Physiological parameter monitoring method and system in disaster emergency scene and electronic equipment

Abstract

The invention relates to the field of physiological parameter monitoring, and discloses a physiological parameter monitoring method and system in a disaster emergency scene and electronic equipment. According to the method, multi-channel physiological signals such as surface electromyogram, electrocardiogram, respiratory frequency and body temperature of a disaster victim are collected through a wearable sensor array, real-time noise reduction and filtering processing are carried out on the multi-channel physiological signals, and environmental noise is removed. The processed signals are fused, multi-signal crosstalk processing is carried out through the frequency division multiplexing technology, crosstalk between different physiological signals is recognized and eliminated, and finally pre-disaster early warning is output based on the processed signals. Various filters used in filtering processing and specific modes of signal fusion and crosstalk processing are also introduced. A corresponding monitoring system adopts the method to process data, electronic equipment can execute the monitoring method, and an effective physiological parameter monitoring scheme is provided for disaster first aid.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of physiological parameter monitoring, in particular a physiological parameter monitoring method, system and electronic device in a disaster first aid scene. BACKGROUND

[0002] In a disaster first aid scene, it is crucial to quickly and accurately obtain physiological parameters of the wounded to formulate treatment strategies. Traditional physiological monitoring methods face multiple challenges in disaster environments, such as complex noise interference. There is a lot of environmental noise in the disaster scene, such as electromagnetic interference, which overlaps with the frequency band of physiological signals, seriously affecting signal quality. For example, in earthquake rescue, surface electromyography (sEMG) is easily overwhelmed by vibration noise from rescue equipment, making it difficult to assess muscle function. Multi-signal crosstalk is intensified, and disasters cause changes in the body position of the wounded and displacement of sensors, making the crosstalk problem between different physiological signals more prominent. For example, the sEMG signal collected by the chest sensor is often disturbed by electrocardiogram (ECG), and traditional filtering methods are difficult to effectively remove ECG interference while preserving the low-frequency components of sEMG. At the same time, disaster first aid requires the monitoring system to provide reliable data in a short time, but existing deep learning models (such as FCN, U-Net) are insufficient in inference speed and anti-interference ability in complex scenarios, resulting in delayed treatment opportunities. SUMMARY

[0003] Therefore, it is necessary to provide a physiological parameter monitoring method and system in a disaster first aid scene to improve the monitoring accuracy of physiological parameters in a disaster first aid scene.

[0004] A physiological parameter monitoring method in a disaster first aid scene, characterized in that it comprises the following steps:

[0005] Collecting multi-channel physiological signals of disaster victims through a wearable sensor array, the physiological signals including surface electromyography (sEMG), electrocardiogram (ECG), respiratory rate signal, and body temperature data;

[0006] Performing real-time noise reduction and filtering processing on the collected physiological signals to remove environmental noise;

[0007] Performing multi-signal crosstalk processing on the multi-channel physiological signals after noise reduction and filtering processing to identify and eliminate crosstalk between different physiological signals;

[0008] Performing signal fusion on the multi-channel physiological signals after crosstalk elimination;

[0009] Outputting a disaster warning based on the fused signals.

[0010] In one embodiment, the filtering processing of the collected physiological signal specifically comprises: removing baseline drift noise of the physiological signal using a filter, and a frequency response of the filter is:

[0011]

[0012] wherein Q(e jω ) is a frequency response of the filter, representing amplitude and phase change values of an input signal at different frequencies ω after passing through the filter; ω is an angular frequency, reflecting frequency characteristics of the signal; j is an imaginary unit; ω c is a cutoff frequency, used to divide a frequency range allowed to pass and blocked by the filter, when 0<∣ω∣<ω c , the filter output is 0, that is, the signal in the frequency range is completely blocked, when ω c <∣ω∣<π, the filter output is 1, and the signal can pass without attenuation.

[0013] In one embodiment, an impulse response of the filter is:

[0014]

[0015] wherein h(n) is an impulse response of the filter, used to describe an output response of the filter to a unit impulse input; n represents an index of a discrete time sequence, used to determine impulse response values at different time instants.

