JP2010148829A - Computer system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a system for easily visualizing the sleeping state of a user. <P>SOLUTION: A sleep monitor system includes a sensor mounted on a human and outputting an acceleration; and a computer for detecting the change of direction of a human body, on the basis of the output acceleration. The system generates an image for evaluating the sleeping state of the user. The computer obtains acceleration information, detects a bedtime, a wake-up time, and roll-over times during the sleep so as to calculate intervals between the respective roll-over times. Besides, the calculated roll-over intervals are time-sequentially arrayed, and the lengths of the roll-over intervals are indicated by shading, so that the image is generated, where the depth of the sleep during sleeping is visualized. Transition of quality in daily sleeping is presented to the user with the rate of the roll-over intervals which are equal to or longer than a predetermined length to the whole sleeping time as an index. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a computer system, and more particularly to a system for evaluating a sleep state by detecting a change in body orientation.

  Sleep is an essential action for a person to survive. There is a great need for better quality sleep to maintain mental and physical health. Many people also suffer from sleep disorders such as insomnia. As a method for measuring the quality of sleep, a method of measuring REM sleep and NREM sleep based on changes in electroencephalogram or eye movement during sleep and analyzing the appearance pattern is common. However, this measurement requires a highly restrictive and large-scale device (for example, polysomnograph). Moreover, it is difficult for non-experts to analyze the measurement results.

  One way to easily estimate the quality of sleep every day at home is to focus on turning over. Rolling over occurs to relieve pain and poor circulation in body parts that are compressed by weight by changing body positions during sleep. Therefore, it is considered that sleep is shallow due to pain during the time before and after turning over. That is, when the frequency of turning over is high, it is suggested that sleep is shallow overall and sleep is not well performed. In addition, when a person is on the floor due to a sleep disorder but remains awake without being able to sleep, the person turns over more frequently.

Therefore, the quality of sleep can be measured by observing the frequency of turning over. For example, a method is known in which body motion frequency and body motion interval during sleep are measured and it is determined that better sleep is obtained when the body motion frequency is low (see, for example, Patent Document 1). Rolling over caused by body movement can be detected using, for example, an air mat laid under the floor. This method has the advantage of not restraining the user.
JP-A-4-109960

  However, the technique described in Patent Document 1 described above has a problem that any body movement of the user is detected. For example, even if the user happens to be scratched and scratched, or even when the floor is vibrated by a third party, the user's body movement is detected, so it is erroneously determined that rolling has occurred. On the other hand, people sometimes turn over slowly. In this case, since the body movement is hardly detected, there is also a problem that it is not determined that the roll has occurred.

  The present invention has been made in view of the above-described problems. A system capable of evaluating a sleep state by reliably detecting a rollover and visualizing a sleep state regardless of the magnitude of body movement. The purpose is to provide.

  A typical example of the present invention is as follows. That is, a computer system comprising a plurality of sensor nodes that are mounted on a user's body and that transmit measured data, and a computer that analyzes data transmitted from the plurality of sensor nodes, the computer system comprising: The sensor node outputs an acceleration every predetermined sampling period, and transmits the output acceleration to the computer. The computer includes a processor and a memory connected to the processor, and the plurality of sensors. The acceleration acquired from the node is acquired, and the processor detects the time when the orientation of the user wearing the sensor node changes based on the change in the acquired acceleration, and the detected body The sleep state of the user is evaluated based on the interval of each time when the orientation of the user changes.

  According to one embodiment of the present invention, the sleep monitor system can reliably detect a turn during sleep and can evaluate a sleep state based on the turn over interval.

  Below, the sleep monitor system of embodiment of this invention is demonstrated using FIGS. 1-12.

  FIG. 1 is a block diagram showing an example of the configuration of a sleep monitor system according to an embodiment of the present invention.

  The sleep monitor system of this embodiment includes a bracelet sensor node 1, a base station 102, and a computer 103.

  The sleep monitor system uses, for example, a bracelet type sensor node 1 including an acceleration sensor as a sensor for measuring an operation (or state) during sleep of a user (observer) of the system. The bracelet type sensor node 1 is worn on the user's arm, outputs acceleration at a predetermined cycle (sampling cycle), and transmits the output acceleration (hereinafter referred to as sensing data) to the base station 102.

  The base station 102 includes an antenna 101 that receives a radio signal, and receives sensing data transmitted from the bracelet type sensor node 1. Note that the base station 102 may receive sensing data from the bracelet type sensor node 1 by wireless or infrared or wired. The base station 102 receives sensing data from the bracelet type sensor node 1 and transfers the received sensing data to a computer (PC) 103.

  The computer 103 includes a totaling unit 200, a totaling information storage unit 300, a rollover analysis unit 400, a sleep information storage unit 500, and a display content generation unit 600, and connects the display unit 104 and the input unit 105.

