Biosignal processing method

Abstract

To provide a method for processing a biological signal, which can perform digitization processing while capturing the characteristics of a gradual rise or fall of the biological signal, and can vectorize the biological signal in a short time. [Solution] A method for processing biological signals including pulse waves, electrocardiograms, electroencephalograms, electrogastrograms, electroenterograms, electromyograms, or respiratory waves, comprising a biological signal acquisition step S1 for acquiring the biological signal, a minute signal application step S2 for applying to the biological signal a minute signal that has a certain amplitude and a certain frequency and is smaller than the biological signal, and acquiring a signal after the minute signal application, an extreme value data acquisition step S3 for acquiring the maximum and minimum values ​​of the signal after the minute signal application, and a statistical processing step S4 for performing statistical processing based on the maximum and minimum values.

Description

[Technical Field]

[0001] The present invention relates to a method for processing a biological signal. [Background technology]

[0002] Human biological signals such as pulse waves, electrocardiograms, electroencephalograms, electrogastrograms, electroenterograms, electromyograms, and respiratory waves contain a wealth of information that indicates the health status of individuals. Taking pulse waves as an example, as shown in Figure 1, the pulse waves exhibit a pattern of repeated instantaneous increases in blood pressure and blood volume in blood vessels, followed by a gradual decrease. Doctors make diagnoses based on the amplitude, frequency, etc. of these signals.

[0003] The above-mentioned biological signals are usually acquired as analog signals, but it is desirable to perform appropriate digitization processing for diagnosis. Here, Patent Document 1 (JP 2011-055253 A) discloses a technology for improving the quantization performance of a biological signal by adding two or more types of noise of appropriate intensities to the analog biological signal. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2011-055253 Summary of the Invention [Problem to be solved by the invention]

[0005] However, even if appropriate quantization is performed on a biological signal as disclosed in Patent Document 1, the information that can be obtained from subsequent analysis is limited based on current knowledge of biological signals. [Means for solving the problem]

[0006] As a result of intensive research, the inventors have discovered that useful information is contained in waveforms showing a gradual rise or fall of a biosignal (hereinafter sometimes simply referred to as the "slow region"). However, it is difficult to handle the slow region of a biosignal using conventional methods (e.g., differentiation and integration, Fourier transform, etc.). In other words, there has been a demand for a biosignal processing technology that can perform digitization processing while capturing the characteristics of the slow region of a biosignal, and vectorize the biosignal in a short period of time.

[0007] Therefore, the present invention has been made to solve the above problems, and aims to provide a method for processing biological signals that can perform digitization processing while capturing the characteristics of the slow region of the biological signal, and vectorize the biological signal in a short period of time.

[0008] The present invention solves the above problems by the solution means described below as one embodiment.

[0009] The disclosed method for processing a biological signal is a method for processing a biological signal including a pulse wave, electrocardiogram, electroencephalogram, electrogastrogram, electroenterogram, electromyogram, or respiratory wave, and requires the steps of: a biological signal acquisition step of acquiring the biological signal; a minute signal application step of applying to the biological signal a minute signal that has a certain amplitude and a certain frequency and is smaller than the biological signal, and acquiring a signal after the minute signal application; an extreme value data acquisition step of acquiring the maximum value and minimum value of the signal after the minute signal application; and a statistical processing step of performing statistical processing based on the maximum value and the minimum value.

[0010] The minute signal preferably has any one of a trigonometric function waveform, a sawtooth waveform, a triangular waveform, a rectangular waveform, and an elliptical waveform.

[0011] Furthermore, it is preferable that the minute signal has a waveform with a frequency, amplitude, and phase that satisfies all of the following conditions 1, 2, and 3. (Condition 1) 50Hz≦(frequency of the above-mentioned minute signal)≦300Hz (Condition 2) (Amplitude of the biological signal) / 1000≦(Amplitude of the minute signal)≦(Amplitude of the biological signal) / 10 (Condition 3) The maximum value and the minimum value appear discontinuously.

[0012] Furthermore, it is preferable to further include a noise removal step, which is a step subsequent to the biosignal acquisition step and prior to the minute signal application step, of removing mechanical noise applied to the biosignal.

