Method and device for processing brain-computer signal of febrile convulsion and computer readable storage medium

Abstract

The invention discloses a febrile convulsion brain-computer signal processing method and device and a computer readable storage medium. The method comprises the following steps: acquiring an electroencephalogram signal, a body temperature signal and a motion signal of a detected person; generating a power spectrum image according to the electroencephalogram signal, the body temperature signal and the motion signal; obtaining power spectrum characteristics according to the power spectrum image; determining a signal classification according to the power spectrum characteristics and the resting state electroencephalogram data of the detected person; and determining a febrile convulsion detection result of the detected person according to the signal classification. According to the embodiment of the invention, the febrile convulsion can be effectively and timely detected under the condition that the body temperature of the patient is increased, the febrile convulsion is early warned, and effective prognosis is carried out on diseases.

Description

Technical Field

[0001] The present application belongs to the field of computer technology and medical technology, and more specifically, to a method, device and computer-readable storage medium for processing febrile convulsion brain-computer signals. Background Art

[0002] Febrile seizures are a common neurological disorder in children, usually occurring in infants and children between 6 months and 5 years old. Febrile seizures are associated with a sharp rise in body temperature, which may be caused by infection, vaccination or other conditions that cause a rise in body temperature. This condition often occurs very suddenly and can cause great distress and stress to the child and family. Although febrile seizures do not lead to long-term health problems in most cases, early warning can undoubtedly greatly reduce the psychological pressure on the family, and measures such as cooling can be taken in advance to prevent the occurrence of convulsions and lead to serious diseases such as epilepsy.

[0003] Therefore, how to effectively and timely detect febrile convulsions when the patient's body temperature is elevated, to effectively warn of febrile convulsions, and to effectively prognose the disease is an issue that medical staff and parents are eager to solve. Summary of the invention

[0004] The purpose of the embodiments of the present application is to provide a method, device and computer-readable storage medium for processing febrile convulsion brain-computer signals, which can realize real-time monitoring of febrile convulsions.

[0005] In a first aspect, an embodiment of the present application provides a method for processing febrile convulsion brain-computer signals, comprising:

[0006] Obtaining the EEG signal, body temperature signal and movement signal of the person being tested;

[0007] Generate a power spectrum image based on EEG signals, body temperature signals and motion signals;

[0008] According to the power spectrum image, power spectrum features are obtained;

[0009] Determine signal classification based on power spectrum characteristics and resting-state EEG data of the person being tested;

[0010] Based on the signal classification, the febrile convulsion detection result of the detected person is determined.

[0011] In a second aspect, the present application provides a method for processing febrile convulsion brain-computer signals, comprising:

[0012] Obtaining the electroencephalogram (EEG) bio-signals, body temperature bio-signals and movement bio-signals of the person being tested;

[0013] Converting the EEG biosignals, body temperature biosignals and motion biosignals into EEG signals, body temperature signals and motion signals that can be recognized by the data analysis end;

[0014] The EEG bio-signals, body temperature bio-signals and motion bio-signals are sent to the data analysis end, so that the data analysis end can generate a power spectrum image based on the EEG signals, body temperature signals and motion signals; based on the power spectrum image, the power spectrum features characterizing the febrile convulsion parameter features of the detected person are obtained; based on the power spectrum features and the resting EEG data of the detected person, the signal classification is determined; based on the signal classification, the febrile convulsion detection result of the detected person is determined.

[0015] In a third aspect, the present application provides a febrile convulsion brain-computer signal processing device, comprising:

[0016] A signal acquisition module is used to obtain the EEG signal, body temperature signal and movement signal of the detected person;

[0017] A power spectrum image generation module is used to generate a power spectrum image according to the EEG signal, the body temperature signal and the motion signal;

[0018] A power spectrum feature acquisition module is used to obtain power spectrum features according to the power spectrum image;

[0019] A signal classification module, used to determine the signal classification according to the power spectrum characteristics and the resting-state EEG data of the detected person;

[0020] The result module is used to determine the febrile convulsion detection result of the detected person according to the signal classification.

[0021] In a fourth aspect, the present application provides an embodiment of the present application further provides a febrile convulsion brain-computer signal processing device, which is applied to a client, including:

[0022] A bio-signal acquisition module is used to obtain the EEG bio-signals, body temperature bio-signals and movement bio-signals of the person being tested;

[0023] A bio-signal conversion module, used to convert EEG bio-signals, body temperature bio-signals and motion bio-signals into EEG signals, body temperature signals and motion signals identifiable by the data analysis end in turn;

[0024] The sending module is used to send EEG bio-signals, body temperature bio-signals and motion bio-signals to the data analysis end, so that the data analysis end can generate a power spectrum image based on the EEG signals, body temperature signals and motion signals; obtain power spectrum features that characterize the febrile convulsion parameter features of the detected person based on the power spectrum image; determine signal classification based on the power spectrum features and the resting EEG data of the detected person; and determine the febrile convulsion detection result of the detected person based on the signal classification.

[0025] In a fifth aspect, an embodiment of the present application further provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable a computer to execute the febrile convulsion brain-computer signal processing method provided in any embodiment of the present application.

[0026] The febrile convulsion brain-computer signal processing method, device and computer-readable storage medium provided in the embodiments of the present application can analyze the EEG signals, body temperature signals and motion signals of the detected person, and determine whether the detected person suffers from febrile convulsion. The implementation complexity is low, so it can be implemented in real time and repeatedly for the detected person, and the test results related to the febrile convulsion of the detected person can be obtained in a short time. It has high accuracy and is conducive to the detection of the health status of the detected person in the home health monitoring scenario. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0028] Figure 1 A schematic diagram of a process flow of a febrile convulsion brain-computer signal processing method provided in an embodiment of the present application;

[0029] Figure 2 A flowchart of a method for processing febrile convulsion brain-computer signals provided as an example of the present application;

[0030] Figure 3 A schematic diagram of the components of a system structure used in a febrile convulsion brain-computer signal processing method in an example of the present application;

[0031] Figure 4 Schematic diagram of the febrile convulsion brain-computer signal processing device provided in an embodiment of the present application.