  The computer 103 includes a processor, a memory, and a storage device (storage medium) not shown. The storage device stores various programs and data. The memory stores various programs executed by the processor. The processor executes various programs stored in the memory. The processor (not shown) of the computer 103 loads, for example, a program stored in a storage device (storage medium) into a memory at a predetermined timing, and executes the loaded program, thereby causing the counting unit 200 to turn over. The processing of the analysis unit 400 and the display content generation unit 600 is realized.

  The totaling unit 200 converts the sensing data transferred by the base station 102 into data per predetermined unit time, and stores the converted data in the totaling information storage unit 300. The turnover analysis unit 400 analyzes the user's operation or state information stored in the total information storage unit 300 for each unit time, and turns the user's turnover during sleep, which will be described later with reference to FIGS. 7 and 8. Generate interval information. In addition, the turnover analysis unit 400 stores the generated information about the turnover interval in the sleep information storage unit 500.

  Note that the sleep monitor system of the present embodiment may observe the sleep states of a plurality of users by using a plurality of bracelet sensor nodes 1. In this case, the identifier of each bracelet type sensor node 1 is given to the sensing data transmitted from the bracelet type sensor node 1. The computer 103 can identify the user by the identifier of the bracelet type sensor node 1 given to the sensing data.

  A user of the sleep monitor system or a person commissioned by the user (such as a doctor) operates the input unit 105 to instruct display processing.

  The display content generation unit 600 generates a screen that visualizes and indexes the sleep state of the user from the sleep information stored in the sleep information storage unit 500 based on an instruction from the user or the like, and displays the generated screen Displayed on the unit 104. The user or the like can observe the sleep state of the user on the screen displayed on the display unit 104.

  In this embodiment, the computer 103 processes the sensing data transmitted from the base station 102, but the server device installed in a remote place (a place away from a place where the user's sleep state is measured) is a base station. The sensing data transmitted from 102 may be processed. In this case, the server device is connected to the base station 102 via a network (not shown) and receives sensing data via the network. The server device may have the same configuration as the computer 103. In this case, the totaling unit 200 of the server device converts the received sensing data into data for each unit time, and stores the converted data in the totaling information storage unit 300 of the server device. The rolling analysis unit 400 of the server device generates sleep information from the data stored in the total information storage unit 300 of the server device, and stores the generated sleep information in the sleep information storage unit 500 of the server device. The computer 103 (client device) may acquire sleep information from the sleep information storage unit 500 of the server device connected to the network. The display content generation unit 600 of the computer 103 (client device) generates a screen that visualizes and indexes the sleep information acquired from the sleep information storage unit 500 of the server device, and displays the generated screen on the display unit 104. Also good.

  With the configuration described above, a user or a person entrusted by the user can observe the sleep state of the user from a remote location.

  The sensor node according to the embodiment of the present invention is, for example, a bracelet type or a wristwatch type sensor. The bracelet type sensor node 1 will be described below.

  FIG. 2 is an explanatory diagram illustrating an example of the configuration of the bracelet type sensor node 1 according to the embodiment of this invention.

  FIG. 2 shows a schematic view (a) of the bracelet type sensor node 1 viewed from the front and a cross-sectional view (b) viewed from the side. The bracelet type sensor node 1 detects the acceleration caused by the body movement of the user (the wearer of the bracelet type sensor node 1) and the gravitational acceleration received by the bracelet type sensor node 1.

  The bracelet type sensor node 1 includes a sensor device 10, a case 11 for storing the sensor device 10, and a band 12 for attaching the case 11 to an arm of a human body.

  The sensor device 10 includes a sensor, a control unit, and the like. The sensor device 10 will be described later with reference to FIG. As the sensor used for the bracelet type sensor node 1, as shown in FIG. 2, an acceleration sensor capable of detecting accelerations of three axes X, Y, and Z orthogonal to each other is preferable.

  The computer 103 can analyze from which direction the user's wrist is receiving gravity by using the three-axis acceleration (gravity acceleration) of X, Y, and Z orthogonal to each other output from the acceleration sensor. That is, in the present embodiment, the user's turn is detected by measuring the change in the wrist angle (posture) of the user.

  In addition, in this embodiment, considering that the user uses the sensor node during the day (for example, the sensor node is used as a pedometer or the like), the sensor node is a bracelet-shaped node that is always easy to wear on the body. However, it may be an ankle-shaped node that is worn on the foot, a belt-shaped node that is worn on the abdomen, or a small portable shape node that is put in a pocket.

  FIG. 3 is a block diagram of the sensor device 10 according to the embodiment of the present invention.

