Brain neural signal processing method and device, electronic equipment and storage medium

By filtering, decomposing and feature extraction of brain nerve signals, pathological recognition results for premature gastric cancer are generated, which solves the problem of low recognition efficiency of premature gastric cancer in the prior art and achieves more accurate pathological recognition.

CN119066469APending Publication Date: 2024-12-03RENMIN HOSPITAL OF WUHAN UNIVERSITY (HUBEI GENERAL HOSPITAL)
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202411173898.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-26
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

The prior art cannot effectively identify different stages of premature gastric cancer, resulting in inefficient identification.

Method used

By obtaining the target brain nerve signal, filtering and decomposing, a feature extraction strategy is used to extract time domain, frequency domain and nonlinear features, splicing feature sets, calculating the minimum distance from standard features, and generating pathological recognition results.

Benefits of technology

It improves the identification efficiency of different stages of premature gastric cancer, provides accurate pathological identification reports, and helps clinical medicine improve diagnostic efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119066469A_ABST
    Figure CN119066469A_ABST
Patent Text Reader

Abstract

The invention discloses a brain neural signal processing method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining a to-be-processed target brain neural signal of a target object, carrying out the feature extraction processing of the target brain neural signal, obtaining a target feature set related to a target type of pathology in the target brain neural signal, the target type pathology comprises a precancerous low-grade pathology, an abnormal hyperplasia type pathology, an early gastric cancer type pathology, a progression stage cancer type pathology and an early gastric cancer postoperative type pathology, performing corresponding processing on the target feature set based on a target processing strategy corresponding to the target type pathology to obtain a target processing result, the target processing result is used for generating a pathology recognition result about whether the target object belongs to the target type pathology, and the pathology recognition result comprises a pathology recognition report. By adopting the embodiment of the invention, the pathological recognition results of different types of pathologies can be recognized through the cranial neural signal, and the recognition efficiency of pathological recognition is effectively improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of medical assistance technology, and in particular to a brain nerve signal processing method, device, electronic equipment and storage medium. Background Art

[0002] EEG signal analysis has played an increasingly important role in brain science research. With the proposal of the brain-stomach axis theory, a large number of studies on the connection between the brain and internal organs have also been carried out. For example, the paper "Brain regulation of gastric dysfunction induced by stress" shows that the accumulation of daily stress in life often leads to gastrointestinal dysfunction. Functional MRI (fMRI) used in patients with stress-related functional dyspepsia and irritable bowel syndrome showed abnormal brain activity, indicating that there is crosstalk between the brain and the stomach under chronic stress (CS) conditions.

[0003] The paper "Vagus nerve signa has an inhibitory influence on the development of peritoneametastasis in murine gastric cancer" shows that the vagus nerve is the only pathway for transmitting parasympathetic nerve signals between the brain and the thoracic and abdominal organs, thereby exerting anti-inflammatory effects through the cholinergic anti-inflammatory pathway. Preserving the vagus nerve may have the advantage of reducing peritoneal recurrence in surgery for advanced gastric cancer.

[0004] In 2016, a team from Harvard Medical School discovered the molecular mechanism of the vagus nerve, which is responsible for sensing "fullness" and controlling nutrient absorption. It has two different types of sensory neurons (GPR65 and GLP1R) distributed in the gastrointestinal tract, which independently perform the two major functions of nutrient absorption and food intake control. The vagus nerve is the longest cranial nerve in the body. It continuously sends sensory information about the internal organs to the brain and plays a vital role. Because it regulates the functions of internal organs, such as digestion, heart rate and breathing. It also helps control reflex actions such as coughing, swallowing and sneezing. This is expected to provide an important target for the development of drugs for weight loss, type 2 diabetes, gastrointestinal disease treatment, etc.

[0005] The paper "Multiple areas of the cerebral cortex influence the stomach" shows that the cortical neurons that affect the parasympathetic output of the stomach originate from the rostral insula and parts of the medial prefrontal cortex, which are related to mutual feeling and emotional control. Correspondingly, the cortical neurons that affect the sympathetic output to the stomach mostly originate from the primary motor cortex, primary somatosensory cortex, and secondary motor cortex, which are related to skeletal motor control and action.

[0006] However, among the many studies on the connection between the brain and internal organs, there is no research on the association mechanism between brain nerve signals and early gastric cancer. As a result, it is impossible to provide effective assistance for the identification of different stages of early gastric cancer in clinical medicine, and it is impossible to improve the recognition efficiency of different stages of early gastric cancer. Summary of the invention

[0007] The purpose of the embodiments of the present invention is to provide a method, device, electronic device and storage medium for processing brain nerve signals to solve the technical problem of being unable to improve the recognition efficiency of different stages of early gastric cancer.

[0008] In a first aspect, an embodiment of the present invention provides a method for processing cranial nerve signals, comprising:

[0009] Acquiring a target cranial nerve signal to be processed from a target object;

[0010] Performing feature extraction processing on the target cranial nerve signal to obtain a target feature set related to a target type pathology in the target cranial nerve signal, wherein the target type pathology includes a precancerous low-grade type pathology, a dysplastic hyperplasia type pathology, an early gastric cancer type pathology, an advanced cancer type pathology, and an early gastric cancer postoperative type pathology;

[0011] Based on the target processing strategy corresponding to the target type pathology, the target feature set is processed accordingly to obtain a target processing result, which is used to generate a pathology recognition result of whether the target object belongs to the target type pathology, and the pathology recognition result includes a pathology recognition report.

[0012] In some embodiments, the step of obtaining a target cranial nerve signal to be processed of the target subject includes:

[0013] Acquiring a target original cranial nerve signal of a target object through a target cranial nerve signal acquisition device;

[0014] Performing filtering processing on the target original cranial nerve signal to obtain a filtered cranial nerve signal;

[0015] Decomposing the filtered cranial nerve signal according to the number of electrode channels and frequency band type of the target cranial nerve signal acquisition device to obtain decomposed cranial nerve signals of different frequency band types in each electrode channel;

[0016] Each of the decomposed cranial nerve signals in each electrode channel is cropped using multiple time windows at different time moments, so as to obtain a target cranial nerve signal in each of the decomposed cranial nerve signals within a time window at a different time moment.

[0017] In some embodiments, the step of performing feature extraction processing on the target cranial nerve signal to obtain a target feature set related to the target type pathology in the target cranial nerve signal includes:

[0018] According to the target signal type of the target cranial nerve signal, a target extraction strategy corresponding to the target signal type is adopted to perform corresponding feature extraction processing on the target cranial nerve signal of each time window in each of the decomposed cranial nerve signals to obtain the window feature of each time window in each of the decomposed cranial nerve signals;

[0019] splicing the window features corresponding to the time windows at the same time moment in each electrode channel to obtain a plurality of splicing feature sets corresponding to different time moments;

[0020] Target features related to the target type pathology are extracted from the spliced ​​feature sets corresponding to multiple different time moments, and all the target features are used as a target feature set.

[0021] In some embodiments, the target signal type includes a first cranial nerve signal type corresponding to a scalp electroencephalogram signal, the target cranial nerve signal includes a first cranial nerve signal, and the target extraction strategy includes a first extraction strategy;

[0022] The step of performing corresponding feature extraction processing on the target brain nerve signal of each time window in each of the decomposed brain nerve signals by adopting a target extraction strategy corresponding to the target signal type according to the target signal type of the target brain nerve signal to obtain the window feature of each time window in each of the decomposed brain nerve signals comprises:

[0023] Determining, according to a first cranial nerve signal type of the target cranial nerve signal, a first extraction strategy corresponding to the first cranial nerve signal type;

[0024] Based on the first extraction strategy, the time domain features, frequency domain features and nonlinear features of the first brain nerve signal in each time window in each of the decomposed brain nerve signals are extracted to obtain the window features of each time window in each of the decomposed brain nerve signals.

[0025] In some embodiments, the target signal type includes a second cranial nerve signal type corresponding to a cortical electroencephalogram signal between the scalp and the cerebral cortex, the target cranial nerve signal includes a second cranial nerve signal, and the target extraction strategy includes a second extraction strategy;

[0026] The step of performing corresponding feature extraction processing on the target brain nerve signal of each time window in each of the decomposed brain nerve signals by adopting a target extraction strategy corresponding to the target signal type according to the target signal type of the target brain nerve signal to obtain the window feature of each time window in each of the decomposed brain nerve signals comprises:

[0027] Determining, according to the second cranial nerve signal type of the target cranial nerve signal, a second extraction strategy corresponding to the second cranial nerve signal type;

[0028] Based on the second extraction strategy, the time domain features, frequency domain features and time-frequency features of the second brain nerve signal in each time window in each of the decomposed brain nerve signals are extracted to obtain the window features of each time window in each of the decomposed brain nerve signals.

