JP4865046B2 - Information processing device - Google Patents

Description

  The present invention relates to an information processing apparatus, and more particularly, the daily life activity level of a subject, particularly an elderly person, is accumulated in the most natural form while ensuring privacy, and the life activity level of the subject is evaluated, and nursing / nursing The present invention relates to an information processing apparatus that can be used for the activities of medical and welfare institutions and can quickly and easily determine the physical condition of a subject.

  In Japan, it is said that the proportion of elderly people over the age of 65 will exceed 30% over the next 50 years. Under such circumstances, measures for elderly people requiring care are currently attracting attention and consideration, but on the other hand, it should not be overlooked that there are many healthy elderly people. Among healthy elderly people, those who live alone are often isolated in terms of space and relationships, and it is very important to understand the health status of these people, but a large number of people There is an overwhelming shortage of human resources to understand and manage the health status of people.

  For this reason, for example, some local governments have introduced a pendant type wireless emergency call system, but this is only used in an emergency and does not grasp the daily health of the elderly. Absent. Service systems that monitor the activities of elderly people by placing various sensors in the home are provided by security companies, etc., but they are popular only for high-income people due to the high cost of system installation and maintenance. is the current situation. Furthermore, Patent Document 1 proposes a system that attaches a sensor to a household electrical appliance and monitors on / off of a hot water supply switch of the electric pot to monitor the safety of a consumer. There is a limit to the location and the amount of information, and in addition, this system cannot be adequately monitored by people who rarely use electrical appliances or who unplug electrical appliances from the outlet at night.

Japanese Patent Laid-Open No. 10-248093

  The present invention has been made in view of such a conventional problem, and can be accumulated in the most natural form while ensuring privacy with respect to the daily life activity of a subject such as an elderly person at a low introduction cost. An object of the present invention is to provide an information processing apparatus that can evaluate a subject's life activity level and can quickly and easily determine the physical condition of the subject.

  According to the present invention, in the information processing apparatus that evaluates the life activity level of the subject from the operation of the home electric device, the analysis evaluation unit that analyzes the operation time information of the predetermined home electric device includes the analysis evaluation unit. Is obtained by analyzing the accumulated operation time information by calculating an aggregate value for each predetermined time interval from the operation time information for one day, and calculating an entropy value of the set of the aggregate values. And a storage unit storing the reference entropy value, and an evaluation unit that compares the entropy value analyzed by the analysis unit with the reference entropy value to evaluate the living activity of the subject. An apparatus is provided.

  Preferably, the information processing apparatus further includes a transmission unit that transmits an evaluation result of the life activity level of the subject by the analysis evaluation unit, and transmits the evaluation result to a care / nursing / medical / welfare organization.

  With the information processing apparatus of the present invention, it is possible to accumulate the daily life activity level of a subject such as an elderly person in the most natural form at a low introduction cost while ensuring privacy, and also accurately evaluate the life activity level of the subject. In addition, the physical condition of the subject can be determined quickly and easily.

It is a schematic block diagram of the life activity level evaluation system of this invention. It is a top view of the house which arrange | positioned various sensors. It is a time chart which shows the use condition of each household appliances. It is a schematic diagram of the analysis evaluation means in this invention. It is a figure which shows the acceleration change at the time of a walk. It is a chaos attractor figure drawn in the two-dimensional number space from the acceleration change at the time of walking. It is a figure which shows the change of an attractor orbital period. It is a figure which shows the change of an attractor orbital period. It is a figure which shows the result of having analyzed the fluctuation of the attractor orbital period by DFA. It is a figure which shows the result of having analyzed the fluctuation of the attractor orbital period by DFA. It is an example of the time series data about the door opening and closing of a refrigerator. It is a histogram which shows the opening / closing frequency | count of the door per hour. It is a figure which shows the example of a daily change of entropy. It is an example of a diary questionnaire. It is a figure which shows the relationship between a physical condition (appetite) and entropy. It is a figure which shows the relationship between a physical condition (body pain) and entropy.

  The present inventors provide a system that can accumulate daily life activities of subjects such as elderly people in the most natural form at a low introduction cost, and further evaluate the life activities of subjects, while ensuring privacy. As a result of earnest examination, we can detect the usage status of household appliances that are used in daily life without discomfort and are widely used, the usage status of gas and water, and the movement status of subjects inside and outside. -It has been found that it suffices to evaluate the living activity level of the subject by accumulating and analyzing such detection results, and has led to the present invention.