[0032] The reference numerals in the above drawings are as follows:

[0033] 31. Data acquisition module; 32. Head-mounted host device; 41. Signal acquisition module; 42. Power spectrum image generation module; 43. Power spectrum feature acquisition module; 44. Signal classification module; 45. Result module. DETAILED DESCRIPTION

[0034] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0035] The present application provides a method for processing febrile convulsion brain-computer signals, such as Figure 1 As shown, including:

[0036] Step S11, obtaining the EEG signal, body temperature signal and movement signal of the detected person;

[0037] Step S12, generating a power spectrum image according to the EEG signal, the body temperature signal and the motion signal;

[0038] Step S13, obtaining power spectrum features according to the power spectrum image;

[0039] Step S14, determining the signal classification according to the power spectrum characteristics and the resting-state EEG data of the detected person;

[0040] Step S15, determining the febrile convulsion detection result of the detected person according to the signal classification.

[0041] In the embodiments of the present application, the EEG signal is the overall reflection of the electrophysiological activity of the brain nerve tissue on the surface of the cerebral cortex. It is the synthesis of the postsynaptic potentials of the brain neurons. It is the electrical signal generated when the brain is active. The EEG signal is a spontaneous biopotential generated by the spontaneous and rhythmic activity of brain cells.

[0042] When generating a power spectrum image based on the EEG signals, body temperature signals and motion signals, the information on the changes in the EEG signals, body temperature signals and motion signals over time is added to the power spectrum image. In this way, in the power spectrum image, the information in the EEG signals, body temperature signals and motion signals can be organically combined, thereby distinguishing febrile convulsions from other diseases with similar symptoms and improving the accuracy of the analysis.

[0043] In the embodiment of the present application, the power spectrum feature may include multiple dimensions, such as spectrum energy, frequency band energy ratio, spectrum change speed, etc. The power spectrum feature can reflect the characteristics of the human body parameters related to febrile convulsions of the detected person in a short period of time, such as the data size and data change characteristics.

[0044] The above-mentioned resting-state EEG data may be data of the detected person when he / she is not having a febrile convulsion, or may be data collected when the detected person is sitting or lying still and when he / she is not having a febrile convulsion.

[0045] Determining the signal classification based on the power spectrum characteristics and the resting-state EEG data of the detected person may include determining the signal classification based on the consistency between the power spectrum characteristics and the resting-state EEG data at different latitudes. Signal classification can be used to indicate whether the currently collected EEG signal, body temperature signal, and motion signal indicate whether the detected person has a febrile seizure with behavioral characteristics (convulsions, etc.), a febrile seizure without behavioral characteristics, and other categories.

[0046] Determining the febrile convulsion detection result for the detected person based on signal classification may include using the signal classification result as the febrile convulsion detection result for the detected person; or determining the febrile convulsion detection result for the detected person based on signal classification and a pre-set detection sensitivity.

[0047] The febrile convulsion brain-computer signal processing method provided in the embodiment of the present application can analyze the EEG signals, body temperature signals and movement signals of the detected person, and determine whether the detected person suffers from febrile convulsion. The implementation complexity is low, so it can be implemented in real time and repeatedly for the detected person, and the test results related to the febrile convulsion of the detected person can be obtained in a short time. It has high accuracy and is conducive to the detection of the health status of the detected person in the home health monitoring scenario.

[0048] In one embodiment, generating a power spectrum image according to the EEG signal, the body temperature signal and the motion signal includes:

[0049] Preprocessing the EEG signal, the body temperature signal and the motion signal in sequence, and obtaining a preprocessed EEG signal, a preprocessed body temperature signal and a preprocessed motion signal accordingly;

[0050] Generate a power spectrum time-frequency diagram according to the preprocessed EEG signal and the preprocessed motion signal;

[0051] According to the preprocessed body temperature signal, at least one pixel information of the power spectrum time-frequency diagram is modified to obtain a power spectrum image.

[0052] In the embodiment of the present application, the EEG signal can be obtained by electrodes, and the motion signal can be obtained by sensors. Generally, the EEG signal and the motion sensor output signal are continuous signals in the time domain. In the present embodiment, the EEG signal and the motion signal can be converted to obtain frequency domain signals, and then the time-frequency domain signals are composed according to the frequency domain signals at different times, that is, the power spectrum time-frequency diagram.

[0053] Modifying at least one pixel information of the power spectrum time-frequency diagram according to the preprocessed body temperature signal may include: converting the body temperature signal into a color pixel value, obtaining a thermal map pixel corresponding to the body temperature signal, and modifying at least one pixel of the power spectrum time-frequency diagram.

[0054] In one embodiment, the EEG signal, the body temperature signal and the motion signal are preprocessed in sequence to obtain the preprocessed EEG signal, the preprocessed body temperature signal and the preprocessed motion signal, including:

[0055] Aligning the EEG signal, the body temperature signal, and the motion signal in sequence with a preset sampling parameter benchmark to obtain an aligned EEG signal, an aligned body temperature signal, and an aligned motion signal;

[0056] The multiple EEG signals, aligned body temperature signals and aligned motion signals are sequentially segmented according to preset time periods to obtain preprocessed EEG signals, preprocessed body temperature signals and preprocessed motion signals accordingly.

[0057] In an embodiment of the present application, the time signal is sliced ​​according to a certain frequency, and the time domain within the slice is Fourier transformed to obtain a frequency domain signal, so that a time-frequency diagram (i.e., a time-frequency diagram) of the signal can be obtained. On the corresponding time-frequency diagram, the body temperature signal is added as a thermal map pixel, and different color blocks represent different temperatures.

[0058] In one embodiment, the EEG signal, the body temperature signal, and the motion signal are aligned in sequence with a preset sampling parameter reference, and the aligned EEG signal, the aligned body temperature signal, and the aligned motion signal are obtained accordingly, including:

[0059] The EEG signal and the motion signal are filtered according to the set filtering frequency to obtain a filtered EEG signal and a preliminarily filtered motion signal;

[0060] Identify the pseudo motion trajectory information in the initially filtered motion signal, filter the signal corresponding to the pseudo motion trajectory information, and obtain a secondary filtered motion signal;

[0061] Based on a preset sampling parameter benchmark, the filtered EEG signal, the secondary filtered motion signal and the body temperature signal are aligned to obtain an aligned EEG signal, an aligned body temperature signal and an aligned motion signal.

