JP6706996B2 - Biological signal processing device, abnormality determination method and program - Google Patents

Description

The present invention relates to a bio-signal processing device that extracts a feature amount of a bio-signal and provides it for analysis processing, and a method and a program for discriminating an abnormality of the bio-signal.

In recent years, an attempt has been made to estimate a user's state by extracting a feature amount from a biological signal acquired using a wearable device such as a Holter electrocardiograph and analyzing the feature amount. However, abnormal signals such as various artifacts and noise may be mixed in the biological signal due to electrode detachment, respiration, body movement, sweating, static electricity, etc. The amount needs to be handled properly.

Therefore, there has been proposed a method of removing or reducing an abnormal signal having a frequency characteristic different from that of the target signal by performing various filtering processes on a biological signal having periodicity in which similar waveforms repeatedly appear like an electrocardiogram ( See, for example, Non-Patent Document 1). However, when the frequency characteristic of the abnormal signal is similar to the target signal, it is difficult to completely remove the abnormal signal by filtering.

On the other hand, as a simple method of removing noise having a frequency characteristic similar to that of a biological signal, a method of determining a measured potential by a threshold has been proposed. However, the magnitude of the amplitude of the biomedical signal measured by the wearable device varies depending on the device configuration, the user, and the environment. Therefore, depending on the set threshold value, there is a risk of over-detection, over-detection, or erroneous detection of the noise region. In order to reduce detection omission, excessive detection, and erroneous detection of the noise region, it is sufficient to set a threshold value suitable for the measurement data each time the measurement of the biomedical signal is completed, but in this way the measurement of the biomedical signal is completed. It is impossible to determine whether or not it is in the noise region.

Further, as another method for discriminating the abnormality of the periodic signal, there is a method using the potential amplitude information of the target signal (for example, Non-Patent Document 2). However, in order to use this method, two pieces of information, that is, "signal period" and "statistic amount in the potential amplitude direction that can be affected by abnormality" are required. Therefore, it is difficult to properly perform the abnormality determination when the information of the statistical amount affected by the abnormality to be determined is not clear.

Further, as another method for determining abnormality of a periodic signal, there is also a method of dividing measurement data into arbitrary lengths and calculating an average amplitude spectrum of frequency components to perform abnormality determination (for example, see Non-Patent Document 3). . However, baseline fluctuations mixed in the electrocardiogram are difficult to be reflected in the average value of the amplitude spectrum, and it may not be possible to determine the abnormality.

Nakai et.al, “Noise Tolerant QRS Detection using Template Matching with Short-Term Autocorrelation”, IEEE EMBC 2014, pp.34-37, 2014 Kana Eguchi et al., “Detection Method of Noise-Induced Region for Wearable Electrocardiograph,” IEICE Tech., Vol.115, No.345, pp.27-32, 2015 Toshiki Kagawa et al., “High-sensitivity wristwatch-type pulse wave measurement sensor that eliminates the effects of noise caused by body movements”, IEICE Transactions on Information and Systems, Vol.J96-D, No.3, pp.743- 752, 2013

As described above, each of the conventionally proposed methods cannot remove an abnormal signal having a frequency characteristic similar to that of a biological signal, may cause an erroneous detection of the abnormal signal depending on the setting of the threshold value, and the potential amplitude direction affected by the abnormality. There is a problem that the abnormality can not be properly discriminated without knowing the statistic of (2), and the baseline fluctuation mixed in the electrocardiogram is difficult to be reflected in the spectrum amplitude average, so that it cannot be discriminated as an abnormal signal.

The present invention has been made in view of the above circumstances, is less susceptible to the influence of frequency characteristics such as baseline fluctuation, and more reliably the area where an abnormal signal is mixed even if the statistical value in the potential amplitude direction is unknown. An object of the present invention is to provide a biological signal processing device, an abnormality determination method, and a program capable of performing determination.