  That is, the major feature of the life activity level evaluation system of the present invention is that the detection signals obtained by the detection means are accumulated by the accumulation means, the signals are stored in time series, and the life of the subject is calculated from the accumulated time series data. The point is that the degree of activity is evaluated by an analytical evaluation means.

  FIG. 1 shows a schematic diagram of a daily activity level evaluation system of the present invention. Signals detected by the various detection sensors 1 disposed in the house or the detection sensor 1 carried by the subject are sent to the accumulating means 2, where they are accumulated and stored as time-series data. The time-series data stored in the accumulating unit 2 is sent to the analyzing / evaluating unit 3 every predetermined time or when a transmission request is made. The analysis evaluation means analyzes and evaluates the collected time series data. The evaluation result by the analysis / evaluation means 3 is sent to the subject himself / herself 5 and the care / nursing / medical / welfare institution 6 as necessary using the transmission means 4, and is used for self-management of health status and activities by the institution. In FIG. 1, it is assumed that the analysis evaluation unit 3 and the transmission unit 4 are mounted on the information processing apparatus. Each means will be described individually below.

  First, as the detection means, for example, home appliances such as an air conditioner, a television, a refrigerator, a microwave oven, and a mat, a conventionally known device such as a vibration sensor, an acceleration sensor, and an infrared sensor can be used, and a plurality of these are used in combination. It is desirable to detect the degree of daily activity in more detail. For example, if home appliances are used as detection means, it is possible to detect on / off of a switch, opening / closing of a door, temperature setting, operation mode, compressor speed, remote control operation, power consumption, entrance / exit, and the like. In addition, if a vibration sensor or acceleration sensor is attached to an indoor member where vibration is transmitted well by indoor movement such as home appliances, toilets, bathrooms, living rooms, bedrooms, Buddhist altars, doors, furniture, floors, etc. It is possible to detect whether vibration or acceleration is detected in minutes, that is, how a person behaves. Furthermore, if a vibration sensor, an acceleration sensor, or a meter is attached to a gas or water pipe, it is also possible to detect the usage status of the gas or water. Further, if an infrared sensor is used as a detection means, it can be detected that a person has entered a specific area. Further, if a portable pedometer (registered trademark) incorporating a vibration sensor, an acceleration sensor, a tilt sensor, or the like is used as a detection means, it is possible to grasp and record the living activity level of an outdoor subject.

  Detection signals detected by the various detection means are sent to the accumulation means, where they are stored as time series data. Although there is no particular limitation on the transmission method from the detection means to the accumulation means, it is preferable to perform transmission using radio or power line conveyance because no new wiring work is required.

  The time series data accumulated and stored in the accumulation means is transmitted to the analysis evaluation means. There is no particular limitation on the transmission method from the accumulation means to the analysis / evaluation means, and a telephone line, CATV line, or the like may be used. In FIG. 1, the accumulating means and the analyzing / evaluating means are separated from each other.

  Next, an analysis / evaluation method in the analysis evaluation means of the present invention will be described. In this method, mobility is calculated, and evaluation is performed based on this value. The mobility here is one index indicating the distance that the subject has moved in a certain period, for example, how much the subject has moved in one day. Hereinafter, this method will be described with reference to the drawings.

  FIG. 2 is a sketch of the house, and a door, a bedroom, a living room, a kitchen, and a bathroom are provided in series for ease of explanation. As detection sensors, entrance: entrance light / door mat, bedroom: bedroom light / bedroom air conditioner, living room: living light / TV / video / air conditioner, kitchen: kitchen light / refrigerator / microwave / rice cooker, bathroom: bathroom They are installed in electric lights and washing machines. For example, when these detection sensors react at least once an hour, the table is shown as “*” as shown in Table 1.

  By analyzing this table, it is possible to grasp the movement of the subject between the rooms. That is, according to Table 1, the subject wakes up around 7 o'clock, moves from the bedroom to the living room, and then moves to the kitchen. Therefore, if you move for one room, you get 1 point, if you move for 2 rooms, 2 points, if you move for 3 rooms, you get 4 points. become. If the average value of mobility of a very active healthy person is 30 points and the mobility of a bedridden person is 1 point, the movement value of 18 points is slightly higher than normal, and you can see your score. This makes it possible to recognize the relative degree of life activity. Also, the calculated movement value is 30 for more than 30 points, 5 for 24 points or more, 3 for 18 points or more, 2 for 12 points or more, 1 for 6 points or more, less than 6 points. May be evaluated in six stages such as “0”. If such a movement value is utilized, a test subject's willingness to act can be heightened so that a movement value may be higher, or it may become a stage one step above.

