JP4754447B2 - Biological analysis apparatus and program - Google Patents

Description

  The present invention relates to a biological analysis apparatus that analyzes biological information, in particular, time-series data such as body movement, heart / beat, and brain waves.

  2. Description of the Related Art Conventionally, a method for diagnosing a patient or performing health care by evaluating time-series data of biological information including body movement (body movement), heartbeat, and the like is well known.

  Taking body motion as an example, the level of sleep or awakening can be estimated by measuring the wrist movement using an acceleration sensor, for example. If you are healthy, there is almost no movement of your wrist during sleep, and you can see more than a certain amount of movement during awakening even if you are resting. However, patients with sleep disorders such as sleep apnea syndrome may experience body movements even though they are sleeping. Also, patients with mental illness such as depression and chronic fatigue syndrome may observe very low body movements despite being awake. May be seen.

  In addition, the amount of exercise can be estimated by attaching an acceleration sensor to the waist instead of the wrist. The lumbar region hardly moves when resting even while awake, but the lumbar region also moves when exercising such as walking. There have been proposals such as attaching acceleration sensors to both the waist and wrist, detecting the resting time during awakening, and collecting blood pressure during daily rest by measuring blood pressure at that time.

  Specifically, the output value of the acceleration sensor attached to the wrist or the like is shown in the figure. FIG. 11A shows time values of activity values as data relating to the body movement of the subject. Here, the activity level (Zero Crossing Data) is the number of times when a threshold is crossed per unit time with respect to the obtained acceleration. For example, if this threshold is crossed “20 times” in “10 seconds”, the activity level is “20”. By determining the activity level, the movement of the subject can be acquired, and the larger the value, the stronger the movement of the subject.

  In the case of FIG. 11A, the solid line 401 indicates the activity, and the broken line 402 indicates the threshold value. In FIG. 11A, the threshold value is crossed “4” times, and the activity level is “4”. In general, the activity level detected from the wrist is a value around “50” when walking or doing some exercise, and is almost “0” when resting or sleeping.

  FIG. 10B shows the heartbeat of the subject. FIG. 10B shows a heartbeat waveform when the heart beats three times. In general, for each pulsation, three upward peaks and two downward peaks are observed. These peaks are named P, Q, R, S, and T in order, and the interval between the most prominent peak R and the next peak R is called “rr-interval”. Strictly speaking, not all of these peaks may be observed or other peaks may appear due to differences in the measurement method and the patient's physical condition.

  Usually, the rr-interval of a person is about 1 second, it becomes smaller when exercising, and becomes larger at rest or sleep. Even if you are not exercising, if rr-interval becomes small, fluctuates extremely unstable, or keeps giving a constant value, you may have a disease in your heart.

The heart rate obtained by subtracting your age from “220” is said to be the highest heart rate that the person has (estimated maximum heart rate), and it is said that it is good to maintain about 70% of that value during exercise. . For example, if you are 40 years old,
(220-40) × 0.7 = 126
It is good to maintain 120-130. In recent years, attention has been paid to a method for diagnosing whether a patient's autonomic nerve is functioning correctly by measuring daily heart rate variability over several days and performing fractal analysis on the heart rate variability.

  And the data regarding such biometric information always fluctuate | varies according to the daily activity condition, a mental state, the symptom of a physical or mental disease.

The discovery of abnormalities from such changes in body movement has attracted attention in the past. For example, a device for detecting an abnormality of a driver from a heartbeat is known (see, for example, Patent Document 1). There is also known an apparatus that calculates a threshold value of an abnormal value from a normal value and discovers an abnormality of a physical symptom with high accuracy (see, for example, Patent Document 2).
Japanese Patent No. 2671394 JP 2005-124718 A

  The biological information of the subject always varies, but depending on the activity status, it may be determined to be abnormal or may be determined to be normal. For example, a heart rate of “140” when you are tense or exercising can be said to be in the normal range because it is caused by mental changes or changes in physical activity. If it is “140”, there is a high possibility that the heart has a disease.