In order to solve the above-mentioned problems, a first aspect of the present invention includes a unit or step for receiving a biological signal including a waveform having periodicity, and the received biological signal including at least one of the waveforms. Means or steps for dividing into a plurality of time regions at a set time interval, and an intersection point when the waveform crosses the origin of the amplitude as a feature amount representing the generation state of the waveform for each of the plurality of time regions. The means or step for calculating the length or the number of intervals, and the calculation by comparing the calculated feature amount with a range of normal feature amount preset based on the periodicity of the waveform of the biological signal, A means or a step for determining whether the obtained characteristic amount is normal corresponding to the waveform of the biological signal or abnormal due to an abnormal signal mixed in the biological signal .

In a second aspect of the present invention, the means or step for calculating the characteristic amount calculates the length or number between the intersections when the waveform crosses the origin of amplitude for each of the plurality of time regions. the by means or step for determining an abnormality, by comparing the length or number of ranges between the length or number between the calculated intersection normal intersection which is set in advance based on the periodicity of the waveform According to the above, the length or number between the calculated intersections is a normal one corresponding to the waveform of the biological signal, or an abnormal one due to an abnormal signal mixed in the biological signal. is there.

A third aspect of the invention, the means for calculating the feature quantity for each of the plurality of time domain, the waveform calculating the length or number between the origin when crossing the Kino intersection amplitude, the The calculated lengths or numbers between the intersections are aggregated on the basis of the information in the amplitude direction of the waveform to calculate at least two types of feature amounts, and the means for discriminating the abnormality determines the calculated at least two types of the abnormalities. By comparing the feature amount with a range of normal feature amount set in advance based on the periodicity of the waveform, the calculated at least two types of feature amounts are normal to each other corresponding to the waveform of the biological signal. In this case, it is determined whether it is an abnormal one or an abnormal one due to an abnormal signal mixed in the biological signal .

A fourth aspect of the invention, the by means for calculating a feature quantity for each of the plurality of time domain, the waveform by calculating the length or number between intersections Kino when crossing the origin of the amplitude, The lengths or numbers between the calculated intersections are aggregated according to the signs of the waveforms to calculate the characteristic amounts on the positive side and the negative side, and by the means for determining the abnormality, the calculated positive side and the negative side. By comparing each feature amount with a preset range of normal feature amounts based on the periodicity of the waveform, the calculated positive and negative feature amounts correspond to the waveform of the biological signal. It is determined whether it is normal or abnormal due to an abnormal signal mixed in the biological signal .

A fifth aspect of the invention, the means for determining the abnormality, when the result of the analysis processing based on the normal and determine characteristic amount is out of the preset allowable range, the analysis A means for changing the discrimination result of the target feature amount to an abnormal one is further provided.

According to the first aspect of the present invention, the following operational effects are exhibited.
That is, in a biological signal having a periodic waveform such as an electrocardiogram, when the biological signal is divided into a plurality of time regions and evaluated at arbitrary intervals, which feature quantity represents the waveform generation state in an ideal state? It is considered that the values are close even in the divided areas. Therefore, as described above, the length or number between the intersections when the waveform intersects the origin of the amplitude is calculated as the feature amount representing the generation state of the waveform for each time domain, and the feature amount is set to the cycle of the waveform. By comparing with the range of the normal characteristic amount set in advance based on the sex, the calculated characteristic amount is a normal one corresponding to the waveform of the biological signal, or an abnormal signal mixed in the biological signal. by determining whether abnormal due, without using the information of the potential amplitude direction can be abnormal signal, such as artifacts and noise are dynamically determined whether or not included in the sensor data.

According to the second aspect of the present invention, the length or number between the intersections is calculated when the waveform crosses the origin of the amplitude, and the length or the number between the intersections is calculated as the feature amount indicating the generation state of the waveform. Is compared with the normal range to determine the abnormality, so that the abnormality can be determined by a relatively simple process.

According to the third aspect of the present invention, the calculated lengths or numbers between the intersections are aggregated in the amplitude direction of the waveform to calculate at least two types of feature values, and the lengths or the numbers between the intersections are normal. The abnormal range is determined by comparison with the range. Therefore, the abnormality determination is performed based on the feature amount that also includes the information in the amplitude direction of the waveform, so that the abnormality determination can be performed more accurately.