[0029] In some embodiments, the target signal type includes a third cranial nerve signal type corresponding to an electroencephalographic signal in the cerebral cortex, the target cranial nerve signal includes a third cranial nerve signal, a fourth cranial nerve signal, and a fifth cranial nerve signal, and the target extraction strategy includes a third extraction strategy;

[0030] The step of performing corresponding feature extraction processing on the target brain nerve signal of each time window in each of the decomposed brain nerve signals by adopting a target extraction strategy corresponding to the target signal type according to the target signal type of the target brain nerve signal to obtain the window feature of each time window in each of the decomposed brain nerve signals comprises:

[0031] Determining, according to a third cranial nerve signal type of the target cranial nerve signal, a third extraction strategy corresponding to the third cranial nerve signal type;

[0032] Based on the third extraction strategy, the time domain features, frequency domain features and time-frequency features of the second brain nerve signal in the time window of each decomposed brain nerve signal are extracted, the area irregularity features and the center of mass deviation features of the third brain nerve signal in the time window of each decomposed brain nerve signal are extracted, and the discharge density features of the fourth brain nerve signal in the time window of each decomposed brain nerve signal are extracted to obtain the window features of each time window in each decomposed brain nerve signal.

[0033] In some embodiments, the step of performing corresponding processing on the target feature set based on the target processing strategy corresponding to the target type pathology to obtain a target processing result includes:

[0034] Based on the target processing strategy corresponding to the target type pathology, obtaining the target standard feature corresponding to the target type pathology;

[0035] Calculating the minimum distance between the target feature corresponding to all time moments in each electrode channel in the target feature set and the target standard feature;

[0036] Calculate and obtain the number of targets whose minimum distance is not greater than the target standard threshold, and obtain the target processing result;

[0037] The cranial nerve signal processing method further includes:

[0038] In a case where the target processing result indicates that the target quantity is greater than a target threshold, generating a pathology recognition result indicating that the target object belongs to the target type pathology;

[0039] Otherwise, a pathology recognition result is generated indicating that the target object does not belong to the target type of pathology.

[0040] In a second aspect, an embodiment of the present invention provides a brain nerve signal processing device, including:

[0041] An acquisition module, used for acquiring a target cranial nerve signal to be processed from a target object;

[0042] A feature extraction module is used to perform feature extraction processing on the target cranial nerve signal to obtain a target feature set related to a target type pathology in the target cranial nerve signal, wherein the target type pathology includes a precancerous low-grade type pathology, a dysplastic hyperplasia type pathology, an early gastric cancer type pathology, an advanced cancer type pathology, and an early gastric cancer postoperative type pathology;

[0043] A processing module is used to perform corresponding processing on the target feature set based on the target processing strategy corresponding to the target type pathology to obtain a target processing result, wherein the target processing result is used to generate a pathology recognition result of whether the target object belongs to the target type pathology, and the pathology recognition result includes a pathology recognition report.

[0044] In a third aspect, an embodiment of the present invention provides an electronic device, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any one of the above-mentioned brain nerve signal processing methods when executing the computer program.

[0045] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any one of the above-mentioned brain nerve signal processing methods are implemented.

[0046] The embodiments of the present invention provide a method, device, electronic device and storage medium for processing brain nerve signals. The method performs feature extraction processing on the target brain nerve signals of the target object to obtain a target feature set corresponding to the target brain nerve signals, and uses a target processing strategy corresponding to the target type pathology to perform corresponding processing on the target feature set to obtain a target processing result, so as to determine whether the target object belongs to the pathology recognition result of the target type pathology based on the target processing result, thereby effectively improving the recognition efficiency of pathology recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 is a flow chart of a method for processing cranial nerve signals provided by an embodiment of the present invention;

[0048] Figure 2 It is a flow chart of a method for obtaining a target cranial nerve signal of a target object provided by an embodiment of the present invention;

[0049] Figure 3 It is a flow chart of a method for extracting features of brain nerve signals provided by an embodiment of the present invention;

[0050] Figure 4 is a schematic diagram of a third cranial nerve signal provided by an embodiment of the present invention;

[0051] Figure 5a It is a splicing schematic diagram of the window features of the non-implantable EEG signal provided by an embodiment of the present invention;

[0052] Figure 5b It is a splicing schematic diagram of the window features of the semi-implantable EEG signal provided by an embodiment of the present invention;

[0053] Figure 5c It is a splicing schematic diagram of the window features of the implantable EEG signal provided by an embodiment of the present invention;

[0054] Figure 6 is a schematic diagram of extracting target features related to target type pathology provided by an embodiment of the present invention;

[0055] Figure 7 It is a schematic diagram of a process for processing a target feature set provided by an embodiment of the present invention;

[0056] Figure 8a-8f It is a schematic diagram of a display interface of the execution process of the cranial nerve signal processing method provided by an embodiment of the present invention in a terminal;

[0057] Fig. 9 is a structural schematic diagram of a brain nerve signal processing device provided by an embodiment of the present invention;

[0058] Fig.10 is a structural schematic diagram of an electronic device provided by an embodiment of the present invention;

[0059] Fig.11 This is another structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0060] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0062] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0063] Among the related technologies, EEG signal analysis has occupied an increasingly important position in brain science research. With the proposal of the brain-stomach axis theory, a large number of studies on the connection between the brain and internal organs have also been carried out.

[0064] However, among the many studies on the connection between the brain and internal organs, there are no studies on the association mechanism between brain nerve signals and early gastric cancer, and it is impossible to speculate on the different stages of early gastric cancer based on brain nerve signals.

[0065] In order to infer different stages of early gastric cancer through brain nerve signals, the embodiment of the present invention provides a brain nerve signal processing method, see Figure 1 , Figure 1 is a flowchart of a method for processing cranial nerve signals provided by an embodiment of the present invention, the method comprising steps 101 to 103;

[0066] Step 101, obtaining a target brain nerve signal to be processed of a target object.

[0067] Among them, the target object provided in this embodiment can be any object from which brain nerve signals can be collected, such as a patient or an experimental animal, and the target brain nerve signal can be any brain nerve signal corresponding to the patient or the experimental animal.

[0068] The target cranial nerve signals provided in this embodiment can be classified according to the different collection locations of the cranial nerve signals, and may at least include scalp EEG signals at the scalp position, cortical EEG signals between the scalp and the cerebral cortex at the position between the scalp and the cerebral cortex, and internal EEG signals of the cerebral cortex at the position inside the cerebral cortex.

[0069] Specifically, the scalp EEG signal can be obtained through non-implantable means, such as EEG (Electroencephalography) obtained with an EEG cap. The EEG signal is obtained by recording the changes in potential on the scalp. It reflects the synchronous activity of a large range of neural networks. When a large number of neurons are active at the same time, they generate electric currents, which propagate between the scalp and brain tissue and can be detected by electrodes on the scalp. The EEG signal is conducted and filtered by tissues such as the skull and cerebrospinal fluid, and finally forms a waveform recorded on the EEG. EEG signals usually include rhythmic oscillations of multiple frequencies, such as α waves, β waves, θ waves and δ waves, etc. These rhythms reflect the neural activity states of different brain regions.

[0070] Cortical EEG signals can be obtained through semi-implantation, such as ECoG signals obtained by semi-implanted electrodes. ECoG signals can obtain local field potentials LFP (Local Field Potential). LFP is generated by the collective activity of local neurons. When a cluster of neurons simultaneously emits action potentials, the currents they generate will propagate in the surrounding tissues. The propagation of these currents in brain tissue will cause changes in the electric potential in the extracellular fluid, forming LFP signals. LFP mainly reflects the overall situation of local neuronal activity, including the excitation and inhibition states of neurons, synaptic transmission, and synchronization of neuronal networks.

[0071] The EEG signals inside the cerebral cortex can be implanted, for example, in conjunction with a microelectrode array, stereoelectroencephalography (SEEG), deep brain stimulation (DBS), etc. to obtain neuronal discharge signals LFP or spike signals in the subcortical or deep brain tissue.

[0072] In some embodiments, see Figure 2 , Figure 2 FIG. 1 is a flow chart of a method for obtaining a target cranial nerve signal of a target object provided by an embodiment of the present invention. Figure 2 As shown, the step of obtaining the target cranial nerve signal to be processed of the target object provided in this embodiment may include steps 201 to 204;

[0073] Step 201, acquiring a target original cranial nerve signal of a target object through a target cranial nerve signal acquisition device.

[0074] Since the target brain nerve signals provided by this embodiment may include scalp EEG signals at the scalp position, cortical EEG signals between the scalp and the cerebral cortex at the position between the scalp and the cerebral cortex, and internal EEG signals of the cerebral cortex at the position inside the cerebral cortex, this embodiment may use brain nerve signal acquisition equipment corresponding to the brain position for different brain positions.

