of the pixel in the electrical distribution map are not the same as the coordinates of the detection point, then the specified value is assigned as the pixel value of the pixel.

[0058] Optionally, the surface condition detection module is further configured to:

[0059] Load the urine moisture sensing detection model;

[0060] The electrical distributor is input into the urine wetness detection model to detect the probability that the urine wetness sensing pad is in a wet state;

[0061] The surface state of the urine-sensing pad is determined to be wet based on the changing trends of multiple probabilities.

[0062] Optionally, the surface condition detection module is further configured to:

[0063] Add windows to multiple of the probabilities;

[0064] During the sliding of the window, the average value of the probabilities within the window is calculated;

[0065] If the difference between the current average value and the previous average value is greater than a preset change threshold, then the surface state of the urine wetness sensor pad is determined to be wet.

[0066] Optionally, the body state recognition module is further used for:

[0067] Load a state detection model; the state detection model includes a first convolutional module, a second convolutional module, a third convolutional module, a first head structure, and a second head structure;

[0068] The EEG signal was converted by performing continuous wavelet transformation to obtain the original EEG features;

[0069] The original EEG features are input into the first convolutional module to extract the first candidate EEG features;

[0070] The first candidate EEG features are input into the first head structure for binary classification to obtain the user's state of consciousness.

[0071] If the state of consciousness is conscious, then the head posture data is converted into a two-dimensional posture matrix;

[0072] The first candidate EEG features are input into the second convolution module to extract the second candidate EEG features;

[0073] The pose matrix is ​​input into the third convolutional module to extract pose features;

[0074] The second candidate EEG features are fused with the posture features to form a multimodal feature;

[0075] The multimodal features are input into the second head structure for binary classification to obtain the user's movement state.

[0076] Optionally, the urine-wet alarm module is also used for:

[0077] If the state of consciousness is impaired or the movement state is non-turning, an alarm signal indicating that urine is wet is generated.

[0078] The alarm signal is sent to the designated nursing management device.

[0079] A third aspect of the present invention provides an electric nursing bed, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the urine wetness detection method of the electric nursing bed as described in the first aspect above.

[0080] A fourth aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method for detecting urine wetness in an electric nursing bed as described in the first aspect above.

[0081] A fifth aspect of the present invention provides a computer program product that, when run on a computer, causes the computer to perform the method for detecting urine wetness in an electric nursing bed as described in the first aspect above.

[0082] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0083] In this embodiment, the electric nursing bed is equipped with a urine-sensing pad on its surface and a headband that receives electrical signals collected from various detection points on the urine-sensing pad. The surface state of the pad is detected based on the temporal and spatial distribution of these electrical signals. If the surface is wet, the headband is activated to collect the user's electroencephalogram (EEG) signals and head posture data. The user's physical condition is identified based on the EEG signals and head posture data. A urine-sensing alarm is then triggered based on this physical condition. This embodiment detects urine wetness by utilizing the temporal and spatial distribution of electrical signals on the pad, increasing the amount of information involved and improving accuracy. Furthermore, the use of the user's EEG signals and head posture data for correction effectively reduces the probability of false alarms and minimizes interference with the user and caregivers. Attached Figure Description

[0084] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0085] Figure 1 This is a schematic diagram of a method for detecting urine wetness in an electric nursing bed provided in an embodiment of the present invention;

[0086] Figure 2 This is a schematic diagram of the structure of a urine-sensing pad provided in an embodiment of the present invention;

[0087] Figure 3 This is a schematic diagram of the structure of a state detection model provided in an embodiment of the present invention;

[0088] Figure 4 This is a schematic diagram of a urine detection device for an electric nursing bed provided in an embodiment of the present invention;

[0089] Figure 5This is a schematic diagram of an electric nursing bed provided in an embodiment of the present invention. Detailed Implementation

[0090] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of the present invention. However, those skilled in the art will recognize that the present application may be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted to avoid unnecessary detail that could obscure the description of the present application.

