CN113261979B - Tinnitus identification system based on electroencephalogram signals - Google Patents
Tinnitus identification system based on electroencephalogram signals Download PDFInfo
- Publication number
- CN113261979B CN113261979B CN202110810822.6A CN202110810822A CN113261979B CN 113261979 B CN113261979 B CN 113261979B CN 202110810822 A CN202110810822 A CN 202110810822A CN 113261979 B CN113261979 B CN 113261979B
- Authority
- CN
- China
- Prior art keywords
- electroencephalogram
- tinnitus
- value
- signal
- electroencephalogram signal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention provides a tinnitus identification system based on electroencephalogram signals, which comprises an electroencephalogram signal acquisition device and an identification device, wherein the electroencephalogram signal acquisition device is electrically connected with the identification device; the electroencephalogram signal acquisition device is used for acquiring multi-path electroencephalogram signals of auditory areas on two sides of the brain of the tested object and sending the electroencephalogram signals to the identification device; the recognition device is used for calculating a connectivity characteristic value between every two of the electroencephalogram signals, inputting the connectivity characteristic value into a recognition model based on a BP neural network to obtain tinnitus coefficient values of two ears, and outputting the tinnitus coefficient values; the connectivity characteristic value comprises at least one of a Pearson correlation coefficient, a phase locking value and a transfer entropy; the tinnitus identification system based on the electroencephalogram signals is beneficial to improving the tinnitus identification speed and the accuracy of identification results.
Description
Technical Field
The invention relates to the technical field of electroencephalogram signal processing, in particular to a tinnitus identification system based on electroencephalogram signals.
Background
With the change of eating habits, psychological stress is increased due to the increase of industrialization degree, the influence of environmental noise and the overuse of earphones are caused, and the incidence rate of tinnitus is increased year by year and the tinnitus tends to be younger. In the global context, tinnitus occurs in 10-20% of the general population, with 10% of patients suffering from symptoms such as vexation, anxiety, depression, cognitive dysfunction, insomnia, stress and emotional failure, which results in a substantial reduction in the quality of life of the patient.
Tinnitus is a relatively subjective feeling, and no standard tinnitus diagnosis standard exists at present in clinical detection. Electroencephalogram (EEG), Magnetoencephalogram (MEG) and Functional Magnetic Resonance Imaging (FMRI) are used as a non-invasive functional imaging means, and are rapidly developed and widely applied to auxiliary diagnosis of tinnitus. Generally, when the electroencephalogram signal is used for tinnitus recognition, a Support Vector Machine (SVM) algorithm is adopted to match with the time-frequency domain characteristics of the electroencephalogram signal for recognition, the recognition speed is low, and the accuracy of a recognition result is low.
Disclosure of Invention
In view of the defects of the prior art, the embodiment of the application aims to provide a tinnitus identification system based on an electroencephalogram signal, and the tinnitus identification speed and the accuracy of an identification result are improved.
The embodiment of the application provides a tinnitus identification system based on electroencephalogram signals, which comprises an electroencephalogram signal acquisition device and an identification device, wherein the electroencephalogram signal acquisition device is electrically connected with the identification device;
the electroencephalogram signal acquisition device is used for acquiring multi-path electroencephalogram signals of auditory areas on two sides of the brain of the tested object and sending the electroencephalogram signals to the identification device;
the recognition device is used for calculating a connectivity characteristic value between every two of the electroencephalogram signals, inputting the connectivity characteristic value into a recognition model based on a BP neural network to obtain tinnitus coefficient values of two ears, and outputting the tinnitus coefficient values;
the connectivity characteristic value comprises at least one of a Pearson correlation coefficient, a phase-locked value and a transition entropy.
According to the tinnitus identification system based on the electroencephalogram signals, connectivity characteristic values between every two electroencephalogram signals in an auditory area are calculated, and identification models based on a BP neural network are input to obtain tinnitus coefficient values of two ears, and medical personnel can judge the tinnitus state of a detected object according to the tinnitus coefficient values, so that effective assistance is provided for the identification of the tinnitus state; the effective dimensionality reduction of the original signal can be realized by analyzing the connectivity characteristics, the speed of recognition and analysis is higher compared with the mode of directly recognizing according to time domain characteristics in the prior art, and in addition, the accuracy of a recognition result is higher when a recognition model based on a BP neural network is used for analysis and processing compared with the mode of analyzing and processing by using a recognition model algorithm based on a Support Vector Machine (SVM).
Preferably, the identification device is further configured to pre-process the electroencephalogram signal, and the pre-processing includes at least one of the following processing steps:
carrying out interpolation processing on the missing data;
deleting abnormally fluctuated signal segments;
carrying out 50Hz power frequency noise reduction processing and 0.5-90Hz band-pass filtering processing on the electroencephalogram signals;
performing baseline correction processing and re-reference processing on the electroencephalogram signals;
and carrying out ICA filtering processing on the electroencephalogram signals.