[0016] In one embodiment, the collected physiological signal is subjected to real-time noise reduction and filtering processing, comprising: removing noise using a notch filter, and a transfer function of the notch filter is:

[0017]

[0018] wherein H(z) is a transfer function of the filter based on z transform, used to describe a relationship between an input and an output of the filter, z is a complex variable, z -1 represents that a signal is delayed by one sampling period in time; z -2 represents that a signal is delayed by two sampling periods in time; is a zero point position of the filter on a complex plane, used to determine attenuation characteristics of the filter to a specific frequency signal; is a pole position of the filter on the complex plane; ω0 is an angular frequency corresponding to an interference frequency, representing frequency characteristics to be suppressed; r is a radius of the pole, used to adjust a notch bandwidth of the filter, 0<r<1, the closer r is to 1, the narrower the notch bandwidth is.

[0019] In one embodiment, the collected physiological signals are subjected to real-time noise reduction and filtering processing, including: removing electromyographic noise by using a filter, and the output signal of the filter is:

[0020]

[0021] wherein y(t) is the output signal of the filter at time t; x(t-k) is the sampling value of the input signal at time t-k; b k is the filter coefficient, k = 0, 1, 2, …, M, and M is the order of the filter.

[0022] In one embodiment, the signal fusion step includes: performing time synchronization processing on the filtered physiological signals, and fusing the time-synchronized physiological signals according to a preset signal fusion algorithm.

[0023] Before fusing the filtered physiological signals, the method further includes: collecting facial action unit data by video acquisition, and fusing the physiological signals and the facial action unit data after synchronization; first calculating the frame number of the video sequence, and sampling the physiological signals to make them consistent with the video frame number, so as to realize one-to-one correspondence between the physiological signals and the facial action unit data, and then fusing.

[0024] In one embodiment, the multi-channel physiological signals subjected to noise reduction and filtering processing are subjected to multi-signal crosstalk processing to identify and eliminate crosstalk between different physiological signals, including: using frequency division multiplexing technology, using a specific transconductance amplifier shown in the following formula to assign different modulation frequencies to each physiological signal, and realizing frequency modulation of the multi-channel physiological signals.

[0025]

[0026] In the back-end processing, a demodulation frequency corresponding to the modulation frequency is used for demodulation to identify and eliminate crosstalk between different physiological signals.

[0027] Meanwhile, frequency modulation is performed according to the following method:

[0028] f chN +(2×BW SIG +BW GB )≤3×f ch1

[0029]

[0030] So that the signals of different frequency channels do not overlap.

[0031] Wherein G mB is the transconductance of the transconductance amplifier.

[0032] g m1 gm

[0033] R F Rf

[0034] r n0 Rd

[0035] g mA gm

[0036] g mB gm

[0037] f ch1 and f chN are the lowest and highest modulation frequencies, respectively, used to determine the range of frequency channels;

[0038] BW SIG is the signal bandwidth value, representing the frequency range of the physiological signals;

[0039] BW GB is the guard band interval value, used to avoid interference between adjacent frequency channels;

[0040] N is the maximum number of channels, representing the number of physiological signals that the system can handle simultaneously.

[0041] In one embodiment, the fused signal is input into a trans-impedance amplifier (TIA). To ensure that the TIA can accurately process the fused signal, the input dynamic range of the TIA is adjusted according to the following formula:

[0042] (I ch1 +I ch2 +…+I chN )×α TIA ≤V max

[0043]

[0044] where I ch1 , I ch2 , …, I chN are the output currents of each channel, representing the intensity of each physiological signal;

[0045] α TIA is the current-voltage gain of the TIA, determining the proportion of current signal converted to voltage signal;

[0046] V maxis the maximum input voltage of the TIA;

[0047] I ch is the output current of a single channel;

[0048] N is the maximum number of channels, representing the number of physiological signal channels that can be processed.

[0049] The application also provides a physiological parameter monitoring system in a disaster first aid scene, the system comprising a wearable sensor array, and the system adopts the physiological parameter monitoring method in a disaster first aid scene as described above for data processing.

[0050] An electronic device, comprising:

[0051] one or more processors;

[0052] a memory;

[0053] one or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs are configured to execute any of the above physiological parameter monitoring methods in a disaster first aid scene.