[0091] The technical solution of the present invention will be illustrated below through specific embodiments.

[0092] Reference Figure 1 The diagram illustrates a method for detecting urine wetness in an electric nursing bed according to an embodiment of the present invention, which may specifically include the following steps:

[0093] Step 101: Receive electrical signals collected at various detection points in the urine-sensing pad.

[0094] like Figure 2 As shown, the electric nursing bed is covered with a urine-sensing pad, on which a flexible printed circuit (FPC) is installed. The flexible printed circuit has multiple detection points that can collect electrical signals such as resistance and capacitance.

[0095] Both motors are connected to the controller of the electric nursing bed, which supports control via brain-computer interface, remote control, etc. The controller is connected to the nursing management system to realize the network management of multiple electric nursing beds.

[0096] The controller includes components such as a display screen, and is responsible for providing the user interface for various services and performing calculations according to the user's operations.

[0097] The headband includes components such as a brain-computer interface, a posture sensor, and a Bluetooth module.

[0098] The brain-computer interface includes modules such as electrodes, filtering units, amplification units, and AD conversion units.

[0099] For example, the electrodes include types such as signal electrodes, reference electrodes, and ground electrodes. According to the 10-20 system distribution standard, the signal electrode is attached to the forehead at position FP2, the reference electrode is attached to position AF8, and the ground electrode is attached to position AF7.

[0100] The filtering and amplification unit is used to filter out noise such as radio frequency interference signals and 50Hz common-mode signals.

[0101] The amplification unit is used to amplify signals such as electroencephalogram (EEG) signals and electrooculogram (EOG) signals.

[0102] The AD conversion unit is used to convert the amplified signal into an analog signal.

[0103] The attitude sensor has a built-in nine-axis gyroscope and motion processor. The nine-axis gyroscope consists of a three-axis accelerometer, a three-axis magnetometer, and a three-axis gyroscope. The motion processor performs attitude calculations on the nine-axis data and outputs the user's head attitude data.

[0104] The Bluetooth module includes a processing unit and a communication unit. The processing unit connects to components such as the brain-computer interface and posture sensors, and is responsible for driving the programs of the brain-computer interface and posture sensors, acquiring EEG signals, EEG signals, and head posture data. The communication unit connects to the controller, and is responsible for communicating with the controller, uploading data such as EEG signals, EEG signals, and head posture data to the controller for processing.

[0105] In this embodiment, once the urine-sensing pad has completed initialization, it can continuously receive electrical signals collected at various detection points within the urine-sensing pad.

[0106] Step 102: Detect the surface condition of the urine-sensing pad based on the distribution information of electrical signals in time and space.

[0107] Under normal circumstances, the characteristics of urine wetting in time and space are significantly different from those of other objects containing liquid components (such as the user's sweat, medicine, or food). Therefore, the distribution information of electrical signals in time and space can be analyzed based on deep learning or machine learning technology to detect the surface state of the urine wetting pad.

[0108] In one embodiment of the present invention, step 102 may include the following steps:

[0109] Step 1021: For the same moment, convert the electrical signal into a two-dimensional electrical distribution map based on the spatial distribution information of the detection points.

[0110] At the same time, electrical signals from various detection points in the wetness sensor pad can be received. The electrical signals can be converted into a two-dimensional graph based on the spatial distribution information of the detection points, which is called the electrical distribution graph.

[0111] In practice, the wetness sensing pad is usually a rectangular or other standard shape. Assuming that the wetness sensing pad is laid flat and slight wrinkles are negligible, a two-dimensional coordinate system is constructed for the wetness sensing pad.

[0112] Query the coordinates of each detection point in the space (coordinate system) of the urine wetness sensor pad.

[0113] Based on the initial size diagram of the wetness sensor pad, a two-dimensional electrical distribution map is constructed. That is, the size of the electrical distribution map and the size of the wetness sensor pad are proportionally mapped to each other.

