CN112842357B

CN112842357B - Brain physiological data processing method, device and storage medium

Info

Publication number: CN112842357B
Application number: CN201911173534.3A
Authority: CN
Inventors: 张振宇; 许娟; 王星; 傅玲; 蒋建慧
Original assignee: Alibaba Health Information Technology Ltd
Current assignee: Alibaba Health Information Technology Ltd
Priority date: 2019-11-26
Filing date: 2019-11-26
Publication date: 2024-04-09
Anticipated expiration: 2039-11-26
Also published as: CN112842357A

Abstract

In the embodiment of the application, the brain physiological data is divided into a plurality of data segments by acquiring the brain physiological data, the data types corresponding to the data segments are identified, and then the target data corresponding to the brain physiological data is determined based on the data types corresponding to the data segments, so that the normal and useful brain physiological data is effectively extracted, and most of the non-significant data segments of the extracted brain physiological data are filtered, so that the quality of the brain physiological data is ensured, and the accuracy and the efficiency of analyzing the brain physiological data are improved.

Description

Brain physiological data processing method, device and storage medium

Technical Field

The present disclosure relates to the field of computer technologies, and in particular, to a method and an apparatus for processing brain physiological data, and a storage medium.

Background

Electroencephalogram is the overall reflection of brain nerve cell electrophysiological activity on the surface of cerebral cortex or scalp, contains a large amount of physiological and disease information, and along with the rapid development of medical technology, the research on brain electrical signals is also becoming more and more extensive. When an electroencephalogram (EEG) is studied, since the electroencephalogram is an electrophysiological signal with very weak intensity, generally in the order of microvolts, and since the manner of electroencephalogram leads (which means that the placement position of a detection electrode in a human body and the connection manner of the detection electrode and an amplifier when the electroencephalogram is recorded) is more easily interfered, common interference is divided into: eye movement disturbances, electrocardiographic disturbances, myoelectric disturbances, power frequency disturbances and high frequency artefact disturbances, etc., which are commonly referred to in medicine as artefacts or artifacts. Therefore, in analyzing the electroencephalogram signals, attention must be paid to activities other than the electroencephalogram activities mixed in the electroencephalogram recording to avoid the influence of the artifact data on the accuracy of the electroencephalogram signal analysis and recognition.

Disclosure of Invention

Aspects of the present application provide a method, apparatus, and storage medium for processing brain physiological data, which are used to remove an artifact portion in brain physiological data, and improve quality and efficiency of analysis of brain physiological data.

In a first aspect, an embodiment of the present application provides a method for processing brain physiological data, including:

acquiring brain physiological data;

dividing the brain physiological data into a plurality of data segments;

identifying the data type corresponding to each of the plurality of data segments;

at least two consecutive data of a first type are determined among the plurality of data segments, the first type being non-artifact data.

In a second aspect, an embodiment of the present application provides a device for processing brain physiological data, including:

the first acquisition module is used for acquiring brain physiological data;

a first partitioning module for partitioning the brain physiological data into a plurality of data segments;

the first identification module is used for identifying the data types corresponding to the data fragments respectively;

and the first determining module is used for determining at least two continuous first type data in the plurality of data fragments, wherein the first type data are non-artifact data.

In a third aspect, an embodiment of the present invention provides an electronic device, including: a memory, a processor; wherein the memory is configured to store one or more computer instructions, wherein the one or more computer instructions, when executed by the processor, implement the method for processing brain physiological data in the first aspect.

In a fourth aspect, an embodiment of the present invention provides a computer storage medium storing a computer program, where the computer program causes a computer to implement the method for processing brain physiological data in the first aspect.

In the embodiment of the application, the brain physiological data is divided into a plurality of data fragments by acquiring the brain physiological data, the data types corresponding to the data fragments are identified, and then at least two continuous first type data are determined in the data fragments based on the data types corresponding to the data fragments, wherein the first type data are non-artifact data, so that the normal and useful brain physiological data are effectively extracted, and most of the non-significant data fragments of the extracted brain physiological data are filtered, namely, artifact parts in the brain physiological data are removed, so that the quality of the brain physiological data is ensured, and the accuracy and the efficiency of analyzing the brain physiological data are improved.

In a fifth aspect, embodiments of the present application provide a model training method, including:

acquiring a plurality of brain physiological data, wherein the plurality of brain physiological data correspond to a plurality of preset data types;

the plurality of brain physiological data is supervised trained to obtain a classification model for identifying a data type of the brain physiological data.

In a sixth aspect, an embodiment of the present application provides a model training apparatus, including:

the second acquisition module is used for acquiring a plurality of brain physiological data, and the brain physiological data correspond to a plurality of preset data types;

and the training module is used for performing supervised training on the plurality of brain physiological data to obtain a classification model for identifying the data type of the brain physiological data.

In a seventh aspect, an embodiment of the present invention provides an electronic device, including: a memory, a processor; wherein the memory is configured to store one or more computer instructions that, when executed by the processor, implement the model training method of the fifth aspect described above.

In an eighth aspect, an embodiment of the present invention provides a computer storage medium storing a computer program, where the computer program causes a computer to implement the model training method in the fifth aspect.

In the embodiment of the application, the plurality of brain physiological data are acquired, the acquired plurality of brain physiological data correspond to a plurality of preset data types, and then the plurality of brain physiological data are subjected to supervised training, so that a classification model for identifying the data types of the brain physiological data can be accurately and effectively obtained, and after the classification model is obtained, the data types of the brain physiological data can be identified by using the classification model, and the practicability of the method is improved.

In a ninth aspect, an embodiment of the present application provides a method for processing data, including:

acquiring data to be processed;

dividing the data to be processed into a plurality of data fragments;

In a tenth aspect, an embodiment of the present application provides a processing apparatus for brain physiological data, including:

the third acquisition module is used for acquiring data to be processed;

the third dividing module is used for dividing the data to be processed into a plurality of data fragments;

The third identification module is used for identifying the data types corresponding to the data fragments respectively;

and a third determining module, configured to determine at least two consecutive first type data among the plurality of data segments, where the first type data is non-artifact data.

In an eleventh aspect, an embodiment of the present invention provides an electronic device, including: a memory, a processor; wherein the memory is configured to store one or more computer instructions, wherein the one or more computer instructions, when executed by the processor, implement the method of processing data in the ninth aspect described above.