  Symptoms may be evaluated by evaluating heart rate and activity data only during rest or light exercise, but in the past, living in a specific controlled environment such as a laboratory, The only way to obtain these data is to create a state of rest in front of the measurer. However, when acquiring data in the laboratory, it has been necessary to restrain the subject for a long time, and there has been a problem that it takes time and cost.

  Also, when acquiring data in the laboratory or acquiring data in front of the measurer, some subjects may become more nervous than usual or conversely create a calm state. There was something that could not be taken correctly.

  In addition, data could not be collected in a controlled environment such as a laboratory if symptoms occurred when taking specific daily actions or at specific locations.

  In order to solve the above problems, the present invention acquires biological information in time series, and correctly evaluates biological information by dividing the biological information acquired in time series by time using a statistical method. It is providing the biological analysis apparatus which can be performed.

In order to solve the above-described problem, the biological analysis apparatus of the present invention analyzes biological information acquired in time series by biological information acquisition means that acquires biological information of a subject in time series, and the biological information acquisition means. A bioanalyzing device, and a bioanalytical apparatus comprising: an evaluation unit that evaluates a state of a subject based on a result analyzed by the bioanalytical unit, wherein the bioanalytical unit is acquired by the biometric information acquiring unit The time-series data is divided into the time-series data of the living body before the time τ and after τ in the data, and for each τ, the statistical difference between the data before τ and the data after τ is calculated. And time division analysis means for analyzing the biological information by dividing the data at the time when the difference is maximum, and the evaluation means is in accordance with each time divided by the time division analysis means. It is a means for evaluating time-series data of biological information.

In the biological analysis apparatus of the present invention, the biological information is an activity level of body movement or a heartbeat.

The biological analysis apparatus according to the present invention further includes an input unit that inputs the behavior or psychological state of the subject, and the behavior or psychological state of the subject input by the input unit and the evaluation evaluated by the evaluation unit. It is characterized by being stored in association.

The biological analysis apparatus of the present invention, as the statistical method, which comprises using the t-test.

Further, the biological analysis apparatus of the present invention is characterized in that a χ ² test is used as the statistical method.

  According to the present invention, the biological information of the subject acquired in time series can be analyzed by dividing the time using a statistical method, and the time series data of the biological information can be evaluated according to each time. . And based on the evaluated content, the user of the bioanalytical apparatus can know and analyze the situation when the subject's symptoms worsen, and can perform treatment and diagnosis. Specific effects are as follows.

  First, biological time-series data can be divided using a comparatively easy and clear calculation method by using a statistical method. In particular, the division using the heart rate, the activity level, and the electroencephalogram allows the user to make a diagnosis while correlating with the behavior and mental state at that time.

  Furthermore, it is possible to divide the time series data more clearly by dividing it into appropriate time, comparing the time series data before and after that time using statistical methods, and determining the time when the difference is the maximum. become able to do.

  In addition, by dividing the divided time using the same method, it became possible to determine the time when the behavior changed finely. By dividing the divided time within a specific time as the start time and repeating the same operation with the divided time as the start time, it is possible to determine the time when the behavior changed finely even with long-term behavior changes. I can do it.

  In addition, the time before the divided time and the time after the divided time are divided by the same method, and the same operation is repeated, so that it is possible to estimate the time when the behavior has changed with high accuracy.

  In addition, by entering the behavior and psychological state later at the divided time, it becomes possible to accurately grasp the situation where the biological time series data to be evaluated is acquired, and more accurately the time series data of the biological body Can be evaluated.

Further, as a statistical method used for the division, a t-test or a χ ² test can be used to clearly divide with high accuracy.

  Next, a most preferred example of the biological information device to which the present invention is applied will be described with reference to the drawings. FIG. 1 shows a biological analysis apparatus 10 when the present invention is applied. The biological analysis device 10 includes a heartbeat data acquisition device 100, an acceleration data acquisition device 110, and a biological information analysis device 120.