  The sensor device 10 includes a wireless communication unit (RF) 2, a microcomputer 3, a real time clock (RTC) 4, an acceleration sensor 6, a battery 7, and a switch 8.

  The wireless communication unit 2 includes an antenna 5 and transmits sensing data to the base station 102. The microcomputer 3 controls the sensor 6 and the wireless communication unit 2. The real time clock 4 is a timer that periodically starts the microcomputer 3. The acceleration sensor 6 detects the acceleration of three axes orthogonal to X, Y, and Z. The battery 7 supplies power to each unit. The switch 8 is a switch for controlling the supply of electric power to the sensor 6.

  In the sensor device 10, the switch 8 and the sensor 6 may be connected to the ground by a bypass capacitor C1. The bypass capacitor C1 removes noise. By controlling the switch 8, the sensor device 10 can reduce the number of times of charging / discharging the bypass capacitor C1 and reduce power consumption.

  The microcomputer 3 includes a CPU 34, a ROM 33, a RAM 32, an interrupt control unit 35, an A / D converter 31, a serial communication interface (SCI) 36, a parallel interface (PIO) 37, and an oscillation unit (OSC) 30.

  The CPU 34 is a processor that executes arithmetic processing. The ROM 33 stores a program executed by the CPU 34 and the like. The RAM 32 stores data and the like. The interrupt control unit 35 processes an interrupt to the CPU 34 based on a signal (timer interrupt) from the RTC 4. The A / D converter 31 converts the analog signal output from the sensor 6 into a digital signal. The serial communication interface 36 transmits and receives serial signals from the wireless communication unit 2. The parallel interface 37 controls the wireless communication unit 2 and the switch 8. The oscillation unit 30 supplies a clock to each unit of the microcomputer 3.

  Each part of the microcomputer 3 described above is connected through a system bus 38. The RTC 4 outputs an interrupt signal (timer interrupt) to the interrupt control unit 35 and outputs a reference clock to the SCI 36 at a predetermined cycle. The PIO 37 controls the switch 8 in accordance with a command from the CPU 34 and controls the supply of power to the sensor 6.

  The bracelet type sensor node 1 starts the microcomputer 3 at a predetermined sampling period (for example, 0.05 seconds), acquires sensing data from the sensor 6, and specifies the bracelet type sensor node 1 in the acquired sensing data. An identifier and a time stamp are assigned. The sensor device 10 transmits the sensing data with the identifier and the time stamp to the base station 102. Note that the control of the bracelet type sensor node 1 may be the same as the control described in Japanese Patent Application Laid-Open No. 2008-59058, for example. The bracelet type sensor node 1 may store the continuously acquired sensing data in the RAM 32 and collectively transmit the stored sensing data to the base station 102 at a predetermined timing.

  Below, the outline | summary of a process of the sleep monitor system of this embodiment is demonstrated.

  FIG. 4 is a flowchart showing processing of the sleep monitor system according to the embodiment of the present invention.

  First, in step S <b> 1, the base station 102 transfers the sensing data received from the bracelet type sensor node 1 to the computer 103.

  Next, in step S <b> 2, the totaling unit 200 of the computer 103 performs sensing transmitted from the bracelet type sensor node 1 at a predetermined cycle (for example, 1 minute) for each user (identifier of the bracelet type sensor node 1). The data is aggregated, and the aggregated sensing data is stored in the aggregate information storage unit 300.

  Specifically, the totaling unit 200 continuously acquires sensing data (for example, a sampling period of 0.05 seconds) acquired for each identifier of the bracelet type sensor node 1, for example, in unit time (for example, 1 minute). The X-axis average acceleration value 303, the Y-axis average acceleration value 304, and the Z-axis average acceleration value 305 per unit time (for example, 1 minute) are calculated. In addition, the counting unit 200 calculates the exercise frequency 306 by a method described later with reference to FIG. The totaling unit 200 displays the calculated X-axis acceleration average value 303, Y-axis acceleration average value 304, Z-axis acceleration average value 305, exercise frequency 306, bracelet type sensor node 1 identifier 301, and date / time 302 as sensing data. Is stored in the total information storage unit 300.

  In the present embodiment, for example, the sampling cycle is 0.05 seconds and the unit time is 1 minute. However, the sampling cycle is sufficiently longer than the time of the user's turning action (the action of changing the body orientation). It may be the same as the unit time for obtaining the average.

  Next, in step S <b> 3, the turnover analysis unit 400 calculates the daily bedtime and wake-up time for each user using the exercise frequency 306 from the total information of the sensing data stored in the total information storage unit 300. .

  Several methods for determining the bedtime and the wake-up time from the exercise frequency 306 are known. For example, the Cole method (Cole RJ, Kripke DF, Gruen W, Mullaney DJ, Gillin JC. Automatic sleep / wakeidentification activity activity. Sleep 1992; 15: 491-469) may be applied.