[0075] Specifically, when the target brain nerve signal to be acquired is a scalp EEG signal at the scalp position, the target brain nerve signal acquisition device may be a non-implantable signal acquisition device, such as an EEG cap. When the target brain nerve signal to be acquired is a cortical EEG signal between the scalp and the cerebral cortex at a position between the scalp and the cerebral cortex, the target brain nerve signal acquisition device may be a semi-implantable signal acquisition device, such as a flexible electrode. When the target brain nerve signal to be acquired is an internal EEG signal of the cerebral cortex at a position inside the cerebral cortex, the target brain nerve signal acquisition device may be an implantable signal acquisition device, such as a microelectrode array device, a stereo EEG acquisition device, or a deep brain electrical stimulation device.

[0076] The target original cranial nerve signal provided in this embodiment is a signal that has not been preprocessed, and the target cranial nerve signal provided in this embodiment is a cranial nerve signal obtained after the target original cranial nerve signal has been preprocessed. Specifically, the preprocessing process provided in this embodiment can refer to steps 1012 to 1014;

[0077] Step 202: filter the target original cranial nerve signal to obtain a filtered cranial nerve signal.

[0078] In this embodiment, the filtering process provided in this embodiment mainly includes denoising, bad track removal and interpolation processing on the target original brain nerve signal.

[0079] Specifically, when the target original brain nerve signal is a scalp EEG signal collected by a non-implantable signal acquisition device, this embodiment can use regression methods, blind source separation algorithms, such as independent component analysis algorithm (ICA), adaptive filtering such as least squares (RLS), least mean squares (LMS) (EMD), Empirical Mode Decomposition) algorithm, and wavelet transform (WT) in the denoising process; in the process of bad channel removal and interpolation, bad channel detection can use automatic detection implemented by the pyprep package, and then perform neighborhood interpolation on the removed channels through their adjacent channels.

[0080] When the target original brain nerve signal is a cortical EEG signal collected by a semi-implantable signal acquisition device, this embodiment can use a time domain-based filtering method, such as median filtering, mean filtering, Kalman filtering, etc. during the filtering process; it can also use a frequency domain-based filtering method, such as low-pass filtering, high-pass filtering, band-pass filtering, etc.; in the process of bad channel elimination, this embodiment mainly eliminates abnormal signal channels directly.

[0081] When the target original brain nerve signal is an internal EEG signal of the cerebral cortex collected by an implantable signal acquisition device, this embodiment may use bandpass filtering during the denoising process to eliminate interference from low-frequency field potentials and high-frequency noise.

[0082] Step 203, decomposing the filtered brain nerve signal according to the number of electrode channels and frequency band type of the target brain nerve signal acquisition device to obtain decomposed brain nerve signals of different frequency band types in each electrode channel.

[0083] In this embodiment, since the target brain nerve signal acquisition device provided in this embodiment has different numbers of electrode channels, each electrode channel has different frequency bands. Therefore, after filtering the target original brain nerve signal, this embodiment also needs to decompose the filtered brain nerve signal according to different numbers of electrode channels and different frequency bands.

[0084] As an optional embodiment, the signal acquisition device provided in this embodiment may be provided with a plurality of electrodes, and each electrode may correspond to a channel. Assuming that the non-implantable signal acquisition device provided in this embodiment has a total of M1 channels, and the non-implantable EEG signals collected by each channel may include 5 frequency bands, such as delta (<4Hz), theta (4-7Hz), alpha (8-13Hz), beta (14-31Hz) and gamma (>32Hz), after an EEG signal is decomposed according to the channel and frequency band, there are 5×M1 decomposed brain nerve signals.

[0085] Assuming that the semi-implantable signal acquisition device provided in this embodiment has M2 channels, the semi-implantable EEG signal collected by each channel may include 1 frequency band, for example, the 60Hz to 200Hz frequency band. After an ECoG signal is decomposed according to the frequency band, there are M2 decomposed brain nerve signals.

[0086] Assuming that the implantable signal acquisition device provided in this embodiment has a total of M3 channels, and the semi-implantable EEG signals collected by each channel can include three signals: LFP signal, Spike original signal, and Spike discharge number signal. In this embodiment, the amplitude threshold can be set first to obtain the LFP signal and the Spike signal, and then a dimensionality reduction algorithm such as the PCA algorithm is used to reduce the dimension and extract features of the Spike signal, and then a clustering algorithm such as the DBSCAN algorithm is used to cluster the extracted features. In this way, the neuron classification situation can be obtained, and the filtered implantable signal can be decomposed into three signals: LFP signal, Spike original signal, and Spike discharge number signal. Finally, after an implantable EEG signal is decomposed, 3×M3 decomposed brain nerve signals can be obtained.

[0087] Step 204, using multiple time windows at different time moments to crop each of the decomposed cranial nerve signals in each electrode channel, to obtain a target cranial nerve signal in each of the decomposed cranial nerve signals within a time window at a different time moment.

[0088] After obtaining the decomposed cranial nerve signals, each decomposed cranial nerve signal needs to be window-cropped to obtain the target cranial nerve signals provided in this embodiment.

[0089] Specifically, assuming that each signal needs to be cropped into m time windows, after window cropping the decomposed brain nerve signals of 5×M1 non-implanted EEG signals, 5×M1×m time windows of non-implanted EEG signals can be obtained. ijk , where i represents the i-th channel, j represents the j-th frequency band, and k represents the k-th time window of the j-th frequency band of the i-th channel; after window clipping the decomposed brain nerve signals of M2 semi-implantable EEG signals, M2×m time windows B of semi-implantable EEG signals can be obtained. ik , where i represents the i-th channel, and k represents the k-th time window of the i-th channel; after window clipping the decomposed brain nerve signals of 3×M3 implantable EEG signals, 3×M3×m time windows Q of implantable EEG signals can be obtained. ijk , where i represents the i-th channel, j represents the j-th signal (LFP signal, original Spike signal (retain the Spike signal fragment in the original signal, and set the signal amplitude of other intervals to 0), Spike discharge number signal), and k represents the k-th time window of the j-th signal of the i-th channel.

[0090] Among them, the brain nerve signal in the time window obtained after clipping is the target brain nerve signal provided in this embodiment.

[0091] Step 102, performing feature extraction processing on the target cranial nerve signal to obtain a target feature set related to the target type pathology in the target cranial nerve signal.

[0092] In this embodiment, the target type pathology provided in this embodiment may include precancerous low-grade type pathology, dysplasia type pathology, early gastric cancer type pathology, advanced cancer type pathology, and early gastric cancer postoperative type pathology.

[0093] Different brain nerve signals correspond to different feature extraction strategies. For details, please refer to Figure 3 , Figure 3 FIG. 1 is a flow chart of a method for extracting features of brain nerve signals provided by an embodiment of the present invention. Figure 3 As shown, the step of performing feature extraction processing on the target cranial nerve signal to obtain a target feature set related to the target type pathology in the target cranial nerve signal provided in this embodiment may include steps 301 to 303;

[0094] Step 301, according to the target signal type of the target brain nerve signal, adopt the target extraction strategy corresponding to the target signal type, perform corresponding feature extraction processing on the target brain nerve signal of each time window in each of the decomposed brain nerve signals, and obtain the window feature of each time window in each of the decomposed brain nerve signals.

[0095] In one embodiment, the target signal type provided in this embodiment may include a first cranial nerve signal type corresponding to a scalp EEG signal, that is, a non-implantable EEG signal type, the target cranial nerve signal includes a first cranial nerve signal, and the target extraction strategy may be a first extraction strategy. Then step 301 provided in this embodiment may specifically include: determining a first extraction strategy corresponding to the first cranial nerve signal type according to the first cranial nerve signal type of the target cranial nerve signal; based on the first extraction strategy, extracting the time domain features, frequency domain features and nonlinear features of the first cranial nerve signal in each time window in each of the decomposed cranial nerve signals to obtain the window features of each time window in each of the decomposed cranial nerve signals.

[0096] Among them, the first extraction strategy for the first cranial nerve signal provided in this embodiment is: extracting the time domain characteristics, frequency domain characteristics and nonlinear characteristics of the first cranial nerve signal in each of the time windows in each of the decomposed cranial nerve signals. Specifically, for the first cranial nerve signal of the non-implantable EEG signal type, its corresponding time domain characteristics and frequency domain characteristics may include Hjorth parameters-activity characteristics (activity, mobility, complexity), maximum power spectrum frequency characteristics, and maximum power spectrum density characteristics; nonlinear characteristics may include correlation dimension characteristics, Lyapunov exponent characteristics, and singular spectrum entropy characteristics.