  The heartbeat data acquisition device 100 includes a heartbeat sensor 102 and a transmission circuit 104, and transmits the heartbeat data of the subject acquired by the heartbeat sensor 102 to the biological information analysis device 120 via the transmission circuit 104. Here, the method by which the heart rate sensor 102 acquires the heart rate data of the subject is known, and the description thereof is omitted.

  The acceleration data acquisition device 110 includes an acceleration sensor 112 and a transmission circuit 114, and transmits acceleration data acquired by the acceleration sensor 112 to the biological information analysis device 120 via the transmission circuit 104. Here, a method in which the acceleration sensor 112 detects acceleration and acquires it as acceleration data is well known, and the description thereof is omitted.

  Further, methods for transferring each data from the heart rate data acquisition device 100 and the acceleration data acquisition device 110 to the biological information analysis device 120 include wired transfer, wireless transfer using radio waves and infrared rays, and a method of transferring data through the living body. Although there are various methods, the present embodiment assumes data transfer using radio waves that are not complicated for the user and are widely spread.

  The biological information analysis device 120 is a device that performs analysis based on the heart rate data received from the heart rate data acquisition device 100 and the activity data received from the acceleration data acquisition device 110. The biological information analysis device 120 includes a receiving circuit 122, a memory 124, a biological analysis circuit 126, and a display device 128.

  The outline of the operation of the biological information analysis apparatus 120 will be described. First, the reception circuit 122 receives the heartbeat data acquired by the heartbeat data acquisition apparatus 100 and the acceleration data acquired by the acceleration data acquisition apparatus 110. Then, the received data is stored in the memory 124.

  The biological information analysis device 120 reads heartbeat data and acceleration data stored in the memory 124 and analyzes the data in the biological analysis circuit 126. Although details will be described later, rr-interval is calculated from heartbeat data, and activity is calculated from acceleration data based on a predetermined divided time. Then, the result of analyzing the data is stored in a memory, and the result is displayed on the display device 128. The display device is assumed to be a liquid crystal capable of displaying text and pictures, an organic EL display, or the like. For example, if it is necessary to urgently notify the user, a call may be made by vibration or sound.

  FIG. 2 is an example of an appearance when the biological analysis apparatus 10 is incorporated into a mobile phone. For example, the mobile phone 201 is foldable, and the parts are divided into a part 202 and a part 203. Here, the cellular phone 201 is provided with the biological analysis device 10 first, and can process various biological signals. Therefore, the mobile phone 201 includes a receiving device that receives biometric information.

  1 has been described as including the memory 124 and the display device 128 separately, these may be shared with the memory, the display device, etc. originally possessed by the mobile phone. By doing so, a significant increase in cost can be avoided and higher portability can be maintained. It should be noted that there is no significant difference from a general mobile phone except for the portion related to the present invention. A part 202 has a number of keys for operating a mobile phone, and a part 203 is a mobile phone display portion.

  Usually, when not in use, it is closed as shown in FIG. 2 (a), and the user usually keeps it in a pocket or the like. When the user uses a mobile phone, confirms the result of the biological analysis device, or inputs his / her information, the state shown in FIG. 2 (a) is passed through the state shown in FIG. 2 (b). ) The mobile phone is opened until it is used.

  FIG. 3 shows an example when the product of the present invention is mounted. The user wraps the heart rate data acquisition device 301 around the chest, the acceleration data acquisition device 302 with the built-in acceleration sensor around the wrist, and puts the mobile phone 303 with the built-in bioanalyzer in the pocket of the pants. In the present embodiment, it is assumed that the heartbeat data acquisition device is wound around the chest, but the present invention is not limited to this location, and for example, a method of acquiring from a wrist, finger, ear, or the like may be used.

  Further, in the present embodiment, the acceleration data acquisition device 302 is wound around the wrist, but various locations such as the waist, chest, and feet can be considered as the location for acquiring body acceleration data. In this embodiment, it demonstrates as acquiring the acceleration of the wrist often used, for example by the determination of sleep.