  Here, the Cole method is to calculate the determination value by applying the number of zero crosses per minute (the number of times the acceleration passes through 0G) for a few minutes before and after sleep introduction to the Cole discriminant. It is a method of determining whether it is an awake state or a sleep state based on the determined value. The exercise frequency 306 during the sleep time is much less than the exercise frequency during the awake (wake) time. In this embodiment, the user's bedtime and wake-up time are calculated by applying the exercise frequency 306 to the Cole discriminant.

  Next, in step S <b> 4, the rolling analysis unit 400 calculates the X-axis acceleration average value 303, the Y-axis acceleration average value 304, and the Z-axis for each user from the aggregation information of the sensing data stored in the aggregation information storage unit 300. Using the axial acceleration average value 305, the sleep time during the sleep time obtained in step S3 (that is, the time from the bedtime to the wake-up time) is detected. A method for detecting the turnover time will be described later with reference to FIG.

  Next, in step S5, the turnover analysis unit 400 calculates the turnover interval 504 from the bedtime detected in step S3, each turnover time detected in S4, and the wake-up time detected in S3. In addition, the turnover analysis unit 400 calculates an average value of the exercise frequency 306 during each turnover interval as the average exercise frequency 505. The rolling analysis unit 400 stores the result calculated in steps S3 and S4 in the sleep information storage unit 500 as the sleep information of the user. The sleep information stored in the sleep information storage unit 500 will be described later with reference to FIG.

  Finally, in step S <b> 6, the user or the person entrusted by the user instructs the display processing operation from the input unit 105. When the display content generation unit 600 receives an operation from a user or the like, the display content generation unit 600 acquires the sleep information of the user stored in the sleep information storage unit 500, and the sleep state of the user is visualized from the acquired sleep information. A screen that can evaluate the sleep state of the user is generated, and the generated screen is displayed on the display unit 104.

  The above is the outline of the processing of the sleep monitor system of the present embodiment. Below, the detailed process of each part of a sleep monitor system is demonstrated.

  FIG. 5 is an explanatory diagram illustrating an example of a format of user operation information stored in the total information storage unit 300 according to the embodiment of this invention.

  The totaling unit 200 totals the sensing data output by the bracelet type sensor node 1 at a predetermined sampling period (for example, 0.05 seconds), and calculates the average acceleration value of each axis per unit time (for example, one minute). calculate. Also, the motion frequency per unit time (1 minute) is calculated from the sensing data every 0.05 seconds. The method of calculating the exercise frequency will be described later with reference to FIG. The totaling unit 200 stores the calculated acceleration average value of each axis and the exercise frequency in the totaling information storage unit 300 as totaling information (user's motion information) of sensing data.

5 includes, for example, the identifier 301 of the bracelet type sensor node 1, the date and time 302 divided by unit time, the X-axis acceleration average value 303, the Y-axis acceleration average value 304, and the Z-axis acceleration average value. 305 and exercise frequency 306. Here, the unit of each acceleration average value 303, 304, 305 is gravitational acceleration “G” (1G = 9.80665 meters / second ² ). The unit of the exercise frequency 306 is “times / minute”. Below, the calculation method of the exercise frequency 306 is demonstrated.

  FIG. 6 is an explanatory diagram illustrating a method of calculating the exercise frequency 306 according to the embodiment of this invention.

  The sensing data output from the bracelet type sensor node 1 includes X-, Y-, and Z-axis accelerations. First, the totalization unit 200 calculates a scalar amount of acceleration of three axes orthogonal to each other in X, Y, and Z for each sampling period. The scalar quantity is calculated by the following equation, where the acceleration of each axis is Xg, Yg, Zg.

Scalar amount = (Xg ² + Yg ² + Zg ² ) ^1/2
Next, the totaling unit 200 performs a filter (band pass filter) process on the calculated scalar quantity. The totaling unit 200 removes noise components by extracting only a predetermined frequency band (for example, 0.1 Hz to 5 Hz) from the calculated scalar quantity.

  Next, the totaling unit 200 detects the number of times (the number of zero crossings) that the scalar amount of acceleration crosses a predetermined threshold (for example, 0G or 0.05G) after the filtering process, as shown in FIG. Note that the number of times the scalar amount of acceleration exceeds a predetermined threshold may be set as the number of zero crossings. The counting unit 200 sets the number of zero crosses detected per unit time (1 minute) as the exercise frequency 306 per unit time (1 minute).

  The exercise frequency 306 indicates the amount of movement of the user during sleep regardless of the direction of the user's wrist (direction of receiving gravity acceleration). The exercise frequency 306 is used to detect a bedtime and a wake-up time in step S3 of FIG. Moreover, you may utilize for evaluation of the quality of sleep.