[0062] According to the set filtering frequency, the EEG signal and the motion signal are filtered to obtain a filtered EEG signal and a preliminarily filtered motion signal. In this step, signals with important frequencies can be retained.

[0063] In one embodiment, the preset sampling parameter benchmark includes a sampling frequency; based on the preset sampling parameter benchmark, the filtered EEG signal, the secondary filtered motion signal and the body temperature signal are aligned to obtain an aligned EEG signal, an aligned body temperature signal and an aligned motion signal, including:

[0064] For the signals in the filtered EEG signals, the secondary filtered motion signals and the body temperature signals, the signals whose acquisition frequency is lower than the highest acquisition frequency of the signal acquisition end are padded so that the acquisition frequency of the padded signals is consistent with the highest acquisition frequency;

[0065] According to the complemented signals, aligned EEG signals, aligned body temperature signals and aligned motion signals are obtained.

[0066] In another embodiment, the preset sampling parameter reference may further include sampling time.

[0067] In one embodiment, generating a power spectrum time-frequency diagram according to the preprocessed EEG signal and the preprocessed motion signal includes:

[0068] Performing short-time Fourier transform on the preprocessed EEG signal and the preprocessed motion signal to obtain power spectrum data of the preprocessed EEG signal and the preprocessed motion signal;

[0069] According to the power spectrum data, the power spectrum time-frequency diagram is obtained.

[0070] In one embodiment, the preprocessed EEG signal includes multiple channels; the preprocessed motion signal includes multiple channels; each EEG signal corresponds to a power spectrum time-frequency diagram, and each motion signal corresponds to a power spectrum time-frequency diagram.

[0071] In one embodiment, 10 electrodes are used to collect EEG signals to obtain 10-channel EEG signals. A 6-axis sensor is used to collect motion signals. After obtaining 6-axis motion signals, 16 Spectrogram images can be obtained. Image features representing power spectrum information are extracted as frequency power spectrum features to classify and identify the combined signal.

[0072] In one embodiment, at least one pixel information of the power spectrum time-frequency diagram is modified according to the preprocessed body temperature signal to obtain a power spectrum image, including:

[0073] According to the preprocessed body temperature signal, the corresponding thermal value color block is determined;

[0074] Add the thermal value color block to the power spectrum time-frequency diagram to obtain the power spectrum image.

[0075] In one embodiment, the power spectrum feature includes at least one of spectrum energy, frequency band energy ratio, and spectrum change speed.

[0076] In one embodiment, the febrile convulsion brain-computer signal processing method further includes:

[0077] Acquire resting-state EEG data when such data exist;

[0078] In the absence of resting-state EEG data, a resting-state EEG data acquisition instruction is sent to the signal acquisition end; and data fed back by the signal acquisition end in response to the resting-state EEG data acquisition instruction is received.

[0079] In one embodiment, determining the signal classification based on the power spectrum characteristics and the resting-state EEG data of the detected person includes:

[0080] Input the power spectrum features and resting-state EEG data into the classification model to obtain output data;

[0081] Based on the output data, determine the signal classification.

[0082] In one embodiment, the febrile convulsion brain-computer signal processing method further includes:

[0083] Obtaining a power consumption mode indication signal;

[0084] When the power consumption mode indication signal indicates a low power consumption mode, obtaining a power spectrum time-frequency diagram according to the EEG signal and the motion signal;

[0085] Determine the febrile convulsion detection result of the detected person according to the power spectrum time-frequency diagram, the body temperature signal and the preset classification threshold;

[0086] When the power consumption mode indication signal indicates a non-low power consumption mode, a step of generating a power spectrum image according to the electroencephalogram signal, the body temperature signal and the motion signal is performed.

[0087] Generally speaking, in order to ensure low power consumption and long-term operation of the equipment and avoid the inconvenience caused to the patient by frequent charging, a low power consumption mode can be adopted when running the febrile convulsion brain-machine signal processing method provided by the embodiment of the present application on a local processor. The febrile convulsion brain-machine signal processing method implemented on the local processor is basically consistent with the steps of implementing the method on the non-local processor in data preprocessing, signal slicing and other links, and the step of data cleaning can be added. After completing the data cleaning, the EEG signal and the motion signal are subjected to fast Fourier transform, and time-frequency-power spectrum analysis is performed to obtain a power spectrum time-frequency diagram, and the temperature data is not processed separately. The obtained temperature data and power spectrum time-frequency diagram are directly input into a pre-designed threshold filter, and numerical comparison and screening are performed to determine whether there is a febrile convulsion attack accompanied by behavioral characteristics (such as convulsions, etc.), a febrile convulsion attack without behavioral characteristics, or a normal signal.

[0088] Among them, the setting of the above-mentioned threshold filter and the design of the key threshold will be set based on the power spectrum energy of the patient's resting EEG data, the typical febrile convulsion epilepsy attack and the power spectrum energy of non-epileptic attack samples marked by clinical conditions, etc. Generally speaking, the power spectrum energy of the patient's resting EEG data is used as the basic value. During a febrile convulsion epilepsy attack, the abnormal energy release of EEG activity causes a significant increase in the power spectrum in most frequency bands. Therefore, the resting energy can be increased by 10%, 20%, or n% as the threshold in the threshold filter. The setting of the threshold in the threshold filter can also be set differently according to the measurement channel, frequency band, etc. For example, for the scheme of collecting EEG signals through 10 channels, the thresholds corresponding to the EEG signals of 10 channels in the threshold filter can be set according to the change amplitude of the EEG signals of 10 channels during febrile convulsions. In comparison, the method of using the power spectrum energy threshold determination to make an alarm may have problems such as false positives and missed detections, but it can meet the requirements of low power consumption and real-time performance. In other scenarios besides home health monitoring, the appropriate application mode can be selected as needed.