[0114] For the same moment, if the coordinates of a pixel in the electrical distribution map match the coordinates of the detection point (i.e., there is a mapping relationship), then the intensity value of the electrical signal (especially the normalized intensity value) is assigned as the pixel value of the pixel.

[0115] If the coordinates of the pixel in the electrical distribution map are not the same as the coordinates of the detection point, then the specified value (such as 0 or 255) will be assigned as the pixel value of the pixel.

[0116] Step 1022: Stack the multi-frame electrical distribution maps at multiple times according to time to obtain a three-dimensional electrical distribution volume.

[0117] In this embodiment, multiple frames of electrical distribution maps at multiple times are stacked in chronological order to form a three-dimensional electrical distribution body.

[0118] Step 1023: Detect the surface condition of the urine-sensing pad based on the electrical distribution body.

[0119] In this embodiment, the distribution information of the electrical distributor in time and space can be analyzed based on deep learning technology or machine learning technology, thereby detecting the surface state of the urine wetness sensing pad.

[0120] In a practical implementation, a urine wetness detection model can be loaded, and the electrical distributor can be input into the urine wetness detection model to detect the probability that the urine wetness sensing pad is in a wet state.

[0121] The urine-sensing detection model can be used to fine-tune a pre-trained model (three-dimensional convolutional neural network) using urine-wet samples (i.e., electrical distribution bodies), or it can be used to train an independently constructed model using urine-wet samples (i.e., electrical distribution bodies). This embodiment does not impose any restrictions on this.

[0122] Generally, a lightweight model is selected for urine wetness detection to improve the speed of urine wetness detection. To achieve this, the results of multiple tests can be accumulated to obtain a sequence of multiple probabilities arranged in chronological order. The surface state of the urine wetness sensor pad is determined based on the changing trends of multiple probabilities. The multiple probabilities are smoothed to reduce the impact of probability fluctuations, thereby improving the accuracy of urine wetness detection.

[0123] For example, a sliding window is added for multiple probabilities; while sliding the window according to a preset step size, the average value of the probabilities within the window is calculated.

[0124] Sort the average values ​​in order.

[0125] If the difference between the current average value and the previous average value is greater than the preset change threshold, it indicates that the instantaneous change in probability is large, and the surface state of the urine wetness sensor pad is determined to be wet.

[0126] Step 103: If the surface is wet, the headband is invoked to collect EEG signals and head posture data from the user.

[0127] In this embodiment, when the user is lying on the electric nursing bed, the headband can be worn on the user's head. According to predetermined rules (such as timed events), the brain-computer interface in the headband is called at various times to collect the user's electroencephalogram (EEG) and electrooculogram (EOG) signals. The posture sensor in the headband is also called to collect the user's head posture data, which makes it easier to identify the user's operating intentions and control the electric nursing bed.

[0128] Generally, the process of controlling a virtual cursor using an EEG interface is as follows:

[0129] S1. Initialize the virtual cursor.

[0130] The user lies on the electric nursing bed, adjusts their head posture, and faces the center of the controller (especially the display screen). The virtual cursor is moved to the center area to complete the initial position calibration. At the same time, the user's initial head posture data is recorded, which includes the user's head angles in three directions (X0, Y0, Z0).

[0131] S2, Calculate the rotation vector

[0132] When a user's head moves, new head pose data is generated, which includes the user's head angles in three directions (X, Y, Z). t ,Y t Z t ), calculate the user rotation vector (X,Y,Z) of the new head pose data relative to the initial head pose data, where X=X t -X0, Y=Y t -Y0, Z=Z t -Z0.

[0133] S3, Move the virtual cursor

[0134] The direction and distance of the virtual cursor movement are calculated based on the user's rotation vector (X,Y,Z) relative to the initial head pose data.