According to the fourth aspect of the present invention, the calculated lengths or numbers between the intersections are aggregated for each sign of the amplitude of the waveform to calculate the positive-side and negative-side characteristic amounts, and the positive-side and negative-side characteristic amounts are calculated. The feature amount is compared with the normal range to determine the abnormality. In general, a biological signal such as an electrocardiogram is an asymmetrical signal with respect to the potential amplitude direction, so if the length or number between the intersections of the waveforms is classified according to the sign of the potential amplitude, the characteristics unique to the biological signal become more prominent. Can be extracted. For this reason, it becomes possible to perform more accurate abnormality determination with respect to a biological signal in which a waveform that is asymmetric with respect to the potential amplitude direction is generated as in the electrocardiogram.

According to the fifth aspect of the present invention, when the result of the analysis process based on the feature amount that is determined to be normal is outside the preset allowable range, the feature amount that is the target of the analysis process is The determination result is changed abnormally. That is, the quality of the analysis result is fed back to the feature amount abnormality determination result and corrected. Therefore, it is possible to further improve the reliability of the feature amount abnormality determination result.

That is, according to the present invention, it is difficult to be influenced by the frequency characteristics such as the baseline fluctuation, and the biological signal that can more reliably determine the region in which the abnormal signal is mixed even if the statistical value in the potential amplitude direction is unknown. A processing device, an abnormality determination method, and a program can be provided.

The block diagram which shows the function structure of the biological signal processing apparatus which concerns on one Embodiment of this invention. 3 is a flowchart showing a processing procedure and processing contents by the biological signal processing apparatus shown in FIG. 1. FIG. 3 is a diagram for explaining a feature amount calculation operation by the biological signal processing device shown in FIG. 1. FIG. 3 is a diagram for explaining an abnormality determination operation by the biological signal processing device shown in FIG. 1. The figure which shows the effect by the biological signal processing apparatus shown in FIG. 1 compared with a prior art.

Hereinafter, embodiments according to the present invention will be described with reference to the drawings.
[One Embodiment]
(Constitution)
FIG. 1 is a block diagram showing a functional configuration of a biological signal processing device according to an embodiment of the present invention.
The biological signal processing device 1 is composed of, for example, a wearable terminal, a smartphone, a tablet type terminal, a personal computer, or a server device, and receives sensor data transmitted from the sensor 2. The sensor 2 is, for example, a Holter electrocardiograph, measures an electrocardiographic signal at a predetermined sampling period, and transmits it as sensor data.

The biological signal processing device 1 includes a control unit 10, an input/output interface unit 20, and a storage unit 30. The input/output interface unit 20 has a wireless interface and has a function of receiving sensor data transmitted from the sensor 2 and a function of transmitting information indicating an analysis result described later. As the wireless interface, for example, an interface adopting a low power wireless data communication standard such as Bluetooth (registered trademark) is used. In addition to the wireless interface, it is possible to use a wired interface that receives the sensor data transmitted from the sensor 2 via a wired cable. Further, the sensor 2 may be built in the biological signal processing device 1.

The storage unit 30 uses, for example, an SSD (Solid State Drive) memory as a storage medium. As a storage area for realizing this embodiment, the storage unit 30 includes a sensor data storage unit 31, a pre-processing unit, as well as a program storage unit. The rear sensor data storage unit 32, the divided data storage unit 33, and the evaluation feature amount storage unit 34 are provided.

The control unit 10 has a CPU (Central Processing Unit), and as its processing functions, a sensor data reception processing unit 11, a sensor data preprocessing unit 12, a sensor data division processing unit 13, and a feature amount calculation processing unit 14. And an abnormality determination processing unit 15 and an analysis processing unit 16. All of these processing functions are realized by causing the CPU to execute an application program stored in the program storage unit.

The sensor data reception processing unit 11 takes in the sensor data transmitted from the sensor 2 via the input/output interface unit 20, and temporarily stores the received sensor data in the sensor data storage unit 31.