[0089] In one example of the present application, the implementation process is as follows: Figure 2 As shown, the following steps S21-S223 are included.

[0090] Among them, after initialization, step S21 is executed: determining whether there is resting-state EEG data.

[0091] When a patient (person being tested) is diagnosed with a risk of febrile seizures, professional medical care providers can provide the patient with a brain-computer interface system for febrile seizure health monitoring, allowing the patient's family to monitor the patient's febrile seizures in real time in a home environment or other environment where professional medical care is lacking. Figure 3 As shown, the patient can wear the data acquisition module 31 corresponding to the brain-computer interface system for febrile convulsion health monitoring on the brain and other parts of the body to acquire and analyze the patient's EEG signals, body temperature signals and movement signals in real time. When the brain-computer interface system for febrile convulsion health monitoring is first used, the patient's resting EEG data such as the resting EEG under normal conditions can be collected, and the resting EEG data will be used as the background signal of the individual patient in the subsequent algorithm.

[0092] In step S22, resting-state EEG data (resting-state EEG signals) are collected under normal conditions.

[0093] If it is determined that there is no resting-state EEG data corresponding to the current person to be detected (patient), the resting-state EEG data of the patient is collected when the patient is not in a febrile convulsion state.

[0094] In step S23, the data is stored in the Flash Memory static storage block.

[0095] Combination Figure 3 The resting-state EEG data can be stored for a long time in the fixed Flash Memory space of the head-mounted host device 32 of the brain-computer interface system for febrile convulsion health monitoring, corresponding to the identity code of the person to be detected.

[0096] In step S24, real-time EEG acquisition is performed, that is, EEG biological signals are acquired in real time.

[0097] When resting-state EEG data is available, the patient wears the device for long-term, continuous real-time monitoring, during which EEG signals, body temperature signals, and movement signals will be collected in real time.

[0098] In step S25, real-time temperature collection is performed, that is, body temperature biological signals are collected in real time.

[0099] In step S26, motion sensor data collection: The patient's motion biological data can be obtained through the sensor.

[0100] As an example, the EEG signal sampling rate can be configured to be 100-2000 Hz, the body temperature sampling rate can be configured to be 0.1-1 Hz, and the motion sensor sampling rate can be configured to be 1000 Hz.

[0101] In step S27, the data is packaged.

[0102] After completing the fixed-duration (eg, 1 second) acquisition, the data of the EEG signal, the body temperature signal, and the motion signal are packaged to obtain packaged data.

[0103] In step S28, the data is wirelessly transmitted or buffered in a flash memory.

[0104] In step S29, the cloud database is uploaded.

[0105] In step S210, the received data is stored in a flash memory data block, and can be stored in a first-in-first-out manner.

[0106] In combination with steps S28, S29, and S210, the data of the EEG signal, body temperature signal, and motion signal are uploaded to the cloud database via a wireless connection, or cached in the local Flash Memory. When the data of the EEG signal, body temperature signal, and motion signal are cached in the local Flash Memory, due to the limitation of the memory of the storage unit, the data will be stored in a FIFO (First In First Out) mode.

[0107] In step S211 , the algorithm is executed in the cloud or locally.

[0108] In this step, it is determined whether to execute the cloud algorithm or the local algorithm to avoid repeated calculations on the same data at both ends. At the same time, data calculation and processing can be realized in real time on different signal processing ends based on the scenario of algorithm operation and the difference in hardware resources.

[0109] In step S212, the cloud algorithm runs.

[0110] The algorithm executed in the cloud can include the following process. After the computing node receives the EEG signal, body temperature signal and motion signal data that are packaged and transmitted back in real time, it first pre-processes the data. In the data pre-processing stage, the EEG signal and motion signal will be filtered, denoised and artifacts removed.

[0111] Among them, the design of the filter needs to consider the need for Spectrogram feature selection and configure the corresponding filter characteristics to retain the components of important frequencies.

[0112] In the noise reduction stage, the main focus is on noise reduction and filtering of industrial frequency interference caused by the environment to obtain high-quality original signals.

[0113] In the artifact removal phase, the motion artifacts in the EEG are mainly filtered out. In this phase, motion data can play an auxiliary role to achieve the purpose of completely removing motion artifacts.

[0114] After completing data preprocessing, time alignment is performed on different signals. At the same time, due to the different sampling rates of different signals, the sampling rates of different signals need to be padded in the alignment process. Generally speaking, the highest sampling signal is used as the benchmark to padded the data of the low sampling rate signal. For example, the temperature sampling rate is much lower than that of EEG and motion signal sampling rates. At the same time, temperature change is a slow variable. For missing data, the previous sampling result can be used as the basis for direct filling and padding.

[0115] After completing the above operations, the signal can be time sliced, usually according to the length of 1s, 5s, 10s, 15s, etc. Selecting different time length slices will result in different delays in the febrile seizure alarm. The slice length can be configured according to actual needs during use.

[0116] After completing the time slice, the EEG signal and the motion signal are short-time Fourier transformed. In this example, the EEG signal includes 10 channels, and the motion signal generally samples a 6-axis motion sensor including 6-channel signals. The 16-channel signal is short-time Fourier transformed to obtain 16 pixel images of the power spectrum time-frequency diagram. Finally, the body temperature signal is filled in the 16 power spectrum time-frequency diagrams in the form of thermal value blocks to obtain 16 complete Spectrogram images. Thereafter, features are extracted from the generated Spectrogram image. Generally speaking, the features include spectral energy, frequency band energy ratio, spectral change speed, etc. Finally, the classification algorithm is used to train and classify with known patient resting EEG data, clinically annotated typical febrile convulsion epilepsy seizures, and non-epileptic seizure samples to complete the identification of febrile convulsion epilepsy seizures. Among them, the classification algorithm can use algorithm frameworks such as support vector machine (SVM, Support Vector Machine), artificial neural network (ANN, Artificial Neural Network), and convolutional neural network (CNN, Convolutional Neural Network). The final data results are generally divided into: febrile seizures with behavioral characteristics (convulsions, etc.), febrile seizures without behavioral characteristics, and normal signals. In addition, with the iteration and upgrade of hardware, the above algorithm can also be completed on an embedded platform based on FPGA (Field Programmable Gate Array, Field Programmable Gate Array). Therefore, the above algorithm process can also be performed locally in real time on a microprocessor platform using FPGA.