  FIG. 7 is an explanatory diagram illustrating an example of a format of sleep information stored in the sleep information storage unit 500 according to the embodiment of this invention.

  The rolling analysis unit 400 analyzes the operation information stored in the total information storage unit 300 and stores the analyzed result in the sleep information storage unit 500 as sleep information. The sleep information includes the identifier 501 of the bracelet type sensor node 1, the start time 502, the end time 503, the turnover interval 504, the average exercise frequency 505, and the evaluation index 506.

  The start date and time 502 is the time when the rollover occurred. The end date and time 503 is the time when the rollover occurred next and the rollover interval ended. It should be noted that the first start date and time 502 in one (overnight) sleep is the bedtime. The last end date and time 503 is a wake-up time. In addition, the end date and time 503 of each turnover interval during sleep is equal to the start date and time 502 of the next turnover interval.

  The turnover interval 504 is a turnover time interval that is calculated by the difference between the time when the turnover occurred (start time 502) and the time when the turnover occurred next (end time 503). The average exercise frequency 505 is a value obtained by averaging the exercise frequency 306 per unit time (1 minute) in one turnover interval from the start date and time 502 to the end date and time 503. The evaluation index 506 is an index indicating the quality of sleep for each turnover interval 504. A method for calculating the evaluation index 506 will be described later with reference to FIG.

  FIG. 8 is an explanatory diagram illustrating an example of a method for detecting the turnover time according to the embodiment of this invention.

  The turnover analysis unit 400 detects the turnover time using the user's operation information stored in the total information storage unit 300. The detected turn time is stored in the sleep information storage unit 500 as the start date and time 502 and the end date and time 503 of the turn shown in FIG.

  The acceleration sensor 6 outputs acceleration in the directions of three axes orthogonal to X, Y, and Z. The output acceleration includes gravity acceleration and acceleration caused by the user's body movement. The acceleration sensor 6 outputs gravitational acceleration in the directions of the three axes X, Y, and Z that are orthogonal to each other in a stationary state.

  Since the value of the gravitational acceleration output for each of the X, Y, and Z axes varies depending on the direction in which the acceleration sensor 6 receives the gravitational acceleration, the acceleration sensor 6 can detect the inclination of the acceleration sensor 6 itself. That is, the acceleration sensor 6 attached to the user's body can detect a change in the orientation of the user's body during sleep. The acceleration sensor 6 outputs a positive or negative value centered on 0G in each of the X, Y, and Z axis directions. In addition, when one or two axes are horizontal with respect to a vertically downward direction that receives gravity, the value of gravitational acceleration in the direction of the axis is zero.

  Therefore, when the positive or negative value of one of the output values of the acceleration sensor 6 in the X, Y, and Z axis directions is reversed, the turnover analysis unit 400 has a larger direction of gravity that the bracelet type sensor node 1 receives. It is determined that the user has changed, that is, the user has turned over. Here, the unit of acceleration output by the acceleration sensor 6 is “G”.

  A broken line from 801 to 807 shown in FIG. 8 is one of the X-axis average acceleration value 303, the Y-axis average acceleration value 304, and the Z-axis average acceleration value 305 per unit time in time series. Shows changes. Here, a region where the acceleration average value is +0.1 G or more is A, a region where the acceleration average value is between +0.1 and −0.1 G is B, and a region where the acceleration average value is −0.1 G or less is C. .

  The turning analysis unit 400 detects the time when the average acceleration value changes from the region A to the region C or from the region C to the region A, and sets the detected time as the turning time. When the average acceleration value indicates a value in the vicinity of 0 G, the positive and negative values of the average acceleration value are inverted even when the inclination of the bracelet type sensor node 1 slightly changes. Therefore, the turn analysis unit 400 ignores the change in the average acceleration value in the region B so that a slight change is not erroneously determined to be turned over.

  In the present embodiment, the rollover analysis unit 400 detects the rollover of the user based on a change in average acceleration per unit time that is longer than the sampling period, instead of a change in acceleration output every sampling period. The turning operation is not necessarily simple but complicated. For this reason, when it is attempted to detect a turn by means of an acceleration output with a sampling period of 0.05 seconds, which is extremely shorter than the time of the turn operation, for example, the sign of the acceleration is repeated many times during the turn operation. Is reversed.

  Accordingly, the turnover analysis unit 400 may erroneously determine that a slight change in acceleration is a turnover. On the other hand, when the average acceleration value per unit time (for example, 1 minute) is used, the complicated motion of the turning operation is averaged and smoothed, so that the problem of erroneous determination does not occur.