[0097] In a second embodiment, the target signal type provided in this embodiment may include a second brain nerve signal type corresponding to a cortical EEG signal, that is, a semi-implantable EEG signal type. The target brain nerve signal includes a second brain nerve signal. The target extraction strategy may be a second extraction strategy. Then step 301 provided in this embodiment may specifically include: determining a second extraction strategy corresponding to the second brain nerve signal type according to the second brain nerve signal type of the target brain nerve signal; based on the second extraction strategy, extracting the time domain features, frequency domain features and time-frequency features of the second brain nerve signal in each time window in each of the decomposed brain nerve signals to obtain the window features of each time window in each of the decomposed brain nerve signals.

[0098] Among them, the second extraction strategy for the second cranial nerve signal provided in this embodiment is: extracting the time domain features, frequency domain features and time-frequency features of the second cranial nerve signal in each of the time windows in each of the decomposed cranial nerve signals. Specifically, for the second cranial nerve signal of the semi-implantable EEG signal type, its corresponding time domain features may include the mean, variance, poles, peak, kurtosis and other features obtained by statistics; frequency domain features may include centroid frequency, mean square frequency, root mean square frequency, frequency variance, frequency standard deviation and other features; time-frequency features may include wavelet coefficients, wavelet packet energy, Hilbert spectrum, marginal spectrum and other features.

[0099] In a third embodiment, the target signal type provided in this embodiment may include a third cranial nerve signal type corresponding to the EEG signal inside the cerebral cortex, that is, an implantable EEG signal type. The target cranial nerve signal also includes a third cranial nerve signal, a fourth cranial nerve signal and a fifth cranial nerve signal. The target extraction strategy may be a third extraction strategy. Then step 301 provided in this embodiment may specifically include: determining a third extraction strategy corresponding to the third cranial nerve signal type according to the third cranial nerve signal type of the target cranial nerve signal; based on the third extraction strategy, extracting the time domain features, frequency domain features and time-frequency features of the third cranial nerve signal in the time window of each of the decomposed cranial nerve signals, extracting the area irregularity features and the center of mass deviation features of the fourth cranial nerve signal in the time window of each of the decomposed cranial nerve signals, and extracting the discharge density features of the fifth cranial nerve signal in the time window of each of the decomposed cranial nerve signals, to obtain the window features of each of the time windows in each of the decomposed cranial nerve signals.

[0100] Among them, the third extraction strategy for the third cranial nerve signal, the fourth cranial nerve signal, and the fifth cranial nerve signal provided in this embodiment is: extracting the time domain characteristics, frequency domain characteristics, and time-frequency characteristics of the third cranial nerve signal in each time window in each of the decomposed cranial nerve signals, extracting the area irregularity characteristics and the center of mass deviation characteristics of the fourth cranial nerve signal, and extracting the discharge density characteristics of the fifth cranial nerve signal.

[0101] Specifically, for the third cranial nerve signal of the implantable EEG signal type, its corresponding time domain characteristics may include the mean, variance, poles, peak, kurtosis and other characteristics obtained by statistics; the frequency domain characteristics may include the centroid frequency, mean square frequency, root mean square frequency, frequency variance, frequency standard deviation and other characteristics; the time-frequency characteristics may include wavelet coefficients, wavelet packet energy, Hilbert spectrum, marginal spectrum and other characteristics.

[0102] For the fourth cranial nerve signal of the implanted EEG signal type, the corresponding area irregularity characteristic can be S Δ = std(S A ,S B ,S C ), the centroid deviation degree characteristic can be H Δ = std(H A ,H B ,H C ). Among them, see Figure 4 , Figure 4 is a schematic diagram of a third cranial nerve signal provided by an embodiment of the present invention, such as Figure 4 As shown, H A ,H B,H C The amplitude heights of the centroids of the three parts of area A, area B, and area C are taken as true values, such as the curvature deviation in Indicates the maximum curvature of the boundary curves of area A, area B, and area C.

[0103] For the fifth cranial nerve signal of the implantable EEG signal type, the corresponding discharge density feature can be the discharge density in each time window of the Spike discharge number signal: Among them, N D Indicates the number of discharges, N Y Represents the number of neurons in each time window.

[0104] According to the above-mentioned embodiment provided by the present invention, the window characteristics of each time window in each decomposed cranial nerve signal can be obtained.

[0105] Step 302: splicing the window features corresponding to the time windows at the same time moment in each electrode channel to obtain a plurality of splicing feature sets corresponding to different time moments.

[0106] After obtaining the window features of each time window, this embodiment also needs to splice the window features of the time window at the same time moment in each electrode channel, so as to obtain multiple splicing feature sets at multiple different time moments in each electrode channel.

[0107] For details, please also refer to Figure 5a , Figure 5b as well as Figure 5c , Figure 5a is a splicing schematic diagram of the window features of the non-implantable EEG signal provided by an embodiment of the present invention, Figure 5b is a splicing schematic diagram of the window features of the semi-implantable EEG signal provided by an embodiment of the present invention, Figure 5c is a splicing schematic diagram of the window features of the implantable EEG signal provided by the embodiment of the present invention.

[0108] like Figure 5a As shown, for the first cranial nerve signal of the non-implantable EEG signal type, this embodiment performs feature splicing on the features extracted from the first cranial nerve signal in the time window at the same time moment in all frequency bands of each channel, that is, splicing the window features of the time window at the kth time moment in the 5 frequency bands in the i-th channel to obtain the spliced ​​feature T ik , and then all the splicing features T ik As a collection of concatenated features.

[0109] like Figure 5bAs shown, for the second cranial nerve signal of the semi-implantable EEG signal type, this embodiment performs feature splicing on the features extracted from the second cranial nerve signal in the time window at the same time moment in each channel, that is, splicing the window features of the time window at the kth time moment in a frequency band in the i-th channel to obtain the spliced ​​feature T Bik , and then all the splicing features T Bik As a collection of concatenated features.

[0110] like Figure 5c As shown, for the third, fourth, and fifth cranial nerve signals of the implantable EEG signal type, this embodiment performs feature splicing on the features extracted from the third, fourth, and fifth cranial nerve signals in the time window at the same time moment in each channel, that is, splicing the window features of the time window at the kth time moment in the i-th channel to obtain the spliced ​​feature T Brik , and then all the splicing features T Brik As a collection of concatenated features.

[0111] Step 303: extract target features related to the target type pathology from the spliced ​​feature sets corresponding to multiple different time moments, and use all the target features as a target feature set.

[0112] In this embodiment, in order to improve the recognition accuracy of the target type pathology, after obtaining the splicing feature set, this embodiment also needs to extract the target features related to the target type pathology from the splicing feature sets at different time moments. Specifically, this embodiment can use a machine learning model trained to identify splicing features of different types of pathologies to identify the splicing feature sets corresponding to non-implanted EEG signals, semi-implanted EEG signals, and implanted EEG signals, so as to identify the target features corresponding to the non-implanted EEG signals, semi-implanted EEG signals, and implanted EEG signals, respectively, and then use all the target features corresponding to each EEG signal type as the target feature set.

[0113] Among them, this embodiment can use a large number of different types of pathology (including patients with and without precancerous low-level pathology, patients with and without dysplasia pathology, patients with and without early gastric cancer pathology, patients with and without progressive cancer pathology, patients with and without postoperative early gastric cancer pathology) corresponding to different EEG signal types (including non-implanted EEG signals, semi-implanted EEG signals and implanted EEG signals) as a training set, and add labels to the training set, for example, the label of the splicing feature of the corresponding type of pathology can be 1, and the label of the splicing feature of the patient without the corresponding type of pathology can be 0. Then use the training set to train the initial model such as SVM, random forest RF, KNN and other models, and use the label to guide the training of the model until the model converges. In this way, a machine learning model that can identify the splicing features of different types of pathologies can be obtained.

[0114] For details, see Figure 6 , Figure 6 is a schematic diagram of extracting target features related to target type pathology provided by an embodiment of the present invention, such as Figure 6 As shown, by using the trained machine learning model provided in the above embodiment, target features related to the target type pathology can be extracted from the spliced ​​feature sets corresponding to multiple different time moments, and all the target features can be used as the target feature set.

[0115] Among them, the target feature set for non-implantable EEG signals includes: the feature set T of "precancerous low-grade" type pathology F1 、The characteristic set of "dysplasia" type pathology T F2 、Feature set of "early gastric cancer" type pathology T F3 , the characteristic set of "advanced cancer" type pathology T F4 、Feature set of "early gastric cancer after surgery" type pathology T F5 .

[0116] The target feature set for the semi-implantable EEG signal includes: the feature set T of the “low-grade precancerous” type pathology B1 、The characteristic set of "dysplasia" type pathology T B2 、Feature set of "early gastric cancer" type pathology T B3 , the characteristic set of "advanced cancer" type pathology T B4 、Feature set of "early gastric cancer after surgery" type pathology T B5 .