[0117] In step S213, the local algorithm runs.

[0118] In step S214, the result is fed back in real time.

[0119] In the cloud or Figure 3 The head-mounted host device 32 (local edge computing) shown runs the corresponding algorithm to detect febrile convulsions in real time and mark the data of the EEG signal, body temperature signal, and movement signal of the febrile convulsion state.

[0120] In step S215, the alarm is marked with a time stamp. Further, the time of the alarm event is recorded.

[0121] In step S216, a buzzer sounds an alarm. For example, a wearable headscarf with electrodes and a buzzer is used to alarm the patient and the patient's guardian, so that the patient or his guardian can handle the febrile seizure in a timely manner.

[0122] In step S217, the doctor client issues an alarm prompt.

[0123] In step S218, the patient client issues an alarm prompt.

[0124] When the person to be detected has a febrile convulsion, an alarm operation is performed through steps S216-S218. Thereafter, the alarm operation may include providing an alarm reminder issued by a buzzer set on the host end of the head mounted device, and pushing a real-time reminder to the medical terminal and the client configured for the patient.

[0125] In step S219, it is determined whether the network data has been updated. If so, the process proceeds to step S223. Otherwise, the process proceeds to step S220.

[0126] In addition to being stored locally in the brain-computer interface system for febrile seizure health monitoring, resting-state EEG data can also be stored in the cloud. Local updates to resting-state EEG data will trigger automatic synchronization of network data, that is, when the network is idle, resting-state EEG data will be automatically uploaded to the cloud for updating, and cloud algorithms will be called for computing nodes deployed in the cloud. If the resting-state EEG data is not synchronized to the network due to lack of network connection or other reasons during local updates, the local update records of the resting-state EEG data will be updated to the network cloud when network conditions are suitable.

[0127] In step S220, it is determined whether there is a network connection. If so, the process proceeds to step S221. Otherwise, the process repeats this step.

[0128] In step S221, it is determined whether the network is occupied. If so, this step is repeated. Otherwise, the process proceeds to step S222.

[0129] In step S222, the cloud database is uploaded.

[0130] In step S223, the automatic uploading of resting-state EEG data is completed.

[0131] Through the above steps S220-S223, the resting-state EEG data can be updated.

[0132] The embodiment of the present application also provides a febrile convulsion brain-computer signal processing method, which is applied to a client, including:

[0133] Obtaining the electroencephalographic biosignals (electroencephalographic physiological signals), body temperature biosignals (body temperature physiological signals) and motion biosignals (motion physiological signals) of the person being tested;

[0134] Converting the EEG biosignals, body temperature biosignals and motion biosignals into EEG signals, body temperature signals and motion signals that can be recognized by the data analysis end;

[0135] The EEG bio-signals, body temperature bio-signals and motion bio-signals are sent to the data analysis end, so that the data analysis end can generate a power spectrum image based on the EEG signals, body temperature signals and motion signals; based on the power spectrum image, the power spectrum features characterizing the febrile convulsion parameter features of the detected person are obtained; based on the power spectrum features and the resting EEG data of the detected person, the signal classification is determined; based on the signal classification, the febrile convulsion detection result of the detected person is determined.

[0136] In one embodiment, the EEG bio-signals include those collected by electrodes arranged at positions O2, T6, T4, F8, Fp2, Fp1, F7, T3, T5, and O1 in the 10-20 system of the electroencephalogram; the motion bio-signals include angular velocity bio-signals and acceleration bio-signals.

[0137] The 10-20 system of electrode placement is the standard electrode placement method specified by the International Electroencephalography Society. The distance from the midpoint of the frontal pole to the root of the nose and the distance from the occipital point to the external occipital protuberance each account for 10% of the total length of this line, and the remaining points are separated by 20% of the total length of this line, so it is named the 10-20 system.

[0138] The electrode positions of the 10-20 system are mainly based on the skull and do not vary due to differences in individual head circumference or head shape. The electrode positions in the 10-20 system are described as follows:

[0139] Anterior-posterior sagittal line: Draw a line from the root of the nose to the external occipital tuberosity. On this line, mark 5 points from front to back, named in sequence: frontal pole midpoint (Fpz), frontal midpoint (Fz), central point (Cz), vertex (Pz), occipital point (Oz). The distance from the frontal pole midpoint to the root of the nose and the distance from the occipital point to the external occipital tuberosity each account for 10% of the total length of this line, and the remaining points are separated by 20% of the total length of this line;

[0140] Horizontal position: draw a line from the left ear prepoint (the depression at the root of the zygomatic arch in front of the tragus) through the central point to the right ear prepoint, and mark the left temporal middle (T3), right temporal middle (T4), left central (C3), and right central (C4) symmetrically on the left and right sides of this line. The distance between T3 and T4 points and the ear prepoint is 10% of the total length of this line, and the remaining points (including Cz point) are separated by 20% of the total length of this line;

[0141] Lateral position: Take the left and right lines from Fpz point to the occipital point through T3 and T4 points, and mark the left frontal pole (Fp1), right frontal pole (Fp2), left anterior temporal (F7), right anterior temporal (F8), left posterior temporal (T5), right posterior temporal (T6), left occipital (O1), and right occipital (O2) points symmetrically from front to back on the left and right lines. The distance from Fp1 and Fp2 points to the midpoint of the frontal pole (Fpz) and the distance from O1 and O2 points to Oz point each account for 10% of the total length of this line, and the remaining points (including T3 and T4) are separated by 20% of the total length of this line;

[0142] The remaining left frontal (F3) and right frontal (F4) points are located between Fp1, Fp2 and C3, C4 points respectively; the left parietal (P3) and right parietal (P4) points are located between C3, C4 and O1, O2 points respectively.