  In FIG. 8, the average acceleration value in a certain axial direction is a positive value from 801 to 802 and belongs to the region A. Next, the average acceleration value changes from 803 to 804 due to a change in body orientation, and reverses from a positive value to a negative value. However, since the acceleration average value between 803 and 804 belongs to the region B, the turn analysis unit 400 determines that the change during this time is not turn over. The average acceleration value is a positive value between 805 and 806 and belongs to the region A. Next, the acceleration average value changes from 806 to 807 and reverses from a positive value to a negative value. Further, the acceleration average value of 807 belongs to the region C. Therefore, the rollover analysis unit 400 determines that the change from 806 to 807 is rollover, and determines that the time 808 is the rollover time.

  If 805 is in the region C, the turning analysis unit 400 ignores the change from the point 803 to the point 804 belonging to the region B and turns the change from the 802 in the region A to 805 in the region C. And the time at point 805 is set as the turnover time. Moreover, the rollover analysis unit 400 determines whether there is a rollover by using all acceleration average values in the X, Y, and Z axis directions. In other words, the rollover analysis unit 400 detects the rollover time when rollover is determined in any axial direction.

  Conventionally, turning over has been determined by the body movement of the user. However, the body motion is not necessarily a roll-over, for example, when a sleeping person slightly scratches his face or when the floor is shaken by a third party. For this reason, according to the conventional method, the detected body movement may be erroneously determined as turning over.

  In the present embodiment, turning over is determined by detecting a change in body orientation, not a user's body movement. When turning over, the body direction (the direction of the user's body) changes greatly, and the direction in which the body receives gravity changes greatly. Therefore, it is possible to detect turning over by detecting a change in the direction of receiving gravity. In order to detect turning over accurately, it is preferable to detect a change in body orientation (body orientation) rather than exercise frequency.

  A method for calculating the evaluation index 506 will be described below. The rolling analysis unit 400 determines an evaluation index 506 that determines the quality of sleep based on the correspondence table shown below.

  FIG. 9 is an explanatory diagram illustrating an example of a correspondence table for generating an evaluation index from a turnover interval according to the embodiment of this invention.

  FIG. 9 shows the correspondence between the rollover interval 504 and the evaluation index 506.

  The turnover interval 504 is classified into classes classified by a predetermined separation value. The evaluation index 506 is an evaluation value given to each class. In this embodiment, for example, when the turnover interval 504 is less than 5 minutes, the evaluation index 506 is “1”, and when the turnover interval 504 is 5 minutes or more and less than 15 minutes, the evaluation index 506 is “2”. When the turnover interval 504 is 15 minutes or more and less than 30 minutes, the evaluation index 506 is “3”, and when the turnover interval 504 is 30 minutes or more, the evaluation index 506 is “4”.

  Evaluation index 506 indicates the quality of sleep during the turnover interval 504. As described above, the sleep state is deeper when the frequency of turning over is lower, that is, the quality of sleep is higher. Therefore, the evaluation index 506 is higher when the turnover interval 504 is longer. On the other hand, the evaluation index 506 is lower when the turnover interval 504 is shorter.

  Note that the turnover analysis unit 400 evaluated the sleep state based on the turnover interval 504, but the sleep state may be evaluated using the average exercise frequency 505 in addition to the turnover interval 504.

  As described above, the turnover analysis unit 400 classifies each turnover interval 504 into four classes that are divided by predetermined delimiter values and are assigned evaluation indexes “1” to “4”. The turnover analysis unit 400 may further determine that the user (the person to be observed) is sleeping gently if the average exercise frequency 505 within each turnover interval 504 is lower than a predetermined threshold. In this case, for example, “0.5” may be added to the evaluation index 506 of the turnover section 504 determined to be sleeping quietly among the turnover intervals 504 to which each evaluation index 506 is assigned.

  In addition, you may measure a user's brain wave, a pulse rate, or a respiration rate with a sensor or apparatus different from the sleep monitor system of this embodiment. In this case, the evaluation index 506 may be set using the measurement result described above in addition to the turnover interval 504 and the average exercise frequency 505.

  FIG. 10 is an explanatory diagram illustrating an example of a display screen generated by the display content generation unit 600 according to the embodiment of this invention.

  The display content generation unit 600 acquires the sleep information of the user stored in the sleep information storage unit 500, visualizes the sleep state of the user from the acquired sleep information, and evaluates the sleep state of the user. A screen that can be generated is generated, and the generated screen is displayed on the display unit 104.