  Note that the heart rate data acquisition device and the acceleration data acquisition device may be configured separately, or are housed in the same housing and acquire both heart rate and acceleration at the wrist as a mounting location of the sensor-equipped device. Also good. A mobile phone with a built-in direct acceleration sensor may be used.

  Next, a mechanism when the biological analysis circuit 126 analyzes the received biological information will be described with reference to the drawings. The biological analysis circuit 126 divides the activity data by time using a statistical method, and analyzes the biological information. FIG. 4 is a diagram showing the degree of activity at a certain time. The horizontal axis represents time, and the vertical axis represents wrist activity (Zero Crossing Data). Here, a solid line 501 indicates the activity of the subject, a dotted line 502 indicates a parameter calculated by a statistical method (t test), and a broken line 503 indicates a parameter calculated by a statistical method (Kolmogorov-Smirnov test: ks test). is there.

The activity shown in FIG. 4 is the body movement of the subject who has continued walking for a while from the start of data acquisition and has taken some work while sitting at the seat at time t ₁ . That is, until the time t ₁ , on average, a high activity level of around 50 is shown, but this time shows that walking is always continued. After time t _1, the activity level has changed up and down, but it is a state where the user stops walking and is working, showing a high value when moving his hand and a low value when resting.

In analyzing physical activity, it is necessary to know what kind of activity was continued from what time to what time, but it is difficult to find the changing point only by the level of activity. The change can be predicted to some extent by visually observing the graph, but there is no clear judgment criterion, and when there is a large amount of data, the judgment cannot be caught by visual observation. In the present invention, the time t ₁ when the activity has changed is estimated by taking a statistical method (t test, ks test) described in one of the following two methods.

ただし、μ_leftはτ₀［秒］〜τ［秒］間のＺｅｒｏ Ｃｒｏｓｓｉｎｇ Ｄａｔａの平均値であり、μ_rightは時刻τ［秒］〜τ_end［秒］間のＺｅｒｏ Ｃｒｏｓｓｉｎｇ Ｄａｔａの平均値である。ここでＳｄ（τ）を以下に従って求める。

ただし、ｓ_leftはτ₀［秒］〜τ［秒］間のＺｅｒｏ Ｃｒｏｓｓｉｎｇ Ｄａｔａの標準偏差であり、ｓ_rightはτ［秒］〜τ_end［秒］間のＺｅｒｏ Ｃｒｏｓｓｉｎｇ Ｄａｔａの標準偏差である。また、Ｎ_leftはτ₀［秒］〜τ［秒］間のＺｅｒｏ Ｃｒｏｓｓｉｎｇ Ｄａｔａの数であり、Ｎ_rightはτ［秒］〜τ_end［秒］間のＺｅｒｏ Ｃｒｏｓｓｉｎｇ Ｄａｔａの数である。いま、１０秒に一度の間隔でＺｅｒｏ Ｃｒｏｓｓｉｎｇ Ｄａｔａを取得するとした場合、Ｎ_leftは（τ？τ₀）／１０となる。 [Method by t test]
First, within a specified range (from time τ ₀ [second] to time τ _end [second]), a t-test value ttest (τ) at time τ [second] is calculated as follows.

Here, μ _left is an average value of Zero Crossing Data between τ ₀ [seconds] to τ [seconds], and μ _right is an average value of Zero Crossing Data between times τ [seconds] to τ _end [seconds]. . Here, Sd (τ) is obtained according to the following.

Here, s _left is the standard deviation of Zero Crossing Data between τ ₀ [seconds] and τ [seconds], and s _right is the standard deviation of Zero Crossing Data between τ [seconds] and τ _end [seconds]. N _left is the number of Zero Crossing Data between τ ₀ [sec] and τ [sec], and N _right is the number of Zero Crossing Data between τ [sec] and τ _end [sec]. Now, if Zero Crossing Data is acquired at an interval of once every 10 seconds, N _left is (τ? Τ ₀ ) / 10.