Preferably, the connectivity characteristic value includes a pearson correlation coefficient, and the identification device calculates the pearson correlation coefficient according to the following formula:
wherein the content of the first and second substances,
is the Pearson correlation coefficient between the ith path of electroencephalogram signal and the kth path of electroencephalogram signal,
for the ith electroencephalogram signal at tThe value of the sample at the time of day,
is the average value of the ith electroencephalogram signal,
is the standard deviation of the ith EEG signal,
is the sampling value of the kth electroencephalogram signal at the time t,
is the average value of the k-th path of electroencephalogram signal,
the standard deviation of the kth path of electroencephalogram signal is obtained; t is the total number of the brain electrical signal sampling values.
Preferably, the connectivity characteristic values include phase lock values, and the identifying means calculates the phase lock values according to the following formula:
wherein the content of the first and second substances,
is the phase locking value between the ith electroencephalogram signal and the kth electroencephalogram signal,
the phase value of the ith electroencephalogram signal at the time t,
the phase value of the kth electroencephalogram signal at the time T is shown, and T is the total number of sampling values of the electroencephalogram signals.
Preferably, the connectivity feature value includes a transition entropy, and the identifying means calculates the transition entropy according to the following formula:
wherein the content of the first and second substances,
the entropy of the transition from the ith electroencephalogram signal to the kth electroencephalogram signal,
is the sampling value of the ith electroencephalogram signal at the time t,
is the sampling value of the kth path of electroencephalogram signal at the time t, p is a joint probability distribution value,
is the joint probability distribution of the intersection of the sampling value of the kth electroencephalogram signal at the time of t +1 and the sampling value of the ith electroencephalogram signal at the time of t and the sampling value of the kth electroencephalogram signal at the time of t,
the joint probability distribution of the intersection of the sampling value of the kth electroencephalogram signal at the time t +1 and the sampling value of the kth electroencephalogram signal at the time t; t is the total number of the brain electrical signal sampling values.
Preferably, the identification device is further used for judging the tinnitus generation position according to the tinnitus coefficient value and outputting tinnitus condition information.
Preferably, the electroencephalogram signal acquisition device comprises a frame-type hood, and a plurality of electrodes are respectively arranged at the positions, corresponding to auditory areas on two sides of a brain, of the left side and the right side of the frame-type hood; the electrode comprises a connecting structure connected with the frame-type head cover, an electrode head used for collecting electroencephalogram signals and an elastic body connected between the connecting structure and the electrode head.
Preferably, the frame-type head cover is a grid-shaped head cover and comprises a plurality of cross-connected transverse strips and longitudinal strips, connecting rings are arranged at the connecting positions of the transverse strips and the longitudinal strips, and the connecting structures of the electrodes are fixedly connected with the connecting rings.
Preferably, the connecting structure comprises a sleeve fixedly connected with the connecting ring, and an inner cylinder inserted in the sleeve; the inner cylinder is detachably connected with the sleeve; the elastic body is fixedly connected with the inner cylinder.
Preferably, the inner cylinder is in threaded connection with the sleeve, or the inner cylinder is in snap connection with the sleeve.
Has the advantages that:
according to the tinnitus identification system based on the electroencephalogram signals, connectivity characteristic values between every two electroencephalogram signals in an auditory area are calculated, the tinnitus coefficient values of two ears are obtained by inputting an identification model based on a BP neural network, and medical personnel can judge the tinnitus condition of a detected object according to the tinnitus coefficient values, so that effective assistance is provided for the identification of the tinnitus condition; the effective dimensionality reduction of the original signal can be realized by analyzing the connectivity characteristics, the speed of recognition and analysis is higher compared with the mode of directly recognizing according to time domain characteristics in the prior art, and in addition, the accuracy of a recognition result is higher by using a recognition model based on a BP neural network for analysis processing compared with using a recognition model algorithm based on a Support Vector Machine (SVM) for analysis processing; therefore, the tinnitus identification system is beneficial to improving the tinnitus identification speed and the accuracy of the identification result.
Drawings
Fig. 1 is a schematic structural diagram of a tinnitus identification system based on an electroencephalogram signal provided in an embodiment of the present application.
Fig. 2 is a schematic structural diagram of an electroencephalogram signal acquisition device in the system for identifying tinnitus based on electroencephalogram signals provided by the embodiment of the application.
Fig. 3 is a schematic structural diagram of a frame-type hood in the system for tinnitus identification based on electroencephalogram signals provided by the embodiment of the application.
Fig. 4 is a schematic structural diagram of a first electrode provided in an embodiment of the present application.
Fig. 5 is a schematic structural diagram of a second electrode provided in an embodiment of the present application.
Fig. 6 is a schematic diagram of a decoupling strand of an inner cylinder in a second electrode according to an embodiment of the present disclosure.
Fig. 7 is a schematic diagram of the distribution positions of the electrodes on the head of a person.
Description of reference numerals: 1. an electroencephalogram signal acquisition device; 2. an identification device; 201. an upper computer; 202. an output device; 3. a frame-type hood; 301. a horizontal bar; 302. longitudinal strips; 303. a connecting ring; 304. connecting holes for binding bands; 305. reinforcing ribs; 4. an electrode; 401. a connecting structure; 402. an electrode tip; 403. an elastomer; 404. a sleeve; 405. an inner barrel; 406. an annular projection; 407. a contraction groove; 408. a limit stop ring; 409. a guide slope.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the accompanying drawings are illustrative only for the purpose of explaining the present invention, and are not to be construed as limiting the present invention.