The sensor data preprocessing unit 12 reads the received sensor data from the sensor data storage unit 31 and performs filtering processing on the sensor data to remove or reduce frequency components other than a preset band. Then, the preprocessed sensor data is stored in the postprocessed sensor data storage unit 32.

The sensor data division processing unit 13 reads the pre-processed sensor data from the pre-processed sensor data storage unit 32, and divides the sensor data into a plurality of regions (sections) at a preset constant time interval A. Then, a process of saving the divided sensor data in the divided data storage unit 33 is performed.

The feature amount calculation processing unit 14 reads sensor data from the divided data storage unit 33 for each of the divided sections, calculates a feature amount from the sensor data, and stores the feature amount in the evaluation feature amount storage unit 34. Specifically, for each divided section, the average length of the interval (intersection interval) between the points at which a plurality of amplitude signals included in the section intersect the origin (0 V) is calculated for each of the positive and negative signs, and the calculated The calculation result is stored in the evaluation feature amount storage unit 34 in association with the identification information of the divided section.

The abnormality determination processing unit 15 reads the feature amount from the evaluation feature amount storage unit 34 for each of the divided sections, and determines whether the divided section is normal or abnormal based on the characteristic amount. Then, the information indicating the determination result is stored in the evaluation feature amount storage unit 34 in association with the identification information of the divided section.

The analysis processing unit 16 excludes the characteristic amount of the divided section determined to be abnormal from the characteristic amounts of the divided sections stored in the evaluation characteristic amount storage unit 34, and determines the characteristic of the divided section determined to be normal. A predetermined analysis process is performed on the amount. Then, the information indicating the analysis result is transmitted from, for example, the input/output interface unit 20 to a terminal or server (not shown).

(motion)
Next, the abnormal signal determination operation by the device configured as described above will be described. FIG. 2 is a flowchart showing the processing procedure and processing contents.

(1) Reception of sensor data When the sensor 2 is attached to a predetermined part of the user and the measurement operation is started, an electrocardiographic signal is measured by the sensor 2 at a predetermined sampling cycle, and the measured signal is transmitted as sensor data. It

On the other hand, in the biological signal processing apparatus 1, the sensor data reception processing unit 11 monitors the start of receiving sensor data in step S10, and when the sensor data is received by the input/output interface unit 20, the sensor data reception processing unit 11 fetches the sensor data and executes the step. The sensor data is stored in the sensor data storage unit 31 in S11.

(2) Preprocessing of Sensor Data In the biological signal processing device 1, when the reception of the sensor data is started, the sensor data preprocessing unit 12 reads the sensor data from the sensor data storage unit 31 in step S12. Then, the sensor data is filtered to remove or reduce high-frequency and low-frequency components other than the frequency band of the waveform that repeatedly appears in the electrocardiogram. Then, the sensor data after the preprocessing is stored in the post-preprocessing sensor data storage unit 32 in step S13.
Note that the sensor data preprocessing unit 12 may perform processing for removing or reducing a waveform that exceeds a preset amplitude value for the sensor data.

(3) Area Division of Sensor Data In the biological signal processing device 1, the sensor data division processing unit 13 next reads the preprocessed sensor data from the preprocessed sensor data storage unit 32 in step S14. Then, the pre-processed sensor data is divided into a plurality of time regions (sections) at a preset time interval A, and the divided sensor data is associated with the identification information of the division sections in step S15 and the divided data storage unit Save to 33.

Here, the time interval A is set as follows, for example. That is, as one of the waveforms repeatedly appearing in the electrocardiogram, there is the QRS complex (R wave) which is an electrical activity due to contraction (depolarization) of the ventricle. The heartbeat interval is calculated from the interval between adjacent R waves. The heart rate of a healthy person is set to 40 to 240 [bpm] as described in Non-Patent Document 1. That is, in the case of a healthy person, the QRS complex is measured once every 0.25 to 1.5 [sec]. Therefore, the time interval A is set to A=1.5 [sec] based on this heartbeat interval. By setting the time interval A for division in this way, it becomes possible to include at least one bio-signal related to the R wave in one division area (section).