  The horizontal axis of the graph 1001 is time, and the vertical axis is the observation date. A graph 1001 is a band graph that displays the turnover time and the turnover interval 504 in time series. Each rollover interval 504 is distinguished by shading based on the evaluation index 506. A region 1002 is a turnover interval 504 where sleep is deepest and the evaluation index 506 is “4”. A region 1003 is a turnover interval 504 where the evaluation index 506 is “3”. A region 1004 is a turnover interval 504 where the evaluation index 506 is “2”. A region 1005 is a turnover interval 504 where sleep is shallowest and the evaluation index 506 is “1”. That is, the longer band indicates better sleep quality (deep sleep).

  Moreover, in the graph 1001, the sleep state every night is displayed. With this graph 1001, the user or the like can visually understand the transition of the bedtime at night, the wake-up time, and the depth of sleep during sleep. For example, when the band is long, the sleep time is long, and when the band is generally dark, it indicates that the sleep is deep.

  Here, the quality of sleep (sleep depth) is shown by shading, but it may be shown by a plurality of colors.

  FIG. 11 is an explanatory diagram illustrating another example of the display screen generated by the display content generation unit 600 according to the embodiment of this invention.

  In the graph 1101, similarly to the graph 1001, each turning interval is distinguished by shading, and the transition of the sleep state is shown. However, the horizontal axis of the graph 1101 is not the time but the elapsed time since going to bed.

  The graph 1101 allows the user or the like to compare the transition of the sleep state for each day based on the bedtime. This makes it easier for users and the like to find sleep state characteristics such as poor sleep, awakening in the middle, and awakening in the early morning.

  In addition, the graph 1102 shows the transition of the ratio of sleep time in which the turnover interval 504 is equal to or longer than a predetermined time to the total sleep time. For example, on September 1, the ratio of sleep time in which the turnover interval 504 is “30 minutes” or more (in other words, sleep time in which the evaluation index is “4” or more) to the total sleep time is about 50%. %. The graph 1102 allows a user or the like to observe a change in sleep quality.

  Note that the graph 1102 shows, for example, the ratio of sleep time in which the turnover interval 504 is “15 minutes” or more (in other words, sleep time in which the evaluation index is “3” or more) to the total sleep time. Good.

  Moreover, the user etc. can observe a sleep state more effectively by displaying the sleep state for several months on a graph. In the graph 1101, the transition of the sleeping state is displayed with the bedtime as a reference, but the transition of the sleeping state may be displayed with the wake-up time as a reference, for example. Moreover, although the quality of sleep (sleep depth) was shown here with the shading, you may show with several colors.

  FIG. 12 is an explanatory diagram illustrating another example of the display screen generated by the display content generation unit 600 according to the embodiment of this invention.

  The graph 1201 shows the relationship between each sleep time and the deep sleep occupancy rate for the user (person) and other users. The horizontal axis of the graph 1201 is the sleep time, and the vertical axis is the deep sleep occupation ratio (the ratio of the sleep time in which the turnover interval 504 is equal to or longer than the predetermined time to the total sleep time). A black circle shown in the graph 1201 is a measurement result of the user (person). White circles are measurement results of users other than the user.

  When the measurement result belongs to the upper left area of the graph 1201, it indicates that the user is deeply asleep even if the sleep time is short. On the other hand, when the measurement result belongs to the lower right portion of the graph 1201, it indicates that the sleep is shallow and the sleep is forever. Thereby, the user (person) can relatively compare his / her sleep state with the sleep state of others and know the position of his / her own measurement result. Note that the center point of the graph 1201 (the intersection of two center lines indicated by broken lines) may be set to match the average value of all displayed users.

  In addition, in the graph 1201 of this embodiment, the measurement results of a plurality of other users are displayed in addition to the measurement results of the user (person), but the measurement results of the user (person) are displayed together. May be. Thereby, the user himself / herself can know the tendency of his / her sleep state.

  As described above, according to the present embodiment, the sleep monitor system detects a change in the orientation of the body during sleep from the sensing data output by the sensor attached to the body, and determines that a roll has occurred. Can do. In addition, by using the sleep time interval (turnover interval) as an indicator of the sleep state, the user's sleep state can be visualized, and a screen on which the user's sleep state can be evaluated can be generated. As a result, the user or the like can quantitatively evaluate the sleep state.

  In the embodiment of the present invention, instead of body movement, a change in the body direction is detected, and the change in the body direction is determined to be turned over. Therefore, the roll over can be detected regardless of the magnitude of the body movement. Moreover, even when a slow turnover occurs or when a moment of body movement is missed, the turnover can be reliably detected by a change in the direction of the body after turning over. In addition, since turning over is detected using the average acceleration value per unit time, a detailed operation during sleeping is not erroneously determined as turning over.

  Further, the present invention can be applied to a computer system that observes and evaluates a person's sleep state. In particular, as a part of health management, it is suitable for a sleep monitor system in which a user can easily check a sleep state at home every day.