[0054] The above physiological parameter monitoring method in a disaster first aid scene comprehensively collects multi-channel physiological signals of disaster victims including surface electromyography, electrocardiogram, respiratory rate and body temperature data by using a wearable sensor array, and uses various filtering techniques to process different noises. For example, a filter with a specific frequency response is used to remove baseline drift noise, a second-order infinite impulse response notch filter is used to suppress noise, and an n-point moving average filter is used to remove electromyographic noise, thereby improving signal quality. The filtered physiological signals are time-synchronized and fused with facial action unit data, and each physiological signal is assigned a modulation frequency by means of frequency division multiplexing technology. The backend uses a corresponding demodulation frequency to eliminate crosstalk, reasonably plans the frequency range to avoid signal overlap, and improves monitoring accuracy. The input dynamic range of the transimpedance amplifier (TIA) is adjusted by the above method to ensure that it accurately processes the fused signals and improves system stability and reliability. The monitoring method is applied to a monitoring system and an electronic device in a disaster first aid scene, which can obtain physiological parameters in time for disaster warning, assist rescue decision-making, improve rescue efficiency, and is suitable for various complex disaster scenes, and has good application prospects. BRIEF DESCRIPTION OF DRAWINGS

[0055] Figure 1 is an example of a physiological parameter monitoring method flowchart in a disaster first aid scene;

[0056] Figure 2 is a block diagram of an example of an electronic device. DETAILED DESCRIPTION

[0057] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not to limit the present application.

[0058] In the embodiments of the present application, the words "first", "second", etc. are used to distinguish the same or similar items with basically the same function, "at least one" means one or more, and "multiple" means two or more, for example, multiple objects refer to two or more objects. The words "include" or "contain" and the like mean that the information appearing before "include" or "contain" covers the information listed after "include" or "contain" and its equivalents, and does not exclude other information. In the embodiments of the present application, "and / or" means that there can be three kinds of relationships, and the character " / " generally means that the objects before and after it are in an "or" relationship.

[0059] Reference Figure 1 As shown in the figure, Figure 1 is an example of a physiological parameter monitoring method implementation flowchart in a disaster first aid scene, as shown in Figure 1 In the figure, multi-channel physiological signals of disaster victims are collected by a wearable sensor array, and the physiological signals include surface electromyography signals, electrocardiogram signals, respiratory rate signals and body temperature data;

[0060] The collected physiological signals are subjected to real-time noise reduction and filtering processing to remove environmental noise;

[0061] The multi-channel physiological signals subjected to noise reduction and filtering processing are subjected to multi-signal crosstalk processing to identify and eliminate crosstalk between different physiological signals;

[0062] The multi-channel physiological signals subjected to crosstalk elimination are subjected to signal fusion;

[0063] Disaster warning is output based on the fused signals.

[0064] The technical solutions of the above embodiments effectively remove noise and crosstalk through a multi-stage signal processing flow, improve physiological signal quality and analysis accuracy, and improve monitoring comprehensiveness and reliability by integrating multi-sensor data.

[0065] As an embodiment, the collected physiological signals are subjected to real-time noise reduction and filtering processing, including: removing baseline drift noise of the physiological signals using a filter, and the frequency response of the filter is:

[0066]

[0067] Wherein, Q(e jωω is the frequency response of the filter, representing the amplitude and phase changes of the input signal at different frequencies ω after passing through the filter; ω is the angular frequency, reflecting the frequency characteristics of the signal; j is the imaginary unit; ω c The cutoff frequency is used to divide the frequency range that the filter allows to pass through and blocks, when 0 < |ω| < ω c When ω = 0, the filter output is 0, meaning that signals within that frequency range are completely blocked. c When |ω| < π, the filter output is 1, and the signal can pass through without attenuation.

[0068] The technical solution of the above embodiment removes baseline drift noise through a filter, which can accurately and effectively filter out low-frequency baseline drift noise (usually around 0.5Hz) while ensuring that high-value high-frequency components such as P waves, QRS groups and T waves in the ECG signal are not affected, thereby significantly improving the accuracy of subsequent feature extraction and analysis of the ECG signal.

[0069] As an example, the impulse response of the filter is:

[0070]

[0071] Where h(n) is the impulse response of the filter, which describes the output response of the filter to a unit impulse input; n represents the index of the discrete time sequence, which is used to determine the impulse response value at different times.

[0072] The technical solution of the above embodiments, from a discrete-time perspective, treats the physiological signal input to the filter (which can be viewed as a superposition of a series of discrete pulse sequences). The filter processes the signal at different times according to h(n), and selects signals of different frequency components by performing convolution operations between the pulse response and the input signal in the discrete system. High-frequency signals that conform to the filtering frequency response characteristics are allowed to pass through, while low-frequency signals are suppressed, thereby achieving the purpose of removing baseline drift noise and retaining effective physiological signal components, laying the foundation for subsequent accurate analysis of physiological parameters.