  Next, another analysis / evaluation method in the analysis evaluation means of the present invention will be described. As this analysis / evaluation method, for example, in the simplest case, if the detection means has not detected the subject's activity for a certain period of time or if the detection of the activity has become extremely small, an abnormality will occur in the subject. For example, a method for evaluating the safety of a subject can be considered.

  As another example, the daily life activity pattern of the subject is accumulated and the daily life activity pattern is calculated and stored in the storage unit as the reference data, and the detected time series data is the reference data. When a pattern different from the data is shown, there is a cluster analysis method for evaluating a subject's health condition as an abnormality has occurred. For example, create a time chart for each home appliance as shown in Fig. 3, how many times the refrigerator doors are opened and closed on average per day, how many hours on average the TV is turned on per day, and the light in the living room -Average data such as whether the light is extinguished is extracted and stored, and the detected data is compared with this average data to evaluate whether the subject's condition has changed. If the detected data is significantly different from the average data, for example, the refrigerator door was opened and closed on average 10 times a day but only opened a few times, or the TV is not turned on at all, and the light in the living room is turned off If not, it is evaluated that some abnormality has occurred in the subject, and the fact is notified to the outside.

  As an analysis / evaluation method that has advanced one step further, the analysis result associated with the daily activity level is stored in the storage unit as reference data, and the analysis result of the time series data is compared with the reference data to match or It is also possible to estimate and evaluate the life activity of the subject in more detail from the approximate reference data. A schematic diagram of this method is shown in FIG. The time series data sent from the accumulating means 2 to the analyzing / evaluating means 3 is first analyzed by the analyzing unit 31. A specific analysis method will be described later. And the analysis result linked | related with a test subject's life activity level is memorize | stored in the memory | storage part 32 as reference | standard data. On the other hand, the analysis result of the time series data sent sequentially is compared with the reference data, and the life activity level of the subject is estimated / evaluated from the reference data that matches or approximates, and is output by the output unit 34.

  First, a case where a nonlinear analysis method is used as a time series data analysis method will be described. Since information about a living body such as behavior is generally nonlinear data, a nonlinear analysis method can be used as an analysis method. Among them, chaos analysis widely used as a nonlinear data analysis technique is most preferably used as the time-series data analysis method. This chaos analysis includes various analysis methods such as fractal dimension analysis, Lyapunov exponent analysis, and chaos attractor analysis, and chaotic attractor analysis is suitable as an analysis method related to biological information. Among them, a method of measuring the orbital period of the chaotic attractor and analyzing the degree of daily activity from the fluctuation of the orbital period is preferable.

  This analysis method will be specifically described below. FIG. 5 shows the acceleration data of walking by the subject actually measured. FIG. 5 is a diagram showing the time-dependent change in acceleration by walking with the horizontal axis representing time and the vertical axis representing acceleration. FIG. 5A shows acceleration data during health, and FIG. 5B shows when physical condition is poor. Acceleration data. As can be understood from these figures, the difference is often not clearly understood even if the waveforms of acceleration data are compared.

  Next, FIG. 6 shows a chaos attractor drawn in a two-dimensional number space based on this data. FIG. 6A shows a chaos attractor when healthy, and FIG. 6B shows a chaos attractor when physical condition is poor. If the pattern matching method can be used to evaluate the shape of the chaos attractor, the evaluation may be performed at this stage, but the difference between the two may be unclear as in this figure. In some cases, the pattern matching method cannot be applied.

  Therefore, the average value of the constituent elements of the chaotic attractor shown in FIG. 6 was obtained, and the orbital period of the chaotic attractor was measured by converting the chaotic attractor into the polar coordinate with the average value coordinate point as the origin. Specifically, a horizontal line S is drawn in the right direction from the point O of the chaotic attractor, and the time until the attractor trajectory rotates 360 ° counterclockwise and returns to the reference line is defined as the trajectory period. did.

  The measurement results of the attractor orbit period are shown in FIG. FIG. 7A shows the attractor trajectory period when healthy, and FIG. 7B shows the attractor trajectory period when physical condition is poor. Although acceleration data and chaos attractor were not shown, FIG. 8 shows the orbital period of the chaos attractor on the day before the subject loses physical condition. As can be seen by comparing (a), (b) and FIG. 8 in FIG. 7, a difference in walking data due to a difference in physical condition, which was not understood by the chaotic attractor pattern matching method, can be found from fluctuations in the orbital period of the attractor. It is. Therefore, if the fluctuation of the attractor trajectory period in various states is stored in the storage unit as the reference data, the fluctuation characteristics of the reference data and the measured fluctuation characteristics are sequentially compared to detect the coincidence. It is possible to determine the state, the above-described example, health, a sign of poor physical condition, and poor physical condition.