[0117] The target feature set for the implanted EEG signal includes: the feature set T of the “low-grade precancerous” type pathology Q1 、The characteristic set of "dysplasia" type pathology T Q2、Feature set of "early gastric cancer" type pathology T Q3 , the characteristic set of "advanced cancer" type pathology T Q4 、Feature set of "early gastric cancer after surgery" type pathology T Q5 .

[0118] Step 103, based on the target processing strategy corresponding to the target type pathology, the target feature set is processed accordingly to obtain a target processing result, and the target processing result is used to generate a pathology recognition result of whether the target object belongs to the target type pathology.

[0119] The pathology recognition result provided in this embodiment may include a pathology recognition report, and the pathology recognition report can help clinical medicine effectively improve the recognition of target type pathology.

[0120] In this embodiment, after obtaining the target feature sets of different EEG signal types, this embodiment can adopt a target processing strategy corresponding to the target type pathology, perform corresponding processing on the target feature set, and obtain a target processing result, and the target processing result can be used to generate a pathology identification result of whether the target object belongs to the target type pathology. Therefore, by adopting the brain nerve signal processing method provided in the embodiment of the present invention, it is possible to realize the pathology identification results of different types of pathologies through brain nerve signals, thereby effectively improving the recognition efficiency of pathology identification in clinical medicine.

[0121] For details, see Figure 7 , Figure 7 is a flow chart of processing a target feature set provided by an embodiment of the present invention, such as Figure 7 As shown, the target processing strategy based on the target type pathology corresponding to the target feature set provided in this embodiment performs corresponding processing to obtain the target processing result, which may specifically include steps 401 to 403;

[0122] Step 401 : acquiring target standard features corresponding to the target type pathology based on the target processing strategy corresponding to the target type pathology.

[0123] In this embodiment, the target processing strategy provided in this embodiment is used to process the target feature set to obtain a target processing result that can be used to generate a pathology recognition result of whether the target object belongs to the target type pathology. The target standard feature provided in this embodiment is a representation of having a target type pathology, which is used to compare with the target feature in the target feature set, so as to determine whether the target feature is identified as having a target type pathology.

[0124] Among them, this embodiment can obtain a target feature set of the same type of EEG signal type (which can be any one of non-implanted EEG signals, semi-implanted EEG signals and implanted EEG signals) corresponding to a large number of patients with the same type of pathology (which can be any one of the precancerous low-level pathology, dysplasia pathology, early gastric cancer pathology, progressive cancer pathology, or postoperative pathology of early gastric cancer) and then average all the target features in the target feature set, and use the average value as the target standard feature of the selected type of pathology under the selected EEG signal type. In this way, standard features of different types of pathologies under different EEG signal types can be obtained.

[0125] Step 402, calculating the minimum distance between the target feature corresponding to all time moments in each electrode channel in the target feature set and the target standard feature.

[0126] In this embodiment, since the target standard features provided in this embodiment are representations of suffering from target type pathology, this embodiment can use a similarity calculation method to calculate the minimum distance between the target features corresponding to all time moments in each electrode channel in the target feature set and the target standard features, so as to determine the number of target features in the target feature set that are similar to the target standard features, thereby determining whether the brain nerve signals corresponding to the target feature set belong to the target type pathology.

[0127] Specifically, the target feature set T of the “dysplasia” type pathology of non-implantable EEG signals is F2 For example, we can calculate the target feature set T by sliding the time windows at different time moments from the first time window to the last time window. F2 The target feature T of the i-th channel and the j-th window in all time windows of each electrode channel F2ij , and the distance d between the standard feature T2 of the “dysplasia” type pathology under the non-implanted EEG signal type F2ij , the calculation formula is The L2 standard feature is the length of the T2 vector, L F2 is the target feature set T F2 Then, determine the minimum distance d that exists in each channel at the jth window j =min(d F2ij ).

[0128] Step 403, calculate and obtain the number of targets whose minimum distance is not greater than the target standard threshold, and obtain the target processing result.

[0129] In this embodiment, the target standard threshold provided in this embodiment can be used as a measure of whether the target feature in the target feature set is similar to the target standard feature. Specifically, the target feature set T of the "dysplasia" type pathology of the non-implantable EEG signal is F2 For example, the target standard threshold can be calculated by using a large number of target feature sets of non-"dysplasia" type pathology patients and a large number of target feature sets of "dysplasia" type pathology patients, and the target standard feature similarity is calculated, and the receiver operating characteristic curve (ROC) ROC-F2 is plotted, and the optimal target standard threshold τ is determined at the same time. F2 .

[0130] Similarly, the above method can be used to obtain the curve ROC-F1 and target standard threshold τ corresponding to the non-implantable EEG signal under the "low-grade precancerous" type pathology. F1 , the corresponding curve ROC-F3 under the pathology of "early gastric cancer" type, the target standard threshold τ F3 , the corresponding curve ROC-F4 under the pathology of "advanced cancer" type, the target standard threshold τ F4 , the corresponding curve ROC-F5 and target standard threshold τ under the pathology of "early gastric cancer after surgery" F5 The curve ROC-B1 and target standard threshold τ of the semi-implantable EEG signal under the pathology of "low-grade precancerous" type are obtained. B1 , the corresponding curve ROC-B2 under the pathological condition of "dysplasia", the target standard threshold τ B2 , the corresponding curve ROC-B3 under the pathology of "early gastric cancer" type, the target standard threshold τ B3 , the corresponding curve ROC-B4 under the pathology of "advanced cancer" type, the target standard threshold τ B4 , the corresponding curve ROC-B5 under the pathology of "early gastric cancer after surgery" and the target standard threshold τ B5 The curve ROC-Q1 and target standard threshold τ of the implanted EEG signal under the pathology of "low-grade precancerous" type are obtained. Q1 , the corresponding curve ROC-Q2 under the pathological condition of "dysplasia", the target standard threshold τ Q2 , the corresponding curve ROC-Q3 under the pathology of "early gastric cancer" type, the target standard threshold τ Q3 , the corresponding curve ROC-Q4 under the pathology of "advanced cancer" type, the target standard threshold τ Q4 , the corresponding curve ROC-Q5 and target standard threshold τ under the pathology of "early gastric cancer after surgery" Q5 .

[0131] In this way, when the curves and target standard thresholds corresponding to EEG signals of different EEG signal types under different types of pathologies are obtained, the number of targets whose minimum distance is not greater than the target standard threshold can still be accurately calculated and obtained after obtaining the EEG signals of different EEG signal types, and the target processing result can be obtained, so as to accurately generate a pathology recognition result of whether the target object belongs to the target type pathology based on the target processing result, so as to improve the recognition efficiency of whether the target object belongs to the target type pathology in clinical medicine.

[0132] In some embodiments, after obtaining the target processing result, a pathology recognition result of whether the target object belongs to the target type pathology can also be generated according to the target processing result. Specifically, the brain nerve signal processing method provided in this embodiment can also include: when the target processing result indicates that the target number is greater than the target threshold, generating a pathology recognition result that the target object belongs to the target type pathology; otherwise, generating a pathology recognition result that the target object does not belong to the target type pathology.

[0133] Among them, the number of targets whose minimum distance provided by this embodiment is not greater than the target standard threshold can be used to indicate the degree to which the target feature set belongs to the target type pathology, wherein the target number can be set according to the number of different time windows, for example, 50%, 60% or 70% of the time windows are used as the target number, so the specific value of the target value is set by different application scenarios and is not specifically limited here.

[0134] As an optional embodiment, when the curves and target standard thresholds corresponding to EEG signals of different EEG signal types under different types of pathologies are obtained, in order to improve the accuracy of the target processing results and to improve the subsequent recognition accuracy of whether the target object belongs to the target type of pathology according to the target processing results, this embodiment can compare the area under the ROC curve AUC (Area Under Curve) corresponding to EEG signals of different EEG signal types for the selected type of pathology to determine which type of EEG signal has the highest recognition accuracy for identifying the selected type of pathology, so that the corresponding brain nerve signal acquisition equipment can be used to acquire brain nerve signals. Similarly, this embodiment can also compare the area under the ROC curve AUC corresponding to different types of pathologies for the selected type of EEG signal to determine which type of pathology has the highest recognition accuracy.

[0135] It should be noted that, in specific implementation, the EEG signal type and the target type of pathology may be selected in coordination according to actual conditions to improve the recognition accuracy of the target type of pathology.

[0136] In the related art, the identification and screening of early gastric cancer is usually carried out through traditional gastric disease examinations in the gastroenterology department. In order to improve the efficiency of the identification and screening of early gastric cancer in the related art, this embodiment can collect the patient's cranial nerve signals through other departments equipped with nerve signal acquisition equipment, and execute the cranial nerve signal processing method provided by the embodiment of the present invention, so as to identify and screen early gastric cancer through cranial nerve signals, and set a corresponding identification report, and choose whether to send it to the gastroenterology department to determine whether further examination and pathological screening are needed, thereby effectively improving the efficiency of the identification and screening of early gastric cancer.