[0073] As an example, the acquired physiological signals are subjected to real-time noise reduction and filtering, including: noise removal using a filter, wherein the transfer function of the filter is:

[0074]

[0075] Where H(z) is the transfer function of the filter based on the z-transform, used to describe the relationship between the filter input and output, and z is a complex variable. -1 This indicates that the signal is delayed by one sampling period in time; z -2 This indicates that the signal is delayed by two sampling periods in time; is the zero point position of the filter on the complex plane, used to determine the attenuation characteristics of the filter to a specific frequency signal; is the pole position of the filter on the complex plane; ω0is the angular frequency corresponding to the interference frequency, representing the frequency characteristics that need to be suppressed; r is the radius of the pole, used to adjust the notch bandwidth of the filter, 0 < r < 1, the closer r is to 1, the narrower the notch bandwidth.

[0076] The technical solutions of the above embodiments use a second-order infinite impulse response notch filter to remove noise, can construct a deep attenuation notch for a frequency of 50Hz or 60Hz, effectively eliminate interference, and at the same time, minimize the influence on adjacent frequency ECG signal components. By adjusting the pole radius r, the notch bandwidth can be flexibly controlled, the best balance between suppressing interference and maintaining signal integrity is achieved, and the detection capability of weak waveform signals (such as P wave and T wave) in the ECG signal is improved.

[0077] As an embodiment, the collected physiological signals are subjected to real-time noise reduction and filtering processing, including: the collected physiological signals are subjected to real-time noise reduction and filtering processing, and an electromyographic noise is removed by using a filter, and an output signal of the filter is:

[0078]

[0079] Wherein, y(t) is an output signal of the filter at time t; x(t-k) is a sampling value of the input signal at time t-k; b k is a filter coefficient, k = 0, 1, 2, …, M, and M is the order of the filter.

[0080] In the above embodiment, the electromyographic (EMG) noise is removed, which can effectively smooth the high-frequency electromyographic noise. By performing average operation on multiple sampling points, the influence of random noise can be significantly reduced, while the main waveform signal characteristics of the ECG signal are retained. Moreover, the filter implementation is simple, the computational complexity is low, the signal can be quickly processed in the real-time monitoring system, the consumption of system resources is reduced, and the real-time and accuracy of monitoring are ensured.

[0081] As an embodiment, the signal fusion specifically includes: performing time synchronization processing on the multiple-channel physiological signals, and fusing the signals after time synchronization according to a preset signal fusion algorithm;

[0082] Before fusing the multiple physiological signals subjected to filtering processing, the method further includes: collecting video of a facial action unit data, and fusing the physiological signals and the facial action unit data after synchronization; first, the number of frames of the video sequence is calculated, and the physiological signals are sampled to be consistent with the number of video frames, so that the physiological signals and the facial action unit data are one-to-one corresponding, and then the fusion is performed.

[0083] In the above embodiment, the physiological signals are fused with the facial action unit data after synchronization, which fully combines the advantages of the two kinds of data in pain recognition. The facial action unit can intuitively reflect the expression changes related to pain, and the physiological signal can reflect the physiological response inside the body caused by pain. Synchronous fusion makes them complement each other, greatly improving the accuracy of pain recognition. Experiments show that after fusion, the recognition accuracy of physiological indicators in all subject experiments is significantly improved, and the effect is significantly improved compared with using physiological signals alone.

[0084] As an embodiment, the multi-channel physiological signals after noise reduction and filtering processing are subjected to multi-signal crosstalk processing, and the crosstalk between different physiological signals is recognized and eliminated, including: using frequency division multiplexing technology, using a specific transconductance amplifier shown in the following formula to assign different modulation frequencies to each physiological signal, and realizing frequency modulation of the multi-channel physiological signals;

[0085]

[0086] In the back-end processing, the demodulation frequency corresponding to the modulation frequency is used for demodulation, so as to recognize and eliminate the crosstalk between different physiological signals;

[0087] At the same time, frequency modulation is carried out according to the following way:

[0088] f chN +(2×BW SIG +BW GB )≤3×f ch1

[0089]

[0090] So that the signals of different frequency channels do not overlap;