[0143] The present application also provides a febrile convulsion brain-computer signal processing device, such as Figure 4 As shown, including:

[0144] The signal acquisition module 41 is used to obtain the EEG signal, body temperature signal and movement signal of the detected person;

[0145] A power spectrum image generating module 42, used to generate a power spectrum image according to the EEG signal, the body temperature signal and the motion signal;

[0146] A power spectrum feature acquisition module 43 is used to obtain power spectrum features according to the power spectrum image;

[0147] A signal classification module 44, used to determine the signal classification according to the power spectrum characteristics and the resting-state EEG data of the detected person;

[0148] The result module 45 is used to determine the febrile convulsion detection result of the detected person according to the signal classification.

[0149] In one embodiment, the power spectrum image generation module includes:

[0150] A preprocessing unit, used to preprocess the EEG signal, the body temperature signal and the motion signal in sequence, and obtain a preprocessed EEG signal, a preprocessed body temperature signal and a preprocessed motion signal accordingly;

[0151] A power spectrum time-frequency diagram unit, used for generating a power spectrum time-frequency diagram according to the preprocessed EEG signal and the preprocessed motion signal;

[0152] The modification unit is used to modify at least one pixel information of the power spectrum time-frequency diagram according to the preprocessed body temperature signal to obtain a power spectrum image.

[0153] In one embodiment, the pre-processing unit is configured to:

[0154] Aligning the EEG signal, the body temperature signal, and the motion signal in sequence with a preset sampling parameter benchmark to obtain an aligned EEG signal, an aligned body temperature signal, and an aligned motion signal;

[0155] The multiple EEG signals, aligned body temperature signals and aligned motion signals are sequentially segmented according to preset time periods to obtain preprocessed EEG signals, preprocessed body temperature signals and preprocessed motion signals accordingly.

[0156] In one embodiment, the pre-processing unit is configured to:

[0157] The EEG signal and the motion signal are filtered according to the set filtering frequency to obtain a filtered EEG signal and a preliminarily filtered motion signal;

[0158] Identify the pseudo motion trajectory information in the initially filtered motion signal, filter the signal corresponding to the pseudo motion trajectory information, and obtain a secondary filtered motion signal;

[0159] Based on a preset sampling parameter benchmark, the filtered EEG signal, the secondary filtered motion signal and the body temperature signal are aligned to obtain an aligned EEG signal, an aligned body temperature signal and an aligned motion signal.

[0160] In one embodiment, the pre-processing unit is configured to:

[0161] For the signals in the filtered EEG signals, the secondary filtered motion signals and the body temperature signals, the signals whose acquisition frequency is lower than the highest acquisition frequency of the signal acquisition end are padded so that the acquisition frequency of the padded signals is consistent with the highest acquisition frequency;

[0162] According to the complemented signals, aligned EEG signals, aligned body temperature signals and aligned motion signals are obtained.

[0163] In one embodiment, the power spectrum time-frequency diagram unit is configured to:

[0164] Performing short-time Fourier transform on the preprocessed EEG signal and the preprocessed motion signal to obtain power spectrum data of the preprocessed EEG signal and the preprocessed motion signal;

[0165] According to the power spectrum data, the power spectrum time-frequency diagram is obtained.

[0166] In one embodiment, the preprocessed EEG signal includes multiple channels; the preprocessed motion signal includes multiple channels; each EEG signal corresponds to a power spectrum time-frequency diagram, and each motion signal corresponds to a power spectrum time-frequency diagram.

[0167] In one embodiment, the modification unit is configured to:

[0168] According to the preprocessed body temperature signal, the corresponding thermal value color block is determined;

[0169] Add the thermal value color block to the power spectrum time-frequency diagram to obtain the power spectrum image.

[0170] In one embodiment, the power spectrum feature includes at least one of spectrum energy, frequency band energy ratio, and spectrum change speed.

[0171] In one embodiment, the febrile convulsion brain-computer signal processing device further includes:

[0172] A resting-state EEG data acquisition module, used to acquire resting-state EEG data when there is resting-state EEG data;

[0173] The resting-state EEG data acquisition module is used to send a resting-state EEG data acquisition instruction to the signal acquisition end when there is no resting-state EEG data; and receive data fed back by the signal acquisition end in response to the resting-state EEG data acquisition instruction.

[0174] In one embodiment, the signal classification module includes:

[0175] A model unit, used for inputting power spectrum features and resting-state EEG data into a classification model to obtain output data;

[0176] The output data processing unit is used to determine the signal classification according to the output data.

[0177] In one embodiment, the febrile convulsion brain-computer signal processing device further includes:

[0178] A power consumption mode indication signal module, used for obtaining a power consumption mode indication signal;

[0179] The power spectrum acquisition module is used to obtain the power spectrum time-frequency diagram according to the EEG signal and the motion signal when the power consumption mode indication signal indicates the low power consumption mode; when the power consumption mode indication signal indicates the non-low power consumption mode, drive the power spectrum image generation module

[0180] The threshold module is used to determine the febrile convulsion detection result of the detected person based on the power spectrum time-frequency diagram, the body temperature signal and the preset classification threshold.

[0181] The present application also provides a febrile convulsion brain-computer signal processing device, which is applied to a client and includes:

[0182] A bio-signal acquisition module is used to obtain the EEG bio-signals, body temperature bio-signals and movement bio-signals of the person being tested;

[0183] A bio-signal conversion module, used to convert EEG bio-signals, body temperature bio-signals and motion bio-signals into EEG signals, body temperature signals and motion signals identifiable by the data analysis end in turn;

[0184] The sending module is used to send EEG bio-signals, body temperature bio-signals and motion bio-signals to the data analysis end, so that the data analysis end can generate a power spectrum image based on the EEG signals, body temperature signals and motion signals; obtain power spectrum features that characterize the febrile convulsion parameter features of the detected person based on the power spectrum image; determine signal classification based on the power spectrum features and the resting EEG data of the detected person; and determine the febrile convulsion detection result of the detected person based on the signal classification.

[0185] An embodiment of the present application also provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable a computer to execute the febrile convulsion brain-computer signal processing method provided in any embodiment of the present application.