[0137] For details, please refer to Figure 8a-8f , Figure 8a-8f A schematic diagram of a display interface showing the execution process of the brain nerve signal processing method provided by this embodiment in a terminal, such as Figure 8a-8e As shown, Figure 8a The interface for selecting the type of EEG signal is shown, that is, determining the EEG signal acquisition method required for the target object; Figure 8b A schematic diagram of the processing process of non-implantable EEG signals is shown. Figure 8b , shows the target feature set of the non-implantable EEG signal, and whether the target feature 1-target feature N is abnormal (whether it belongs to the target type pathology); Figure 8c The schematic diagram of the processing process of the semi-implantable EEG signal is shown. Figure 8c , which shows the target feature set of the semi-implantable EEG signal, and whether the target feature 1 to the target feature N are abnormal (whether they belong to the target type pathology); Figure 8d A schematic diagram of the processing process of implantable EEG signals is shown. Figure 8d , which shows the target feature set of the implanted EEG signal, and whether the target feature 1 to the target feature N are abnormal (whether they belong to the target type pathology); Figure 8e The display interface of the pathology identification result (pathology identification report) of whether the target object belongs to the target type pathology is generated according to the target processing result. Figure 8e The options of generating and sending reports are also shown, as well as identifying the report preview area, so as to facilitate the different needs of the target objects; Figure 8f for Figure 8e The display page formed after the pathology recognition report preview area is expanded in the display page, which displays the pathology recognition report. Figure 8f Only one identification report template of a pathology identification report is shown.

[0138] in, Figure 8b , Figure 8c as well as Figure 8dIt can be used to identify any one of the "before low-grade precancerous" type pathology, "dysplasia" type pathology, "early gastric cancer" type pathology, "advanced cancer" type pathology, and "early gastric cancer after surgery" type pathology.

[0139] It should be noted that the display interface diagram of the execution process of the cranial nerve signal processing method in the terminal provided in this embodiment is not limited to the display interface diagram provided in the above embodiment. Figures 8a to 8f The form shown may also be other display interface diagrams that can represent the execution process of the brain nerve signal processing method provided by this embodiment in the terminal, and no specific limitation is made here.

[0140] The above completes the detailed description of a method for processing cranial nerve signals provided by this embodiment.

[0141] In summary, an embodiment of the present invention provides a method for processing cranial nerve signals, which includes obtaining a target cranial nerve signal to be processed by a target object, performing feature extraction processing on the target cranial nerve signal, and obtaining a target feature set related to a target type pathology in the target cranial nerve signal, wherein the target type pathology includes a precancerous low-grade type pathology, a dysplastic hyperplasia type pathology, an early gastric cancer type pathology, an advanced cancer type pathology, and an early gastric cancer postoperative type pathology, and based on a target processing strategy corresponding to the target type pathology, the target feature set is processed accordingly to obtain a target processing result, and the target processing result is used to generate a pathology recognition result of whether the target object belongs to the target type pathology, and the pathology recognition result includes a pathology recognition report. By adopting the embodiment of the present invention, pathology recognition results of different types of pathologies can be identified through cranial nerve signals, and the recognition efficiency of pathology recognition can be effectively improved.

[0142] According to the method described in the above embodiment, this embodiment will be further described from the perspective of a brain nerve signal processing device. The brain nerve signal processing device can be implemented as an independent entity or integrated into an electronic device, such as a terminal. The terminal may include a mobile phone, a tablet computer, etc.

[0143] See also Fig. 9 , Fig. 9 is a structural schematic diagram of a brain nerve signal processing device provided by an embodiment of the present invention, such as Fig. 9 As shown, the brain nerve signal processing device 500 provided in the embodiment of the present invention includes: an acquisition module 501, a feature extraction module 502 and a processing module 503;

[0144] The acquisition module 501 is used to acquire the target cranial nerve signal to be processed of the target object.

[0145] In this embodiment, the acquisition module 501 provided in this embodiment is specifically used to: acquire the target original brain nerve signal of the target object through the target brain nerve signal acquisition device; filter the target original brain nerve signal to obtain a filtered brain nerve signal; decompose the filtered brain nerve signal according to the number of electrode channels and frequency band type of the target brain nerve signal acquisition device to obtain decomposed brain nerve signals of different frequency band types in each electrode channel; use multiple time windows at different time moments to crop each of the decomposed brain nerve signals in each electrode channel to obtain the target brain nerve signal in the time window at different time moments in each of the decomposed brain nerve signals.

[0146] The feature extraction module 502 is used to perform feature extraction processing on the target brain nerve signal to obtain a target feature set related to the target type pathology in the target brain nerve signal, and the target type pathology includes precancerous low-grade type pathology, dysplasia type pathology, early gastric cancer type pathology, advanced cancer type pathology, and early gastric cancer postoperative type pathology.

[0147] In this embodiment, the feature extraction module 502 provided in this embodiment is specifically used for: according to the target signal type of the target brain nerve signal, adopting the target extraction strategy corresponding to the target signal type, performing corresponding feature extraction processing on the target brain nerve signal of each time window in each of the decomposed brain nerve signals, and obtaining the window features of each time window in each of the decomposed brain nerve signals; splicing the window features corresponding to the time windows at the same time moment in each electrode channel to obtain a set of spliced ​​features corresponding to multiple different time moments; extracting the target features related to the target type pathology from the spliced ​​feature sets corresponding to multiple different time moments, and taking all the target features as the target feature set.

[0148] In one embodiment, the target signal type provided in this embodiment includes a first cranial nerve signal type corresponding to a scalp EEG signal, the target cranial nerve signal includes a first cranial nerve signal, and the target extraction strategy includes a first extraction strategy. The feature extraction module 502 provided in this embodiment is also specifically used to: determine a first extraction strategy corresponding to the first cranial nerve signal type according to the first cranial nerve signal type of the target cranial nerve signal; based on the first extraction strategy, extract the time domain features, frequency domain features and nonlinear features of the first cranial nerve signal in each time window in each of the decomposed cranial nerve signals to obtain the window features of each time window in each of the decomposed cranial nerve signals.

[0149] In a second embodiment, the target signal type provided in this embodiment includes a second cranial nerve signal type corresponding to the cortical EEG signal between the scalp and the cerebral cortex, the target cranial nerve signal includes a second cranial nerve signal, and the target extraction strategy includes a second extraction strategy. The feature extraction module 502 provided in this embodiment is also specifically used to: determine the second extraction strategy corresponding to the second cranial nerve signal type according to the second cranial nerve signal type of the target cranial nerve signal; based on the second extraction strategy, extract the time domain features, frequency domain features and time-frequency features of the second cranial nerve signal in each time window in each of the decomposed cranial nerve signals to obtain the window features of each time window in each of the decomposed cranial nerve signals.

[0150] In a third embodiment, the target signal type provided in this embodiment includes a third cranial nerve signal type corresponding to the electroencephalographic signal inside the cerebral cortex, the target cranial nerve signal includes a third cranial nerve signal, a fourth cranial nerve signal and a fifth cranial nerve signal, the target extraction strategy includes a third extraction strategy, and the feature extraction module 502 provided in this embodiment is further specifically used to: determine a third extraction strategy corresponding to the third cranial nerve signal type according to the third cranial nerve signal type of the target cranial nerve signal; based on the third extraction strategy, extract the time domain features, frequency domain features and time-frequency features of the third cranial nerve signal in the time window of each of the decomposed cranial nerve signals, extract the area irregularity features and the center of mass deviation features of the fourth cranial nerve signal in the time window of each of the decomposed cranial nerve signals, and extract the discharge density features of the fifth cranial nerve signal in the time window of each of the decomposed cranial nerve signals, so as to obtain the window features of each of the time windows in each of the decomposed cranial nerve signals.

[0151] The processing module 503 is used to perform corresponding processing on the target feature set based on the target processing strategy corresponding to the target type pathology to obtain a target processing result. The target processing result is used to generate a pathology recognition result of whether the target object belongs to the target type pathology. The pathology recognition result includes a pathology recognition report.

[0152] In this embodiment, the processing module 503 provided in this embodiment is specifically used to: obtain the target standard features corresponding to the target type pathology based on the target processing strategy corresponding to the target type pathology; calculate the minimum distance between the target features corresponding to all time moments in each electrode channel in the target feature set and the target standard features; calculate and obtain the number of targets whose minimum distance is not greater than the target standard threshold, and obtain the target processing result.