  As a method of analyzing the fluctuation of the attractor trajectory, conventionally known analysis methods such as detrended fluctuation analysis (DFA), frequency conversion such as Fourier transform, wavelet analysis, and multifractal analysis can be used. However, DFA is desirable because it can be analyzed objectively and accurately.

  The outline of the analysis method by DFA is as follows. First, the fluctuation system is divided by a predetermined window size, and the waveform is linearly approximated for each window. Then, the absolute value of the deviation from the linear approximation is integrated, and this integrated value is set as the magnitude of fluctuation. The vertical axis is the magnitude of fluctuation, the vertical axis is the window size, the magnitude of fluctuation with respect to each window size is plotted, and the slope and y-intercept are used as indicators for state determination / prediction.

  FIG. 9 shows the result of analyzing the fluctuation of the attractor trajectory period shown in FIGS. 7A and 7B by DFA. In FIG. 9, the horizontal axis represents the window size and the vertical axis represents the fluctuation, and the healthy time and the poor physical condition are indicated by a solid line and a broken line, respectively. The broken line at the time of poor physical condition has a smaller slope than the solid line at the time of health. Therefore, in this case, the health and poor physical condition can be evaluated mainly from the change in the slope of the line.

  FIG. 10 shows the result of analyzing the fluctuation of the attractor trajectory period by DFA using the vibration data during walking as time series data. The vibration during walking was measured using a vibration switch (manufactured by Koshinsha). Further, since the output of the vibration switch is a digital signal of “1” and “0”, the signals are averaged and smoothed to draw a chaotic attractor. Similar to FIG. 9, FIG. 10 shows the healthy time and the poor physical condition with a solid line and a broken line, respectively, with the horizontal axis representing the window size and the vertical axis representing the fluctuation. Since the solid line at the time of health and the broken line at the time of poor physical condition are different from each other in y-intercept and inclination, it is possible to determine health and poor physical condition from this difference.

  Next, a case where entropy calculated from the usage frequency of home appliances every predetermined time is used as a method for analyzing the degree of daily activity will be described. Entropy is a value that quantifies the degree of variation of samples in a statistical distribution, and here indicates whether home appliances are used diffusely or intensively. The inventors of the present invention installed detection means on various home appliances in the homes of 100 subjects (single persons), and investigated the usage status for one month. As a result, we obtained new knowledge that there is a close correlation between entropy about the use of home appliances and the physical condition of the subject, and found that the entropy can be used as one index representing the health condition of the subject. Hereinafter, the case where a refrigerator is used as a home appliance will be described as an example.

  A door open / close detector is provided in the refrigerator, and when the door of the refrigerator is opened / closed, the date / time of opening / closing and an “open” or “closed” signal are sent to the analysis unit 31 (shown in FIG. 4). The door open / close detector and the analysis unit are connected by, for example, a power line, and data is sent from the door open / close detector to the analysis unit based on a network constructed on the power line. FIG. 11 shows an example of time-series data sent to the analysis unit. In FIG. 11, for example, the first line shows that the refrigerator door was opened at 6: 9: 14 on July 31, 2000, and the next line at 6: 9: 18 Indicates that the door has been closed.

Next, FIG. 12 shows a histogram obtained by integrating the number of times the refrigerator door is opened and closed at predetermined time intervals. FIG. 12 shows the cumulative number of doors opened and closed every hour, with the horizontal axis representing time and the vertical axis representing the cumulative number of doors opened and closed. The entropy about the opening and closing of the door is obtained from the door opening and closing frequency data. Entropy E is calculated from the following equation, where n _i (i = 1, 2,..., 23) is the number of times the door is opened and closed at each time, and N is the total number of times the door is opened and closed per day. In the entropy calculation, in the above example, the sum of the number of times of opening and closing every hour is taken, but it goes without saying that various time intervals such as 1 minute, 15 minutes, and 3 hours may be taken.

  An example of the change in entropy for each day calculated in this way is shown in FIG. FIG. 13 is a diagram showing changes in entropy for each day, with the vertical axis representing entropy and the horizontal axis representing date, and it can be seen that the entropy changes daily.