It is a block diagram which shows the example of a structure of the sleep monitor system of embodiment of this invention. It is explanatory drawing which shows the example of a structure of the bracelet type sensor node of embodiment of this invention. It is a block diagram of the sensor apparatus of embodiment of this invention. It is a flowchart which shows the process of the sleep monitor system of embodiment of this invention. It is explanatory drawing which shows the example of the format of the user's operation | movement information stored in the total information storage part of embodiment of this invention. It is explanatory drawing which shows the method of calculating the exercise frequency of embodiment of this invention. It is explanatory drawing which shows the example of the format of the sleep information stored in the sleep information storage part of embodiment of this invention. It is explanatory drawing which shows the example of the method of detecting the turning time of embodiment of this invention. It is explanatory drawing which shows the example of the conversion table for producing | generating an evaluation index | exponent from the turnover interval of embodiment of this invention. It is explanatory drawing which shows the example of the display screen which the display content generation part of embodiment of this invention produces | generates. It is explanatory drawing which shows another example of the display screen which the display content generation part of embodiment of this invention produces | generates. It is explanatory drawing which shows another example of the display screen which the display content generation part of embodiment of this invention produces | generates.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Bangle sensor node 102 Base station 103 Computer 104 Display part 105 Input part 200 Total part 300 Total information storage part 400 Rolling-back analysis part 500 Sleep information storage part 600 Display content generation part

Claims (10)

A computer system comprising a plurality of sensor nodes that are mounted on a user's body and that transmit measured data, and a computer that analyzes data transmitted from the plurality of sensor nodes,
The plurality of sensor nodes are:
Acceleration is output every predetermined sampling period
Sending the output acceleration to the computer;
The calculator is
A processor and a memory connected to the processor;
Obtaining accelerations transmitted from the plurality of sensor nodes;
The processor is
Based on the acquired change in acceleration, detects the time when the orientation of the user wearing the sensor node has changed,
A computer system characterized by evaluating a sleep state of the user based on an interval of each time when the detected body orientation changes. The plurality of sensor nodes are:
An acceleration sensor that detects positive or negative acceleration according to the user's movement and body orientation;
Outputting acceleration detected by the acceleration sensor at each predetermined sampling period;
The processor is
Calculate the average value of the acceleration per unit time,
From the time when the calculated average value of acceleration changes from a value greater than or equal to the first reference value greater than 0 to a value less than or equal to the second reference value less than 0, and a value less than or equal to the second reference value The time when the user changes to a value greater than or equal to the first reference value is determined as the user's turnover time,
Based on each set turnover time, each turnover interval is calculated,
The computer system according to claim 1, wherein the sleep state of the user is evaluated based on the calculated turnover intervals. The acceleration sensor detects acceleration in three directions substantially orthogonal to each other,
The processor is
Calculating an average value of the accelerations in the three directions per predetermined unit time;
The time when any of the calculated average values of accelerations in the three directions has changed from a value greater than or equal to the first reference value to a value less than or equal to the second reference value, and from a value less than or equal to the second reference value The computer system according to claim 2, wherein a time at which the value has changed to a value equal to or greater than a first reference value is determined as the turnover time. The acceleration sensor detects acceleration in three directions substantially orthogonal to each other,
The processor is
Calculate the magnitude of acceleration in three substantially orthogonal directions for each predetermined sampling period,
The number of times that the calculated acceleration exceeds the third reference value within a predetermined time is determined as an exercise frequency,
The computer system according to claim 2, wherein the sleep state of the user is evaluated based on the determined exercise frequency. The processor is
The calculated turn-over intervals are classified into a plurality of times according to the time,
The calculated turnover intervals are arranged on the time axis based on the start time or the end time of each turnover interval,
The computer system according to claim 2, wherein data for displaying the arranged turning intervals so that the classification of the turning intervals is distinguishable is created.   The computer system according to claim 5, wherein the processor arranges the turnover intervals on a time axis using the same time axis for each measurement date.   6. The computer system according to claim 5, wherein the processor arranges the turn-over intervals on the time axis so that the start times of the first turn-off intervals for each measurement date are at the same position on the time axis. The processor is
The calculated turn-over intervals are classified into a plurality of times according to the time,
The total number of turn-over intervals that are evaluated as having good sleep among the intervals between turn-ups classified into a plurality of the above-mentioned sleep-sleeps is the ratio of the sleep time until the user wakes up and wakes up. Set to occupancy,
The computer system according to claim 2, wherein data indicating a relationship between the occupancy rate of the deep sleep and the one sleep time is created.   The computer system according to claim 1, wherein the sampling period is sufficiently longer than an operation time in which the orientation of the user's body changes.   The computer system according to claim 1, wherein the sensor node is a bracelet type and is attached to a user's arm.

Images