[0153] Optionally, the brain nerve signal processing device 500 provided in this embodiment may further include a generation module;

[0154] The generating module is used to generate a pathology recognition result indicating that the target object belongs to the target type pathology when the target processing result indicates that the target quantity is greater than a target threshold; otherwise, generate a pathology recognition result indicating that the target object does not belong to the target type pathology.

[0155] During specific implementation, the above modules and / or units can be implemented as independent entities, or can be arbitrarily combined to be implemented as the same or several entities. The specific implementation of the above modules and / or units can refer to the previous method embodiments. The specific beneficial effects that can be achieved can also refer to the beneficial effects in the previous method embodiments, which will not be repeated here.

[0156] Also, see Fig.10 , Fig.10 1 is a schematic diagram of a structure of an electronic device provided by an embodiment of the present invention, and the electronic device may be a mobile terminal such as a smart phone, a tablet computer, or the like. Fig.10 As shown, the electronic device 600 includes a processor 601 and a memory 602. The processor 601 is electrically connected to the memory 602.

[0157] The processor 601 is the control center of the electronic device 600. It uses various interfaces and lines to connect various parts of the entire electronic device. It executes various functions of the electronic device 600 and processes data by running or loading applications stored in the memory 602 and calling data stored in the memory 602, thereby monitoring the electronic device 600 as a whole.

[0158] In this embodiment, the processor 601 in the electronic device 600 will load the instructions corresponding to the processes of one or more applications into the memory 602 according to the following steps, and the processor 601 will run the application stored in the memory 602, thereby implementing any step in the brain nerve signal processing method provided in the above embodiment.

[0159] The electronic device 600 can implement the steps in any embodiment of the brain nerve signal processing method provided in the embodiments of the present invention, and therefore can achieve the beneficial effects that can be achieved by any brain nerve signal processing method provided in the embodiments of the present invention. Please refer to the previous embodiments for details and will not be repeated here.

[0160] See also Fig.11 , Fig.11 is another structural schematic diagram of an electronic device provided by an embodiment of the present invention, such as Fig.11 As shown, Fig.11The electronic device 700 is a block diagram of a specific structure of an electronic device provided by an embodiment of the present invention, and the electronic device can be used to implement the brain nerve signal processing method provided in the above embodiment. The electronic device 700 can be a mobile terminal such as a smart phone or a laptop computer.

[0161] The RF circuit 710 is used to receive and send electromagnetic waves, realize the mutual conversion between electromagnetic waves and electrical signals, and thus communicate with a communication network or other devices. The RF circuit 710 may include various existing circuit elements for performing these functions, such as antennas, radio frequency transceivers, digital signal processors, encryption / decryption chips, user identity module (SIM) cards, memories, etc. The RF circuit 710 can communicate with various networks such as the Internet, corporate intranets, wireless networks, or communicate with other devices through wireless networks. The above-mentioned wireless networks may include cellular telephone networks, wireless local area networks, or metropolitan area networks. The above-mentioned wireless networks can use various communication standards, protocols and technologies, including but not limited to Global System for Mobile Communication (GSM), Enhanced Data GSM Environment (EDGE), Wideband Code Division Multiple Access (WCDMA), Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Wireless Fidelity (Wi-Fi) (such as Institute of Electrical and Electronics Engineers standards IEEE 802.11a, IEEE 802.11b, IEEE802.11g and / or IEEE802.11n), Voice over Internet Protocol (VoIP), Worldwide Interoperability for Microwave Access (Wi-Max), other protocols for email, instant messaging and short messages, and any other suitable communication protocols, even those that have not yet been developed.

[0162] The memory 720 can be used to store software programs and modules, such as the program instructions / modules corresponding to the brain nerve signal processing method in the above-mentioned embodiment. The processor 780 executes various functional applications and processes brain nerve signals by running the software programs and modules stored in the memory 720.

[0163] The memory 720 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 720 may further include a memory remotely arranged relative to the processor 780, and these remote memories may be connected to the electronic device 700 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0164] The input unit 730 can be used to receive input digital or character information, and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control. Specifically, the input unit 730 may include a touch-sensitive surface 731 and other input devices 732. The touch-sensitive surface 731, also known as a touch display screen or touch pad, can collect user touch operations on or near it (such as operations performed by users using fingers, styluses, or any other suitable objects or accessories on or near the touch-sensitive surface 731), and drive corresponding connection devices according to a pre-set program. Optionally, the touch-sensitive surface 731 may include a touch detection device and a touch controller. Among them, the touch detection device detects the user's touch orientation, detects the signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts it into touch point coordinates, and then sends it to the processor 780, and can receive and execute commands sent by the processor 780. In addition, the touch-sensitive surface 731 can be implemented in various types such as resistive, capacitive, infrared, and surface acoustic wave. In addition to the touch-sensitive surface 731, the input unit 730 may further include other input devices 732. Specifically, the other input devices 732 may include, but are not limited to, one or more of a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a trackball, a mouse, a joystick, and the like.

[0165] The display unit 740 can be used to display information input by the user or information provided to the user and various graphical user interfaces of the electronic device 700, which can be composed of graphics, text, icons, videos and any combination thereof. The display unit 740 may include a display panel 741, and optionally, the display panel 741 may be configured in the form of LCD (Liquid Crystal Display), OLED (Organic Light-Emitting Diode), etc. Further, the touch-sensitive surface 731 may cover the display panel 741, and when the touch-sensitive surface 731 detects a touch operation on or near it, it is transmitted to the processor 780 to determine the type of the touch event, and then the processor 780 provides corresponding visual output on the display panel 741 according to the type of the touch event. Although in the figure, the touch-sensitive surface 731 and the display panel 741 are implemented as two independent components to implement input and output functions, in some embodiments, the touch-sensitive surface 731 and the display panel 741 can be integrated to implement input and output functions.

[0166] The electronic device 700 may also include at least one sensor 750, such as a light sensor, a motion sensor, and other sensors. Specifically, the light sensor may include an ambient light sensor and a proximity sensor, wherein the ambient light sensor may adjust the brightness of the display panel 741 according to the brightness of the ambient light, and the proximity sensor may generate an interrupt when the flip cover is closed or closed. As a type of motion sensor, the gravity acceleration sensor can detect the magnitude of acceleration in each direction (generally three axes), and can detect the magnitude and direction of gravity when stationary, which can be used for applications that identify the posture of the mobile phone (such as horizontal and vertical screen switching, related games, magnetometer posture calibration), vibration recognition related functions (such as pedometer, tapping), etc.; as for other sensors such as gyroscopes, barometers, hygrometers, thermometers, infrared sensors, etc. that can also be configured in the electronic device 700, they will not be repeated here.

[0167] The audio circuit 760, the speaker 761, and the microphone 762 can provide an audio interface between the user and the electronic device 700. The audio circuit 760 can transmit the electrical signal converted from the received audio data to the speaker 761, which is converted into a sound signal for output; on the other hand, the microphone 762 converts the collected sound signal into an electrical signal, which is received by the audio circuit 760 and converted into audio data, and then the audio data is output to the processor 780 for processing, and then sent to another terminal through the RF circuit 710, or the audio data is output to the memory 720 for further processing. The audio circuit 760 may also include an earplug jack to provide communication between an external headset and the electronic device 700.

[0168] The electronic device 700 can help the user receive requests, send information, etc. through the transmission module 770 (such as a Wi-Fi module), which provides the user with wireless broadband Internet access. Although the transmission module 770 is shown in the figure, it is understandable that it is not a necessary component of the electronic device 700 and can be omitted as needed without changing the essence of the invention.

[0169] The processor 780 is the control center of the electronic device 700. It uses various interfaces and lines to connect various parts of the entire mobile phone. By running or executing software programs and / or modules stored in the memory 720, and calling data stored in the memory 720, it executes various functions of the electronic device 700 and processes data, thereby monitoring the electronic device as a whole. Optionally, the processor 780 may include one or more processing cores; in some embodiments, the processor 780 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application programs, etc., and the modem processor mainly processes wireless communications. It can be understood that the above-mentioned modem processor may not be integrated into the processor 780.

[0170] The electronic device 700 also includes a power supply 790 (such as a battery) for supplying power to various components. In some embodiments, the power supply can be logically connected to the processor 780 through a power management system, so that the power management system can manage charging, discharging, and power consumption management. The power supply 790 can also include any components such as one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, and power status indicators.

[0171] Although not shown, the electronic device 700 also includes a camera (such as a front camera, a rear camera), a Bluetooth module, etc., which will not be described in detail here. Specifically in this embodiment, the display unit of the electronic device is a touch screen display, and the mobile terminal also includes a memory, and one or more programs, one or more of which are stored in the memory, and are configured to be executed by one or more processors to implement any step of the brain nerve signal processing method provided in the above embodiment.