[0135] The distance moved is: S X =aX,S Y =bY, where a and b are movement coefficients determined by the controller based on the display resolution and control window size. When the user turns their head to the left, S... X When the value is positive, the virtual cursor moves S to the left.X When the user turns their head to the right, S X When the value is negative, the virtual cursor shifts S to the right. X When the user looks up, S Y When the value is positive, the virtual cursor moves upward by S. Y When the user looks down, S Y When the value is negative, the virtual cursor moves down by S. Y .

[0136] In this embodiment, if the surface of the urine-sensing pad is detected to be wet, the EEG signals and head posture data collected by the headband at this time can be reused.

[0137] Step 104: Identify the user's physical state based on EEG signals and head posture data.

[0138] In this embodiment, the user's physical condition can be identified based on EEG signals and head posture data to supplement urine detection and further reduce the possibility of false detection.

[0139] In practical implementation, the physical state includes the state of consciousness and the state of movement; the state of consciousness is a binary category, including conscious and impaired consciousness (i.e., not conscious), and the state of movement is a binary category, including turning and not turning.

[0140] Therefore, a loadable state detection model can be used; such as Figure 3 As shown, the state detection model includes a first convolutional module, a second convolutional module, a third convolutional module, a first head structure, and a second head structure.

[0141] Among them, the first convolution module, the second convolution module, and the third convolution module are all modules that encapsulate convolution-related structures (such as convolutional layers, pooling layers, etc.). They can be used to fine-tune the structure for extracting features from third-party pre-trained models (such as ResNet (residual network)) or to train independently built structures.

[0142] Both the first-head and second-head structures are modules used for binary classification, containing fully connected layers, activation functions, and other structures.

[0143] The EEG signal was transformed by performing a continuous wavelet transform (CWT) to obtain the original EEG features (time-frequency map).

[0144] The original EEG features are input into the first convolution module to extract features in the time-frequency domain, thus obtaining the first candidate EEG features.

[0145] The first candidate EEG features are input into the first head structure for binary classification to obtain the user's state of consciousness.

[0146] If the consciousness state is conscious, functions such as Reshape are used to convert the head pose data into a two-dimensional pose matrix.

[0147] The first candidate EEG features are input into the second convolutional module to continue extracting time-frequency domain features, thus obtaining the second candidate EEG features.

[0148] The pose matrix is ​​input into the third convolution module to extract features, thus obtaining pose features.

[0149] The second candidate EEG features and pose features are fused into multimodal features using functions such as Concat (concatenation) and Add (addition).

[0150] The multimodal features are input into the second head structure for binary classification to obtain the user's movement state.

[0151] In this embodiment, the state detection model has a simple structure, low computational load, and fast response speed, making it suitable for correcting urine wetness detection.

[0152] Step 105: Execute the wetness alarm operation according to your physical condition.

[0153] In this embodiment, a wetness alarm can be triggered at an appropriate time based on the user's physical condition.

[0154] In practice, if the user is in a state of impaired consciousness or is in a state of full consciousness and is not turning around, the user will have little impact on the wetness sensor pad and the probability of the user wetting the pad is high. In this case, an alarm signal indicating that the pad is wet will be generated and sent to the designated nursing management device. The nursing management device will issue an alarm through sound and light or provide a prompt on the user interface so as to notify the nursing staff to perform subsequent cleaning.

[0155] Nursing management equipment includes alarms, nurse workstations, and so on.

[0156] Conversely, when the user is conscious and turning around, the wetness sensor may experience wrinkles due to the soft pad, potentially leading to false alarms. Therefore, the wetness detection should continue.