The ttest (τ) calculated by the above calculation method is a value indicating how much the behavior is different before the time τ and after the time τ. Therefore, ttest (τ) is calculated from τ ₀ to τ _end , and the place where ttest is maximized is the time when the behavior changes most in the interval τ _{0 to} τ _end .

ただし、Ｓ_leftはτ₀［秒］〜τ［秒］間の累積分布関数を示し、Ｓ_rightは時刻τ［秒］〜τ_end［秒］間の累積分布関数を示す。そして、図７は、は図４のデータの累積分布関数を示している。ｋｓ（τ）は、それぞれの累積分布関数の差の最大値であり、図７で矢印で示した大きさにあたる。この差ｋｓ（τ）を以下の式に当てはめ、ｋｓ検定値Ｐ（τ）を計算する。

ただし、Ｎ_leftはτ₀［秒］〜τ［秒］間のＺｅｒｏ Ｃｒｏｓｓｉｎｇ Ｄａｔａの数であり、Ｎ_rightはτ［秒］〜τ_end［秒］間のＺｅｒｏ Ｃｒｏｓｓｉｎｇ Ｄａｔａの数であり、Ｑ_ks（λ）は次のように定義する。

上記の計算方法で計算したＰ（τ）は、時刻τ以前と時刻τ以降で、どれだけ行動に差があったかを示す値である。したがって、Ｐ（τ）をτ₀〜τ_endまで計算し、Ｐ（τ）が最大となるところが、区間τ₀〜τ_endで最も行動が変化した時点となる。 [Method by ks test]
First, within the specified range (from time τ ₀ [second] to time τ _end [second]), the ks statistic ks (τ) at time τ [second] is calculated as follows.

Here, S _left indicates a cumulative distribution function between τ ₀ [second] and τ [second], and S _right indicates a cumulative distribution function between time τ [second] and τ _end [second]. FIG. 7 shows the cumulative distribution function of the data of FIG. ks (τ) is the maximum value of the difference between the respective cumulative distribution functions and corresponds to the size indicated by the arrow in FIG. This difference ks (τ) is applied to the following equation to calculate a ks test value P (τ).

Here, N _left is the number of Zero Crossing Data between τ ₀ [sec] and τ [sec], N _right is the number of Zero Crossing Data between τ [sec] and τ _end [sec], and Q _ks (Λ) is defined as follows.

P (τ) calculated by the above calculation method is a value indicating how much the behavior is different before and after time τ. Therefore, P (τ) is calculated from τ ₀ to τ _end , and the point where P (τ) is maximized is the time when the behavior changes most in the interval τ _{0 to} τ _end .

The value of ttest (τ) is plotted by the above method with respect to the activity level (solid line 501) in FIG. 4 is a dotted line 502 in FIG. 4, and the value of P (τ) is plotted in FIG. 4 is a broken line 503. From this, it can be seen that P (τ) and ttest (τ) are maximized at time t ₁ when the subject moves from walking to work, regardless of which calculation method is used. By taking the above method, the time t ₁ when the behavior changes can be calculated.

  Even if the method using the t test or the method using the ks test is used, a point where the behavior has changed can be found with high accuracy, so that there is no significant difference in the results. However, the t-test method can be obtained with the same degree of accuracy even if the value to be compared has a large variation. However, if the method to be compared varies, the accuracy varies depending on the variation. The t-test is considered more suitable when it varies widely from “0” to “100”, such as body movement, and the ks test when concentrated around 1 second, such as rr-interval. Is considered appropriate and should be used according to the situation.

  In addition, although FIG. 4 picked up the time of changing from one action to another action only once, a person's action changes many times a day. Two methods for detecting the time of such change will be described with reference to FIGS. First, the first method is a method that can be used to know a point that has changed over time, and is effective when it is mounted on a portable device or the like that is effective in real time, and will be described with reference to FIG. The second method is a method of searching for the time points where the behavior changes in descending order, and is effective when high accuracy is required, and will be described with reference to FIG.