[0186] The embodiments of the present invention described above are combinations of elements and features of the present invention. Unless otherwise mentioned, elements or features may be considered as optional. Each element or feature may be put into practice without being combined with other elements or features. In addition, embodiments of the present invention may be constructed by combining some elements and / or features. The order of operations described in the embodiments of the present invention may be rearranged. Some configurations of any one embodiment may be included in another embodiment, and may be replaced by the corresponding configuration of another embodiment. It is obvious to those skilled in the art that claims that do not have a clear reference relationship to each other in the attached claims may be combined into embodiments of the present invention, or may be included as new claims in the amendment after submitting the present invention.

[0187] In a firmware or software configuration, the embodiments of the present invention may be implemented in the form of modules, procedures, functions, etc. The software code may be stored in a memory unit and executed by a processor. The memory unit is located inside or outside the processor and may send data to and receive data from the processor via various known means.

[0188] Various aspects of the systems and methods described herein may be implemented as functions programmed into any of a variety of circuits, including programmable logic devices (PLDs), such as field programmable gate arrays (FPGAs), programmable array logic (PAL) devices, electronic programmable logic and storage devices, standard cell-based devices, and application-specific integrated circuits (ASICs). Some other possibilities for implementing these aspects of the system include: microcontrollers with memory, such as electronically erasable programmable read-only memory (EEPROM), embedded microprocessors, firmware, software, etc. In addition, these aspects of the system may be embodied in microprocessors with software-based circuit simulation, discrete logic (sequential and combinational), custom devices, fuzzy (neural) logic, quantum devices, and combinations of any of the above various device types. Of course, the underlying device technology may be provided in a variety of component types, such as metal oxide semiconductor field effect transistor (MOSFET) technology such as complementary metal oxide semiconductor (CMOS), bipolar technology such as emitter coupled logic (ECL), polymer technology (e.g., silicon conjugated polymer and metal conjugated polymer metal structure), hybrid analog and digital, etc.

[0189] The various functions or processes disclosed herein may be described as data and / or instructions embodied in various computer-readable media in terms of their behavior, register transfers, logic components, transistors, geometric layouts, and / or other characteristics. Computer-readable media that may contain such formatted data and / or instructions include, but are not limited to, various forms of non-volatile storage media (e.g., optical, magnetic, or semiconductor storage media) and carrier waves, which may be used to transmit such formatted data and / or instructions via wireless, optical, or wired signal media, or any combination thereof. When received in any of a variety of circuits (e.g., computers), such data and / or instructions may be processed by a processing entity (e.g., one or more processors).

[0190] The above description of the illustrated embodiments of the systems and methods is not intended to be exhaustive or to limit the systems and methods to the precise forms disclosed. Although specific embodiments and examples of the system components and methods are described herein for illustrative purposes, it will be appreciated by those skilled in the art that various equivalent modifications may be made within the scope of the systems, components, and methods. The teachings of the systems and methods provided herein may be applied to other processing systems and methods, not just to the above-described systems and methods.

[0191] Those skilled in the art will appreciate that many variations and / or modifications may be made to the invention shown in the specific embodiments without departing from the spirit or scope of the invention as broadly described. Therefore, the present embodiments are to be considered in all respects as illustrative and not restrictive. In addition, the present invention includes any combination of features described for the different embodiments (including those in the abstract), even if the feature or combination of features is not explicitly specified in the claims or the detailed description of the embodiments.

[0192] Generally, in the following claims, the terms used should not be construed to limit the systems and methods to the specific embodiments disclosed in the specification and the claims, but should be construed to include all processing systems that operate under the claims. Therefore, the systems and methods are not limited by the present disclosure, but rather the scope of the systems and methods is determined entirely by the claims.

[0193] Unless the context clearly requires otherwise, throughout the specification and claims, the words "include", "including", "comprising", etc. should be interpreted in an inclusive sense and not in an exclusive or exhaustive sense; that is, in the sense of "including but not limited to". Words using the singular or plural number also include the singular or plural number, respectively. In addition, "herein", "herein below", "above", "hereafter" and words of similar meaning refer to this application as a whole and not to any particular part of this application. When the word "or" is used in a list involving two or more items, the word "or" includes all of the following interpretations of the word: any item in the list, all items in the list, and any combination of items in the list.

[0194] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "plurality" is more than two, unless otherwise clearly and specifically defined.

[0195] The above are only preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for processing brain-computer signals for febrile convulsions, comprising: Obtaining the EEG signal, body temperature signal and movement signal of the person being tested; generating a power spectrum image according to the EEG signal, the body temperature signal and the motion signal; According to the power spectrum image, obtaining power spectrum features; Determining signal classification according to the power spectrum characteristics and the resting-state EEG data of the detected person; A febrile seizure detection result for the detected person is determined according to the signal classification.

2. The method according to claim 1, wherein: The step of generating a power spectrum image according to the EEG signal, the body temperature signal and the motion signal comprises: Preprocessing the EEG signal, the body temperature signal and the motion signal in sequence to obtain a preprocessed EEG signal, a preprocessed body temperature signal and a preprocessed motion signal respectively; Generate a power spectrum time-frequency diagram according to the preprocessed EEG signal and the preprocessed motion signal; At least one pixel information of the power spectrum time-frequency diagram is modified according to the preprocessed body temperature signal to obtain the power spectrum image.

3. The method according to claim 2, wherein: The preprocessing of the EEG signal, the body temperature signal and the motion signal in sequence to obtain a preprocessed EEG signal, a preprocessed body temperature signal and a preprocessed motion signal accordingly includes: Aligning the EEG signal, the body temperature signal, the motion signal and a preset sampling parameter reference in sequence, and obtaining an aligned EEG signal, an aligned body temperature signal and an aligned motion signal accordingly; The multiple EEG signals, aligned body temperature signals and aligned motion signals are sequentially segmented according to preset time periods to obtain the preprocessed EEG signals, preprocessed body temperature signals and preprocessed motion signals accordingly.