[0157] In this embodiment, the electric nursing bed is equipped with a urine-sensing pad on its surface and a headband that receives electrical signals collected from various detection points on the urine-sensing pad. The surface state of the pad is detected based on the temporal and spatial distribution of these electrical signals. If the surface is wet, the headband is activated to collect the user's electroencephalogram (EEG) signals and head posture data. The user's physical condition is identified based on the EEG signals and head posture data. A urine-sensing alarm is then triggered based on this physical condition. This embodiment detects urine wetness by utilizing the temporal and spatial distribution of electrical signals on the pad, increasing the amount of information involved and improving accuracy. Furthermore, the use of the user's EEG signals and head posture data for correction effectively reduces the probability of false alarms and minimizes interference with the user and caregivers.

[0158] It should be noted that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0159] Reference Figure 4 The diagram illustrates a urine detection device for an electric nursing bed according to an embodiment of the present invention. The electric nursing bed is covered with a urine-sensing pad and is equipped with a headband. The device may specifically include the following modules:

[0160] The electrical signal receiving module 401 is used to receive electrical signals collected at various detection points in the urine wetness sensing pad;

[0161] Surface state detection module 402 is used to detect the surface state of the urine wetness sensor pad based on the distribution information of the electrical signal in time and space.

[0162] User data acquisition module 403 is used to call the headband to collect EEG signals and head posture data from the user if the surface is wet.

[0163] The body state recognition module 404 is used to recognize the user's body state based on the electroencephalogram (EEG) signal and the head posture data.

[0164] The urine wetness alarm module 405 is used to perform a urine wetness alarm operation based on the said physical condition.

[0165] In one embodiment of the present invention, the surface state detection module 402 is further configured to:

[0166] For the same moment, the electrical signal is converted into a two-dimensional electrical distribution map based on the spatial distribution information of the detection points;

[0167] The electrical distribution maps of multiple times are stacked in time to obtain a three-dimensional electrical distribution volume;

[0168] The surface condition of the urine-sensing pad is detected based on the electrical distributor.

[0169] In one embodiment of the present invention, the surface state detection module 402 is further configured to:

[0170] Query the coordinates of the detection points distributed in the space of the urine-sensing pad;

[0171] A two-dimensional electrical distribution map is constructed based on the dimensions of the urine-sensing pad;

[0172] At any given moment, if the coordinates of a pixel in the electrical distribution map match the coordinates of the detection point, then the intensity value of the electrical signal is assigned as the pixel value of that pixel.

[0173] If the coordinates of the pixel in the electrical distribution map are not the same as the coordinates of the detection point, then the specified value is assigned as the pixel value of the pixel.

[0174] In one embodiment of the present invention, the surface state detection module 402 is further configured to:

[0175] Load the urine moisture sensing detection model;

[0176] The electrical distributor is input into the urine wetness detection model to detect the probability that the urine wetness sensing pad is in a wet state;

[0177] The surface state of the urine-sensing pad is determined to be wet based on the changing trends of multiple probabilities.

[0178] In one embodiment of the present invention, the surface state detection module 402 is further configured to:

[0179] Add windows to multiple of the probabilities;

[0180] During the sliding of the window, the average value of the probabilities within the window is calculated;

[0181] If the difference between the current average value and the previous average value is greater than a preset change threshold, then the surface state of the urine wetness sensor pad is determined to be wet.

[0182] In one embodiment of the present invention, the body state recognition module 404 is further configured to:

[0183] Load a state detection model; the state detection model includes a first convolutional module, a second convolutional module, a third convolutional module, a first head structure, and a second head structure;

[0184] The EEG signal was converted by performing continuous wavelet transformation to obtain the original EEG features;

[0185] The original EEG features are input into the first convolutional module to extract the first candidate EEG features;

[0186] The first candidate EEG features are input into the first head structure for binary classification to obtain the user's state of consciousness.

[0187] If the state of consciousness is conscious, then the head posture data is converted into a two-dimensional posture matrix;

[0188] The first candidate EEG features are input into the second convolution module to extract the second candidate EEG features;

[0189] The pose matrix is ​​input into the third convolutional module to extract pose features;

[0190] The second candidate EEG features are fused with the posture features to form a multimodal feature;

[0191] The multimodal features are input into the second head structure for binary classification to obtain the user's movement state.