  On the other hand, the daily physical condition of the subject during the survey period was simultaneously investigated using a diary questionnaire shown in FIG. 14, and the relationship between entropy and the physical condition of the subject was examined. Table 2 shows an example in which the entropy of each day is divided and entered for each physical condition (good, normal, bad) of the subject obtained from the questionnaire, and FIG. 15 shows a graph of this. In FIG. 15, the upper and lower sides of a quadrilateral with triangular notches on the left and right sides of the center indicate values of 25% from the top and 25% from the bottom of the mother data.

  It can be seen from Table 2 and FIG. 15 that the entropy is small when the appetite is good and normal, and the entropy tends to be large when there is no appetite (bad). Therefore, based on the data in Table 2, the correlation between entropy and physical condition was verified by hypothesis testing means in statistics. Table 3 shows the results. The values in Table 3 are values generally indicating probabilities called “p-values”, and were actually obtained when it was assumed that there was no correlation between the subject's physical condition such as facial color and mood and entropy. This is the probability of obtaining a survey result. That is, when the p-value is sufficiently small, the hypothesis that there is no correlation is rejected, and both are correlated. Therefore, the smaller the value in Table 3, the stronger the correlation between the subject's physical condition and entropy.

According to Table 3, it can be seen that there is a strong correlation between entropy about the frequency of opening and closing the refrigerator door, appetite, and sleep. For example, when the time interval is 15 minutes, an increase in entropy and a decrease in appetite have a p-value of 5.67 × 10 ⁻⁶ , indicating a high correlation. Similarly, an increase in entropy and a decrease in sleep have a p-value of 4.79 × 10 ⁻⁹ , indicating that there is an extremely high correlation. Table 3 (the same applies to Tables 4, 6, and 7) shows the results of analysis of p-values for the entire category such as “good”, “normal”, and “bad”, but “normal” and “bad” Some correlations show a strong correlation. For example, there is a strong correlation between the increase in entropy when the time interval is 1 minute, 15 minutes, and 1 hour and the relationship between appetite and sleep, and the increase between entropy and sleep when the time interval is 3 hours. Show.

  Moreover, when using television as a household appliance, it turned out that there is a correlation between entropy and a specific physical condition like the case where a refrigerator is used. Instead of the number of times of opening and closing the door in the case of a refrigerator, the number of signal transmissions from a remote controller (hereinafter sometimes referred to as “remote control”) is used in the case of a television. That is, the number of transmissions of signals sent from the remote controller to the television such as power on / off, channel switching, volume increase / decrease was used. Calculation of entropy and statistical processing are the same as in the case of the refrigerator, and are omitted here. FIG. 16 shows the correlation between the obtained entropy and physical condition (body pain). Table 4 shows the results of verifying the correlation between entropy and physical condition by hypothesis testing means in statistics.

According to Table 4, when the time interval is 1 minute, the p value between the entropy and the pain with respect to the number of signal transmissions of the television is 4.93 × 10 ⁻⁷ , indicating that there is a high correlation.

  In addition, the daily entropy is accumulated for a predetermined period, and the average during that period is obtained and compared with the standard entropy. It can also be grasped. As an example, Table 5 shows the results of conducting a “POMS test” for 78 subjects after a 29-day experiment and examining the relationship between two physical condition items of “depression” and “fatigue” and the entropy of the number of times the refrigerator door is opened and closed. .

  According to Table 5, regarding “depression” by “POMS test”, 43 of the subjects were less likely and 34 were more likely. The average entropy of subjects with a low probability of “depression” was 5.32, while the average entropy of subjects with a high probability of “depression” was 5.57. That is, if the average entropy is high, there is a high possibility that the state is “depressed”. Therefore, when the correlation between entropy and “depression” was verified by the statistical processing, the p value was as small as 0.0152, and it was recognized that there was a correlation between the two.

  Regarding “fatigue”, it was found that 44 of the subjects were less likely to be fatigued and 33 were more likely to be fatigued by the “POMS test”. The average entropy of subjects who are less likely to be fatigued was 5.33, while the average entropy of subjects who were likely to be fatigued was 5.55. From this, it is considered that if the average entropy is high, the possibility of fatigue is high. As described above, when the correlation between entropy and “fatigue” was verified by statistical processing, the p value was 0.0344, which was a relatively small value, and it was recognized that there was a correlation between the two.