[First method]
FIG. 5A shows an example of behavior in which the subject starts sleeping and continues to act with low activity for a while after getting up at time t ₂ and then starts walking for movement at time t ₃ . As described above, a method for detecting a change in behavior when the behavior changes twice will be described below.

The point in time when the behavior has changed within 30 minutes from the beginning of the data is examined by the same method as in FIG. FIG. 5B shows the body movement data for 30 minutes from the top of FIG. 5A by a solid line and the ttest (τ) value by a dotted line. Then, ttest (τ) takes the maximum value at time t ₂ , and it can be understood that it is the time when the first action change occurred.

Next, starting from time t ₂ , the time point when the behavior has changed within 30 minutes is similarly examined. FIG. 5C shows the body movement data for 30 minutes from time t ₂ in FIG. 5A by a solid line and the ttest (τ) value by a dotted line. Similarly, at time t ₃ , the ttest (τ) value shows the maximum value, and it can be seen that this is the time when the second action change occurred. A similar technique can be executed with data of a longer time and can be calculated in real time.

  In the present embodiment, the time point when the behavior changes is searched for within 30 minutes, but it is not necessary to limit to 30 minutes. If you need to see more detailed changes in behavior, you can search in less than 30 minutes. Conversely, if you want to see rough changes in behavior, you can search in more than 30 minutes. You may modify it accordingly.

In the description of FIGS. 5A to 5C, it is determined whether or not there is a change in behavior at the time when the ttest (τ) value becomes maximum. It may be a criterion for judging whether or not there is a change in behavior at the maximum value. ttest (τ) indicates a difference in behavior before time τ and after time τ, and even if it is a peak, the behavior does not necessarily change. For example, time t ₂ is the time when the behavior changes from sleep to rest and the difference in behavior is small, but time t ₃ is the time when the behavior changes from rest to walking, The difference is big. Each peak value is about 30 at time t ₂ and about 90 at time t ₃ , indicating a value proportional to the difference in behavior.

  Therefore, if it is not necessary to distinguish between sleep and rest, it is assumed that there is a change in behavior for the first time when the peak value is, for example, 60 or more. You can see changes.

[Second method]
FIG. 6A is the same data as FIG. 5A, and the value of ttest (τ) is calculated based on the values from the beginning to the end of the data and is indicated by a dotted line. Then, ttest (τ) has taken the largest value at time t _3, the body movement shown in FIG. 6 (a) has changed most markedly at time t _3. Here, two sections at time t ₃ (section A: top ~t _3, section B: t ₃ ~ last) is divided into. Ttest (τ) is calculated from the data of section A having a time of 30 minutes or more out of the two divided sections, and the next time t ₂ when the change is noticeable is the next time the action changes. And By repeating the method as described above, it is possible to detect a change in behavior when the behavior changes twice or more. In addition, the above method specifies the time of change in the order in which the change of behavior appears prominently, so it is possible to know the time when the behavior has changed with high accuracy.

  In the present embodiment, ttest (τ) is used for a section of 30 minutes or longer, but it is not necessary to limit to 30 minutes. If you need to see more detailed changes in behavior, you can search in less than 30 minutes. Conversely, if you want to see rough changes in behavior, you can search in more than 30 hours. You can correct it.

Moreover, it is good also as a judgment criterion of the area which divides | segments the maximum value of ttest ((tau)) instead of using the judgment time of the area which divides | segments the data continuation time. For example, if it is not necessary to distinguish between sleep and rest, if the method of repeating the division until the peak value becomes 60 or less, for example, it is determined that the division is not divided in FIG. A measure of only time t ₃ can also be taken.

  In addition, although both methods were demonstrated by the method by t test, the same calculation is possible by the method by ks test.