4. The method according to claim 3, wherein: The step of sequentially aligning the EEG signal, the body temperature signal, and the motion signal with a preset sampling parameter reference to obtain an aligned EEG signal, an aligned body temperature signal, and an aligned motion signal includes: Filtering the EEG signal and the motion signal according to a set filtering frequency to obtain a filtered EEG signal and a preliminarily filtered motion signal; Identifying pseudo motion trajectory information in the initially filtered motion signal, filtering a signal corresponding to the pseudo motion trajectory information, and obtaining a secondary filtered motion signal; Based on a preset sampling parameter benchmark, the filtered EEG signal, the secondary filtered motion signal and the body temperature signal are aligned to obtain an aligned EEG signal, an aligned body temperature signal and an aligned motion signal.

5. The method according to claim 4, wherein: The preset sampling parameter benchmark includes a sampling frequency; the filtered EEG signal, the secondary filtered motion signal and the body temperature signal are aligned based on the preset sampling parameter benchmark to obtain an aligned EEG signal, an aligned body temperature signal and an aligned motion signal, including: Complementing the signals whose acquisition frequency is lower than the highest acquisition frequency of the signal acquisition end among the filtered EEG signal, the secondary filtered motion signal and the body temperature signal, so that the acquisition frequency of the complemented signal is consistent with the highest acquisition frequency; According to the padded signals, an aligned EEG signal, an aligned body temperature signal and an aligned motion signal are obtained.

6. The method according to claim 2, wherein: The generating of a power spectrum time-frequency diagram according to the preprocessed EEG signal and the preprocessed motion signal comprises: Performing short-time Fourier transform on the preprocessed EEG signal and the preprocessed motion signal to obtain power spectrum data of the preprocessed EEG signal and the preprocessed motion signal; The power spectrum time-frequency diagram is obtained according to the power spectrum data.

7. The method according to claim 5, wherein: The preprocessed EEG signals include multiple channels; the preprocessed motion signals include multiple channels; each EEG signal corresponds to a power spectrum time-frequency diagram, and each motion signal corresponds to a power spectrum time-frequency diagram.

8. The method according to claim 2, wherein: The step of modifying at least one pixel information of the power spectrum time-frequency diagram according to the preprocessed body temperature signal to obtain the power spectrum image includes: Determine a corresponding thermal value color block according to the preprocessed body temperature signal; The thermal value color block is added to the power spectrum time-frequency diagram to obtain the power spectrum image.

9. The method according to claim 1, wherein: The power spectrum feature includes at least one of spectrum energy, frequency band energy ratio, and spectrum change speed.

10. The method according to claim 1, further comprising: Acquire resting-state EEG data when such data exist; In the absence of resting-state EEG data, a resting-state EEG data acquisition instruction is sent to a signal acquisition end; and data fed back by the signal acquisition end in response to the resting-state EEG data acquisition instruction is received.

11. The method according to claim 1, wherein: The step of determining the signal classification according to the power spectrum characteristics and the resting-state EEG data of the detected person includes: Inputting the power spectrum features and the resting-state EEG data into a classification model to obtain output data; Based on the output data, determine the signal classification.

12. The method according to any one of claims 1 to 11, further comprising: Obtaining a power consumption mode indication signal; When the power consumption mode indication signal indicates a low power consumption mode, obtaining a power spectrum time-frequency diagram according to the EEG signal and the motion signal; Determining a febrile convulsion detection result for the detected person according to the power spectrum time-frequency diagram, the body temperature signal and a preset classification threshold; In the case where the power consumption mode indication signal indicates a non-low power consumption mode, the step of generating a power spectrum image according to the EEG signal, the body temperature signal and the motion signal is performed.

13. A method for processing brain-computer signals for febrile convulsions, comprising: Obtaining the electroencephalogram (EEG) bio-signals, body temperature bio-signals and movement bio-signals of the person being tested; Converting the EEG biosignal, body temperature biosignal and motion biosignal into EEG signals, body temperature signals and motion signals identifiable by a data analysis terminal in sequence; The EEG bio-signals, body temperature bio-signals and motion bio-signals are sent to the data analysis end, so that the data analysis end can generate a power spectrum image based on the EEG signals, the body temperature signals and the motion signals; based on the power spectrum image, power spectrum features characterizing the parameter features of the febrile convulsions of the detected person are obtained; based on the power spectrum features and the resting-state EEG data of the detected person, signal classification is determined; based on the signal classification, a febrile convulsion detection result for the detected person is determined.

14. The method according to claim 13, wherein: The EEG bio-signals include those collected by electrodes arranged at positions O2, T6, T4, F8, Fp2, Fp1, F7, T3, T5, and O1 in the EEG 10-20 system; the motion bio-signals include angular velocity bio-signals and acceleration bio-signals.

15. A febrile convulsion brain-computer signal processing device, comprising: A signal acquisition module is used to obtain the EEG signal, body temperature signal and movement signal of the detected person; A power spectrum image generation module, used to generate a power spectrum image according to the EEG signal, the body temperature signal and the motion signal; A power spectrum feature acquisition module, used for acquiring power spectrum features according to the power spectrum image; A signal classification module, used to determine the signal classification according to the power spectrum characteristics and the resting-state EEG data of the detected person; The result module is used to determine the febrile convulsion detection result of the detected person according to the signal classification.

16. A febrile convulsion brain-computer signal processing device, comprising: A bio-signal acquisition module is used to obtain the EEG bio-signals, body temperature bio-signals and movement bio-signals of the person being tested; A bio-signal conversion module, used to convert the EEG bio-signal, body temperature bio-signal and motion bio-signal into EEG signals, body temperature signals and motion signals identifiable by a data analysis terminal in sequence; A sending module is used to send the EEG bio-signals, body temperature bio-signals and motion bio-signals to the data analysis end, so that the data analysis end can generate a power spectrum image based on the EEG signals, the body temperature signals and the motion signals; obtain power spectrum features characterizing the parameter features of febrile convulsions of the detected person based on the power spectrum image; determine signal classification based on the power spectrum features and the resting EEG data of the detected person; and determine the febrile convulsion detection result of the detected person based on the signal classification.

17. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-14.

Images