In the description of the present invention, it is to be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", and the like, indicate orientations and positional relationships based on those shown in the drawings, and are used only for convenience of description and simplicity of description, and do not indicate or imply that the device or element being referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus, should not be considered as limiting the present invention. Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, features defined as "first", "second", may explicitly or implicitly include one or more of the described features. In the description of the present invention, "a plurality" means two or more unless specifically defined otherwise.
The following disclosure provides embodiments or examples for implementing different configurations of the invention. To simplify the disclosure of the present invention, the components and arrangements of specific examples are described below. Of course, they are merely examples and are not intended to limit the present invention. Furthermore, the present invention may repeat reference numerals and/or letters in the various examples, such repetition is for the purpose of simplicity and clarity and does not in itself dictate a relationship between the various embodiments and/or configurations discussed. In addition, the present invention provides examples of various specific processes and materials, but those of ordinary skill in the art will recognize applications of other processes and/or uses of other materials.
Referring to fig. 1, the tinnitus identification system based on electroencephalogram signals provided by the embodiment of the application comprises an electroencephalogram signal acquisition device 1 and an identification device 2, wherein the electroencephalogram signal acquisition device 1 is electrically connected with the identification device 2;
the electroencephalogram signal acquisition device 1 is used for acquiring multi-path electroencephalogram signals of auditory areas on two sides of the brain of a tested object (refer to fig. 7, which is a distribution position diagram of an electrode 4 for acquiring the electroencephalogram signals on the head of the tested object), and sending the electroencephalogram signals to the recognition device 2;
the recognition device 2 is used for calculating a connectivity characteristic value between every two of the electroencephalogram signals, inputting the connectivity characteristic value into a recognition model based on a BP neural network to obtain tinnitus coefficient values of two ears, and outputting the tinnitus coefficient values;
the connectivity characteristic value comprises at least one of a Pearson correlation coefficient, a phase-locked value and a transition entropy.
The tinnitus identification system based on the electroencephalogram signals obtains the tinnitus coefficient values of two ears by calculating the connectivity characteristic values between every two of the multichannel electroencephalogram signals in the auditory area and inputting an identification model based on a BP neural network, and medical personnel can judge the tinnitus condition of a detected object according to the tinnitus coefficient values (whether the two ears have the tinnitus condition or not and the corresponding tinnitus severity degree can be respectively judged), so that effective assistance is provided for the identification of the tinnitus condition; the effective dimensionality reduction of the original signal can be realized by analyzing the connectivity characteristics, the speed of recognition and analysis is higher compared with the mode of directly recognizing according to time domain characteristics in the prior art, and in addition, the accuracy of a recognition result is higher when a recognition model based on a BP neural network is used for analysis and processing compared with the mode of analyzing and processing by using a recognition model algorithm based on a Support Vector Machine (SVM).
Actually, the tinnitus symptom belongs to the brain nerve dysfunction symptom and is a result caused by brain function network disorder, the connectivity characteristic can represent the relation and the influence degree between different brain areas of the brain, and the forming mechanism of the tinnitus nerve symptom is better met, so that the accuracy of the identification result is higher by analyzing and identifying the connectivity characteristic.
Wherein, the recognition device 2 comprises an upper computer 201 (such as a computer) and an output device 202 electrically connected with the upper computer 201, the output device 202 includes but is not limited to at least one of a display, a voice playing device and a printer, and when the recognition device 2 outputs the tinnitus coefficient value, at least one of the following steps is executed: displaying the tinnitus coefficient value via a display; the tinnitus coefficient value is broadcasted through a voice playing device; the tinnitus coefficient value was printed by a printer.
The sampling frequency of the electroencephalogram signal acquisition device 1 can be set according to actual needs, and in the embodiment, the sampling frequency is 1000 Hz.
In practical application, the acquired electroencephalogram signals may have data loss, signal interference and other conditions, and if the electroencephalogram signals are directly used for analysis processing, the accuracy of the identification result is affected. Therefore, in some preferred embodiments, the recognition device 2 is further configured to pre-process the brain electrical signal, the pre-processing including at least one of the following processing steps:
carrying out interpolation processing on the missing data;
deleting abnormally fluctuated signal segments;
carrying out 50Hz power frequency noise reduction processing and 0.5-90Hz band-pass filtering processing on the electroencephalogram signals;
performing baseline correction processing and re-reference processing on the electroencephalogram signals;
and carrying out ICA filtering processing on the electroencephalogram signals.
When the missing data is interpolated, the value of the missing data may be obtained by performing interpolation calculation using a plurality of data on the front side of the missing data (front side in time series), or the value of the missing data may be obtained by performing interpolation calculation using data on the rear side of the missing data (rear side in time series), or the value of the missing data may be obtained by performing interpolation calculation using data on the front side of the missing data and data on the rear side. The interpolation calculation may be performed by primary interpolation or secondary interpolation.