  As is clear from the above, the entropy of each subject in the normal state (normal) is calculated in advance and stored as reference entropy. From the detection data sent every day or the detection data within a predetermined period, The physical condition of the subject can be determined by comparing the calculated entropy with the reference entropy. In addition, although the refrigerator and the television are used as the home appliances in the embodiment, the home appliances are not limited to these, and conventionally known home appliances can be used. In addition, by calculating a plurality of entropies from the frequency of use of a plurality of home appliances, the physical condition of the subject can be determined more specifically.

  Next, as a method for analyzing the degree of life activity, a case will be described in which at least one time of the first use time, the last use time, and the average use time of the home appliance is used in one day. Similar to the case where entropy is used as an analysis method, the detection means are arranged in various home appliances in the home of 100 subjects (single person), and the usage situation is investigated over one month. A new finding that there is a close correlation between the time and the physical condition of the subject was obtained, and it was found that this use time can be used as one index representing the health condition of the subject. Hereinafter, a case where a refrigerator and a television are used as home appliances will be described as an example.

  The refrigerator is provided with a door open / close detector, and when the refrigerator door is opened and closed, the time when the door is opened and closed and an “open” or “closed” signal are detected. In the case of a television, a signal transmission detector is provided on the remote control to detect the time when the remote control transmits a signal to the television. And the time when the signal transmission of the refrigerator door and the remote control was first activated in the day, the time when it was last activated, and the average time T calculated from the following formula were recorded, respectively, as shown in FIG. We investigated the relationship between the physical condition of the subjects surveyed in the diary questionnaire and the detected times. The results are shown in Table 6 and Table 7. Tables 6 and 7 show the p values calculated by statistical processing, as in Table 3.

  From Table 6, it can be seen that, for example, the time when the refrigerator door is first opened, the time when the refrigerator door is closed last, and the average time have a strong correlation with the appetite. Also from Table 7, it can be seen that there is a strong correlation between appetite and the time at which the remote controller of the television first transmitted the signal, the time at which the signal was last transmitted, and the average time, for example. Therefore, the physical condition of each subject can be determined by storing each reference time in the normal state (normal) of each subject and comparing the time detected every day or the average time within a predetermined period with the reference time. . In addition, although the refrigerator and the television are used as the home appliances in the embodiment, the home appliances are not limited to these, and conventionally known home appliances can be used. Moreover, the physical condition of the subject can be determined more specifically by calculating use times of a plurality of home appliances.

  Specifically, when entropy or the above-mentioned use time is used as the analysis method of the life activity level evaluation system of the present invention, in FIG. 4, when each subject is in the normal state (normal) from the detection data sent every day. If necessary, the entropy or use time is calculated by the analysis unit 31 and stored in the storage unit 32, and the entropy or use time calculated by the analysis unit 32 from the sent detection data is used as the reference entropy or the storage time of the storage unit 32. The physical condition of the subject is evaluated by comparing the reference time with the evaluation unit 33.

  The life activity level of the subject evaluated in this way by the analysis / evaluation means is transmitted to the subject himself / herself and to a care / nursing / medical / welfare institution through a communication means to improve the daily life of the subject and the activity of the institution. It is preferable to make use of it. In this case, the evaluated life activity level may be transmitted to the requester via the communication means in response to a request from the subject himself / herself, his / her family, the institution or the like.

DESCRIPTION OF SYMBOLS 3 Analysis evaluation means 4 Transmission means 31 Analysis part 32 Storage part 33 Evaluation part

Claims (2)

A detection means for detecting a subject's activity; an accumulation means for receiving and storing a detection signal from the detection means; and an analysis evaluation means for analyzing the detection signal stored in the accumulation means to evaluate the physical condition of the subject. And
The detection means is a household electrical appliance;
The analysis evaluation means includes
The number of predetermined operations of the household electrical appliance for each predetermined time interval is calculated from time-series data indicating the predetermined operation of the household electrical appliance for one day, and the entropy value is calculated based on the following formula 1 An analysis unit for calculating
A storage unit for storing a reference entropy value obtained by analyzing a detection signal accumulated in advance ;
An information processing apparatus comprising: an evaluation unit that evaluates the physical condition of a subject by comparing the entropy value analyzed by the analysis unit and the reference entropy value.

However,
E: Entropy
n _i (i = 1, 2,...): The number of predetermined operations of the household electrical appliance at each predetermined time interval
N: Total number of predetermined operations of household electrical appliances per day
It is. The information processing apparatus according to claim 1, further comprising: a transmission unit that transmits the evaluation result of the physical condition of the subject by the analysis evaluation unit, and transmits the evaluation result to a care / nursing / medical / welfare organization.

Images