(4) Calculation of feature amount The biological signal processing apparatus 1 subsequently reads the sensor data from the divided data storage unit 33 for each of the divided sections under the control of the sensor data division processing unit 13 in step S16. The process of calculating the characteristic amount from the sensor data is executed as follows.

That is, first, the point where the electrocardiographic waveform intersects the origin (0 V) is detected. Then, the intervals between the detected intersections are calculated, and the results are classified into the positive side and the negative side according to the sign of the signal amplitude of the electrocardiographic waveform. For example, assuming that the sensor data shown in FIG. 3 is obtained, in the divided section F1, the intersection intervals Pos11, Neg11, Pos12, Neg12, Pos21 from the intersection of the electrocardiographic waveform included in the divided section F1 and the origin. , Neg21, Pos22, Neg22 are calculated. Then, these intersection intervals are divided into positive-side intersection intervals Pos11, Pos12, Pos21, Pos22 and negative-side intersection intervals Neg11, Neg12, Neg21, Neg22.

Subsequently, the sensor data division processing unit 13 calculates the average length of the positive-side intersection intervals Pos11, Pos12, Pos21, Pos22 and the average length of the negative-side intersection intervals Neg11, Neg12, Neg21, Neg22, respectively. Then, the calculated average length Posavr of the intersection intervals on the positive side and the calculated average length Negavr of the intersection intervals on the negative side are set as the characteristic amount of the sensor data in the divided section F1, and are associated with the identification information of the divided section F1 in step S17. And stores it in the evaluation feature amount storage unit 34.

Note that other values such as the standard deviation may be calculated instead of the average length as long as they reflect the information about the positive-side and negative-side intersection intervals, and this may be used as the feature amount.

(5) Discrimination of Abnormality Under the control of the abnormality discrimination processing unit 15, the biological signal processing apparatus 1 discriminates whether or not the characteristic amount of the divided section is abnormal in each of the divided sections as follows in step S18. To do. That is, the positive side intersection distance average length and the negative side intersection distance average length are read from the evaluation feature amount storage unit 34 for each divided section, and the feature amount represented by these values is read as the positive side intersection distance average length. Is plotted on the X-axis, and the average length of the negative side intersection distance is plotted on the Y-axis on a two-dimensional plane. Then, by calculating the distance between the center of gravity of the normal data (correct value set) obtained in the past arbitrary period B and the plotted point, and determining whether or not the distance is within a predetermined range, It is evaluated whether the feature amount of each divided section is normal. The arbitrary period B is set to a length that is an integral multiple (for example, 6 times) of the divided section A.

The abnormality determination process will be described in more detail with reference to FIG.
That is, first, n points corresponding to an arbitrary period B in which each is exclusive are set as a correct value set, and these n center of gravity Go(x _COG , y _COG ) are calculated. Further, when the distance between each point of the correct value set and the center of gravity Go(x _COG , y _COG ) is represented by d, the average μ of the distances between each point of the correct value and the center of gravity Go is expressed by equations (1) and (2). , The standard deviation σ is calculated. In FIG. 4, the average μ of the above distances is shown as an area E1.

Next, the distance di between a certain point (xi, yi) and the center of gravity Go(x _COG , y _COG ) of the correct answer set is calculated by the equation (3). Then, the calculated distance di is evaluated by the equation (4). Here, m is an arbitrary integer.

As a result of the above evaluation, when a certain point (xi, yi) satisfies the relationship shown in equation (4), that is, when the certain point (xi, yi) is included in the area E1 or E2 shown in FIG. 15 regards a certain point (xi, yi) and the division section corresponding to the point as "normal". On the other hand, when the certain point (xi, yi) does not satisfy the relationship shown in the equation (4), that is, when the certain point (xi, yi) is not included in any of the areas E1 and E2 shown in FIG. The abnormality determination processing unit 15 determines that the divided section corresponding to the certain point (xi, yi) is "abnormal".

The correct value set is updated when the certain point (xi, yi) is determined to be normal, or when it is different from any of the normal and already existing correct value sets. For example, the oldest point belonging to the correct value set is updated to the point (xi, yi) determined to be normal, and the center of gravity Go(x _COG , y _COG ) and the distance average are calculated based on the updated points. Recalculate μ and distance standard deviation σ.