When the connectivity characteristic value is calculated, it is required to ensure that the electroencephalograms are in one-to-one correspondence in time, so when deleting an abnormally fluctuating signal segment, if an abnormally fluctuating signal segment which needs to be deleted exists in one electroencephalogram, the signal segment of the corresponding time period needs to be deleted for each electroencephalogram.
In some examples, the specific step of deleting the abnormally fluctuating signal segment includes:
calculating the standard deviation of the electroencephalogram signal as a first standard deviation;
selecting a plurality of data in the electroencephalogram signal by a sliding window to calculate a standard deviation as a second standard deviation by taking N as a window length and 1 as a stepping length; n is a preset positive integer value and is greater than 1;
every time a second standard deviation is obtained through calculation, whether the ratio of the second standard deviation to the first standard deviation exceeds a first preset deviation threshold value or not is judged, and if the ratio exceeds the first preset deviation threshold value, the corresponding sliding window data section is judged to be an abnormal sliding window data section;
and deleting a signal segment from the initial data to the end data by taking the first data of the first abnormal sliding window data segment in the continuous abnormal sliding window data segments as the initial data and the last data of the last abnormal sliding window data segment in the continuous abnormal sliding window data segments as the end data.
The value N and the first preset deviation threshold may be set according to actual needs.
In other examples, the specific step of deleting the abnormally fluctuating signal segment includes:
calculating the standard deviation of the electroencephalogram signal as a first standard deviation;
averagely dividing the electroencephalogram signal into M first signal segments, and calculating the standard deviation of each first signal segment to serve as a third standard deviation; m is a preset positive integer value and is greater than 2;
taking a first signal segment with the ratio of a third standard deviation to the first standard deviation exceeding a second preset deviation threshold value as a target signal segment;
circularly executing the following steps until the length of the target signal segment does not exceed the preset length: dividing each target signal segment into two second signal segments respectively, and calculating the standard deviation of each second signal segment as a fourth standard deviation; taking a second signal segment with the ratio of a fourth standard deviation to the first standard deviation exceeding a second preset deviation threshold as a new target signal segment;
and deleting all target signal segments.
The value M, the second preset deviation threshold value and the preset length can be set according to actual needs.
In practical applications, the specific step of deleting the abnormally fluctuating signal segment is not limited thereto.
The specific steps of performing power frequency noise reduction processing and band-pass filtering processing on the electroencephalogram signals refer to a processing method in the prior art.
The steps of performing the baseline correction processing and the re-reference processing on the electroencephalogram signal can refer to the processing method in the prior art.
The baseline correction is used for preventing the influence caused by data drift. Sometimes, for some reasons, the collected data will show a gradual and slow upward drift tendency, and after the segments are divided, the starting point of each segment of data will not be in the same place due to the influence of the upward drift, so that the absolute amplitude of the segment of data will also become high. The baseline correction can correct the influence of the drift, and each piece of data has a more or less starting point.
Wherein, the definition of reference: the value of the acquired electroencephalogram signal is actually the potential difference between the position of the electrode and the reference electrode. Because of the reference of different positions, certain influence is caused to the data. For example, when cz or the head center reference is used for recording, the potential difference recorded by the electrode points closer to the cz point is naturally very small, while the potential difference recorded by the electrode points farther away is naturally larger, and the difference between the large potential and the small potential is not generated by cognitive activities but generated by the recording mode. In analyzing data, it is sometimes necessary to convert the position of a reference point, and thus, re-referencing processing is required.
The specific steps of ICA filtering processing on the electroencephalogram signals can refer to the processing method in the prior art. The ICA filtering treatment can filter the electro-ocular components, the electrocardio components and other artifact components, so that the electroencephalogram signal is more pure.
In some embodiments, the connectivity characteristic values include pearson correlation coefficients, which the identification apparatus 2 calculates according to the following formula:
wherein the content of the first and second substances,
is the Pearson correlation coefficient between the ith path of electroencephalogram signal and the kth path of electroencephalogram signal,
is the sampling value of the ith electroencephalogram signal at the time t,
is the average value of the ith electroencephalogram signal,
is the standard deviation of the ith EEG signal,
is the sampling value of the kth electroencephalogram signal at the time t,
is the average value of the k-th path of electroencephalogram signal,
the standard deviation of the kth path of electroencephalogram signal is obtained; t is the total number of the brain electrical signal sampling values.
The linear relation between the two electroencephalogram signals is measured to be a continuous number from-1 to 1 by the Pearson correlation coefficient, the PCC values are-1 and 1 respectively corresponding to complete negative linear correlation and positive linear correlation, and the PCC value is zero to indicate that the two electroencephalogram signals are not related. The PCC value belongs to a non-directional connectivity characteristic value, and only one PCC value exists between two electroencephalogram signals.
In some embodiments, the connectivity characteristic values comprise phase lock values, and the identification device 2 calculates the phase lock values according to the following formula:
wherein, the phase locking value between the ith electroencephalogram signal and the kth electroencephalogram signal,
the phase value of the ith electroencephalogram signal at the time t,
the phase value of the kth electroencephalogram signal at the time T is shown, and T is the total number of sampling values of the electroencephalogram signals.