  FIG. 8 shows the physical activity of a subject wearing the product of the present invention. FIG. 8A is the same as the body movement data shown in FIGS. 5 and 6 and shows the activity on the vertical axis and the time on the horizontal axis. In FIG. 8B, the vertical axis represents the rr-interval of the heartbeat, and the horizontal axis represents time on the same scale as FIG. 8A. rr-interval means that the larger the value, the smaller the heart rate that beats per unit time.

  The times indicated by broken lines in FIGS. 8A and 8B are times divided using the first method described in FIGS. 5A to 5C. At this divided time, the subject is changing his / her behavior. For example, time t82 indicates the time when the exercise changes to rest. Comparing FIG. 8 (a) and FIG. 8 (b), the interval from time t81 to t82 changes from the active state to the exercise state, and the activity activity is very high. Accordingly, rr-interval is The value is lower than the time. This indicates that the heart rate has increased due to exercise, and when evaluating the subject's rr-interval, it is necessary to perform the evaluation on the premise that the subject is exercising.

  On the other hand, each of the sections from time t83 to t84, time t85 to t86, and time t87 is almost the same level, and the activity level is low, indicating that the subject is resting. However, when looking at the change in rr-interval, only the interval from time t85 to t86 is lower than the surrounding values, indicating a decrease in average during exercise. Since this may represent any sign of heart disease, the subject's rr-interval is evaluated assuming that it is at rest. However, since it is not possible to determine whether the disease is caused by the activity level or heart rate alone, the situation where the subject is placed is entered at the same time.

  In the present embodiment, as an example, when the activity of the calculated section is an average of “15” or less and the rr-interval is less than “700”, what is being done in the section is input by the mobile phone. . For example, if the rr-interval is low as a result of remembering forgetfulness and irritability, there is no problem. However, if there is no psychological change, but the rr-interval is low, there is no heart disease. there is a possibility. In addition, if the rr-interval is low due to being in a train or a crowd, there is a high possibility of a psychosomatic disorder such as a panic disorder.

  Specifically, the operation of the biological analysis apparatus 120 will be described using the operation flow of FIG. First, heart rate data is acquired from the heart rate data acquisition device 100, and acceleration data is acquired from the acceleration data acquisition device 110 (step S100). Subsequently, based on the acquired data, the biological analysis circuit 126 calculates the activity level and rr-interval of the subject (step S102). Here, any of the above-described methods is used as a method of calculating the activity level and rr-interval.

  Subsequently, the condition of the subject is evaluated. Specifically, first, it is determined whether or not the activity level calculated in step S102 is an average “15” or less (step S104). Here, when the average of activity is “15” or less (step S104; Yes), it is subsequently determined whether or not rr-interval is “700” or less (step S106). When the average of rr-interval is “700” or less (step S106; Yes), a screen for inputting the state of the subject is displayed on the display device 128, and the state of the subject is input (step S108). The input state of the subject, the calculated activity, and rr-interval are stored in the memory 124 in association with each other.

  The user of the biological analysis apparatus 10 wears the product of the present invention according to FIG. The subject's body movement (activity) and rr-interval are periodically sent to the mobile phone 303, and each time the mobile phone 303 receives data, it analyzes the time series data by the method described above, and the rr-interval Divide time series data. If there is a possibility of abnormality for each of the divided rr-interval data (for example, the section from time t85 to time 86 in FIG. 8A), what was being done at that time and what kind of place Asked questions such as whether they were in the hospital or what psychological state they were in. FIG. 9 is an example of a question asking what was being done, and is selected from “meal”, “exercise”, “movement by vehicle”, and “others”. If, for example, “movement by vehicle” is selected and the rr-interval shows a low value even though the user is resting, there is a possibility of motion sickness or panic disorder. Further, by repeating the question and repeating the analysis of rr-iterval and body movement, it becomes possible to make a detailed diagnosis with higher accuracy. The diagnosis result obtained here can be used later in the medical institution for diagnosis, and can immediately issue a warning to the medical institution on the spot. It is also possible to perform self-treatment (cognitive behavioral therapy) on the spot.