In a twelfth aspect, an embodiment of the present invention provides a computer storage medium storing a computer program that causes a computer to implement the method for processing data in the ninth aspect when executed.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute an undue limitation to the application. In the drawings:

FIG. 1 is a schematic diagram of a brain physiological data processing system according to an exemplary embodiment of the present application;

Fig. 2a is a schematic diagram of an application scenario of a method for processing brain physiological data according to an exemplary embodiment of the present application;

FIG. 2b is a flow chart of a method of processing brain physiological data according to an exemplary embodiment of the present application;

FIG. 3 is a schematic diagram I of determining target data corresponding to the brain physiological data according to an exemplary embodiment of the present application;

FIG. 4 is a second schematic diagram of determining target data corresponding to the brain physiological data according to an exemplary embodiment of the present application;

FIG. 5 is a flowchart of a method for processing brain physiological data according to another exemplary embodiment of the present application;

FIG. 6 is a flow chart of a method of processing brain physiological data according to yet another exemplary embodiment of the present application;

FIG. 7 is a flow chart of a model training method according to an exemplary embodiment of the present application;

FIG. 8 is a schematic diagram of a device for processing brain physiological data according to an exemplary embodiment of the present application;

FIG. 9 is a schematic structural diagram of an electronic device corresponding to the apparatus for processing brain physiological data according to the embodiment shown in FIG. 8;

FIG. 10 is a schematic diagram of a model training apparatus according to an exemplary embodiment of the present application;

FIG. 11 is a schematic structural diagram of an electronic device corresponding to the model training apparatus provided in the embodiment shown in FIG. 10;

FIG. 12 is a flow chart of a method of processing data according to an exemplary embodiment of the present application;

FIG. 13 is a schematic diagram of a data processing apparatus according to an exemplary embodiment of the present application;

fig. 14 is a schematic structural diagram of an electronic device corresponding to the data processing apparatus provided in the embodiment shown in fig. 13.

Detailed Description

For the purposes, technical solutions and advantages of the present application, the technical solutions of the present application will be clearly and completely described below with reference to specific embodiments of the present application and corresponding drawings. It will be apparent that the described embodiments are only some, but not all, of the embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.

In the analysis process of the original brain physiological data, the brain physiological data often includes pseudo-difference data or artifact data, for example: the brain physiological data comprises eye movement interference data, electrocardio interference data and the like, and when the brain physiological data comprises pseudo-difference data or artifact data, the existing pseudo-difference data or artifact data can influence the accuracy of analyzing and identifying the brain physiological data.

The embodiment of the application provides a processing method of brain physiological data, which can realize detection and removal operation of pseudo-difference data, specifically, after obtaining the brain physiological data, based on the long-range characteristics of the brain physiological data, the brain physiological data can be divided into a plurality of data segments, and data types corresponding to the data segments are identified, for example: based on the data types corresponding to all the data segments, alpha waves and artifact waves can extract target data for the whole brain physiological data, wherein the target data is normal and useful brain physiological data, most of non-significant data segments are filtered from the target data, and then the target data can be used as new brain physiological data for analysis processing operation, so that the quality of the brain physiological data and the analysis efficiency of the brain physiological data are improved.

The following describes in detail the technical solutions provided by the embodiments of the present application with reference to the accompanying drawings.

FIG. 1 is a schematic diagram of a brain physiological data processing system according to an exemplary embodiment of the present application. As shown in fig. 1, the brain physiological data processing system 100 may include: electroencephalogram acquisition device 101 and server 102. It should be noted that the above brain physiological data may include: electroencephalogram signals, cerebral blood flow, intracranial pressure, and the like, wherein an electroencephalogram signal refers to an electrical signal obtained when brain activity is recorded; cerebral blood flow refers to the flow of blood through a certain cross-sectional area of cerebral blood vessels per unit time; intracranial pressure refers to the pressure of cerebrospinal fluid within the cranial cavity. For convenience of explanation, the brain physiological data will be described by taking an electroencephalogram signal as an example.

Specifically, the electroencephalogram acquisition device 101 is configured to acquire electroencephalogram signals of the human body 104, and specifically, the electroencephalogram acquisition device 101 may be connected with a plurality of detection electrodes (for example, the plurality of detection electrodes include a detection electrode 103a, a detection electrode 103b and a detection electrode 103 c), and the detection electrodes may be distributed according to an electrode placement method of the international 10-20 system. During the acquisition, the detection electrodes 103a, 103b, 103c may be fixed at various positions of the head of the human body 104. After the interface 105 of the electroencephalogram acquisition apparatus 101 is connected with the detection electrode 103a, the detection electrode 103b, and the detection electrode 103c, the electroencephalogram acquisition apparatus 101 can acquire an electroencephalogram signal of the human body 104. In addition, the electroencephalogram acquisition device 101 may be further connected with a display device 106, and after acquiring the electroencephalogram signals of the human body 104, the electroencephalogram signals may be displayed in real time through the display device 106. Of course, the specific shape and structure of the electroencephalogram acquisition apparatus 101 is merely exemplary, and may further include a helmet, a hat, bedding, a packaged portable detection electrode, and other various devices, which are not limited herein.

Server 102 refers to a device that can provide computing processing services in a network virtual environment, and generally refers to a server that utilizes a network for information planning. In a physical implementation, the server 102 may be any device capable of providing computing services, responding to service requests, and performing data processing, for example: may be a conventional server, cloud host, virtual center, etc. The server 102 is constructed primarily from processors, hard disks, memory, system buses, etc., similar to a general purpose computer architecture.

In this example, an electroencephalogram acquisition device 101 is configured to acquire an electroencephalogram signal, and transmit the electroencephalogram signal to a server 102; the server 102 is configured to receive an electroencephalogram signal, and divide the electroencephalogram signal into a plurality of data segments; then, the data types corresponding to the data segments are identified, and target data corresponding to the electroencephalogram signal is determined, at this time, the determined target data includes at least two continuous first type data in the data segments, the data types of the at least two continuous first type data are non-pseudo-difference data, and it is understood that the pseudo-difference data refer to interference signals existing in the electroencephalogram signal, for example: eye movement interference, electrocardiographic interference, myoelectric interference, power frequency interference, high frequency artifact interference, and the like. The non-pseudo-difference data is data in the brain electrical signal, which is not an interference signal, for example: alpha wave data, beta wave data, and the like.

In some examples, after the at least two consecutive first type data are acquired, the at least two consecutive first type data may be processed, for example: the method can perform data abnormality detection processing, disease prediction processing, abnormal discharge signal detection processing and the like, so that processing results such as data abnormality detection results, disease prediction results, abnormal discharge signal detection results and the like of the brain electrical signals can be obtained, and the prevention, early diagnosis and early intervention of brain diseases based on the brain electrical signals are effectively realized based on the obtained processing results.

In some examples, when the server 102 divides the electroencephalogram signal into a plurality of data segments, the server 102 can obtain a sliding window length and a window overlap ratio between adjacent windows; based on the sliding window length and the window overlapping rate, the electroencephalogram signal is divided into a plurality of data fragments by utilizing a sliding window algorithm.

In some examples, when server 102 identifies a data type corresponding to each of the plurality of data segments, server 102 may identify the data type corresponding to each of the plurality of data segments according to a convolutional neural network model; wherein the convolutional neural network model is trained to identify a plurality of brain electrical signals of a preset data type.

Wherein the plurality of preset data types may include at least one of: alpha wave data, artifact wave data, other types of wave data. Note that other types of wave data may refer to any type of data other than alpha wave data and artifact wave data, such as: beta wave data, theta wave data, delta wave data, and the like.

In some examples, the server 102 may obtain: data locations of at least two consecutive first type data in the electroencephalogram signal; acquiring and determining lead positions corresponding to at least two continuous first type data respectively; and storing the data position, the lead position and the at least two continuous first type data in an associated mode.

In some examples, when the server 102 obtains data locations of at least two consecutive first type data in the electroencephalogram signal, the server 102 can obtain a data length of the electroencephalogram signal; based on the data length, the sliding window length and the window overlap ratio, data positions of at least two continuous first type data in the electroencephalogram signals are determined.

In some examples, when the at least two consecutive first type data include a data segment having a data type that is alpha wave data, server 102 may further determine a duty cycle of the data segment having a data type that is alpha wave data in the at least two consecutive first type data; and when the duty ratio is smaller than a preset threshold value, correcting at least two continuous first type data.

In some examples, when the server 102 corrects at least two consecutive first type data, the server 102 may obtain type data of a first data segment of the at least two consecutive first type data; and deleting the first data segment when the type data of the first data segment is other type wave data.

In some examples, when the server 102 corrects at least two consecutive first type data, the server 102 may obtain type data of a last bit data segment of the at least two consecutive first type data; and deleting the last data segment when the type data of the last data segment is other type wave data.

In some examples, when the server 102 acquires the electroencephalogram signals, the server 102 can acquire the electroencephalogram signals of all channels acquired by the electroencephalogram acquisition device; and filtering the electroencephalogram signals of each channel to obtain the electroencephalogram signals.

In some embodiments, the electroencephalogram acquisition device 101 can be in a network connection with the server 102, which can be a wireless or wired network connection. If the electroencephalogram acquisition device 101 is in communication connection with the server 102, the network system of the mobile network may be any one of 2G (GSM), 2.5G (GPRS), 3G (WCDMA, TD-SCDMA, CDMA2000, UTMS), 4G (LTE), 4g+ (lte+), wiMax, etc.

The following describes details of a procedure for processing an electroencephalogram signal by the server 102 in connection with the method embodiment.

Fig. 2a is a schematic diagram of an application scenario of a method for processing brain physiological data according to an exemplary embodiment of the present application; FIG. 2b is a flow chart of a method of processing brain physiological data according to an exemplary embodiment of the present application; referring to fig. 2 a-2 b, a method 200 for processing brain physiological data according to an embodiment of the present application may be performed by the server 102, where the method 200 includes the following steps:

S201: brain physiological data is acquired.

S202: the brain physiological data is divided into a plurality of data segments.

S203: a data type corresponding to each of the plurality of data segments is identified.

S204: at least two consecutive data of a first type are determined among the plurality of data segments, the first type being non-artifact data.

The following is a detailed description of the above steps:

s201: brain physiological data is acquired.

Wherein the brain physiological data may include: electroencephalogram signals, cerebral blood flow, intracranial pressure, and the like, and an electroencephalogram signal will be described below as an example of brain physiological data. The acquired electroencephalogram signal may be data sent to the server by the electroencephalogram acquisition device, and the electroencephalogram signal may be a real-time digital electroencephalogram signal data stream, for example: 01001011010, 1011100011, etc. When the electroencephalogram signal is acquired by using the electroencephalogram acquisition apparatus, the signal sampling rate and the number of leads of the electroencephalogram acquisition apparatus are not limited, for example: the lead number of the electroencephalogram acquisition equipment can comprise a plurality of number specifications of 8 leads, 24 leads, 64 leads and the like; and, formats of the acquired brain electrical signals include, but are not limited to: electroencephalogram (EEG) format, fractal image format (Fractal Image Format FIF), european data format (European Data Format EDF) format, and so forth.

In some examples, the electroencephalogram signal may be data after being subjected to filtering processing, that is, after the electroencephalogram acquisition device acquires electroencephalogram signals of a plurality of channels through a plurality of detection electrodes, the electroencephalogram signals of all channels may be subjected to filtering processing, specifically, the electroencephalogram signals of each channel may be input to a filter, and independent bandpass filtering processing is performed on the electroencephalogram signals of each channel through the filter, so as to filter out some high-frequency signals and/or some low-frequency signals included in the electroencephalogram signals, for example, the frequency of the bandpass filter may be 0.5-40Hz, and at this time, after the electroencephalogram signals are subjected to filtering processing by the bandpass filter, the low-frequency signals lower than 0.5Hz and the high-frequency signals higher than 40Hz included in the electroencephalogram signals may be filtered, thereby obtaining the electroencephalogram signals after being subjected to filtering processing. It will be appreciated that the frequency of the band-pass filter is not limited to the above examples, and those skilled in the art can set the frequency according to the specific application scenario, for example: in one example, the frequency of the band pass filter may be set to 0.2-40Hz.

After the electroencephalogram signal after the filtering processing is obtained, the electroencephalogram signal acquisition equipment can send the electroencephalogram signal to the server, and at the moment, the server can receive the electroencephalogram signal after the filtering processing.

In some examples, the electroencephalogram signal may also be data which is not subjected to filtering processing, that is, after the electroencephalogram acquisition device acquires the electroencephalogram signals of a plurality of channels through a plurality of detection electrodes, the electroencephalogram signals of the channels are directly sent to the server, at this time, the server may acquire the electroencephalogram signals of all channels acquired by the electroencephalogram acquisition device, then, the electroencephalogram signals of each channel may be subjected to filtering processing, specifically, the electroencephalogram signals of each channel may be input to a filter, and independent bandpass filtering processing is performed on the electroencephalogram signals of each channel through the filter, so that some high-frequency signals and/or some low-frequency signals included in the electroencephalogram signals may be filtered.

After acquiring the electroencephalogram signal, the electroencephalogram signal may be divided into a plurality of data segments according to a preset rule, in this embodiment, the specific number of the data segments is not limited, and a person skilled in the art may set according to specific application requirements and design requirements, for example: the number of data fragments may be 8, 10 or 12, etc.

In addition, the specific implementation manner of dividing the electroencephalogram signal into a plurality of data segments is not limited in this embodiment, and a person skilled in the art may set arbitrarily according to a specific application scenario, for example: the segment start positions corresponding to the plurality of data segments may be configured, and the electroencephalogram signal may be divided into the plurality of data segments by the segment start positions corresponding to the plurality of data segments.

In some examples, dividing the electroencephalogram signal into the plurality of data fragments may include: acquiring the length of a sliding window and the window overlapping rate between adjacent windows; based on the sliding window length and the window overlapping rate, the electroencephalogram signal is divided into a plurality of data fragments by utilizing a sliding window algorithm.

The sliding window length and the window overlapping rate between adjacent windows can be configured according to specific application scenes and application requirements, and in some examples, the sliding window length and the window overlapping rate can be determined through a convolutional neural network model, wherein the convolutional neural network model is used for identifying electroencephalogram signals with various preset data types; specifically, the sliding window initial value and the step size information may be set based on the convolutional neural network model, for example: the initial value of the sliding window is 0.5s, the step length information is 0.1, and then the convolutional neural network model can be utilized to analyze and identify the data segment acquired based on the initial value of the sliding window, and the accuracy of analysis and identification is acquired; when the accuracy is smaller than a preset threshold, the initial value of the sliding window can be adjusted based on the step size information, the data segments acquired based on the adjusted sliding window length are analyzed and identified by utilizing the convolutional neural network model, the accuracy of analysis and identification is acquired, the operation is repeated, and then the sliding window length with higher identification accuracy can be determined as the target sliding window length, so that the specific implementation process of the sliding window length is realized.

In addition, the specific implementation process of determining the window overlapping rate through the convolutional neural network model is similar to the specific implementation process of determining the sliding window length, and is not described herein.

In some examples, the sliding window length and window overlap ratio may also be configured to integrate the length of the electroencephalogram signal, for example: when the electroencephalogram signal length is 10s, the sliding window length can be 2s, the window overlapping rate can be 50%, and at the moment, adjacent windows overlap for 1s. When the electroencephalogram signal length is 20s, the sliding window length can be 3s, the window overlapping rate can be 50%, and at the moment, adjacent windows overlap for 1.5s. Note that the window overlap ratio is less than 100%.

After the sliding window length and the window overlap rate are obtained, the electroencephalogram signal may be divided into a plurality of data segments using a sliding window algorithm, for example: when the electroencephalogram signal length is 10s, the sliding window length is 2s and the window overlapping rate is 50%, at this time, the electroencephalogram signal can be divided into 9 data fragments by utilizing a sliding window algorithm, and the lengths of all the data fragments are equal, so that the electroencephalogram signal is divided into a plurality of fixed-length data fragments.

After the plurality of data segments are acquired, the plurality of data segments may be analyzed to identify a data type corresponding to each of the plurality of data segments, where one data segment may correspond to one data type, and the data types corresponding to the plurality of data segments may include: alpha wave data, artifact wave data, other types of wave data.

In some examples, identifying the data type to which each of the plurality of data segments corresponds may include: identifying data types corresponding to the data segments according to the convolutional neural network model; wherein the convolutional neural network model is trained to identify a plurality of brain electrical signals of a preset data type. Specifically, the plurality of preset data types includes at least one of: alpha wave data, artifact wave data, other types of wave data.

The convolutional neural network model is a model which is trained in advance and is used for identifying electroencephalogram signals of various preset data types. When training the convolutional neural network model, it may include: acquiring a plurality of electroencephalogram signals, wherein the plurality of electroencephalogram signals correspond to a plurality of preset data types; at this time, the plurality of preset data types include: alpha wave data, artifact wave data, other types of wave data. And then, performing supervised training on the plurality of electroencephalogram signals to obtain the convolutional neural network model for identifying the data type of the electroencephalogram signals.

Specifically, when the convolutional neural network is utilized to identify the data type corresponding to each of the plurality of data segments, probability information of the data of the preset type corresponding to each data segment can be obtained; and then determining the data type corresponding to the data fragment by using the preset type information with the maximum probability information. For example: the preset type data comprises alpha wave data, artifact wave data and other types of wave data, wherein for one data segment, the probability L1 of the alpha wave data is 70%, the probability L2 of the artifact wave data is 10%, and the probability L3 of the other types of wave data is 20%, wherein L1+L2+L3=1; as can be seen from comparison, the relationship between the three probabilities is L1> L3> L2, and therefore, the data type of the data segment is determined to be alpha wave data.

In some examples, after identifying the data type to which each of the plurality of data segments corresponds, the plurality of data segments may be added with type-marking information based on the data type to which each of the plurality of data segments corresponds. For example: the length of the electroencephalogram signal is 10s, the length of the sliding window is 2s, the window overlapping rate is 50%, namely, the overlapping rate is 1s, at this time, the whole electroencephalogram signal can be divided into 9 data segments, and after the convolutional neural network model identifies the data types corresponding to the data segments, the following steps are assumed: the alpha wave data is marked with 0, the artifact wave data is marked with 1, and the other type wave data is marked with 2, at which time a type marking sequence corresponding to 9 data segments is 1,2,0,0,0,1,1,1,2.

After the data types corresponding to the data segments are acquired, determining at least two continuous first type data in the data segments based on the data types corresponding to the data segments, wherein the first type data is non-pseudo-difference data, the two continuous data segments refer to two adjacent data segments, and the non-pseudo-difference data can comprise alpha wave type data and other type wave data; for ease of illustration, at least two consecutive data of a first type may be referred to as target data.

For example, the alpha wave data is identified as 0, the artifact wave data is identified as 1, the other types of wave data are identified as 2, the electroencephalogram signal is divided into a plurality of data segments, namely a segment a, a segment b, a segment c, a segment d, a segment e and a segment f in sequence, as shown in fig. 3, if the data types corresponding to the data segments are: 0. 1, 2, 0, 1, 0, and by the data types corresponding to the data segments, it can be determined that the target data is two consecutive segments c and d. As shown in fig. 4, if the data types corresponding to the data segments are: 1. 0, 2, 0, 1, 2, the target data can be determined to be three consecutive segments b, c, and d by the data types corresponding to the data segments.

According to the brain physiological data processing method, the brain physiological data are obtained and divided into the data fragments, the data types corresponding to the data fragments are identified, and then at least two continuous first type data are determined in the data fragments based on the data types corresponding to the data fragments, wherein the first type data are non-artifact data, so that normal and useful brain physiological data are effectively extracted, and most of non-significant data fragments of the extracted brain physiological data are filtered, namely artifact parts in the brain physiological data are removed, so that the quality of the brain physiological data is guaranteed, and the accuracy and the efficiency of analysis of the brain physiological data are improved.

Fig. 5 is a flowchart of a method for processing brain physiological data according to another exemplary embodiment of the present application. On the basis of the above embodiment, with continued reference to fig. 5, the method in this embodiment includes:

s501: data locations of the at least two consecutive first type data in the brain physiological data are acquired.

S502: the method further comprises obtaining a determination of a lead position corresponding to each of the at least two consecutive first type data.

S503: and storing the data position, the lead position and the at least two continuous first type data in an associated mode.

Specifically, one electroencephalogram signal may include one or more target data, and each target data is at least two continuous first type data, so that in order to facilitate extraction operation on the at least two continuous first type data, position information of the at least two continuous first type data in the electroencephalogram signal may be acquired first; in some examples, acquiring the data locations of at least two consecutive first type data in the electroencephalogram signal may include: acquiring the data length of the brain electrical signal; based on the data length, the sliding window length and the window overlap ratio, data positions of at least two continuous first type data in the electroencephalogram signals are determined.

It should be noted that, after identifying at least two consecutive first type data included in the electroencephalogram signal, since there is a window overlapping rate between adjacent windows, that is, one data segment may include partial data in a preceding data segment and partial data in a subsequent data segment, artifact wave data in the preceding data segment included in the first data segment needs to be removed; similarly, for the last data segment in at least two continuous first type data, artifact wave data in the last data segment needs to be removed, and then the data positions of the at least two continuous first type data in the electroencephalogram signal can be obtained.

For example, assume that alpha wave data is identified as 0, artifact wave data is identified as 1, and other types of wave data are identified as 2; the length of the electroencephalogram signal is 10s, the length of the sliding window is 2s, the window overlapping rate is 50%, namely, the overlapping rate is 1s, at this time, the whole electroencephalogram signal can be divided into 9 data fragments, the type marking sequences corresponding to the 9 data fragments are (1, 2,0, 1 and 2), and the positions of 9 data in the electroencephalogram signal are respectively judged to be 1-2s, 2-3s, 3-4s, 4-5s, 5-6s, 6-7s, 7-8s, 8-9s and 9-10s.

By determining the data types corresponding to the 9 data segments respectively as (2, 0), and removing the artifact data included in the first data segment and the last data segment for the determined at least two continuous first type data, it may be determined that the data segment corresponding to the 3 rd to 5 th s is at least two continuous first type data in the whole 10s electroencephalogram signal, that is, the data positions of the at least two continuous first type data in the electroencephalogram signal are obtained.

In addition, when at least two continuous first type data are acquired, the positions of leads corresponding to the at least two continuous first type data are acquired and determined, wherein the leads refer to the placement position of the detection electrode on the human body and the connection mode of the detection electrode and the amplifier when the electroencephalogram signals are recorded. It will be appreciated that one probe electrode may correspond to one lead location and that multiple probe electrodes may correspond to multiple different lead locations.

Specifically, when acquiring the brain electrical signal, the brain electrical signal includes lead identification information, and the lead position can be determined through the lead identification information, for example: after determining at least two continuous first type data, acquiring the lead identification information corresponding to the at least two continuous first type data as F1 and F3, and determining that the lead positions corresponding to the at least two continuous first type data are F1 leads and F3 leads according to the lead identification information.

After the data position and the lead position are acquired, at least two continuous first type data can be determined in the electroencephalogram signal, so that the accuracy and the reliability of determining the at least two continuous first type data included in the target data are realized.

In some examples, after the data locations and the lead locations of the at least two consecutive first type data are obtained, the data locations, the lead locations, and the at least two consecutive first type data may be stored in association, for example: the data locations and lead locations of at least two consecutive data of the first type may be written to a preset file. When the corresponding data fragments are required to be read, searched, checked and extracted, the stored data positions and the lead positions can be directly positioned to at least two continuous first type data, so that the operation of reading and extracting the at least two continuous first type data in the electroencephalogram signals is realized.

Fig. 6 is a flow chart of a method for processing brain physiological data according to yet another exemplary embodiment of the present application. On the basis of the above embodiment, referring to fig. 6, when at least two consecutive data segments of the first type data include a data segment of which the data type is alpha wave data, the method in this embodiment may further include:

s601: the duty ratio of a data segment of which the data type is alpha wave data in at least two consecutive first type data is determined.

S602: and when the duty ratio is smaller than a preset threshold value, correcting at least two continuous first type data.

When at least two consecutive first type data include a data segment (hereinafter referred to as a "specific data segment") whose data type is alpha wave data, the duty ratio of the specific data segment in the at least two consecutive first type data may be obtained, for example, assuming that the alpha wave data is identified as 0, the artifact wave data is identified as 1, and the other type wave data is identified as 2; when the at least two consecutive first type data are 2,0, then it may be determined that the specific data segment has a 3/4=75% ratio in the at least two consecutive first type data; when at least two consecutive first type data are 2,0, then it may be determined that the specific data segment has a 1/2=50% ratio in the at least two consecutive first type data.

After the duty cycle is acquired, the duty cycle may be compared with a preset threshold, where the preset threshold is a minimum limit value of at least two consecutive first type data including a specific data segment, and in different application scenarios, a data range of the preset threshold may be different, for example: the preset threshold may be 50%, 60%, or 70%, etc. When the duty ratio is compared with a preset threshold in an analysis mode, if the duty ratio is larger than or equal to the preset threshold, the specific data fragments included in at least two continuous first type data at the moment meet the preset requirement. If the duty ratio is smaller than the preset threshold value, the specific data fragments included in the at least two continuous first type data are fewer, and at this time, the at least two continuous first type data can be corrected.

Specifically, the embodiment is not limited to a specific manner of implementing the correction of at least two consecutive first type data, and a person skilled in the art may set the correction according to a specific application scenario and design requirements, and in some examples, the correction of at least two consecutive first type data may include: acquiring type data of a first data segment in at least two continuous first type data; and deleting the first data segment when the type data of the first data segment is other type wave data.

For example, assuming that the preset threshold is 60%, when at least two consecutive first type data are 2, 0, and the specific data segment has a ratio of 2/4=50% in the at least two consecutive first type data, the at least two consecutive first type data may be corrected due to 50% < 60%. Specifically, the data type of the first data segment in the at least two continuous first type data can be identified, the data type of the first data segment is 2 as a result of the identification, namely, the data type of the first data segment is other type wave data, at this time, the first data segment in the at least two continuous first type data can be deleted, so that modified data segments of 0, 2 and 0 are obtained, at this time, the ratio of a specific data segment in the at least two continuous first type data is 2/3=66.67%, and the ratio of the specific data segment in the at least two continuous first type data is 66.67% to 60%, and at this time, the at least two continuous first type data meet the preset requirement.

In other examples, modifying at least two consecutive data of the first type may include: acquiring type data of last bit data fragments in at least two continuous first type data; and deleting the last data segment when the type data of the last data segment is other type wave data.

For example, assuming that the preset threshold is 60%, when at least two consecutive first type data are 0, 2, 0, and 2, at this time, the ratio of the specific data segment in the at least two consecutive first type data is 2/4=50%, because 50% <60%, the above at least two consecutive first type data may be corrected, specifically, the data type of the last data segment in the at least two consecutive first type data may be identified, the result of the identification is that the data type of the last data segment is 2, that is, the data type of the last data segment is other type wave data, at this time, the last data segment in the at least two consecutive first type data may be deleted, so as to obtain corrected data segments of 0, 2, and 0, at this time, the ratio of the specific data segment in the at least two consecutive first type data is 2/3=66.67%, and the ratio of the specific data segment in the at this time is 66.67% >60%, where the at least two consecutive first type data satisfies the preset requirements.

It will be appreciated that the implementation of modifying at least two consecutive first type data is not limited to the above description, and those skilled in the art may also implement other ways, for example, the data types of the first data segment and the last data segment in at least two consecutive first type data may be identified at the same time, and when the data types of the first data segment and the last data segment are all other types of wave data, the first data segment and the last data segment may be deleted at the same time. Alternatively, the data segments of other types of wave data included in at least two consecutive first type data may be directly identified, and when at least two adjacent data segments are all other types of wave data, any one or several data segments may be deleted. Of course, other implementation manners may be adopted by those skilled in the art, so long as the ratio of the specific data segment in at least two consecutive first type data can meet the preset requirement, and no further description is given here.

In this embodiment, by determining the duty ratio of the data segment with the data type of alpha wave data in at least two continuous first type data, when the duty ratio is smaller than the preset threshold, the at least two continuous first type data are modified, so that the duty ratio of the data segment with the data type of alpha wave data in the at least two continuous first type data is effectively ensured, and the accuracy and reliability of analyzing and processing the brain physiological data are further improved.

FIG. 7 is a flow chart of a model training method according to an exemplary embodiment of the present application. The model training method 700 provided in the embodiments of the present application may be performed by a model training device, and it may be understood that the model training device may be implemented as software, or a combination of software and hardware, and the model training device may be disposed in the server 102 when specifically applied. Specifically, the model training method 700 includes the following steps:

s701: a plurality of brain physiological data is acquired, the plurality of brain physiological data corresponding to a plurality of preset data types.