[0192] In one embodiment of the present invention, the urine-wet alarm module 405 is further configured to:

[0193] If the state of consciousness is impaired or the movement state is non-turning, an alarm signal indicating that urine is wet is generated.

[0194] The alarm signal is sent to the designated nursing management device.

[0195] The present invention provides a urine wetness detection device for an electric nursing bed. By using the urine wetness detection device for an electric nursing bed, the various steps in the aforementioned urine wetness detection method embodiments for electric nursing beds can be realized.

[0196] It should be noted that the module division in the various electric nursing bed wetness detection devices provided in the above embodiments is illustrative and only represents a logical functional division. In actual implementation, other division methods may also be used. Furthermore, the functional modules in the various embodiments of this invention can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0197] If the integrated module is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the technical solution of the embodiments of the present invention can be embodied in the form of a computer program product, which is stored in a computer storage medium and includes several instructions to cause an electric nursing bed or processor to execute all or part of the steps of the methods in the various embodiments of the present invention. The aforementioned computer storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0198] Furthermore, the urine wetness detection device for the electric nursing bed provided in the above embodiments and the urine wetness detection method embodiment for the electric nursing bed belong to the same concept, and the specific implementation process can be found in the method embodiment, which will not be repeated here.

[0199] Reference Figure 5 The diagram illustrates an electric nursing bed provided by an embodiment of the present invention. Figure 5 As shown, the electric nursing bed in this embodiment of the invention includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps in the above-described embodiment of the urine wetness detection method for the electric nursing bed. Alternatively, when the processor executes the computer program, it implements the functions of each module in the above-described embodiment of the urine wetness detection device for the electric nursing bed.

[0200] For example, the computer program may be divided into one or more modules, which are stored in the memory and executed by the processor to complete this application. The one or more modules may be a series of computer program instruction segments capable of performing a specific function, which can be used to describe the execution process of the computer program in the electric nursing bed.

[0201] The electric nursing bed may include computing devices such as desktop computers and cloud servers. The electric nursing bed may include, but is not limited to, processors and memory. Those skilled in the art will understand that... Figure 5 This is merely one example of an electric nursing bed and does not constitute a limitation on electric nursing beds. It may include more or fewer components than shown, or combine certain components, or different components. For example, the electric nursing bed may also include input / output devices, network access devices, buses, etc.

[0202] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0203] The memory can be an internal storage unit of the electric nursing bed, such as a hard drive or RAM. Alternatively, it can be an external storage device, such as a plug-in hard drive, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, etc., provided on the electric nursing bed. Furthermore, the memory can include both internal and external storage units. The memory is used to store the computer program and other programs and data required by the electric nursing bed. The memory can also be used to temporarily store data that has been output or will be output.

[0204] This invention also discloses an electric nursing bed, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the urine wetness detection method of the electric nursing bed as described in the foregoing embodiments.

[0205] This invention also discloses a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method for detecting urine wetness in an electric nursing bed as described in the foregoing embodiments.

[0206] This invention also discloses a computer program product that, when run on a computer, causes the computer to execute the urine detection method for the electric nursing bed described in the foregoing embodiments.

[0207] The embodiments described above are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for detecting urine wetness on an electric nursing bed, characterized in that, The electric nursing bed is equipped with a urine-sensing pad on its bed frame and a headband. The method includes: Receive electrical signals collected at various detection points in the urine-sensing pad; The surface state of the urine-sensing pad is detected based on the distribution information of the electrical signal in time and space; If the surface is wet, the headband is invoked to collect EEG signals and head posture data from the user. The user's physical state is identified based on the electroencephalogram (EEG) signals and the head posture data; The urine-wet alarm will be activated based on the described physical condition.