[0172] In specific implementation, the above modules can be implemented as independent entities, or can be arbitrarily combined and implemented as the same or several entities. The specific implementation of the above modules can be found in the previous method embodiments, which will not be repeated here.

[0173] Those skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructions, or by controlling related hardware through instructions, and the instructions can be stored in a computer-readable storage medium and loaded and executed by a processor. To this end, an embodiment of the present invention provides a storage medium, which stores multiple instructions, and the instructions can be executed by the processor to implement any step of the brain nerve signal processing method provided in the above embodiments.

[0174] The storage medium may include: a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0175] Since the instructions stored in the storage medium can execute the steps in any embodiment of the brain nerve signal processing method provided in the embodiments of the present invention, the beneficial effects that can be achieved by any brain nerve signal processing method provided in the embodiments of the present invention can be achieved. Please refer to the previous embodiments for details and will not be repeated here.

[0176] The above is a detailed introduction to a brain nerve signal processing method, device, electronic device and storage medium provided in the embodiments of the present application. Specific examples are used herein to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and its core idea of ​​the present application. At the same time, for those skilled in the art, according to the idea of ​​the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application. Moreover, for those of ordinary skill in the art, several improvements and modifications can be made without departing from the principles of the present invention, and these improvements and modifications are also regarded as the scope of protection of the present invention.

Claims

1. A method for processing cranial nerve signals, characterized in that: include: Acquiring a target cranial nerve signal to be processed from a target object; Performing feature extraction processing on the target cranial nerve signal to obtain a target feature set related to a target type pathology in the target cranial nerve signal, wherein the target type pathology includes a precancerous low-grade type pathology, a dysplastic hyperplasia type pathology, an early gastric cancer type pathology, an advanced cancer type pathology, and an early gastric cancer postoperative type pathology; Based on the target processing strategy corresponding to the target type pathology, the target feature set is processed accordingly to obtain a target processing result, which is used to generate a pathology recognition result of whether the target object belongs to the target type pathology, and the pathology recognition result includes a pathology recognition report.

2. The cranial nerve signal processing method according to claim 1, characterized in that: The step of obtaining the target cranial nerve signal to be processed of the target object comprises: Acquiring a target original cranial nerve signal of a target object through a target cranial nerve signal acquisition device; Performing filtering processing on the target original cranial nerve signal to obtain a filtered cranial nerve signal; Decomposing the filtered cranial nerve signal according to the number of electrode channels and frequency band type of the target cranial nerve signal acquisition device to obtain decomposed cranial nerve signals of different frequency band types in each electrode channel; Each of the decomposed cranial nerve signals in each electrode channel is cropped using multiple time windows at different time moments, so as to obtain a target cranial nerve signal in each of the decomposed cranial nerve signals within a time window at a different time moment.

3. The cranial nerve signal processing method according to claim 2, characterized in that: The step of performing feature extraction processing on the target cranial nerve signal to obtain a target feature set related to the target type pathology in the target cranial nerve signal includes: According to the target signal type of the target cranial nerve signal, a target extraction strategy corresponding to the target signal type is adopted to perform corresponding feature extraction processing on the target cranial nerve signal of each time window in each of the decomposed cranial nerve signals to obtain the window feature of each time window in each of the decomposed cranial nerve signals; splicing the window features corresponding to the time windows at the same time moment in each electrode channel to obtain a plurality of splicing feature sets corresponding to different time moments; Target features related to the target type pathology are extracted from the spliced ​​feature sets corresponding to multiple different time moments, and all the target features are used as a target feature set.

4. The cranial nerve signal processing method according to claim 3, characterized in that: The target signal type includes a first cranial nerve signal type corresponding to a scalp electroencephalogram signal, the target cranial nerve signal includes a first cranial nerve signal, and the target extraction strategy includes a first extraction strategy; The step of performing corresponding feature extraction processing on the target brain nerve signal of each time window in each of the decomposed brain nerve signals by adopting a target extraction strategy corresponding to the target signal type according to the target signal type of the target brain nerve signal to obtain the window feature of each time window in each of the decomposed brain nerve signals comprises: Determining, according to a first cranial nerve signal type of the target cranial nerve signal, a first extraction strategy corresponding to the first cranial nerve signal type; Based on the first extraction strategy, the time domain features, frequency domain features and nonlinear features of the first brain nerve signal in each time window in each of the decomposed brain nerve signals are extracted to obtain the window features of each time window in each of the decomposed brain nerve signals.

5. The cranial nerve signal processing method according to claim 3, characterized in that: The target signal type includes a second cranial nerve signal type corresponding to a cortical electroencephalogram signal between the scalp and the cerebral cortex, the target cranial nerve signal includes a second cranial nerve signal, and the target extraction strategy includes a second extraction strategy; The step of performing corresponding feature extraction processing on the target brain nerve signal of each time window in each of the decomposed brain nerve signals by adopting a target extraction strategy corresponding to the target signal type according to the target signal type of the target brain nerve signal to obtain the window feature of each time window in each of the decomposed brain nerve signals comprises: Determining, according to the second cranial nerve signal type of the target cranial nerve signal, a second extraction strategy corresponding to the second cranial nerve signal type; Based on the second extraction strategy, the time domain features, frequency domain features and time-frequency features of the second brain nerve signal in each time window in each of the decomposed brain nerve signals are extracted to obtain the window features of each time window in each of the decomposed brain nerve signals.

6. The cranial nerve signal processing method according to claim 3, characterized in that: The target signal type includes a third cranial nerve signal type corresponding to an electroencephalographic signal in the cerebral cortex, the target cranial nerve signal includes a third cranial nerve signal, a fourth cranial nerve signal and a fifth cranial nerve signal, and the target extraction strategy includes a third extraction strategy; The step of performing corresponding feature extraction processing on the target brain nerve signal of each time window in each of the decomposed brain nerve signals by adopting a target extraction strategy corresponding to the target signal type according to the target signal type of the target brain nerve signal to obtain the window feature of each time window in each of the decomposed brain nerve signals comprises: Determining, according to a third cranial nerve signal type of the target cranial nerve signal, a third extraction strategy corresponding to the third cranial nerve signal type; Based on the third extraction strategy, the time domain features, frequency domain features and time-frequency features of the third cranial nerve signal in the time window of each decomposed cranial nerve signal are extracted, the area irregularity features and the center of mass deviation features of the fourth cranial nerve signal in the time window of each decomposed cranial nerve signal are extracted, and the discharge density features of the fifth cranial nerve signal in the time window of each decomposed cranial nerve signal are extracted to obtain the window features of each time window in each decomposed cranial nerve signal.

7. The method for processing cranial nerve signals according to any one of claims 4 to 6, characterized in that: The step of performing corresponding processing on the target feature set based on the target processing strategy corresponding to the target type pathology to obtain a target processing result includes: Based on the target processing strategy corresponding to the target type pathology, obtaining the target standard feature corresponding to the target type pathology; Calculating the minimum distance between the target feature corresponding to all time moments in each electrode channel in the target feature set and the target standard feature; Calculate and obtain the number of targets whose minimum distance is not greater than the target standard threshold, and obtain the target processing result; The cranial nerve signal processing method further includes: In a case where the target processing result indicates that the target quantity is greater than a target threshold, generating a pathology recognition result indicating that the target object belongs to the target type pathology; Otherwise, a pathology recognition result indicating that the target object does not belong to the target type of pathology is generated.

8. A brain nerve signal processing device, characterized in that: include: An acquisition module, used for acquiring a target cranial nerve signal to be processed from a target object; A feature extraction module is used to perform feature extraction processing on the target cranial nerve signal to obtain a target feature set related to a target type pathology in the target cranial nerve signal, wherein the target type pathology includes a precancerous low-grade type pathology, a dysplastic hyperplasia type pathology, an early gastric cancer type pathology, an advanced cancer type pathology, and an early gastric cancer postoperative type pathology; A processing module is used to perform corresponding processing on the target feature set based on the target processing strategy corresponding to the target type pathology to obtain a target processing result, wherein the target processing result is used to generate a pathology recognition result of whether the target object belongs to the target type pathology, and the pathology recognition result includes a pathology recognition report.

9. An electronic device, characterized in that: The electronic device comprises a processor, a memory, and a computer program stored in the memory and executable on the processor, and the processor implements the steps in the method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the method according to any one of claims 1 to 7 are implemented.

Citation Information

Patent Citations

  • Brain wave signal detection method, related device and storage medium

    CN110367976A

  • Diagnosis and treatment system for precancerous lesions of gastrointestinal tract

    CN113331767A

  • Electro-acupuncture system

    KR101761724B1

  • Brain Activity as a Marker of Disease

    US20110112426A1

  • Method for prognosis of type 2 diabetes mellitus after stomach cancer surgery

    WO2022173201A2