[0091] Wherein, G mB is the transconductance of the transconductance amplifier;

[0092] g m1 is the transconductance of the output stage MOS tube in the transconductance amplifier;

[0093] R F is the resistance for the source negative feedback of the output stage MOS tube in the transconductance amplifier, which adjusts the performance of the amplifier through negative feedback;

[0094] r n0 is the drain resistance of the output stage MOS tube in the transconductance amplifier;

[0095] g mA is the transconductance of the negative feedback transconductance stage;

[0096] g mBto output the transconductance of the transconductance stage;

[0097] f ch1 and f chN are the lowest and highest modulation frequencies, respectively, used to determine the range of frequency channels;

[0098] BW SIG is a signal bandwidth value representing the frequency range of the physiological signals;

[0099] BW GB is a guard band interval value used to avoid interference between adjacent frequency channels;

[0100] N is the maximum number of channels, representing the number of physiological signals that the system can process simultaneously.

[0101] In the above embodiment, by using the above transconductance amplifier and frequency design, the crosstalk between multiple signals can be effectively reduced, and the separation degree and accuracy of the signals can be improved. Through accurate frequency modulation and demodulation, each physiological signal can be transmitted in a separate frequency channel, reducing the mutual interference between signals, thereby improving the quality of physiological signal acquisition and providing a reliable data basis for subsequent accurate physiological parameter analysis. Moreover, according to the above method, the system bandwidth resources can be fully utilized to support more physiological signal channels in a limited bandwidth, improving the integration and practicality of the system.

[0102] As an embodiment, the fused signal is input into a transimpedance amplifier (TIA) to ensure that the TIA can accurately process the fused signal. The input dynamic range of the TIA is adjusted according to the following formula:

[0103] (I ch1 +I ch2 +…+I chN )×α TIA ≤V max

[0104]

[0105] where I ch1 , I ch2 , …, I chN are the output currents of each channel, representing the intensity of each physiological signal.

[0106] α TIA is the current-voltage gain of the TIA, determining the proportion of current signal conversion to voltage signal;

[0107] V max is the maximum input voltage of the TIA;

[0108] I ch is the output current of a single channel;

[0109] N is the maximum number of channels, indicating the number of physiological signal channels that can be processed.

[0110] In the above embodiment, the TIA input dynamic range is reasonably designed according to the above method, which can prevent saturation distortion when the backend trans-impedance amplifier processes the fusion signal, ensure signal integrity, and provide protection for subsequent accurate analysis. In actual physiological signal monitoring, it is ensured that weak signals can be effectively amplified, and strong signals will not lose information due to overload. At the same time, the maximum number of physiological signal channels that the system can handle is determined comprehensively according to the above two formulas, and the number of channels is limited from the frequency and circuit two key aspects, which can make the system maintain stable operation under the condition of complex physiological signal input. Reduce system failure caused by signal interference or circuit overload, and ensure the continuous and reliable monitoring of physiological parameters of the system in disaster emergency and other scenes.

[0111] At the same time, the TIA input dynamic range is reasonably designed to enable the system to stably process physiological signal inputs of different intensities, improve the adaptability to various physiological signals, reduce measurement errors caused by changes in signal intensity, and enhance the overall stability of the system. The intensity of human physiological signals will change greatly with the state of human motion, and the above method can ensure stable and reliable monitoring data.

[0112] When the physiological signal includes a surface electromyography signal (sEMG) and the noise includes electrocardiogram signal (ECG) interference, a fractional-based diffusion model is used to denoise the sEMG signal;

[0113] The fractional-based diffusion model is trained in the following manner:

[0114] During the diffusion process, isotropic Gaussian noise is gradually added to the pure sEMG signal x0 from time t-1 to time t, and the noise addition process satisfies the following rules:

[0115]

[0116] In the formula, q(x t |x t-1 ) represents the probability distribution of the current time signal x t-1 given the previous time signal x t ; F represents a Gaussian distribution, β t is the noise scheduling parameter at time t, which controls the intensity of noise addition; is the mean of the Gaussian distribution, which reflects the influence of the previous time signal x t-1 on the mean of the current time signal x t ; I is an identity matrix, indicating that the noise is independently and identically distributed in each dimension; x tis the current time signal after adding noise at time t, x t-1 is the previous time signal.