The phase locking value describes the phase synchronism between the two electroencephalogram signals and is calculated as the average value of absolute values of phase differences, and the value range of PLV is 0 to 1, which respectively represents the complete independence and the complete synchronism of the two electroencephalogram signals. The PLV value belongs to a characteristic value of undirected connectivity, and only one PLV value exists between two paths of electroencephalogram signals.
In some embodiments, the connectivity feature value includes a transition entropy, and the identification apparatus 2 calculates the transition entropy according to the following formula:
wherein the content of the first and second substances,
the entropy of the transition from the ith electroencephalogram signal to the kth electroencephalogram signal,
is the sampling value of the ith electroencephalogram signal at the time t,
is the sampling value of the kth path of electroencephalogram signal at the time t, p is a joint probability distribution value,
is the joint probability distribution of the intersection of the sampling value of the kth electroencephalogram signal at the time of t +1 and the sampling value of the ith electroencephalogram signal at the time of t and the sampling value of the kth electroencephalogram signal at the time of t,
the joint probability distribution of the intersection of the sampling value of the kth electroencephalogram signal at the time t +1 and the sampling value of the kth electroencephalogram signal at the time t; t is the total number of the brain electrical signal sampling values.
The transfer entropy represents the information quantity transferred from one path of electroencephalogram signal to the other path of electroencephalogram signal in a directional flow mode, it can also be understood that TE describes the gain obtained by predicting the other path of electroencephalogram signal through one path of known electroencephalogram signal, and a TE value of 0 represents that no causal relationship exists between the two paths of electroencephalogram signals. The TE value belongs to a directed connectivity characteristic value, and two TE values exist between two paths of electroencephalogram signals.
Compared with the data volume of the original multi-path electroencephalogram signals, the data volume of the three connectivity characteristic values is greatly reduced, the effective dimensionality reduction of the original signals is effectively realized, and the BP neural network-based identification model has less data to be processed, so that the processing speed can be effectively improved. If one of the connectivity characteristic values is used alone for analysis and identification, the PLV is superior to the PCC, and the PCC is superior to the TE in terms of accuracy of identification results.
In some preferred embodiments, the connectivity characteristic values of the Pearson correlation coefficient, the phase locking value and the transfer entropy are simultaneously used for analysis and identification, and the accuracy is high.
Preferably, the recognition device 2 is further configured to determine a tinnitus occurrence position according to the tinnitus coefficient value and output tinnitus condition information. Wherein, different tinnitus coefficient value ranges can be pre-allocated with corresponding tinnitus levels, such as tinnitus levels including no tinnitus, mild tinnitus, moderate tinnitus, severe tinnitus and the like, and whether tinnitus occurs to two ears and the corresponding tinnitus levels are determined according to the ranges of the tinnitus coefficient values of the two ears; thereby the output tinnitus illness state information comprises tinnitus generation position information and corresponding tinnitus grade information; the tinnitus-generating position information includes, for example, "tinnitus on the binaural side", "tinnitus on the left ear side", "tinnitus on the right ear side", and the like. The way of the recognition device 2 outputting the tinnitus disease information includes, but is not limited to, at least one of displaying the tinnitus disease information through a display, broadcasting the tinnitus disease information through a voice broadcasting device, and printing the tinnitus disease information through a printer. Wherein, the tinnitus coefficient value range that different tinnitus grades correspond can be set up by the user according to actual need to can automatic fast output tinnitus degree's judged result.
In some embodiments, as shown in fig. 2 to 6, the electroencephalogram signal acquisition device 1 comprises a frame-type hood 3, wherein a plurality of electrodes 4 are respectively arranged at positions, corresponding to auditory areas on two sides of a brain, of the left side and the right side of the frame-type hood 3; the electrode 4 includes a connection structure 401 connected with the frame type headcap 3, an electrode head 402 for collecting brain electrical signals, and an elastic body 403 connected between the connection structure 401 and the electrode head 402. Wherein the electrode head 402 is connected with the recognition device 2 by an electric wire. When the frame type head cover 3 is used, the frame type head cover 3 is covered on the head of a tested person, the electrode head 402 can be reliably attached to the head under the action of the elastic body 403, and the situation that effective electroencephalogram signals cannot be obtained due to poor contact is avoided.
Preferably, referring to fig. 3, the frame-type hood 3 is a grid-type hood, and includes a plurality of cross-connected cross bars 301 and longitudinal bars 302, a connecting ring 303 is disposed at a connection position of the cross bars 301 and the longitudinal bars 302, and a connecting structure 401 of the electrode 4 is connected and fixed to the connecting ring 303. The frame type hood 3 with the structure has smaller weight and good air permeability, and can improve the wearing comfort.