2. The method according to claim 1, characterized in that, The step of detecting the surface state of the urine-sensing pad based on the temporal and spatial distribution information of the electrical signal includes: For the same moment, the electrical signal is converted into a two-dimensional electrical distribution map based on the spatial distribution information of the detection points; The electrical distribution maps of multiple times are stacked in time to obtain a three-dimensional electrical distribution volume; The surface condition of the urine-sensing pad is detected based on the electrical distributor.

3. The method according to claim 2, characterized in that, The step of converting the electrical signal into a two-dimensional electrical distribution map based on the spatial distribution information of the detection points at the same time includes: Query the coordinates of the detection points distributed in the space of the urine-sensing pad; A two-dimensional electrical distribution map is constructed based on the dimensions of the urine-sensing pad; At any given moment, if the coordinates of a pixel in the electrical distribution map match the coordinates of the detection point, then the intensity value of the electrical signal is assigned as the pixel value of that pixel. If the coordinates of the pixel in the electrical distribution map are not the same as the coordinates of the detection point, then the specified value is assigned as the pixel value of the pixel.

4. The method according to claim 2, characterized in that, The method of detecting the surface state of the urine-sensing pad based on the electrical distributor includes: Load the urine moisture sensing detection model; The electrical distributor is input into the urine wetness detection model to detect the probability that the urine wetness sensing pad is in a wet state; The surface state of the urine-sensing pad is determined to be wet based on the changing trends of multiple probabilities.

5. The method according to claim 4, characterized in that, Determining whether the surface state of the urine-sensing pad is wet based on the changing trends of multiple probabilities includes: Add windows to multiple of the probabilities; During the sliding of the window, the average value of the probabilities within the window is calculated; If the difference between the current average value and the previous average value is greater than a preset change threshold, then the surface state of the urine wetness sensor pad is determined to be wet.

6. The method according to any one of claims 1-5, characterized in that, The physical state includes a state of consciousness and a state of movement; the process of identifying the user's physical state based on the electroencephalogram (EEG) signals and the head posture data includes: Load a state detection model; the state detection model includes a first convolutional module, a second convolutional module, a third convolutional module, a first head structure, and a second head structure; The EEG signal was converted by performing continuous wavelet transformation to obtain the original EEG features; The original EEG features are input into the first convolutional module to extract the first candidate EEG features; The first candidate EEG features are input into the first head structure for binary classification to obtain the user's state of consciousness. If the state of consciousness is conscious, then the head posture data is converted into a two-dimensional posture matrix; The first candidate EEG features are input into the second convolution module to extract the second candidate EEG features; The pose matrix is ​​input into the third convolutional module to extract pose features; The second candidate EEG features are fused with the posture features to form a multimodal feature; The multimodal features are input into the second head structure for binary classification to obtain the user's movement state.

7. The method according to claim 6, characterized in that, The operation of performing the wetness alarm based on the physical condition includes: If the state of consciousness is impaired or the movement state is non-turning, an alarm signal indicating that urine is wet is generated. The alarm signal is sent to the designated nursing management device.

8. A urine detection device for an electric nursing bed, characterized in that, The electric nursing bed is equipped with a urine-sensing pad on its bed frame and a headband. The device includes: An electrical signal receiving module is used to receive electrical signals collected at various detection points in the urine-sensing pad. A surface condition detection module is used to detect the surface condition of the urine wetness sensor pad based on the distribution information of the electrical signal in time and space. The user data acquisition module is used to call the headband to collect EEG signals and head posture data from the user if the surface is wet. A body state recognition module is used to identify the user's body state based on the electroencephalogram (EEG) signals and the head posture data. The urine wetness alarm module is used to perform urine wetness alarm operations based on the said physical condition.

9. An electric nursing bed, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method for detecting urine wetness in an electric nursing bed as described in any one of claims 1-7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the method for detecting urine wetness of the electric nursing bed as described in any one of claims 1-7.

Images