[0117] By continuously repeating this noise adding process (i.e. continuously adding at multiple times), the pure signal can be gradually converted into a noise-dominant signal. One of the goals of subsequent model training is to learn how to reverse this process and recover the original sEMG signal from the noise in order to better use it for electromyography analysis, motion recognition and other related tasks.

[0118] The noise scheduling parameter β t A cosine scheduler is used to generate:

[0119]

[0120] where T is the total number of times, s = 0.008 is an offset parameter for controlling the noise growth rate;

[0121] In the reverse process, the sEMG signal x' contaminated by ECG is defined as x' = x0 + γ × e, where x0 is the pure sEMG signal, e is the ECG interference signal, and γ is the interference intensity coefficient, and γ ∈ [0, 1];

[0122] The contaminated sEMG signal x ′ and the noise scale variable As a condition for the neural network, the Gaussian noise is restored to the pure sEMG signal or segment.

[0123] The above model can learn the noise distribution under different γ ′ , and when the ECG interference intensity changes dynamically (such as electrode displacement caused by movement), it can still maintain stable signal-to-noise ratio improvement. At the same time, by taking x ′ as a condition, the model is exposed to a variety of contaminated samples during training, and has stronger generalization ability for unseen γ values, thereby reducing the false detection rate.

[0124] The step of denoising the sEMG signal using a fractional-based diffusion model includes:

[0125] The noise scale parameter in the noise scale variable and the contaminated sEMG signal x ′ are input as conditions to the fractional-based diffusion model;

[0126] T sampling steps are performed by the fractional-based diffusion model to gradually restore the Gaussian noise to the pure sEMG signal, where each sampling step generates the next sample x t based on the previous sample x t-1 and the condition:

[0127]

[0128] wherein α t = 1 - β t , z ~ N(0, 1) is random noise; ∈ θ (x t , a t ', x') is a neural network's prediction of the noise, which takes x θ , a t ', x' as input and is computed by: t

[0129] ∈ θ (x t , a t ', x') = U Net(x t , Condition(a t ', x')), wherein Condition() is a conditional embedding function that converts a scalar parameter into a feature vector, and U Net is a convolutional neural network architecture representing a noise predictor using Unet structure to predict the noise for the input;

[0130] The neural network architecture of the fractional-based diffusion model comprises an encoder and a decoder, the encoder down-samples the input signal to extract features, and the decoder reconstructs the pure sEMG signal based on the extracted features;

[0131] The encoder contains L = 5 convolutional layers, each with a filter number of C v = C0*2 v / L (C0= 64), and followed by a GELU activation function and a Group Normalization; wherein "v" is a convolutional layer level index variable, which takes a value ranging from 1 to 5 (i.e. L = 5). The decoder contains L = 5 deconvolutional layers, which uses skip connection to concatenate the feature maps of the corresponding layers of the encoder and the decoder in the channel dimension;

[0132] wherein, is the output of the v-th layer of the encoder, is the output of the v+1-th layer of the decoder, is the concatenation of the output of the v-th layer and the output of the v+1-th layer, fusing the features of both; and each deconvolutional layer is followed by a GELU activation function, denotes a transpose convolution operation on the concatenated features, i.e. the output of the v-th layer of the decoder

[0133] ​Through the above model, the noise can be more accurately estimated, the noise suppression capability is improved under the condition of low signal-to-noise ratio, the gradient vanishing problem is reduced, the training stability of the model is maintained in the deep network, and the convergence speed is improved; the encoder-decoder structure of L=5 forms 5-level feature extraction, the shallow layer captures high-frequency electromyographic details (100-500Hz), the deep layer extracts muscle activation patterns (<40Hz), and the feature expression capability is improved.

[0134] Embodiments of an electronic device are described below.

[0135] The present application provides a technical solution of an electronic device to realize physiological parameter monitoring related functions in a disaster first aid scene.

[0136] In one embodiment, the present application provides an electronic device, which comprises:

[0137] one or more processors;

[0138] a memory;

[0139] one or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs are configured for the physiological parameter monitoring method in the disaster first aid scene of any embodiment.

[0140] As Figure 2 shown, Figure 2 is a block diagram of an example electronic device. The electronic device can be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc. Referring to Figure 2 , the apparatus 900 can include one or more of the following components: a processing component 902, a memory 904, a power supply component 906, a multimedia component 908, an audio component 910, an input / output (I / O) interface 912, a sensor component 914, and a communication component 916.