Moreover, when dynamically changing the correct value set, it is preferable to configure so that the points forming the correct value set do not overlap. By doing so, it is possible to prevent the value of the feature amount of the abnormal section from becoming excessively large or small.

Further, in the above processing, the average length of the intersection intervals is calculated for each of positive and negative in the amplitude direction, and the case where the abnormality of the feature amount represented by the calculated value is determined has been described as an example. Therefore, another feature amount may be adopted as long as the feature amount having two different properties can be calculated. For example, without dividing the intersection interval by code, the “average length of the intersection interval” and the “standard deviation of the intersection interval” are calculated, and the above abnormality determination processing is executed using these two as the feature amounts, Abnormality determination may be performed.

The abnormality determination processing unit 15 stores the information indicating the result of the abnormality determination in the evaluation feature amount storage unit 34 in association with the identification information of the divided section in step S19. That is, information indicating the normal/abnormal determination result is added to the feature amount for each divided section. Then, in step S20, it is determined whether or not the reception of the sensor data is completed. If the reception of the sensor data is continued, the process returns to step S10 to repeat the series of processes described above, and if the reception of the sensor data is completed. The process ends.

(6) Analysis of Feature Amount In the biological signal processing device 1, the feature amount stored in the evaluation feature amount storage unit 34 is read by the analysis processing unit 16 during or after the series of abnormality determination processes and predetermined. Analysis processing of. At that time, the analysis processing unit 16 excludes the feature amount to which the information indicating the determination result of “abnormal” is added from the analysis target. Then, the analysis result information is transmitted from, for example, the input/output interface unit 20 to a user terminal (not shown) or a server device operated by a service provider.

When the analysis result of the feature amount of a certain division section is obtained as an analysis result that is significantly different from that of other normal sections, the analysis processing unit 16 feeds back that effect to the abnormality determination processing unit 15. The abnormality determination processing unit 15 rejects the characteristic amount of the corresponding divided section stored in the evaluation characteristic amount storage unit 34 according to the fed-back information, or the characteristic amount of the divided section is “abnormal”. Add information to the effect.

(effect)
As described above in detail, in one embodiment, the sensor data is divided into a plurality of sections at a time interval A whose length is set so as to include at least one heartbeat after performing a predetermined pre-processing, The average length of the points between the points where the electrocardiographic waveform crosses the origin is divided into positive and negative sides for each divided section, and calculated. Whether or not the point represented by the average length is abnormal is determined according to the distance from the center of gravity Go of the correct value set.

Since the electrocardiogram is a biological signal with periodicity, when the electrocardiogram is divided into multiple sections at arbitrary intervals for evaluation, the number of intersections with the origin and the intersection intervals are close to each other in the ideal state. Considered to take a value. Further, since the electrocardiogram is an asymmetric signal with respect to the potential amplitude direction, it is considered that when the intersection intervals are classified according to the sign of the potential amplitude in the same period, the characteristics peculiar to the electrocardiogram become more prominent.

Therefore, by performing the abnormality determination processing as described above, it is possible to dynamically determine whether or not an abnormal signal such as an artifact or noise is included in the sensor data without using the information on the potential amplitude direction. . Then, by adding information indicating the determination result to the data of the feature amount, the feature amount determined as an abnormal section can be excluded from the analysis target, or the heartbeat feature amount analysis result can be given reliability. It becomes possible to improve the accuracy of electrocardiogram analysis.

FIG. 5 is a diagram showing an abnormality determination result by the biological signal processing device according to the embodiment, in comparison with a conventional method. FIG. 5A shows an example of an electrocardiographic signal containing artifacts and noise. When abnormality determination is performed on such an electrocardiographic signal by the method according to the embodiment of the present invention, as shown in FIG. For example, the divided section can be determined as abnormal regardless of the magnitude of the potential amplitude. In particular, it is possible to determine that the base line fluctuation portion Td having an average intersection distance different from that in the normal time is abnormal.

Incidentally, FIG. 5C shows the discrimination result by the conventional method for discriminating the potential amplitude based on the threshold value. As shown in the figure, artifacts or noises whose potential amplitude is larger than the normal electrocardiographic waveform are abnormal. Although it can be discriminated, it is not possible to discriminate that the baseline fluctuation portion Td having a potential amplitude similar to a normal electrocardiographic waveform is abnormal.

Further, in one embodiment, the filtering process is performed in advance on the signal component to be excluded from the target in the preprocessing. That is, the abnormality determination is performed by combining the filtering process in advance and the determination process based on the average length of each intersection of waveforms for each code. For this reason, it is possible to remove or reduce the periodic noise or the like having different frequency bands in the pre-processing, thereby further improving the accuracy of abnormality determination.

[Other Embodiments]
The present invention is not limited to the above embodiment. For example, in the preprocessing, some kind of biological signal processing such as wavelet transform or fast Fourier transform may be performed. As a result, it is possible to perform abnormality determination based on the value obtained by cutting the specific frequency band of various conversion results, thereby reducing the influence of baseline fluctuation and DC noise, and determining only the artifacts derived from electrodes and myoelectricity. It becomes possible to be a target.

Further, in the above-mentioned one embodiment, the case where the abnormal signal discrimination processing is performed in real time has been described as an example, but the abnormal signal discrimination processing is batch-processed every time the sensor data is stored for a predetermined time or after the measurement of the sensor data is completed. May be performed in.

Further, in the above-described embodiment, the average length of the intersection intervals is calculated for each positive and negative sign for each divided section and used as the feature amount. However, for each divided section, the number of the intersection intervals for each positive and negative signs, that is, for the positive and negative signs is calculated. The number of generated amplitude waveforms may be counted, and the counted number may be used as the feature amount. This makes it possible to reduce the processing load due to the calculation of the feature amount.

Further, as the feature amount, the average length or the total value of the area (integrated value) of the amplitude waveform may be calculated for each positive/negative sign for each divided section. By doing so, it is possible to add the statistical value in the amplitude direction to the feature value in addition to the statistical value in the time axis direction of the sensor data, and thus, among the signals whose generation pattern of the amplitude signal in the time axis direction is similar to the normal signal. , It becomes possible to discriminate signals having different amplitudes as abnormal.

Furthermore, in the above-described embodiment, the case where the pre-processing is performed has been described as an example, but the pre-processing need not necessarily be performed. Further, a process of feeding back the analysis result from the analysis processing unit 16 to the abnormality determination processing unit 15 and changing the feature amount of the corresponding divided section stored in the evaluation feature amount storage unit 34 to “abnormal” is not necessarily performed. You don't have to.

In addition, the types of biological signals, the means for acquiring the biological signals, the configuration of the biological signal processing device, the series of processing procedures and processing contents, and the like can be variously modified and implemented without departing from the scope of the present invention.

In short, the present invention is not limited to the above embodiments as they are, and can be embodied by modifying the constituent elements within a range not departing from the gist of the invention in an implementation stage. Further, various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the above embodiments. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, the constituent elements of different embodiments may be combined appropriately.

DESCRIPTION OF SYMBOLS 1... Biological signal processing device, 2... Sensor, 10... Control unit, 11... Sensor data reception processing part, 12... Sensor data preprocessing part, 13... Sensor data division processing part, 14... Feature amount calculation processing part, 15... Abnormality determination processing unit, 16... Analysis processing unit, 20... Input/output interface unit, 30... Storage unit, 31... Sensor data storage unit, 32... Preprocessed sensor data storage unit, 33... Divided data storage unit, 34... Evaluation Feature quantity storage unit.