[0013] Furthermore, it is preferable that two or more of the minute signals are applied in the minute signal application step. [Effects of the Invention]

[0014] According to the present invention, a digitization process is performed while capturing the characteristics of the slow region of a biological signal, and the biological signal can be vectorized in a short time. [Brief explanation of the drawings]

[0015] [Figure 1] FIG. 1 is a graph of a human pulse wave before a minute signal is applied in each embodiment of the present invention. [Figure 2] FIG. 2 is an explanatory diagram showing the "slow region" of the biosignal exemplified in FIG. [Figure 3] FIG. 3 is an explanatory diagram showing an example of a minute signal in each embodiment of the present invention. [Figure 4] FIG. 4 is a graph of a signal after application of a small signal in each embodiment of the present invention. [Figure 5] FIG. 5 is a flowchart of a method for processing a biological signal in each embodiment of the present invention. [Figure 6] FIG. 6 is an explanatory diagram of the extreme value data acquisition step of the signal after the small signal is applied in each embodiment of the present invention. [Figure 7] FIG. 7 is an explanatory diagram of the statistical processing step of the signal after the addition of the small signal in each embodiment of the present invention. [Figure 8]FIG. 8 is a graph showing the power spectrum of the pulse wave graph of FIG. [Figure 9] FIG. 9 is a graph showing the power spectrum of the pulse wave graph of FIG. 1 after mechanical noise has been removed. [Figure 10] FIG. 10 is a graph of the pulse wave of FIG. 1 after mechanical noise has been removed. [Figure 11] 11A to 11C are explanatory diagrams showing how sensors are used in the respective embodiments of the present invention. [Figure 12] FIG. 12 is an explanatory diagram of the extreme value data acquisition step and statistical processing step in each embodiment of the present invention. [Figure 13] Figure 13A is a graph showing the results of statistical processing of data for each physical fitness state of the subject in each embodiment of the present invention, and Figure 13B is a graph showing the results of statistical processing of data for each psychological state of the subject in each embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0016] Hereinafter, the method for processing a biological signal in each embodiment will be described in detail with reference to the drawings. FIG. 1 is a graph of a human pulse wave before application of a small signal in each embodiment of the present invention. FIG. 2 is an explanatory diagram showing the "slow region" of the biological signal exemplified in FIG. 1. FIG. 3 is an explanatory diagram showing an example of a small signal in each embodiment of the present invention. FIG. 4 is a graph of a signal after application of a small signal in each embodiment of the present invention. FIG. 5 is a flowchart of the method for processing a biological signal in each embodiment of the present invention. FIG. 6 is an explanatory diagram showing the extreme value data acquisition process of the signal after application of a small signal in each embodiment of the present invention. FIG. 7 is an explanatory diagram showing the statistical processing process of the signal after application of a small signal in each embodiment of the present invention. Note that in all figures used to explain each embodiment, components having the same function are designated by the same reference numerals, and repeated explanations thereof may be omitted.

[0017] First, the biosignals (human biosignals) handled in each embodiment will be described. The biosignals include pulse waves, electrocardiograms, electroencephalograms, electrogastrograms, electroenterograms, electromyograms, and respiratory waves. For any of the biosignals described above, as shown in FIG. 1, the horizontal axis represents elapsed time, and the vertical axis represents measured values ​​of the biosignals acquired by various sensors. The biosignals typically exhibit a waveform that alternates between a momentary rise in the measured value (hereinafter simply referred to as the "rising region") and a subsequent gradual rise or fall (slow region). Note that FIG. 1 shows a graph of a human pulse wave, with the horizontal axis representing elapsed time and the vertical axis representing arterial wave motion (an index corresponding to blood pressure, blood flow, volume change, etc.). Note that approximately 1,000 pieces of data are acquired per second in FIG. 1. Also, FIG. 2 shows the slow region of the biosignal (particularly, the human pulse wave).

[0018] Examples of the above-mentioned sensors include, but are not limited to, potential sensors, mechanical sensors, optical sensors, and microphones.

[0019] Furthermore, the above-mentioned biological signals are usually acquired as analog signals, but are not limited to this and may be acquired as digital signals.

[0020] Next, the minute signals handled in each embodiment will be described. Signal analysis is generally performed after removing noise (e.g., mechanical noise). However, the inventors have discovered that applying an appropriate minute signal (e.g., an appropriate real-valued minute signal) to the above-described biosignal enables subsequent useful analysis of the biosignal. Specifically, in the method for processing a biosignal in each embodiment, a signal (e.g., a digital signal) having a constant amplitude and a constant period, as shown in FIG. 3, and smaller than the above-described biosignal is applied. Note that the method for applying the minute signal includes not only adding the minute signal to the biosignal but also subtracting the minute signal from the biosignal.