[0141] The processing component 902 usually controls the overall operation of the apparatus 900, such as operations associated with displaying, making phone calls, data communications, camera operations and recording operations.

[0142] The memory 904 is configured to store various types of data to support the operation of the device 900. As static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0143] Power component 906 provides power to various components of device 900.

[0144] Multimedia component 909 includes a screen providing an output interface between device 900 and a user. In some embodiments, the screen can include a liquid crystal display (LCD) and a touch panel (TP). In some embodiments, multimedia component 908 includes a front camera and / or a rear camera.

[0145] Audio component 909 is configured to output and / or input audio signals.

[0146] I / O interface 912 provides an interface between processing component 902 and peripheral interface modules, which can be a keyboard, a click wheel, buttons, and the like. The buttons can include, but are not limited to, a home button, a volume button, a start button, and a lock button.

[0147] Sensor component 914 includes one or more sensors to provide various state assessments for device 900. Sensor component 914 can include a proximity sensor configured to detect the presence of nearby objects without any physical contact.

[0148] Communication component 916 is configured to facilitate wired or wireless communication between device 900 and other devices. Device 900 can access a wireless network based on a communication standard, such as WiFi, an operator network (such as 2G, 3G, 4G, or 5G), or a combination thereof.

[0149] The present application provides a computer readable storage medium to realize the functions of the image synthesis method based on chroma keying. The computer readable storage medium stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, at least one program, code set or instruction set is loaded and executed by the processor to implement the physiological parameter monitoring method in the disaster first aid scene.

[0150] In an exemplary embodiment, the computer readable storage medium can be a non-transitory computer readable storage medium including instructions, such as a memory including instructions. For example, the non-transitory computer readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, and an optical data storage device, etc.

[0151] The above embodiments only express several implementation ways of the present application, and the description is specific and detailed, but it should not be understood as a limitation to the patent scope of the application. It should be pointed out that for ordinary skilled in the art, without departing from the concept of the present application, several modifications and improvements can be made, which are all within the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A method for monitoring physiological parameters in disaster emergency rescue scenarios, characterized in that, include: Multi-channel physiological signals of disaster victims are collected through a wearable sensor array, including surface electromyography signals, electrocardiogram signals, respiratory rate signals, and body temperature data. The collected physiological signals are subjected to real-time noise reduction and filtering to remove environmental noise; Multi-signal crosstalk processing is performed on multi-channel physiological signals after noise reduction and filtering to identify and eliminate crosstalk between different physiological signals. Signal fusion is performed on multi-channel physiological signals after crosstalk elimination; Disaster early warning based on the fused signal output.

2. The method for monitoring physiological parameters in disaster emergency rescue scenarios according to claim 1, characterized in that, The acquired physiological signals are subjected to real-time noise reduction and filtering, including: baseline drift noise removal using a filter on the physiological signals, wherein the frequency response of the filter is: Where, Q(e) jω ω is the frequency response of the filter, representing the amplitude and phase changes of the input signal at different frequencies ω after passing through the filter; ω is the angular frequency, reflecting the frequency characteristics of the signal; j is the imaginary unit; ω c The cutoff frequency is used to divide the frequency range that the filter allows to pass through and blocks, when 0 < |ω| < ω c When ω = 0, the filter output is 0, meaning that signals within that frequency range are completely blocked. c When |ω| < π, the filter output is 1, and the signal can pass through without attenuation.

3. The method for monitoring physiological parameters in disaster emergency rescue scenarios according to claim 2, characterized in that, The impulse response of the filter is: Where h(n) is the impulse response of the filter, which describes the output response of the filter to a unit impulse input; n represents the index of the discrete time sequence, which is used to determine the impulse response value at different times.

4. The method for monitoring physiological parameters in disaster emergency rescue scenarios according to claim 3, characterized in that, The acquired physiological signals are subjected to real-time noise reduction and filtering, including: noise removal using a notch filter, wherein the transfer function of the notch filter is: Among them, \(H(z)\) is the transfer function of the filter based on the z-transform, which is used to describe the relationship between the input and output of the filter. \(z\) is a complex variable, and \(z\) -1 represents that the signal is delayed by one sampling period in time; \(z\) -2 represents that the signal is delayed by two sampling periods in time; is the zero position of the filter in the complex plane, which is used to determine the attenuation characteristics of the filter for specific frequency signals; is the pole position of the filter in the complex plane; \(\omega_0\) is the angular frequency corresponding to the interference frequency, representing the frequency characteristics to be suppressed; \(r\) is the radius of the pole, which is used to adjust the notch bandwidth of the filter, \(0 < r < 1\), and the closer \(r\) is to 1, the narrower the notch bandwidth.