[Modification]
In this embodiment, the time of rr-interval is divided using the activity level of body movement, and the change of rr-interval in daily life is made possible to detect signs and diagnose the medical condition. It is not necessary to limit the degree of activity as biometric information for viewing, and it is also possible to evaluate from a psychological change by using, for example, an electroencephalogram.

  In addition, the bioanalytical apparatus 10 according to the present embodiment can be used not only for finding a sign or diagnosing a medical condition, but also as a memorandum of behavior. For example, it can be determined later how long the presentation has been given based on the time when the behavior or psychology changes from body movement or heartbeat.

  In addition to body movements, heartbeats (or rr-intervals), and electroencephalograms, changes in behavior and psychology can be applied as long as the body features change in a short time. For example, the movement of the eyeball, the number of blinks, the body surface conductivity, the amount of sweat, the body temperature, etc. are closely related to psychological changes and actions, and it is considered that similar effects can be obtained.

It is a figure which shows the block diagram of the biometric information apparatus in this embodiment. It is a figure which shows the external appearance of the mobile telephone incorporating the biological information apparatus in this embodiment. It is a figure which shows the example of mounting | wearing of the biometric information apparatus in this embodiment. It is a figure for demonstrating the method of dividing | segmenting an activity degree into a time series. It is a figure for demonstrating the method of dividing | segmenting an activity degree into a time series. It is a figure for demonstrating the method of dividing | segmenting an activity degree into a time series. It is a figure for demonstrating a statistical method. It is a figure for demonstrating the case where an activity degree and a heart rate are divided | segmented into the time series. It is an operation | movement flow for demonstrating the process of a biometric information analyzer. It is a figure which shows the example of a screen display in this embodiment. It is a figure for demonstrating an activity level and a heart rate.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Biological information apparatus 100 Heartbeat data acquisition apparatus 102 Heartbeat sensor 104 Transmission circuit 110 Acceleration data acquisition apparatus 112 Acceleration sensor 114 Transmission circuit 120 Biological information analysis apparatus 122 Reception circuit 124 Memory 126 Biological analysis circuit 128 Display apparatus

Claims (6)

Biological information acquisition means for acquiring biological information of the subject in time series;
Biological analysis means for analyzing biological information acquired in time series by the biological information acquisition means;
Based on the result analyzed by the biological analysis means, evaluation means for evaluating the state of the subject,
In a biological analysis apparatus equipped with
The biological analysis unit divides the time series data acquired by the biological information acquisition unit into a time series data of the living body before and after a certain time τ in the data, and at each τ, before τ. Calculating a statistical difference between the data and the data after τ, dividing the data at a time when the difference is maximum, and analyzing the biological information;
The biological analysis apparatus characterized in that the evaluation means is means for evaluating time-series data of biological information according to each time divided by the time division analysis means.   The biological analysis apparatus according to claim 1, wherein the biological information is an activity level of body movement or a heartbeat. It further comprises input means for inputting the subject's behavior or psychological state,
The biological analysis apparatus according to claim 1 or 2 , wherein the behavior or psychological state of the subject input by the input unit and the evaluation evaluated by the evaluation unit are stored in association with each other. As the statistical method, the biological analysis device according to any one of claims 1, characterized by using the t-test 3. Wherein a statistical technique, biological analysis device according to any of claims 1 3, which comprises using the chi ² test. On the computer,
A biological information acquisition function for acquiring biological information of the subject in time series,
A biological analysis function for analyzing biological information acquired in time series by the biological information acquisition function;
Based on the result analyzed by the biological analysis function, an evaluation function for evaluating the state of the subject,
Is a program that realizes
The biological analysis function divides the time series data acquired by the biological information acquisition function into a time series data of a living body before and after a certain time τ in the data, and at each τ, before τ Calculates the statistical difference between the data and the data after τ, and has a time division analysis function for analyzing the biological information by dividing the data at the time when the difference is maximum ,
The evaluation function causes a function to evaluate time-series data of biological information according to each time divided by the time division analysis function.

Images