Preferably, the frame-type hood 3 is made of plastic materials such as plastic, rubber or silica gel, so that the hood has good plastic deformation capacity, can be suitable for heads of different sizes to wear and use, and has good applicability. Wherein, both sides of the frame type hood 3 are respectively provided with at least one binding belt connecting hole 304 for connecting a binding belt; in use, the straps are connected between the corresponding strap connecting holes 304 to secure the frame type hood 3 to the head of the subject. Further, in order to increase the strength of the frame-type hood 3 and to prevent the frame-type hood 3 from being too flexible, reinforcing ribs 305 may be provided on the top of the horizontal and vertical bars 301 and 302.
Wherein, the number of the electrodes 4 is generally 8-12, preferably, the number of the electrodes 4 is even, and the number of the electrodes 4 on both sides is the same and the distribution positions are symmetrical. For example, in fig. 2 and 7, 12 electrodes 4 are provided, and 6 electrodes are provided on both sides of each electrode, so that 12 paths of electroencephalogram signals are acquired.
The elastic body 403 may be, but not limited to, a spring, a rubber block, a silicone block, and the like. For example, in fig. 4 and 5, the elastic body 403 is a spring.
In some embodiments, see fig. 4 and 5, the connection structure 401 includes a sleeve 404 fixedly connected to the connection ring 303, and an inner cylinder 405 inserted into the sleeve 404; the inner cylinder 405 is detachably connected with the sleeve 404; the elastic body 403 is fixedly connected to the inner cylinder 405. The sleeve 404 may be connected with the connection ring 303 in an interference fit manner and may be kept fixed by friction, and the sleeve 404 and the connection ring 303 may also be connected and fixed by gluing, fastening, or screwing. Because the inner cylinder 405 is detachably connected with the sleeve 404, the electrode head 402 can be quickly assembled, disassembled and replaced after being damaged. The inner cylinder 405 has a through hole, and the electric wire of the electrode head 402 can pass through the through hole to reach the upper part of the frame-type hood 3 and then be connected with the identification device 2, which is beneficial to avoiding the electric wire from contacting with the head of the person to be tested.
In some embodiments, see fig. 4, the inner barrel 405 is threadably coupled to the sleeve 404. Wherein, the inside of sleeve 404 is provided with the internal thread, is provided with corresponding external screw thread on the outer peripheral face of inner tube 405, the external screw thread with the internal thread meshing.
In other embodiments, see fig. 5 and 6, the inner barrel 405 and the sleeve 404 are snap-fit to each other. The middle part of the inner cylinder 405 is matched with the inner hole of the sleeve 404, the upper part of the inner cylinder 405 is provided with an annular bulge 406 protruding outwards, the top part of the inner cylinder 405 is provided with at least one shrinkage groove 407 extending along the axial direction, and the lower part of the inner cylinder 405 is provided with a limit stop ring 408. When the sleeve 404 is connected with the inner cylinder 405, the annular protrusion 406 extends to the upper side of the sleeve 404 and is clamped with the sleeve 404, and the limiting stop ring 408 abuts against the lower end of the sleeve 404, so that the inner cylinder 405 is fixed. When the inner cylinder 405 is attached and detached, the annular projection 406 is pressed to contract the upper end of the inner cylinder 405, and the inner cylinder 405 can be inserted or extracted.
In this example, four shrinkage slots 407 are provided, and the four shrinkage slots 407 are uniformly arranged along the circumferential direction of the inner cylinder 405; the lower end of the constricting channel 407 is rounded (to reduce concentrated stress). But the number and shape of the contraction groove 407 are not limited thereto.
Preferably, the upper side of the annular protrusion 406 is provided with a guide slope 409, so that it is more convenient to insert the inner cylinder 405 without pressing the upper end of the inner cylinder 405 by hand.
In summary, although the present invention has been described with reference to the preferred embodiments, the above-described preferred embodiments are not intended to limit the present invention, and those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, which are substantially the same as the present invention.
Claims (9)
1. The tinnitus identification system based on the electroencephalogram signals is characterized by comprising an electroencephalogram signal acquisition device (1) and an identification device (2), wherein the electroencephalogram signal acquisition device (1) is electrically connected with the identification device (2);
the electroencephalogram signal acquisition device (1) is used for acquiring multi-path electroencephalogram signals of auditory areas on two sides of the brain of the tested object and sending the electroencephalogram signals to the recognition device (2);
the recognition device (2) is used for calculating a connectivity characteristic value between every two of the multi-path electroencephalogram signals, inputting the connectivity characteristic value into a recognition model based on a BP neural network to obtain tinnitus coefficient values of two ears, and outputting the tinnitus coefficient values;
the connectivity characteristic value comprises at least one of a Pearson correlation coefficient, a phase locking value and a transfer entropy;
the identification device (2) is also used for judging the tinnitus generation position according to the tinnitus coefficient value and outputting tinnitus illness state information.
2. The brain electrical signal-based tinnitus recognition system of claim 1 wherein the recognition device (2) is further configured to pre-process the brain electrical signal, the pre-processing including at least one of the following processing steps:
carrying out interpolation processing on the missing data;
deleting abnormally fluctuated signal segments;
carrying out 50Hz power frequency noise reduction processing and 0.5-90Hz band-pass filtering processing on the electroencephalogram signals;
performing baseline correction processing and re-reference processing on the electroencephalogram signals;
and carrying out ICA filtering processing on the electroencephalogram signals.