Claims (8)

Receiving means for receiving a biological signal including a waveform having periodicity,
Dividing means for dividing the received biomedical signal into a plurality of time regions at time intervals set to include at least one of the waveforms;
For each of the plurality of time regions, as a feature amount indicating the generation state of the waveform, a feature amount calculation means for calculating the length or number between the intersections when the waveform intersects the origin of amplitude,
By comparing the calculated feature amount with a range of normal feature amount set in advance based on the periodicity of the waveform of the biological signal, the calculated feature amount corresponds to the waveform of the biological signal. A biological signal processing device, comprising: an abnormality determining unit that determines whether the normal signal is abnormal or abnormal due to an abnormal signal mixed in the biological signal. The feature amount calculation means calculates, for each of the plurality of time regions, the length or number between the intersections when the waveform intersects the origin of amplitude,
The abnormality determining means compares the calculated length or number of intersections with a range of length or number of normal intersections preset based on the periodicity of the waveform to calculate 2. The length or number between the intersected intersections is discriminated whether it is a normal one corresponding to the waveform of the biological signal or an abnormal one caused by an abnormal signal mixed in the biological signal. Biological signal processing device. Receiving means for receiving a biological signal including a waveform having periodicity,
Dividing means for dividing the received biological signal into a plurality of time regions at time intervals set to include at least one of the waveforms;
For each of the previous SL plurality of time regions, and calculate the length or number of intersections at which the waveform crosses the origin of the amplitude, the length between the calculated intersection or the number the waveform amplitude direction A feature amount calculating unit that aggregates based on information to calculate at least two types of feature amounts;
At least two types of features are pre SL calculated by comparing with a preset normal feature amount ranging, based on the periodicity of the waveform feature amount of at least two the calculated is the An abnormality determining means for determining whether the normal one corresponds to the waveform of the biological signal or the abnormal one caused by the abnormal signal mixed in the biological signal.
A biological signal processing device comprising: Receiving means for receiving a biological signal including a waveform having periodicity,
Dividing means for dividing the received biomedical signal into a plurality of time regions at time intervals set to include at least one of the waveforms;
For each of the previous SL plurality of time regions, the waveform calculating the length or number of intersections when crossing the origin of the amplitude, aggregate length or number between the calculated intersection sign another said waveform And a feature amount calculation means for calculating the feature amount on the positive side and the negative side,
Each feature amount before Symbol calculated positive and negative sides, by comparing with a preset normal feature amount ranging, based on the periodicity of the waveform, the calculated positive and negative sides An abnormality determining means for determining whether each feature amount is normal corresponding to the waveform of the biological signal or abnormal due to an abnormal signal mixed in the biological signal.
A biological signal processing device comprising: The abnormality determination means,
When the result of the analysis processing based on the characteristic amount determined to be normal is out of a preset allowable range, a means for changing the determination result of the characteristic amount targeted for the analysis processing to an abnormal one. The biomedical signal processing apparatus according to claim 1, further comprising: An abnormality determination method executed by a biological signal processing device comprising a processor and storage means,
Receiving a biological signal including a waveform having periodicity,
Dividing the received biomedical signal into a plurality of time domains at time intervals set to include at least one of the waveforms;
For each of the plurality of time regions, a step of calculating the length or the number of intersections when the waveform intersects the origin of the amplitude, as a feature amount indicating the generation state of the waveform,
By comparing the calculated feature amount with a range of normal feature amount set in advance based on the periodicity of the waveform of the biological signal, the calculated feature amount corresponds to the waveform of the biological signal. And a normal signal that is abnormal due to an abnormal signal mixed with the biological signal. An abnormality determination method executed by a biological signal processing device comprising a processor and storage means,
Receiving a biological signal including a waveform having periodicity,
Dividing the received biological signal into a plurality of time regions at time intervals set to include at least one of the waveforms;
For each of the previous SL plurality of time regions, the waveform calculating the length or number of intersections when crossing the origin of the amplitude, the length or number between the calculated intersection in the amplitude direction of the waveform calculating at least two types of features are aggregated,
At least two types of features are pre SL calculated by comparing with a preset normal feature amount ranging, based on the periodicity of the waveform feature amount of at least two the calculated is the A step of determining whether it is a normal one corresponding to the waveform of the biological signal or an abnormal one due to an abnormal signal mixed in the biological signal ;
An abnormality determination method comprising : A program that causes a computer to function as each unit included in the biological signal processing apparatus according to claim 1.