S702: a plurality of brain physiological data is supervised trained to obtain a classification model for identifying a data type of the brain physiological data.

The following is a detailed description of the above steps:

Wherein, a plurality of brain electrical signals that acquire correspond to a plurality of preset data types, and a plurality of preset data types include: alpha wave data, artifact wave data, other types of wave data. In order to facilitate the supervision training of the electroencephalogram signals, in the process of one training, the lengths of the plurality of electroencephalogram signals can be fixed, the specific lengths of the electroencephalogram signals can be adjusted according to feedback information of model training, when the plurality of electroencephalogram signals are acquired, the signal sampling rate and the lead number of the plurality of electroencephalogram signals are correspondingly acquired, and the data format of the acquired plurality of electroencephalogram signals is not limited, for example, the EEG data format comprises, but is not limited to, EEG format, fif format, edf format and the like.

In some examples, to improve the quality and effect of model training, acquiring a plurality of electroencephalogram signals may include: acquiring electroencephalogram signals of all channels acquired by electroencephalogram acquisition equipment; and filtering the electroencephalogram signals of each channel to obtain a plurality of electroencephalogram signals.

Specifically, after the model training device obtains the electroencephalogram signals of all channels collected by the electroencephalogram collecting device, the electroencephalogram signals of each channel can be subjected to filtering processing, and in specific implementation, the electroencephalogram signals of all channels can be input into the filter, and independent band-pass filtering processing operation is performed on the electroencephalogram signals of each channel by using the filter, so that a plurality of electroencephalogram signals subjected to filtering processing can be obtained.

After a plurality of electroencephalograms are acquired, the plurality of electroencephalograms can be subjected to supervision training, and when the monitoring is specifically realized, the plurality of electroencephalograms and a plurality of corresponding preset data types can be input into a convolutional neural network (Convolutional Neural Networks, CNN for short) for learning training, wherein the number of layers of the CNN network is not limited, so that a classification model for identifying the data types of the electroencephalograms can be obtained.

In some embodiments, the method in this embodiment may further include: acquiring the identification accuracy of the classification model for the electroencephalogram signals of the target data types, wherein the target data types are contained in a plurality of preset data types; and when the recognition accuracy is smaller than a preset threshold, generating a submodel of the electroencephalogram signal for recognizing the target data type.

After the classification model for identifying the data type of the electroencephalogram is obtained, the electroencephalogram can be analyzed and identified by the classification model, at the moment, the identification accuracy of the classification model for analyzing and identifying the electroencephalogram can be obtained, and whether the classification model is optimized or not can be judged according to the identification accuracy.

For example, assuming that the preset data type includes the target data type, when the classification model is used for analyzing and identifying the electroencephalogram signals of the target data type, the identification accuracy can be obtained, then the identification accuracy is analyzed and compared with the preset threshold, and when the identification accuracy is greater than or equal to the preset threshold, the classification model is used for analyzing and identifying the electroencephalogram signals of the target data type with higher accuracy. When the recognition accuracy is smaller than a preset threshold, the accuracy of the classification model in analyzing and recognizing the electroencephalogram signals of the target data type is lower, and at the moment, in order to ensure that the electroencephalogram signals of the target data type can be analyzed and recognized, a sub-model for recognizing the electroencephalogram signals of the target data type can be generated.

In some embodiments, generating the sub-model of the electroencephalogram for identifying the target data type may include: and training a sub-model of the electroencephalogram signal for identifying the target data type by taking the electroencephalogram signal of the target data type as a positive sample and taking the electroencephalogram signal of the non-target data type as a negative sample.

It can be understood that the number of the submodels can be one or more, and the submodels can be integrated in the classification model, so that when the electroencephalogram signals are analyzed and identified, the submodels can be utilized to analyze and identify the electroencephalogram signals first, and then the classification model is utilized to analyze and identify the electroencephalogram signals, and the accuracy of identifying the data types of the electroencephalogram signals is effectively improved.

According to the model training method provided by the embodiment, the plurality of brain physiological data are acquired, the acquired plurality of brain physiological data correspond to a plurality of preset data types, and then the plurality of brain physiological data are subjected to supervised training, so that a classification model for identifying the data types of the brain physiological data can be accurately and effectively obtained, and after the classification model is obtained, the data types of the brain physiological data can be identified by utilizing the classification model, and the practicability of the method is further improved.

Fig. 8 is a schematic structural diagram of a brain physiological data processing device according to an exemplary embodiment of the present application. Referring to fig. 8, the present embodiment provides a device 800 for processing brain physiological data, where the brain physiological data may include: electroencephalogram signals, cerebral blood flow, intracranial pressure, and the like, wherein an electroencephalogram signal refers to an electrical signal obtained when brain activity is recorded; cerebral blood flow refers to the flow of blood through a certain cross-sectional area of cerebral blood vessels per unit time; intracranial pressure refers to the pressure of cerebrospinal fluid within the cranial cavity. Specifically, when the electroencephalogram data includes an electroencephalogram signal, the apparatus 800 may be applied to an electroencephalogram acquisition device, where the apparatus 800 includes a first acquisition module 801, a division module 802, an identification module 803, and a determination module 804, the following details are set forth for the functions of each module:

A first acquisition module 801 for acquiring brain physiological data;

a first partitioning module 802 for partitioning brain physiological data into a plurality of data segments;

a first identifying module 803, configured to identify a data type corresponding to each of the plurality of data segments;

a first determining module 804, configured to determine at least two consecutive first type data among the plurality of data segments, where the first type data is non-artifact data.

In some examples, when the first partitioning module 802 partitions the brain physiological data into a plurality of data segments, the first partitioning module 802 is to: acquiring the length of a sliding window and the window overlapping rate between adjacent windows; based on the sliding window length and the window overlapping rate, brain physiological data is divided into a plurality of data segments using a sliding window algorithm.

In some examples, when the first identifying module 803 identifies a data type corresponding to each of the plurality of data segments, the first identifying module 803 is configured to: identifying data types corresponding to the data segments according to the convolutional neural network model; wherein the convolutional neural network model is trained to identify brain physiological data of a plurality of preset data types.

In some examples, the plurality of preset data types includes at least one of: alpha wave data, artifact wave data, other types of wave data.

In some examples, the first determination module 804 may be to: acquiring data positions of at least two continuous first type data in brain physiological data; acquiring and determining lead positions corresponding to at least two continuous first type data respectively; and storing the data position, the lead position and the at least two continuous first type data in an associated mode.

In some examples, when the first determination module 804 obtains data locations of at least two consecutive first type data in the brain physiological data, the first determination module 804 may be configured to: acquiring the data length of brain physiological data; based on the data length, the sliding window length, and the window overlap ratio, data locations of at least two consecutive first type data in the brain physiological data are determined.

In some examples, the at least two consecutive first type data include a data segment with a data type of alpha wave data, and the first determining module 804 in this embodiment is further configured to: determining the duty ratio of a data segment with the data type of alpha wave data in at least two continuous first type data; and when the duty ratio is smaller than a preset threshold value, correcting at least two continuous first type data.

In some examples, when the first determination module 804 corrects at least two consecutive first type data, the first determination module 804 may be configured to perform: acquiring type data of a first data segment in at least two continuous first type data; and deleting the first data segment when the type data of the first data segment is other type wave data.

In some examples, when the first determination module 804 corrects at least two consecutive first type data, the first determination module 804 may be configured to perform: acquiring type data of last bit data fragments in at least two continuous first type data; and deleting the last data segment when the type data of the last data segment is other type wave data.

In some examples, when the first acquisition module 801 acquires brain physiological data, the first acquisition module 801 may be configured to: acquiring brain physiological data of all channels acquired by an electroencephalogram acquisition device; and filtering the brain physiological data of each channel to obtain the brain physiological data.

The apparatus of fig. 8 may perform the method of the embodiment of fig. 1-6, and reference is made to the relevant description of the embodiment of fig. 1-6 for parts of this embodiment that are not described in detail. The implementation process and the technical effect of this technical solution are described in the embodiments shown in fig. 1 to 6, and are not described herein.

In one possible design, the structure of the brain physiological data processing apparatus shown in fig. 8 may be implemented as an electronic device, which may be various devices such as an electroencephalogram acquisition device, a server, and the like. As shown in fig. 9, the electronic device may include: a processor 901 and a memory 902. The memory 902 is used for storing a program for the corresponding electronic device to execute the method for processing brain physiological data provided in the embodiment shown in fig. 1 to 6, and the processor 901 is configured to execute the program stored in the memory 902.

The program comprises one or more computer instructions, wherein the one or more computer instructions, when executed by the processor 901, are capable of performing the steps of:

acquiring brain physiological data;

dividing brain physiological data into a plurality of data segments;

identifying data types corresponding to the plurality of data fragments respectively;

Further, the processor 91 is further configured to perform all or part of the steps in the embodiments shown in fig. 1-6.

The electronic device may also include a communication interface 903 in the structure for the electronic device to communicate with other devices or communication networks.

In addition, an embodiment of the present invention provides a computer storage medium for storing computer software instructions for an electronic device, where the computer storage medium includes a program for executing the method for processing brain physiological data in the method embodiments shown in fig. 1 to 6.

Fig. 10 is a schematic structural diagram of a model training apparatus according to an exemplary embodiment of the present application. Referring to fig. 10, the present embodiment provides a model training apparatus 1000, the model training apparatus 100 may be applied to a server, and the model training apparatus 1000 may include: the second acquisition module 1001 and the training module 1002 are described in detail below for the functions of the respective modules:

the second obtaining module 1001 is configured to obtain a plurality of brain physiological data, where the plurality of brain physiological data corresponds to a plurality of preset data types.

A training module 1002 is configured to perform supervised training on the plurality of brain physiological data to obtain a classification model for identifying a data type of the brain physiological data.

In some examples, when the second acquisition module 1001 acquires a plurality of brain physiological data, the second acquisition module 1001 may be configured to perform: acquiring brain physiological data of all channels acquired by an electroencephalogram acquisition device; and filtering the brain physiological data of each channel to obtain a plurality of brain physiological data.

In some examples, the second acquisition module 1001 and the training module 1002 may also be used to perform the steps of:

a second obtaining module 1001, configured to obtain accuracy of identifying brain physiological data of a classification model for a target data type, where the target data type is included in a plurality of preset data types.

The training module 1002 is configured to generate a sub-model of brain physiological data for identifying a target data type when the identification accuracy is less than a preset threshold.

In some examples, when training module 1002 generates a sub-model of brain physiological data for identifying a target data type, training module 1002 may be configured to perform: taking brain physiological data of the target data type as a positive sample, taking brain physiological data of the non-target data type as a negative sample, and training a sub-model for identifying brain physiological data of the target data type.

The apparatus shown in fig. 10 may perform the method of the embodiment shown in fig. 7, and reference is made to the relevant description of the embodiment shown in fig. 7 for parts of this embodiment not described in detail. The implementation process and the technical effect of this technical solution are described in the embodiment shown in fig. 7, and are not described herein.

In one possible design, the model training apparatus shown in fig. 10 may be implemented as an electronic device, which may be a mobile phone, a terminal device, a server, or other devices. As shown in fig. 11, the electronic device may include: a processor 1101 and a memory 1102. Wherein the memory 1102 is used for storing a program for the corresponding electronic device to execute the model training method provided in the embodiment shown in fig. 7, the processor 1101 is configured to execute the program stored in the memory 1102.

The program comprises one or more computer instructions, wherein the one or more computer instructions, when executed by the processor 1101, are capable of performing the steps of:

a plurality of brain physiological data is supervised trained to obtain a classification model for identifying a data type of the brain physiological data.

Further, the processor 1101 is further configured to perform all or part of the steps in the embodiment shown in fig. 7.

The electronic device may also include a communication interface 1103 in the structure for the electronic device to communicate with other devices or communication networks.

In addition, an embodiment of the present invention provides a computer storage medium for storing computer software instructions for use in an electronic device, where the computer storage medium includes a program for executing the model training method according to the embodiment of the method shown in fig. 7.

Fig. 12 is a flowchart of a method for processing data according to an exemplary embodiment of the present application. Referring to fig. 12, the present embodiment provides a data processing method 1200, where the data processing method 1200 may be performed by a server that obtains data to be processed. Specifically, the method 1200 may include:

S1201: and obtaining data to be processed.

S1202: dividing the data to be processed into a plurality of data fragments.

S1203: and identifying the data type corresponding to each of the plurality of data fragments.

S1204: at least two consecutive data of a first type are determined among the plurality of data segments, the first type being non-artifact data.

In some examples, the data to be processed includes at least one of the following types: brain physiological data, electrocardiosignals, gastric electrical signals. Specifically, when the data to be processed includes an electrocardiograph signal, the embodiment provides a processing method of the electrocardiograph signal, and a specific implementation process, an implementation principle and an implementation effect of the processing method are similar to those of the processing method of brain physiological data, and specific reference may be made to the above description, so that details are not repeated. Similarly, when the data to be processed includes a gastric electric signal, the present embodiment provides a processing method of a gastric electric signal, and the specific implementation process, implementation principle and implementation effect of the processing method are similar to those of the processing method of brain physiological data, and specific reference may be made to the above description, and details are not repeated herein.

It can be understood that the data to be processed in this embodiment is not limited to the above illustrated data, and may further include other types of data, and those skilled in the art may perform any setting according to specific application requirements and design requirements, so long as the removal of the artifact data included in the data to be processed can be achieved, and the target data corresponding to the data to be processed may be obtained, which is not described herein again.

It should be noted that the method in this embodiment may further include the method of the embodiment shown in fig. 1 to fig. 6, and reference is made to the relevant description of the embodiment shown in fig. 1 to fig. 6 for a part of this embodiment that is not described in detail. The implementation process and the technical effect of this technical solution are described in the embodiments shown in fig. 1 to 6, and are not described herein.

Fig. 13 is a schematic structural diagram of a data processing apparatus according to an exemplary embodiment of the present application. Referring to fig. 13, the present embodiment provides a data processing apparatus 1300, where the data processing apparatus 1300 may be applied to a server, and specifically, the processing apparatus 1300 may include: the third acquiring module 1301, the third dividing module 1302, the third identifying module 1303 and the third determining module 1304 are described in detail below for the functions of the respective modules:

A third acquiring module 1301, configured to acquire data to be processed;

a third dividing module 1302, configured to divide the data to be processed into a plurality of data segments;

a third identifying module 1303, configured to identify data types corresponding to the plurality of data segments respectively;

a third determining module 1304 is configured to determine at least two consecutive data of a first type among the plurality of data segments, where the first type is non-artifact data.

In some examples, the data to be processed includes at least one of the following types: electroencephalogram signals, electrocardiosignals and gastric electrical signals.

The apparatus shown in fig. 13 may perform the method of the embodiment shown in fig. 12, and reference is made to the relevant description of the embodiment shown in fig. 12 for parts of this embodiment not described in detail. The implementation process and the technical effect of this technical solution are described in the embodiment shown in fig. 12, and are not described herein.

In one possible design, the structure of the data processing apparatus shown in fig. 13 may be implemented as an electronic device, which may be a mobile phone, a terminal device, a server, or other devices. As shown in fig. 14, the electronic device may include: a processor 1401 and a memory 1402. Wherein the memory 1402 is for storing a program for a corresponding electronic device to execute the processing method of data provided in the embodiment shown in fig. 12 described above, the processor 1401 is configured to execute the program stored in the memory 1402.

The program comprises one or more computer instructions, wherein the one or more computer instructions, when executed by the processor 1401, are capable of performing the steps of:

and obtaining data to be processed.

Dividing the data to be processed into a plurality of data fragments.

And identifying the data type corresponding to each of the plurality of data fragments.

Further, the processor 1401 is further configured to perform all or part of the steps in the embodiment shown in fig. 12.

The electronic device may also include a communication interface 1403 in its structure for the electronic device to communicate with other devices or communication networks.

Furthermore, an embodiment of the present invention provides a computer storage medium storing computer software instructions for an electronic device, which includes a program for executing the processing method of data in the embodiment of the method shown in fig. 12.

In addition, in some of the above embodiments and the flows described in the drawings, a plurality of operations appearing in a specific order are included, but it should be clearly understood that the operations may be performed out of the order in which they appear herein or performed in parallel, the sequence numbers of the operations such as 201, 202, 203, etc. are merely used to distinguish between the various operations, and the sequence numbers themselves do not represent any order of execution. In addition, the flows may include more or fewer operations, and the operations may be performed sequentially or in parallel. It should be noted that, the descriptions of "first" and "second" herein are used to distinguish different messages, devices, modules, etc., and do not represent a sequence, and are not limited to the "first" and the "second" being different types.

The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.

From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by adding necessary general purpose hardware platforms, or may be implemented by a combination of hardware and software. Based on such understanding, the foregoing aspects, in essence and portions contributing to the art, may be embodied in the form of a computer program product, which may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable multimedia data computing device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable multimedia data computing device, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable multimedia data computing device to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable multimedia data computing device to cause a series of operational steps to be performed on the computer or other programmable device to produce a computer implemented process such that the instructions which execute on the computer or other programmable device provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In one typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.

Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.

Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method of processing brain physiological data, comprising:

acquiring brain physiological data;

dividing the brain physiological data into a plurality of data segments;

determining at least two continuous first type data in the plurality of data fragments, wherein the first type data are non-pseudo-difference data, and the at least two continuous first type data comprise data fragments with data types of alpha wave data;

determining the duty ratio of the data segment of which the data type is alpha wave data in the at least two continuous first type data;

When the duty ratio is smaller than a preset threshold value, acquiring type data of a first data segment and type data of a last data segment in the at least two continuous first type data;

deleting the first data segment from the at least two continuous first type data if the type data of the first data segment is other type wave data;

and deleting the last bit data segment from the at least two continuous first type data if the type data of the last bit data segment is other type wave data.

2. The method of claim 1, wherein the dividing the brain physiological data into a plurality of data segments comprises:

acquiring the length of a sliding window and the window overlapping rate between adjacent windows;

based on the sliding window length and the window overlap ratio, dividing the brain physiological data into a plurality of data segments using a sliding window algorithm.

3. The method of claim 1, wherein the identifying the data type to which each of the plurality of data segments corresponds comprises:

identifying the data types corresponding to the data segments according to the convolutional neural network model; wherein the convolutional neural network model is trained to identify brain physiological data of a plurality of preset data types.

4. A method according to claim 3, wherein the plurality of preset data types comprises at least one of: alpha wave data, artifact wave data, other types of wave data.

5. The method according to claim 2, wherein the method further comprises:

acquiring data positions of the at least two continuous first type data in the brain physiological data;

acquiring and determining lead positions corresponding to the at least two continuous first type data respectively;

and storing the data position, the lead position and the at least two continuous first type data in an associated mode.

6. The method of claim 5, wherein acquiring the data location of the at least two consecutive first type data in the brain physiological data comprises:

acquiring the data length of the brain physiological data;

based on the data length, the sliding window length, and the window overlap ratio, a data location of the at least two consecutive first type data in the brain physiological data is determined.

7. The method of any one of claims 1-6, wherein the acquiring brain physiological data comprises:

Acquiring brain physiological data of all channels acquired by an electroencephalogram acquisition device;

and filtering the brain physiological data of each channel to obtain the brain physiological data.

8. A method according to claim 3, wherein the convolutional neural network model is trained by:

9. The method of claim 8, wherein the plurality of preset data types comprises: alpha wave data, artifact wave data, other types of wave data.

10. The method of claim 8, wherein the acquiring a plurality of brain physiological data comprises:

and filtering the brain physiological data of each channel to obtain the plurality of brain physiological data.

11. The method of claim 8, wherein the method further comprises:

Acquiring the recognition accuracy of the classification model for brain physiological data of a target data type, wherein the target data type is contained in the plurality of preset data types;

and when the identification accuracy is smaller than a preset threshold, generating a sub-model of the brain physiological data for identifying the target data type.

12. The method of claim 11, wherein the generating a sub-model of brain physiological data for identifying the target data type comprises:

and training a sub-model for identifying the brain physiological data of the target data type by taking the brain physiological data of the target data type as a positive sample and taking the brain physiological data of the target data type as a negative sample.

13. A method of processing data, comprising:

acquiring data to be processed;

dividing the data to be processed into a plurality of data fragments;

14. The method of claim 12, wherein the step of determining the position of the probe is performed,

the data to be processed comprises at least one of the following types: electroencephalogram signals, electrocardiosignals and gastric electrical signals.

15. A processing device of brain physiological data, comprising a memory and a processor;

the memory is used for storing a computer program;

the processor is configured to execute the computer program for:

acquiring brain physiological data;

dividing the brain physiological data into a plurality of data segments;

16. A computer readable storage medium storing a computer program, which when executed by one or more processors causes the one or more processors to implement the steps in the method of any of claims 1-7.

17. A data processing device comprising a memory and a processor;

the memory is used for storing a computer program;

the processor is configured to execute the computer program for:

acquiring data to be processed;

dividing the data to be processed into a plurality of data fragments;

18. A computer readable storage medium storing a computer program, which when executed by one or more processors causes the one or more processors to implement the steps of the method of any of claims 13-14.