3. The brain-electrical-signal-based tinnitus recognition system of claim 1 wherein the connectivity characteristic value includes a pearson correlation coefficient, the recognition device (2) calculating the pearson correlation coefficient according to the formula:
wherein the content of the first and second substances,
is the Pearson correlation coefficient between the ith path of electroencephalogram signal and the kth path of electroencephalogram signal,
is the sampling value of the ith electroencephalogram signal at the time t,
is the average value of the ith electroencephalogram signal,
is the standard deviation of the ith EEG signal,
is the sampling value of the kth electroencephalogram signal at the time t,
is the average value of the k-th path of electroencephalogram signal,
the standard deviation of the kth path of electroencephalogram signal is obtained; t is the total number of the brain electrical signal sampling values.
4. The brain electrical signal-based tinnitus recognition system of claim 1 wherein the connectivity characteristic value comprises a phase lock value, the recognition device (2) calculating the phase lock value according to the formula:
wherein the content of the first and second substances,
is the phase locking value between the ith electroencephalogram signal and the kth electroencephalogram signal,
the phase value of the ith electroencephalogram signal at the time t,
the phase value of the kth electroencephalogram signal at the time T is shown, and T is the total number of sampling values of the electroencephalogram signals.
5. The brain electrical signal-based tinnitus recognition system of claim 1 wherein the connectivity feature value includes a transition entropy, the recognition device (2) calculating the transition entropy according to the following formula:
wherein the content of the first and second substances,
the entropy of the transition from the ith electroencephalogram signal to the kth electroencephalogram signal,
is the sampling value of the ith electroencephalogram signal at the time t,
is the sampling value of the kth path of electroencephalogram signal at the time t, p is a joint probability distribution value,
is the joint probability distribution of the intersection of the sampling value of the kth electroencephalogram signal at the time of t +1 and the sampling value of the ith electroencephalogram signal at the time of t and the sampling value of the kth electroencephalogram signal at the time of t,
the joint probability distribution of the intersection of the sampling value of the kth electroencephalogram signal at the time t +1 and the sampling value of the kth electroencephalogram signal at the time t; t is the total number of the brain electrical signal sampling values.
6. The tinnitus recognition system based on electroencephalogram signals is characterized in that the electroencephalogram signal acquisition device (1) comprises a frame-type hood (3), and a plurality of electrodes (4) are respectively arranged at the positions, corresponding to auditory areas on two sides of the brain, of the left side and the right side of the frame-type hood (3); the electrode (4) comprises a connecting structure (401) connected with the frame-type hood (3), an electrode head (402) used for collecting electroencephalogram signals, and an elastic body (403) connected between the connecting structure (401) and the electrode head (402).
7. The system for tinnitus recognition based on electroencephalogram signals according to claim 6, wherein the frame-type hood (3) is a grid-shaped hood and comprises a plurality of cross-connected horizontal bars (301) and vertical bars (302), a connecting ring (303) is arranged at the joint of the horizontal bars (301) and the vertical bars (302), and the connecting structure (401) of the electrode (4) is fixedly connected with the connecting ring (303).
8. The system for tinnitus identification based on an electroencephalogram signal according to claim 7, wherein the connection structure (401) includes a sleeve (404) fixedly connected with the connection ring (303), and an inner barrel (405) inserted into the sleeve (404); the inner cylinder (405) is detachably connected with the sleeve (404); the elastic body (403) is fixedly connected with the inner cylinder (405).