5. The method for monitoring physiological parameters in disaster emergency rescue scenarios according to claim 4, characterized in that, The acquired physiological signals are subjected to real-time noise reduction and filtering, including: removing electromyographic noise using a filter, wherein the output signal of the filter is: Where y(t) is the output signal of the filter at time t; x(tk) is the sampled value of the input signal at time tk; b k These are the filter coefficients, k = 0, 1, 2, ..., M, where M is the order of the filter.

6. The method for monitoring physiological parameters in disaster emergency rescue scenarios according to claim 5, characterized in that, The signal fusion specifically includes: performing time synchronization processing on the multi-channel physiological signals, and fusing the time-synchronized signals according to a preset signal fusion algorithm; Before fusing multiple filtered physiological signals, the process includes video acquisition of facial motion unit data, synchronizing each physiological signal with the facial motion unit data before fusing; first, the number of frames in the video sequence is calculated, and when sampling the physiological signals, it is made consistent with the number of video frames to achieve a one-to-one correspondence between the physiological signals and the facial motion unit data, and then the fusion is performed.

7. The method for monitoring physiological parameters in disaster emergency rescue scenarios according to claim 6, characterized in that, Multi-signal crosstalk processing is performed on the multi-channel physiological signals after noise reduction and filtering to identify and eliminate crosstalk between different physiological signals. This includes: using frequency division multiplexing technology, employing a specific transconductance amplifier as shown in the following formula, and assigning different modulation frequencies to each physiological signal to achieve frequency modulation of multi-channel physiological signals. In the back-end processing, demodulation is performed using a demodulation frequency corresponding to the modulation frequency, thereby identifying and eliminating crosstalk between different physiological signals. Simultaneously, frequency modulation is performed in the following manner: f chN +(2×BW SIG +BW GB )≤3×f ch1 This ensures that signals from different frequency channels do not overlap; Among them, G mB The transconductance of the transconductance amplifier; g m1 The transconductance of the output stage MOSFET in the transconductance amplifier; R F This is a resistor used for negative feedback of the source of the MOS transistor in the output stage of the transconductance amplifier, which adjusts the performance of the amplifier through negative feedback; r n0 This refers to the drain resistor of the MOSFET in the output stage of the transconductance amplifier. g mA The transconductance of the negative feedback transconductance stage; g mB To output the transconductance of the transconductance stage; f ch1 and f chN These are the lowest and highest modulation frequencies, used to determine the range of the frequency channel; BW SIG This is the signal bandwidth value, representing the frequency range of the physiological signal; BW GB To protect the band gap value, used to avoid interference between adjacent frequency channels; N is the maximum number of channels, representing the number of physiological signals the system can process simultaneously.

8. The method for monitoring physiological parameters in disaster emergency rescue scenarios according to claim 7, characterized in that, The fused signal is input to a transimpedance amplifier. To ensure that the transimpedance amplifier can accurately process the fused signal, the input dynamic range of the transimpedance amplifier is adjusted according to the following formula: (I ch1 +I ch2 +…+I chN )×α TIA ≤V max Among them: I ch1 I ch2 , ..., I chN These represent the output current of each channel, indicating the intensity of each physiological signal; α TIA The current-voltage gain of the transimpedance amplifier determines the ratio at which the current signal is converted into a voltage signal; V max This is the maximum input voltage of the transimpedance amplifier; I ch This refers to the output current of a single channel; N is the maximum number of channels, representing the number of physiological signal channels that can be processed.

9. A physiological parameter monitoring system for disaster emergency rescue scenarios, characterized in that, The physiological parameter monitoring system for disaster emergency rescue scenarios includes a wearable sensor array, and the physiological parameter monitoring system for disaster emergency rescue scenarios uses the physiological parameter monitoring method for disaster emergency rescue scenarios according to any one of claims 1-8 for data processing.

10. An electronic device, characterized in that, The electronic device includes: One or more processors; Memory; One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, the one or more applications being configured to perform a physiological parameter monitoring method in a disaster emergency response scenario according to any one of claims 1-8.

Images