[0021] The minute signal has any of a trigonometric function waveform, a sawtooth waveform, a triangular waveform, a rectangular waveform, and an elliptical waveform, as exemplified in Fig. 3, but is not limited to these and can be appropriately selected depending on the purpose. In particular, when an elliptical waveform is adopted, it becomes possible to transform it into a complex space, a cryptographic space, a space with a changing topology such as a torus, and the like.

[0022] Furthermore, it is preferable that the minute signal has a waveform with a frequency, amplitude, and phase that satisfies all of the following conditions 1, 2, and 3. That is, as the first condition, it is preferable that 50 Hz≦(frequency of the minute signal)≦300 Hz, as the second condition, it is preferable that (amplitude of the biological signal) / 1000≦(amplitude of the minute signal)≦(amplitude of the biological signal) / 10, and as the third condition, it is preferable that the maximum and minimum values ​​(especially the maximum and minimum values ​​in the falling region) of the biological signal to which the minute signal has been added (hereinafter, sometimes simply referred to as the "signal after addition of the minute signal"; see FIG. 4) appear discontinuously.

[0023] By making the minute signal have the waveform described above, it is possible to extract the characteristics of the biosignal of each subject while maintaining the characteristics of the rising and slowing regions.

[0024] Next, the method for processing a biological signal in each embodiment will be described in detail. As shown in the flowchart of Fig. 5, the method for processing a biological signal includes a biological signal acquisition step S1 for acquiring a biological signal, a minute signal application step S2 for applying the above-mentioned minute signal to the biological signal and acquiring a post-minute signal application signal, an extreme value data acquisition step S3 for acquiring maximum and minimum values ​​of the post-minute signal application signal, and a statistical processing step S4 for performing statistical processing based on the maximum and minimum values.

[0025] First, the biosignal acquisition step S1 will be described. The biosignals described above are usually acquired as analog signals, so in the biosignal acquisition step S1, the user uses a digital converter to convert the biosignals into digital signals and acquire them. Alternatively, since the biosignals can also be digitized in the small signal application step S2 described below, in the biosignal acquisition step S1, the user acquires the biosignals as analog signals using the sensor described above.

[0026] Next, the microsignal applying step S2 will be described. In the microsignal applying step S2, the user applies a preset microsignal (particularly, a digital signal) to the biosignal using a control device (such as a PC terminal, server, microcomputer, micromachine, FPGA, etc.) having a memory such as ROM / RAM and a CPU to obtain a post-microsignal applying signal. More specifically, in the microsignal applying step S2, the user applies a microsignal to the biosignal stored in the storage unit of the control device, thereby increasing or decreasing the waveform of the biosignal and obtaining a post-microsignal applying signal having the increased or decreased waveform. More specifically, in the microsignal applying step S2, the user applies a microsignal to the biosignal and adjusts at least one of the amplitude, frequency, and phase of the microsignal until the maximum and minimum values ​​described below no longer appear consecutively, and repeats the microsignal applying step S2.

[0027] In the minute signal applying step S2, two or more minute signals may be applied to the biosignal, thereby enabling more detailed analysis to be performed.

[0028] Next, the extreme value data acquisition step S3 will be described. Here, the post-microsignal-applied signal maintains the characteristics of the original biosignal, namely, the rise in the rising region and the gradual rise or fall in the slow region, but the application of the microsignal causes the rise region and the slow region to each have a waveform that changes with a slight increase or decrease. Therefore, in the extreme value data acquisition step S3, the user acquires the maximum and minimum values ​​from the post-microsignal-applied signal stored in the storage unit of the control device, as shown in FIG. 6. More specifically, in the extreme value data acquisition step S3, the user uses the control device to extract the maximum and minimum values ​​from the post-microsignal-applied signal stored in the storage unit of the control device, extracts the time, the value of the wave (in the case of a pulse wave, arterial blood pressure or volume change), and information on whether the value is a maximum or minimum, and stores the extracted information in the storage unit. The extreme value data acquisition step S3 enables data compression while capturing the characteristics of the biosignal (especially the characteristics of the slow region of the biosignal), and further enables vectorization of the biosignal in a short time.

[0029] Next, the statistical processing step S4 will be described. In the statistical processing step S4, the user uses the control device described above to perform statistical processing on extreme values ​​such as the maximum and minimum values ​​described above, as shown in Fig. 7. More specifically, in the statistical processing step S4, the user calculates the maximum or minimum values, or the average or variance of the maximum and minimum values. Alternatively, in the statistical processing step S4, the user calculates the average or variance of adjacent maximum or minimum values, or adjacent maximum and minimum values, as shown in Fig. 5. The statistical processing step S4 makes it possible to analyze the characteristics of the biosignals of individual subjects in more detail than conventional methods.

[0030] Next, the noise removal step S5 will be described. The biosignal processing method in each embodiment may further include a noise removal step S5, which is a step subsequent to the biosignal acquisition step 1 and a step prior to the small signal application step S2, for removing mechanical noise applied to the biosignal. As shown in the power spectrum graph for each frequency in FIG. 8, the mechanical noise is noise that exhibits properties different from the small signal described above, and is generated by vibrations during measurement or environmental electromagnetic waves. Specifically, FIG. 8 shows that noise of around 60 Hz has been applied. Therefore, in the noise removal step S5, the user uses a control device to apply a predetermined filter to the biosignal to remove the mechanical noise. By performing the noise removal step S5, a pre-application signal from which mechanical noise has been removed can be obtained, as shown in FIGS. 9 and 10, enabling the statistical processing step S4 to be performed with higher accuracy.

[0031] <Example> Next, an example of the biosignal processing method according to each embodiment of the present invention will be described in detail. Figures 11A to 11C are explanatory diagrams of the biosignal acquisition step S1 in this example. Figure 12 is an explanatory diagram of the extreme value data acquisition step S3 and the statistical processing step S4. Figures 13A and 13B are graphs showing the difference in vector length between adjacent extreme value data for each time.

[0032] In this example, a sensor (potential sensor) 10 as shown in Fig. 11A was attached to the fingertip of a subject. Note that the sensor 10 may be attached to the chest or abdomen as shown in Fig. 11B, or may be attached to the side of the head or the like as shown in Fig. 11C.

[0033] Next, the inventors acquired biosignals (pulse waves) using the sensor 10 for each physical condition of the subject (specifically, normal, immediately after bathing, after exercise, and before going to bed) (biological signal acquisition step S1). Furthermore, the inventors acquired biosignals (pulse waves) using the sensor 10 for each psychological state of the subject (mild anxiety, moderate anxiety, and severe anxiety) (biological signal acquisition step S1).

[0034] Next, the inventors removed mechanical noise of around 60 Hz from each of the seven pulse waves (noise removal step S5). Specifically, a predetermined noise removal filter capable of removing mechanical noise of 59 Hz to 61 Hz was applied to each pulse wave.

[0035] Next, the inventors applied the above-mentioned predetermined minute signal to each pulse wave from which mechanical noise had been removed (minute signal application step S2). Specifically, a minute signal of a cosine waveform with a frequency of 100 Hz and an amplitude of 51 was applied (added) to each pulse wave.

[0036] Next, the inventor extracted feature points of each of the signals after the small signal was applied. Specifically, the inventor extracted maximum value data and minimum value data of each of the signals after the small signal was applied (extreme value data acquisition step S3), as shown in FIG.

[0037] Next, the inventors performed predetermined statistical processing on the extracted maximum value data and minimum value data (statistical processing step S4). In order to perform the statistical processing step S4, the slow region of the signal after the small signal is applied is designated as T1, T2, ..., T3, as shown in FIG. k ,…,T N The data corresponding to these times are X1, X2, ..., X k ,…,X N These data are expressed as extreme vectors ((T1,X1),(T2,X2),…,(T k ,X k ),…,(T N ,X N )) and the vector of the difference between adjacent extreme vectors ((T k ,X k )-(T k-1 ,X k-1 )) and the difference vector between adjacent maximum vectors ((T 2m ,X 2m )-(T 2m-2 ,X 2m-2 )), and the vector of differences between adjacent minimum vectors ((T2m-1 ,X 2m-1 )-(T 2m-3 ,X 2m-3 The length of the difference vector was calculated, and the average length m and variance v of the difference vector were calculated for both cases where the difference vector was rising and falling. The results are shown in Table 1 below.

[0038] [Table 1]

[0039] Table 1 shows the results of calculating the mean m and variance v of the difference vector for each subject's physical and psychological state. The arrow pointing to the upper right indicates an upward trend in the difference vector, while the arrow pointing to the lower right indicates a downward trend in the difference vector. The "Daytime" and "Before Bedtime" data in Table 1 are both statistically processed data from a relatively relaxed state, where the parasympathetic nervous system is dominant. The "After Bath" data are statistically processed data from a relatively excited state, where the sympathetic nervous system is dominant after a prolonged immersion in water over 42°C. The "After Exercise" data are statistically processed data from a relatively excited state, where the sympathetic nervous system is dominant after a prolonged jog. The "When Anxious" data are statistically processed data obtained for each level of anxiety.

[0040] From Table 1, it can be seen that the variance value when the difference vector is rising and the average value when the difference vector is falling show results that differ greatly depending on the physical fitness and psychological state of the subject. Furthermore, it is presumed that the sympathetic or parasympathetic dominant states after bathing in this example are similar to those during mild to moderate anxiety, and it was confirmed that the variance value when the difference vector is rising and the average value when the difference vector is falling are similar. Furthermore, it was confirmed that there is a difference between the variance value when the difference vector is rising and the average value when the difference vector is falling between severe anxiety and mild to moderate anxiety.

[0041] Next, the difference vector between adjacent maximum vectors ((T2m ,X 2m )-(T 2m-2 ,X 2m-2 )), and the vector of differences between adjacent minimum vectors ((T 2m-1 ,X 2m-1 )-(T 2m-3 ,X 2m-3 The results of calculating the length of )) are shown in Figures 13A and 13B.

[0042] 13A, it was confirmed that the graphs showed different aspects depending on the physical fitness state. Also, FIG. 13B, it was confirmed that the graphs showed similar aspects depending on the psychological state. Furthermore, it was confirmed that the graphs for "after bathing" and "at mild to moderate anxiety" were particularly similar.

[0043] The above results suggest that the biosignal processing method of the present invention can obtain biosignal characteristics corresponding to the subject's physical fitness and psychological state, which are difficult to obtain using conventional methods, by applying an appropriate minute signal to a human biosignal (pulse wave) and performing a predetermined statistical analysis based on the extreme values ​​of the resulting signal after application. Therefore, by using the biosignal processing method of the present invention, it is possible to visualize the physical fitness and psychological state of the subject, which the subject himself is not aware of, and it is thought that this method will be useful in various situations in daily life.

[0044] The present invention is not limited to the above-described embodiment, and various modifications are possible without departing from the scope of the present invention. [Explanation of symbols]

[0045] 10 sensors S1 Biosignal acquisition process S2 Micro signal application process S3 Extreme value data acquisition process S4 Statistical processing process S5 Noise removal process

Claims

1. A method for processing biological signals including pulse waves, electrocardiograms, electroencephalograms, electrogastrograms, electroenterograms, electromyograms, or respiratory waves, comprising: a biological signal acquiring step of acquiring the biological signal; Next, a minute signal applying step of applying a minute signal having a certain amplitude and a certain frequency and being smaller than the biological signal to the biological signal, and acquiring a signal after applying the minute signal; Next, an extreme value data acquisition step of acquiring the maximum value and the minimum value of the signal after the small signal is applied; Next, a statistical processing step is provided in which statistical processing is performed based on the maximum value and the minimum value. A method for processing a biological signal, characterized by:

2. The minute signal has any one of a triangular waveform, a sawtooth waveform, a triangular waveform, a rectangular waveform, and an elliptical waveform.

2. The method for processing a biological signal according to claim 1,

3. The minute signal has a waveform of frequency, amplitude, and phase that satisfies all of the following conditions 1, 2, and 3:

3. The method for processing a biological signal according to claim 1 or 2, wherein: (Condition 1) 50 Hz≦(frequency of the minute signal)≦300 Hz (Condition 2) (Amplitude of the biological signal) / 1000≦(Amplitude of the minute signal)≦(Amplitude of the biological signal) / 10 (Condition 3) The maximum value and the minimum value appear discontinuously.

4. As a subsequent process of the biological signal acquisition process and a previous process of the minute signal application process, Further comprising a noise removal step of removing mechanical noise imparted to the biological signal.

3. The method for processing a biological signal according to claim 1 or 2, wherein:

5. In the micro-signal applying step, two or more micro-signals are applied.

3. The method for processing a biological signal according to claim 1 or 2, wherein:

Images