9. The brain electrical signal-based tinnitus recognition system of claim 8 wherein either the inner barrel (405) and the sleeve (404) are threaded or the inner barrel (405) and the sleeve (404) are snap-fit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110810822.6A CN113261979B (en) | 2021-07-19 | 2021-07-19 | Tinnitus identification system based on electroencephalogram signals |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110810822.6A CN113261979B (en) | 2021-07-19 | 2021-07-19 | Tinnitus identification system based on electroencephalogram signals |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113261979A CN113261979A (en) | 2021-08-17 |
| CN113261979B true CN113261979B (en) | 2021-10-08 |
Family
ID=77236660
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110810822.6A Active CN113261979B (en) | 2021-07-19 | 2021-07-19 | Tinnitus identification system based on electroencephalogram signals |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113261979B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113712542A (en) * | 2021-08-27 | 2021-11-30 | 中山大学孙逸仙纪念医院 | Tinnitus electroencephalogram convolutional neural network model construction method based on deep learning |
| CN115024719B (en) * | 2022-08-11 | 2022-12-09 | 安徽声讯信息技术有限公司 | Tinnitus detection method and device based on hearing bionics |
| CN116269443A (en) * | 2023-04-03 | 2023-06-23 | 中国人民解放军海军特色医学中心 | Method and system for detecting fatigue of driver during underwater operation of closed cabin |
| CN116439710B (en) * | 2023-04-11 | 2023-10-20 | 中国人民解放军海军特色医学中心 | Ship driver fatigue detection system and method based on physiological signals |
Family Cites Families (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AUPP927599A0 (en) * | 1999-03-17 | 1999-04-15 | Curtin University Of Technology | Tinnitus rehabilitation device and method |
| TW519486B (en) * | 2001-02-05 | 2003-02-01 | Univ California | EEG feedback control in sound therapy for tinnitus |
| JP2013511304A (en) * | 2009-12-02 | 2013-04-04 | ヴェーデクス・アクティーセルスカプ | Method and apparatus for warning a person carrying an EEG assembly |
| DE102012218057A1 (en) * | 2012-10-02 | 2014-04-03 | Forschungszentrum Jülich GmbH | DEVICE AND METHOD FOR INVESTIGATING A NARROW INTERACTION BETWEEN DIFFERENT BRAIN SIZES |
| US9782585B2 (en) * | 2013-08-27 | 2017-10-10 | Halo Neuro, Inc. | Method and system for providing electrical stimulation to a user |
| CN103816007B (en) * | 2013-11-22 | 2016-04-06 | 刘志勇 | A kind of tinnitus treatment Apparatus and method for based on brain electricity frequency domain character indexing algorithm |
| CN104783808B (en) * | 2015-04-09 | 2017-12-05 | 佛山博智医疗科技有限公司 | A kind of tinnitus detection method and its therapeutic equipment |
| CN105411535B (en) * | 2016-01-06 | 2018-03-02 | 四川微迪数字技术有限公司 | A kind of device and method that tinnitus treatment effect is assessed using position location technology |
| US10405112B2 (en) * | 2017-03-31 | 2019-09-03 | Starkey Laboratories, Inc. | Automated assessment and adjustment of tinnitus-masker impact on speech intelligibility during fitting |
| CN109602425B (en) * | 2018-12-03 | 2019-08-20 | 无锡矽太益耳听力科技有限公司 | A kind of tinnitus detection device and treatment system |
| CN109893129B (en) * | 2019-04-10 | 2020-05-05 | 北京航空航天大学 | Dry electrode for adjusting depth by utilizing air bag and wearable electroencephalogram detection device |
| CN111477299B (en) * | 2020-04-08 | 2023-01-03 | 广州艾博润医疗科技有限公司 | Method and device for regulating and controlling sound-electricity stimulation nerves by combining electroencephalogram detection and analysis control |
| CN112656418B (en) * | 2021-03-15 | 2021-06-04 | 季华实验室 | Flexible dry-type brain electrode and using method thereof |
-
2021
- 2021-07-19 CN CN202110810822.6A patent/CN113261979B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN113261979A (en) | 2021-08-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113261979B (en) | Tinnitus identification system based on electroencephalogram signals | |
| Satija et al. | A review of signal processing techniques for electrocardiogram signal quality assessment | |
| Nakamura et al. | Automatic sleep monitoring using ear-EEG | |
| Wu et al. | Automatic epileptic seizures joint detection algorithm based on improved multi-domain feature of cEEG and spike feature of aEEG | |
| Sininger | Auditory brain stem response for objective measures of hearing | |
| KR100719068B1 (en) | Apparatus and method of diagnosing health using cumulative data pattern analysis via fast Fourier transformation of brain wave data measured from frontal lobe | |
| Delgado et al. | Automated auditory brainstem response interpretation | |
| US20130211224A1 (en) | Method and device for removing eeg artifacts | |
| CN106413541B (en) | System and method for diagnosing sleep | |
| US9100758B2 (en) | Sound pressure assessment system, and method and program thereof | |
| Agarwal et al. | Detection of rapid-eye movements in sleep studies | |
| US20200196977A1 (en) | System and method for determining cardiac rhythm and/or respiratory rate | |
| JP2004049838A (en) | Sleep stage discriminating method and sleep stage discriminating device | |
| Orosco et al. | An epileptic seizures detection algorithm based on the empirical mode decomposition of EEG | |
| Grose et al. | Electrophysiological measurement of binaural beats: effects of primary tone frequency and observer age | |
| US9439013B2 (en) | Uncomfortable sound pressure evaluation system, uncomfortable sound pressure evaluation apparatus, uncomfortable sound pressure adjustment apparatus, uncomfortable sound pressure evaluation method, and computer program thereof | |
| US20050177049A1 (en) | System and a method for analyzing ECG curvature | |
| CN110942103A (en) | Training method of classifier and computer-readable storage medium | |
| Magosso et al. | A wavelet based method for automatic detection of slow eye movements: A pilot study | |
| CN111481207A (en) | Sleep posture recognition device and method based on cardiac shock signal | |
| Blum et al. | Evaluation of Riemannian ASR on cEEGrid data: an artifact correction method for BCIs | |
| Cho et al. | Detection of arousals in patients with respiratory sleep disorders using a single channel EEG | |
| Shi et al. | Microstate feature fusion for distinguishing AD from MCI | |
| Vempati et al. | Eog eye blink artifact removal using multivariate variational mode decomposition and pca | |
| KR20170004547A (en) | Method and apparatus for extracting Brain information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |