JP6145916B2 - Sensor device, measurement system, and measurement program - Google Patents

Description

  The present invention relates to a technique for integrating and measuring temporal changes in myoelectric potential and acceleration of a living body, for example, a human being.

  As a technique for detecting human muscle movement, measurement of myoelectric potential is often used. In recent years, the acceleration sensor has been miniaturized, and the acceleration sensor may be used as a method for measuring human movement and behavior. That is, such myoelectric potential measurement and acceleration measurement are used in various fields such as medical treatment, sports, and behavior analysis.

  For example, research on measuring human body movements and biological signals and using them as an interface between humans and machines has been extensively conducted.

  In recent studies, it has been reported that the user's intentions and body movements are identified by attaching sensors to the body and devices attached to the body and measuring joint angles, accelerations, external loads, etc. . In addition, research has been reported that attempts to detect human movements and intentions from neural signals such as surface electromyogram (sEMG) detected from the human skin surface (Non-patent Document 1).

  Or there exists a report about the action recognition apparatus which recognizes a motion or action of a subject using an acceleration sensor. For example, Patent Document 1 discloses a measurement unit that observes a change in state associated with a motion or action of a subject, a feature amount extraction unit that extracts a feature amount in an observation result, and an operation or action that should be recognized by the recognition device. And a recognition means for recognizing the action or behavior of the subject from the feature quantity extracted from the observation result and the stored feature quantity. The state changes observed by the measuring means include acceleration, angular acceleration, velocity, angular velocity, position, rotation, and bending angle, and the subject's biological information such as pulse, blood pressure, body temperature, blood glucose level, breathing, myoelectricity, heart Electricity, blood flow, and electroencephalogram are mentioned. Recognized movements include basic actions such as squatting, running, walking, climbing ladder, going down stairs, control panel operation at high altitude, valve operation at floor, suddenly falling, falling from high altitude , Rest, walking leisurely, walking powerfully, running very quiet, walking normally, walking well, running well, patients falling down, suffering complicated movements and so on.

  In addition, for exercise training, we will provide a mechanism that supports training of how to use the body at rest before starting exercise, such as technology that uses acceleration sensors and myoelectric potential measurement for training in human exercise Has also been reported (for example, Patent Document 2).

  In this training system, the shake evaluation unit identifies the period and amplitude of the shake of each part based on the acceleration of each part of the body of the exerciser detected by the acceleration sensor. At this time, the myoelectric potential is also used to detect human movement. The sound source stores the sound waveform data of the notification sound corresponding to each part where the acceleration sensor is mounted in its own memory, and generates an audio signal indicated by the sound waveform data. The notification sound synthesizing unit emits the audio signal generated by the sound source from the speaker as the notification sound, and changes the sense of distance of the notification sound according to the cycle and amplitude of each part specified by the shake evaluation unit. With such an operation, the system supports training of the target human movement.

  On the other hand, as such an acceleration sensor, a small system capable of detecting not only acceleration but also angular velocity has been developed by applying MEMS (Micro Electro Mechanical Systems) technology (for example, Patent Document 3, (See Patent Document 4).

  When such an acceleration sensor is used for observing human movement and behavior as described above, wireless communication is used to notify the system of the measurement result of the acceleration sensor. In this case, it is necessary to measure the magnitude and direction of acceleration accurately in association with the measurement time. In such a case, several methods have already been developed for time synchronization and time adjustment techniques for a plurality of sensors (see, for example, Patent Document 5 and Patent Document 6).

JP-A-10-113343 JP 2009-233092 A JP 2010-185835 A JP 2011-133244 A JP 2007-174330 A JP 2008-056161 A

Yoshikawa, Mikawa, Tanaka, “Real-time motion identification of hands with support vector machine using myoelectric potential”, IEICE paper, January 2009, J92, No.1, p. 93-103

  In particular, when measuring the movement and behavior of a single human being, when measuring temporal changes in the acceleration and myoelectric potential of each human part accurately, the time synchronization between the measurement data must be accurately taken. There is. However, conventionally, in the measurement of myoelectric potential and the measurement of acceleration, it has been usual to perform a process of integrating data measured by another sensor system on the computer side that processes the measurement data. .

  For example, a sensor for measuring myoelectric potential and a system that measures temporal changes associated with physical quantity movements such as acceleration are generally independent of myoelectric potential and acceleration when measuring acceleration and myoelectricity at the same time. A method of taking data is used. In order to perform recording and processing, it is necessary to synchronize each data measurement time. For each sensor (a microcomputer that controls measurement processing in each sensor), a synchronization signal is sent from a clock generator provided in each sensor. Alternatively, there may be a configuration in which data is wirelessly transmitted at an arbitrary cycle timing to a receiving computer that performs processing for providing time information and integrating measurement data from each sensor.

  In this case, when the measurement data is transmitted wirelessly, the synchronization signal is delayed due to the transmission path or the clock used by each sensor is different. In the case of trigger signal loop reception, synchronization within an arbitrary range is difficult, and a data string including the trigger signal must be received.

  Alternatively, each sensor may be configured to perform synchronization based on the time received by the receiving computer without performing the process for the time. In this case, the processing can be simplified, but the delay in the transmission path is not so small that it does not cause a problem in the data processing, or the delay is constant and does not fluctuate, or the delay can be calculated backwards Only available in certain cases.

  That is, when measuring data from each sensor as described above is transmitted to the system by wireless communication, measurement of myoelectric potential, such as transmission timing deviation, transmission delay, and depending on the situation, occurrence of data retransmission processing, etc. It is sometimes difficult to accurately synchronize the acceleration measurement with the acceleration measurement.

  In particular, when the system tries to measure the movements and activities of multiple people at the same time, the above problems become more serious.

  In addition, for users who are not necessarily skilled in measuring myoelectric potential and measuring acceleration, determine which part of the subject is appropriate to wear an accelerometer or electrodes for measuring myoelectric potential. There is also the problem that it is difficult to do. Furthermore, the user needs to be skilled in how to interpret the acquired data.

  Alternatively, as described above, when measuring data from a sensor is transmitted to the system by wireless communication, it is also important to reduce power consumption on the sensor side.

  The present invention has been made to solve the above-described problems, and an object thereof is to execute myoelectric potential measurement and acceleration measurement accurately in time synchronization with a simple configuration. It is to provide a sensor device and a measurement system capable of performing the above.

  Another object of the present invention is to provide a sensor device and a measurement system capable of reducing power consumption and performing myoelectric potential measurement and acceleration measurement synchronously.

  Still another object of the present invention is to provide a sensor device, a measurement system, and a measurement program capable of mounting a sensor at an appropriate position in measuring myoelectric potential.

  Still another object of the present invention is to provide a sensor device, a measurement system, and a measurement program capable of supporting myoelectric potential measurement and interpretation of acceleration measurement data executed in synchronization.

According to one aspect of the present invention, there is provided a sensor device for measuring a change in physical quantity and myoelectric potential associated with movement of a measurement object, the clock generation means being reset according to an instruction from the outside, and the measurement object being attached to the measurement object The first measuring means for using the signal from the electrode for measuring the myoelectric potential as myoelectric potential data and the second measuring unit for using the change in the physical quantity generated in the sensor device attached to the measurement object as the measuring data The measurement means, the myoelectric potential data from the first measurement means and the measurement data from the second measurement means are acquired in a lump, and the time stamp from the clock generation means is added and transmitted to the outside by wireless communication and control means for, second measuring means, in response to the measurement signal from the first measuring means, Ru is activated by the wake-up signal given from the control means.

According to another aspect of the present invention, the measurement system includes a sensor device for measuring a change in physical quantity and myoelectric potential associated with movement of a measurement target, and the sensor device is reset in response to an instruction from the outside. Clock generation means, a first measurement means for making a signal from an electrode for measuring myoelectric potential attached to a measurement object into myoelectric potential data, and a change in physical quantity generated in a sensor device attached to the measurement object The measurement data from the second measurement means, the myoelectric potential data from the first measurement means, and the measurement data from the second measurement means are collectively acquired, and the time stamp from the clock generation means is obtained. And a control means for transmitting to the outside by wireless communication, giving an instruction to the clock generation means to the sensor device, and outputting myoelectric potential data and measurement data transmitted from the sensor device. Further comprising a data management device including a storage device for storing in association with data and time stamp, the storage device for each muscle to be measured, the mounting position of the electrode and the sensor device for measuring a myoelectric potential And a discrimination data for discriminating a measurement pattern when an electrode is attached to the muscle to be measured and measured, and the data management device presents the image to the user before starting the measurement. Presenting means, and determination means for determining the mounting state based on the determination data with respect to the measurement data at the electrode position currently mounted, and the presenting means displays the determination result of the determination means. you presented.

  Preferably, the discriminating unit discriminates the muscle movement as a transition of a plurality of time-series states based on the measured myoelectric potential data and the measurement data, and the presenting unit displays the transition of the plurality of time-series states. To do.

  Preferably, the second measuring unit is activated by a wake-up signal provided from the control unit in response to the measurement signal from the first measuring unit.

  Preferably, the data management device enters a sleep state after giving an instruction to the clock generation means, and is activated by a wakeup signal.

According to still another aspect of the present invention, the measurement executed by the arithmetic unit of the data management device that controls the measurement operation of the measurement system having the sensor device for measuring the change in physical quantity and the myoelectric potential accompanying the movement of the measurement target The sensor device includes a clock generation unit that is reset in response to an instruction based on the measurement program, and a first signal that is used as myoelectric potential data from an electrode that is attached to the measurement target and that measures the myoelectric potential. From the first measuring means, the second measuring means for using the change in the physical quantity generated in the sensor device attached to the measuring object as measurement data, the myoelectric potential data from the first measuring means, and the second measuring means And control means for acquiring the measurement data in a batch, adding a time stamp from the clock generation means, and transmitting to the outside by wireless communication. Transmitting a signal for instructing the clock generation means to reset to the clock device, receiving myoelectric potential data, measurement data and time stamp transmitted from the sensor device, and storing them in the storage device in association with each other. For each muscle to be measured, the storage device has an electrode for measuring the myoelectric potential and an image showing the mounting position of the sensor device, and an electrode is attached to the muscle to be measured. Discriminating data for discriminating the measurement pattern when the measurement is performed, the measurement program presents an image to the user before starting the measurement and the measurement data at the electrode position currently mounted. Te, based on the discrimination data, the step of determining the mounted state, the step of presenting the determination result to the user, further, Ru is executed to the processing unit

  Preferably, based on the measured myoelectric potential data and the measured data, a step of discriminating the muscle movement as a transition of a plurality of time-series states, and a step of presenting a transition of the plurality of time-series states to the user Further, the arithmetic unit is caused to execute.

  According to the sensor device and the measurement system of the present invention, it is possible to execute the myoelectric potential measurement and the acceleration measurement accurately in time synchronization with a simple configuration.

  Also, according to the present invention, it is possible to reduce power consumption and execute myoelectric potential measurement and acceleration measurement synchronously.

  Further, according to the sensor device, the measurement system, and the measurement program of the present invention, it is possible to mount the sensor at an appropriate position for obtaining accurate measurement data.

  Alternatively, according to the sensor device, the measurement system, and the measurement program of the present invention, it is possible to support the myoelectric potential measurement and the interpretation of the acceleration measurement data that are executed in synchronization.

It is a figure which shows the usage condition of the sensor apparatus main body 100 of this Embodiment, and the computer 1000 for reception. 2 is a functional block diagram for explaining a configuration of a sensor device main body 100. FIG. It is a figure which shows the hardware constitutions of the computer 1000 for reception in the block diagram format. 4 is a flowchart for explaining operations of the sensor device main body 100 and a receiving computer 1000. It is a conceptual diagram which shows the relationship between the sticking position of electrodes 130.1 and 130.2 and a test subject's exercise | movement. It is a figure which shows the relationship between the position where the measurement electrode 130.2 was stuck, and the signal waveform measured. It is a figure which shows an example of the database for the computer 1000 for reception to perform a process. It is a flowchart for demonstrating the flow of a process in the case of performing an adjustment process of an electrode position. It is a flowchart for demonstrating the adjustment process of an electrode position. It is a conceptual diagram which shows the mode of a state transition of a time series. It is a conceptual diagram which shows the result of having extracted the state transition about the measurement result of a myoelectric potential and acceleration about a healthy subject and a patient about a predetermined | prescribed exercise | movement. It is a flowchart for demonstrating the flow of a process in the case of performing an exercise state evaluation process. It is a flowchart for demonstrating the evaluation process of an exercise state. It is a conceptual diagram for demonstrating the structure of wakeup. It is a flowchart for demonstrating the flow of a process in the case of performing a wakeup process.

  Hereinafter, configurations of a sensor device, a measurement system, and a measurement program according to an embodiment of the present invention will be described with reference to the drawings. In the following embodiments, components and processing steps given the same reference numerals are the same or equivalent, and the description thereof will not be repeated unless necessary.

  In addition, as a physical quantity to be measured in synchronism with the measurement of myoelectric potential, an explanation will be given taking acceleration as an example, but as a physical quantity to be measured in synchronism, in addition to acceleration (or in place of acceleration), For example, the structure which measures angular velocity etc. synchronously may be sufficient.

(Embodiment 1)
FIG. 1 is a diagram illustrating how the sensor device main body 100 and the reception computer 1000 according to the present embodiment are used.

  First, FIG. 1A shows an example of electrode arrangement in the case where a myoelectric potential is measured for a measurement site of a subject, for example, a forearm muscle.

  For example, the reference electrode 130.1 for measuring the reference potential is attached in the vicinity of a tendon where no muscle exists, and the measurement electrode 130.2 is attached on the muscle to be measured. The electrodes 130.1 and 130.2 are connected to the sensor device main body 100 by electric wires.

  As shown in FIG. 1B, when measuring myoelectric potential and acceleration for the upper arm and the forearm, the sensor device main body 100 is fixed to the forearm or the upper arm by a band. As will be described later, an acceleration sensor is provided in the sensor device main body 100, and acceleration measurement data and myoelectric potential measurement data are transmitted from the sensor device main body 100 to the reception computer 1000 by wireless communication. Is done. As a wireless communication method, so-called wireless LAN communication may be used, or as a short-distance communication method, a Bluetooth (registered trademark) method may be used.

  As will be described later, the reception computer 1000 performs processing such as storage of measurement data received from the sensor device main body 100 by wireless communication, data display, and data analysis.

  FIG. 2 is a functional block diagram for explaining the configuration of the sensor device main body 100.

  The sensor device main body 100 is measured by wireless communication with the clock interface (clock I / F) 102 for receiving the synchronization signal or time signal from the reception computer 1000 and the wireless communication method as described above with the reception computer 1000. A data input / output interface (data input / output I / F) 106 for transferring data and commands is provided.

  The sensor device main body 100 further includes a clock generation unit 104 that generates a clock signal for controlling measurement timing based on a signal from the clock I / F 102, and a microcomputer 110 that controls the measurement operation of the sensor device main body 100. Is provided. The microcomputer 110 operates based on the clock signal from the clock generation unit 104 and performs a control operation according to a command from the receiving computer according to an internal program.

  Here, the “synchronization signal” is a timing signal for synchronizing the internal clock of the receiving computer with the clock generated by the clock generation unit 104. The “time signal” is a signal including information (time information) representing the time itself sent from the receiving computer 1000, and the clock generation unit 104 is reset by this time information and is represented by the time information. Start measuring time from.

  Both the “synchronization signal” or the “time signal” may be transmitted from the reception computer 1000 to the sensor device main body 100, and either one of the internal clock and the sensor of the reception computer 1000 may be used. The configuration may be such that the internal clock of the apparatus main body 100 is synchronized.

  The microcomputer 110 receives the acceleration measurement data from the acceleration sensor 120 and the signals from the electrodes 130.1 and 130.2 via the sensor interface (sensor I / F) 132, and the myoelectric amplifier 134 amplifies the data. The D-converted measurement data is received and the measurement data is integratedly managed.

  Prior to the start of measurement, the clock generator 104 is reset upon receiving a synchronization signal or time information when synchronizing the internal clock. The reset of the clock generation unit 104 is performed based on a signal from the receiving computer 1000 in a wired or controlled environment with a negligible delay. The microcomputer 110 adds a time stamp generated based on the synchronization signal or time information obtained from the clock generation unit 104 to the measurement data, and transmits the measurement data to the reception computer 1000.

  FIG. 3 is a block diagram showing the hardware configuration of the receiving computer 1000. As shown in FIG.

  As shown in FIG. 3, in addition to the disk drive 2030 and the memory drive 2020, the computer main body 2010 constituting the receiving computer 1000 includes a CPU (Central Processing Unit) 2040 connected to a bus 2050, a ROM ( A memory including a read only memory (RAM) 2060 and a random access memory (RAM) 2070, a non-volatile rewritable storage device such as a hard disk 2080, and a communication interface 2090 for performing communication via a network. Yes. The disc 2030 is loaded with an optical disc such as a CD-ROM 2200. A memory card 2210 is attached to the memory drive 2020.

  As will be described later, when the program of the receiving computer 1000 operates, the database that stores information that is the basis of the operation is described as being stored in the hard disk 2080. However, such a database may be stored in another computer that communicates via a network.

  In FIG. 3, the CD-ROM 2200 is assumed as a medium capable of recording information such as a program installed in the computer main body. However, other media such as a DVD-ROM (Digital Versatile Disc) is used. Alternatively, a memory card or a USB memory may be used. In that case, the computer main body 2200 is provided with a drive device capable of reading these media.

  The main part of the receiving computer 1000 is composed of computer hardware and software executed by the CPU 2040. Generally, such software is stored and distributed in a storage medium such as a CD-ROM 2200, read from the storage medium by a disk drive 2030 or the like, and temporarily stored in the hard disk 2080. Alternatively, when the device is connected to the network, it is temporarily copied from the server on the network to the hard disk 2080. Then, the data is further read from the hard disk 2080 to the RAM 2070 in the memory and executed by the CPU 2040. In the case of network connection, the program may be directly loaded into the RAM and executed without being stored in the hard disk 2080.

  The program for functioning as the receiving computer 1000 does not necessarily include an operating system (OS) that causes the computer main body 2010 to execute functions such as an information processing apparatus. The program only needs to include an instruction portion that calls an appropriate function (module) in a controlled manner and obtains a desired result. How the computer system 2010 operates is well known and will not be described in detail.

  Further, the computer that executes the program may be singular or plural. That is, centralized processing may be performed, or distributed processing may be performed.

  Further, the CPU 2040 may be a single processor or a plurality of processors. That is, it may be a single core processor or a multi-core processor.

  FIG. 4 is a flowchart for explaining operations of the sensor device main body 100 and the receiving computer 1000 described with reference to FIGS.

  Referring to FIG. 4, first, a synchronization signal or a time signal is transmitted from reception computer 1000 to sensor device main body 100 arranged in a wired or extremely short distance (S100). At this time, when a plurality of sensor device main bodies 100 are used for measurement, a plurality of synchronization signals or time signals are simultaneously transmitted.

  When receiving the synchronization signal or the time signal, the sensor device body 100 resets the count operation of the clock generation unit 104 (S102).

  As described above, when measurement is started, the microcomputer 110 sets, for example, the variable n to 1 in order to specify the number of measurements (S106).

  The microcomputer 110 acquires T = clock (t) (clock (t) is the output of the clock generation unit 104) as the measurement time T (S107), and the myoelectric signal amplified and A / D converted by the myoelectric amplifier 134. In addition to receiving (S108), the acceleration sensor 120 receives the acceleration signal (S109) amplified and A / D converted, and acquires sensor information (S110).

  The microcomputer 110 adds a time stamp to the measurement data corresponding to the myoelectric signal and acceleration signal acquired in a lump and transmits the measurement data (S122).

  If the microcomputer 110 has not received the end command from the receiving computer 1000 (N in S132 and S134), the variable n is incremented by 1, the process is returned to step S107, and the measurement is continued.

  If the microcomputer 110 has received the end command (Y in S132 and S134), the microcomputer 110 ends the process (S140).

  On the other hand, when the receiving computer 1000 receives the measurement data from the sensor device main body 100 (S124), the measurement data of the myoelectric potential and acceleration (or angular velocity if necessary) is associated with the time stamp, for example, like the hard disk 2080. (S126). If there is no end instruction from the user, the receiving computer 1000 returns the process to step S124 to receive the measurement data.

  If necessary, the receiving computer 1000 may execute processing for displaying the received measurement data.

  On the other hand, when the end of measurement is instructed (Y in S128), the receiving computer 1000 transmits an end instruction command to the sensor device main body 100 (S130) and ends the processing (S142).

  With the configuration as described above, even if the receiving computer 1000 receives the data with a delay, the receiving computer 1000 can accurately detect the synchronization information (or time information) in the added time stamp. Measurement at the synchronized measurement time becomes possible. Even when a plurality of separate sensors are used at the same time, the respective measurement data can be acquired in synchronization.

(Embodiment 2)
In the first embodiment, even when the sensor device 100 and the receiving computer 1000 exchange measurement data by wireless communication, it is possible to acquire measurement data of myoelectric potential and measurement data such as acceleration in synchronization. A description has been given of a simple configuration.

  In the second embodiment, a configuration in which the user can easily confirm the attachment position of the electrode used for measuring the myoelectric potential in the configuration of the first embodiment will be described.

  Therefore, the hardware configuration of the sensor device 100 and the receiving computer 1000 is the same as that of the first embodiment. As will be described below, functions that are mainly added to the software processing executed by the receiving computer 1000 are described below. Is added.

  That is, as described above, by using the myoelectric sensor, it is possible to directly measure the activity of the muscle fibers of the human body. In recent years, myoelectric sensors have begun to attract attention in non-medical research such as human behavior analysis and sports science. However, anatomical expertise is required to determine the position where electrodes for surface electromyography are to be applied, and skillful know-how is required because the appropriate position varies depending on the physique and other individual differences. It is.

  For example, when the amplitude of the myoelectric signal is not sufficiently obtained, it can be determined whether the attachment position is inappropriate or the amount of contraction of myofiber is insufficient. It is difficult just to see.

  Therefore, in the receiving computer 1000 according to the second embodiment, when a non-professional uses a myoelectric sensor, it supports that an electrode is easily attached at an appropriate position, and actually appropriate data is obtained. It is possible to easily check whether

  FIG. 5 is a conceptual diagram showing the relationship between the attachment positions of the electrodes 130.1 and 130.2 and the movement of the subject.

  In FIG. 5, as an example, a case where the myoelectric potential of the biceps is acquired will be described.

  Referring to FIG. 5, in electrode attachment positions A, B, C, and G, position G represents a position where reference electrode 130.1 is attached, and positions A to C are users as described later. Indicates a position where the measuring electrode 130.2 may be attached.

  In this example, it is assumed that the sensor device main body 100 is worn on the wrist, and the user instructs the receiving computer 1000 to perform an exercise of bending and extending the arm.

  FIG. 6 is a diagram showing the relationship between the position where the measurement electrode 130.2 is attached and the signal waveform to be measured.

  As shown in FIG. 6A, when the measurement electrode 130.2 is attached to the position A and the potential difference between GA is measured, consistency with the acceleration data can be obtained, but the amplitude is sufficient. I can't get it.

  As shown in FIG. 6 (b), when the measurement electrode 130.2 is attached to the position B, there is a contradiction with the acceleration data (stretching / shrinking but opposite phase). From the relationship with acceleration, the receiving computer 1000 can estimate that the measurement electrode 130.2 is attached to the position of the triceps instead of the biceps.

  As shown in FIG. 6C, when the measurement electrode 130.2 is attached to the position C, the compatibility with the acceleration sensor and the amplitude are good, so that the receiving computer 1000 is placed at an appropriate position. It can be estimated that the measurement electrode 130.2 is attached.

  FIG. 7 is a diagram illustrating an example of a database for the reception computer 1000 to execute processing.

  In such a database, information on the measurement site and measurement target muscle, image data indicating the attachment position of the sensor and the electrode, the attachment position of the acceleration sensor, and the exercise instructed to adjust the attachment position to the subject are instructed. Data indicating a correlation between the exercise, myoelectric potential, and acceleration, and information on an instruction for prompting the user to change the attachment position according to the estimated attachment position of the measurement electrode 130.2. Stored in association.

  Here, as data indicating the correlation between the instructed exercise and the myoelectric potential and the acceleration, in FIG. 7, information indicating whether the acceleration and the contraction of the muscle are in phase is associated. However, for example, a typical pattern of acceleration and myoelectric potential with time is machine-learned in advance to form a discriminator. Which of these discriminators is used for determination? Such information may be associated. Such a discriminator has the measurement electrode 130.2 at any position when the measurement electrode 130.2 is attached to the optimal position with respect to the measurement target muscle, other muscles in the vicinity of the measurement target muscle, or other positions. It may be machine-learned so that it is possible to determine whether or not is attached.

  In addition, as the “instruction for prompting the replacement of the pasting position”, for example, when the measurement target muscle is currently the biceps brachii, it is determined that it is pasted on the triceps brachii Remembers the message “The electrode seems to be attached to the triceps now. Please adjust the position of the electrode a little further toward the inside of the arm.” In addition, it may be displayed on the display 2120 of the receiving computer 1000.

  FIG. 8 is a flowchart for explaining the flow of processing when performing the electrode position adjustment processing as described above.

As shown in FIG. 8, compared with the case of FIG. 4 of the first embodiment, in the second embodiment, a synchronization signal or a time signal is transmitted from the receiving computer 1000 to the sensor device main body 100 ( S100) When the sensor device main body 100 receives the synchronization signal or the time signal, the count operation of the clock generation unit 104 is reset (S102), and then the electrode position adjustment processing S104 is executed, and the case of FIG. The process is the same. FIG. 9 is a flowchart for explaining the electrode position adjustment process in FIG.

  Referring to FIG. 9, when a synchronization signal or a time signal is transmitted, receiving computer 1000 displays an image indicating the sensor device main body 100 and electrode mounting positions and the exercise to be performed by the user in accordance with a designation from the user. (S200), and a standby state for a start instruction from the user is entered (S201).

  When the user instructs the start of mounting position adjustment, the receiving computer 1000 transmits a command for instructing the start of mounting position adjustment to the microcomputer 110 (S202).

  When receiving the command for starting the adjustment of the mounting position (S204), the microcomputer 110 sets, for example, a variable n0 to 1 in order to specify the number of times of measurement (S206).

  The microcomputer 110 acquires T = clock (t) as the measurement time T (S208), receives the myoelectric signal (S210) amplified and A / D converted by the myoelectric amplifier 134, and the acceleration sensor 120 amplifies it. The A / D converted acceleration signal (S212) is received and sensor information is acquired (S220).

  The microcomputer 110 transmits the measurement data corresponding to the acquired myoelectric signal and acceleration signal with a time stamp added (S222).

  If the microcomputer 110 has not received the adjustment process end command from the receiving computer 1000 (N in S236 and S238), the microcomputer 110 increments the variable n0 by 1, returns the process to step S208, and continues the measurement.

  If the microcomputer 110 has received a measurement process end command (Y in S236 and S238), the microcomputer 110 ends the process (S240).

  On the other hand, when receiving the measurement data from the microcomputer 110 (S224), the receiving computer 1000 stores the measurement data in a storage device such as the hard disk 2080 (S226).

  Further, the receiving computer 1000 collates whether the measured changes in time of myoelectric potential and acceleration measurement data correspond to the case in which the muscle is attached to the measurement electrode 130.2.

  Here, the receiving computer 1000 determines from the data of the acceleration / angular velocity sensor whether the instructed motion is being performed and whether the mounting position of the acceleration / angular velocity sensor is appropriate. If there is a problem with any of these, a display to that effect may be displayed on the display 2120.

  In checking which muscle corresponds to the case where it is attached to the measurement electrode 130.2, for example, the relationship / feature between the data of myoelectricity and acceleration (or angular velocity if necessary) is calculated. The data is collated with data stored in a database in which previously learned data is stored.

  Such pattern recognition for data collation is pattern recognition of time series data. For example, in the case of recognition of time-series data in the case of speech recognition, recognition by supervised learning of a hidden Markov model (HMM: Hidden Markov Models) by a computer has been put into practical use.

  However, in the case of voice recognition, the target of pattern recognition by the HMM is a scalar time-series pattern called a voice signal, whereas in the case of motion or motion recognition that requires integration of information from multiple sensors. However, not only does the number of dimensions of the pattern increase, but there is also a problem that there is no unified symbol expression corresponding to the pattern.

  Therefore, in order to deal with such a problem, for example, a technique using a continuous distribution type hidden Markov model is disclosed in the following documents, for example.

Reference 1: Tetsuya Inagi, “HMM and human motion recognition, teaching and generation,” Journal of the Robotics Society of Japan, Vol. 29 No. 5, pp. 419-422, 2011
Further, a technique for applying such a continuous distribution type hidden Markov model to more general pattern recognition using unsupervised learning of dynamic time-series data is disclosed in the following documents.

Document 2: Japanese Patent Laid-Open No. 2006-162898 Therefore, it is possible to pattern-recognize the position affixed to the measurement electrode 130.2 and time-series data of myoelectric potential and acceleration (and further, if necessary, angular velocity). .

  In this case, the likelihood for any of the patterns can be calculated at the same time, and this can be used as a score of coincidence for each pattern.

  Alternatively, in the case where muscle movement is decomposed into movement elements and grasped as supervised learning, a recognition graph can be configured by a weighted finite state transducer based on HMM.

  Such weighted finite state transducers are disclosed in, for example, the following documents.

Document 3: JP 2012-48119 A In this case as well, it is possible to calculate a coincidence score for a pattern based on likelihood.

  When it is determined that the matching score obtained as described above is sufficiently high and the pasting position is good (S230), the receiving computer 1000 determines that the electrode pasting position is appropriate, for example. Display is performed, an instruction to end adjustment is transmitted to the microcomputer 110 (S234), and the adjustment process is ended (S242).

  On the other hand, if the coincidence score is not within the allowable amount in step S228, the receiving computer 1000 recognizes that the attachment position of the myoelectric sensor is incorrect, displays that fact, and readjusts the attachment position. (S232), and the process returns to step S224. At this time, if the matching score with the relationship / feature of another pasting position with respect to an appropriate pasting position is more than a certain level, the candidate as the current pasting position that is shifted and its A method for changing from a candidate position to an appropriate position may be displayed together.

  According to such a configuration, even a user with insufficient anatomical expertise can determine the attachment position of the surface electromyogram electrode. In addition, it becomes easy to determine the attachment position of the surface electromyogram electrode at an appropriate position that varies depending on the physique and other individual differences.

(Embodiment 3)
In the first embodiment, even when the sensor apparatus 100 and the receiving computer 1000 exchange measurement data by wireless communication, it is possible to acquire measurement data of myoelectric potential and measurement data such as acceleration in synchronization. The configuration has been described.

  In the third embodiment, in the configuration of the first embodiment, the user can easily confirm the coincidence or difference when the time change of the myoelectric potential is compared with a typical pattern. Will be described.

  That is, the data obtained by the myoelectric sensor is influenced by individual differences and the environment at the time of measurement due to the electrical resistance of the skin and many other parameters. Therefore, when processing and comparing a plurality of data, instead of directly handling raw data, an interval integration method or frequency analysis is used.

  However, these methods do not strictly process time-series change patterns. Therefore, in order to perform a comparison that is more in-depth than a bird's-eye comparison, medical expertise that enables direct judgment by looking at the waveform is required.

  In addition, pattern recognition using a neural network is sometimes used as an automatic process using a computer, but it is used to recognize data obtained based on a series of movements as one action pattern. There are many.

  On the other hand, in the present embodiment, it is necessary to analyze how the state transition is performed in time series in a motion in which a plurality of signals (myoelectricity, acceleration, angular velocity, etc.) cooperate with each other. is there.

  On the other hand, when a change in time-series measurement data is regarded as a transition between predetermined states, as described in the second embodiment, this is expressed by a hidden Markov model, and a state transition sequence is represented by its likelihood. For example, in the technical field such as voice recognition, the process of extracting together is already realized.

  Therefore, in the third embodiment, when a plurality of signals are measured as time series data, the receiving computer 1000 executes processing for recognizing such time series data as transitions between a plurality of states.

  FIG. 10 is a conceptual diagram showing the state transition of such a time series.

  FIG. 10A shows a waveform of one physical quantity among data actually measured for a specific user, and FIG. 10B shows a measurement performed on a plurality of subjects in advance with respect to exercise performed by the user. An ideal waveform for the corresponding physical quantity obtained from the obtained data is shown.

  Here, the ideal waveform and the actually measured waveform are not particularly limited. For example, the above-described “weighted finite state transducer” is used. When comparing the measurement result with an ideal waveform model learned using a large amount of data, the amplitude and time length data cannot be compared as they are, so the state transition and likelihood are compared as a state transition model by a hidden Markov model.・ Calculate the degree of coincidence derived from the posterior probability. In order to develop the measured data into such a “weighted finite state transducer”, it is possible to use not only the method disclosed in Document 3 described above, but also a multilayer perceptron.

  Such a multilayer perceptron is disclosed in, for example, the following documents.

Reference 4: Wen-Yuan Chen, Sin-Horng Chen, Cheng-Jung Lin, "A speech recognition method based on the sequential multi-layer perceptrons", Neural Networks Volume 9, Issue 4, June 1996, pp.655-669
FIG. 10 is a diagram showing a concept in the case where one physical quantity is measured. In practice, this is performed by multidimensional signal processing corresponding to the measurement of a plurality of physical quantities.

  In the example of FIG. 10, the ideal waveform corresponds to the transition of state A → state B → state C, while the actually measured waveform is state A → state B → state X → state Y. An example analyzed to correspond to a transition is shown.

  FIG. 11 is a conceptual diagram showing a result of extracting state transitions for measurement results of myoelectric potential and acceleration (and / or angular velocity) for a healthy subject and a patient for a predetermined exercise, for example.

  In FIG. 11, when comparing the waveforms of a healthy person and a patient, the normalization is performed using the time length from the start to the end of the exercise instructing the time length.

  As shown in FIG. 11, the receiving computer 1000 has a state transition of state A → state B → state C → state D → state E for a healthy person, whereas a patient substitutes for state B. The score (likelihood) when the state Z is replaced is displayed as the maximum. Further, the receiving computer 1000 also displays the fitting result assuming that the state transition itself of the healthy person is fitted in the patient, that is, assuming that the state B is the state B after the state A. Also good. In this case, it is displayed that the period of the state B is shorter than the ideal state and the score is also low.

  Note that such high and low scores may be displayed in different colors when an image as shown in FIG. 11 is displayed on the display.

  Further, in the comparison between the ideal waveform and the actually measured waveform, a method using the “continuous distribution type hidden Markov model” as described above may be used.

  FIG. 12 is a flowchart for explaining the flow of processing when performing the exercise state evaluation processing as described above.

  As shown in FIG. 12, compared with the case of FIG. 4 of the first embodiment, in the third embodiment, after an end instruction is transmitted from the receiving computer 1000 to the sensor device main body 100 (S 130). The process is the same as that in FIG. 4 except that the exercise state evaluation process (S141) is executed.

  FIG. 13 is a flowchart for explaining the exercise state evaluation process in FIG.

  The receiving computer 1000 acquires each data of time change of myoelectricity and acceleration (and further angular velocity if necessary) from a storage device such as the hard disk 2080 (S300), and myoelectricity and acceleration (and further angular velocity if necessary). Is expanded to a weighted finite state transducer (S302).

  Subsequently, the receiving computer 1000 calculates a score of each state by comparison with an ideal state transition according to the result of development (S304). The obtained state transition / each state score is presented to the user alone or with the state transitions of the motion model arranged side by side (S306).

With the configuration as described above, even a user with insufficient advanced and specialized medical knowledge can recognize at which timing a reaction different from that of a healthy person is occurring when performing any exercise. .
(Embodiment 4)
In the first embodiment, even when the sensor device 100 and the receiving computer 1000 exchange measurement data by wireless communication, it is possible to acquire measurement data of myoelectric potential and measurement data such as acceleration in synchronization. A description has been given of a simple configuration.

  In the fourth embodiment, a configuration capable of reducing power consumption as a sensor system in the configuration of the first embodiment will be described.

  Therefore, the hardware configuration of sensor device 100 and receiving computer 1000 is in principle the same as that of the first embodiment. As described below, acceleration sensor 120 and receiving computer are mainly controlled by microcomputer 110. A function is added to the operation program of the microcomputer 110 and the operation program of the receiving computer 1000 so that 1000 is woken up from the sleep state to the active state.

  That is, all the sensors (myoelectric sensors including the myoelectric amplifier 134, acceleration sensor 120, etc.) always measure data, continue to process and record, even when valid data cannot be obtained, In addition to consuming processing resources, wasteful power is consumed.

  When a person exercises, it always involves contraction of muscle fibers. And when muscle fibers contract, muscle discharge always occurs. By receiving this signal by the myoelectric sensor, it is possible to detect the contraction trigger of the myofiber.

  FIG. 14 is a conceptual diagram for explaining such a wake-up configuration.

  When measuring a person's movement using an acceleration / angular velocity sensor, circuits other than the myoelectric sensor and the microcomputer 110 are usually put to sleep, and each sensor is woken up after detecting this contraction trigger. Thereby, the power consumption of the whole circuit can be suppressed.

  That is, an electrode is affixed to enable measurement of muscle fiber contraction ((1) in FIG. 14). Only when the myoelectric data is measured, the myoelectric amplifier is automatically turned on and put into operation ((2) in FIG. 14). The microcomputer 110 continues to acquire measurement data through the myoelectric signal through the myoelectric amplifier ((3) in FIG. 14).

  The sensors for acceleration, angular velocity, etc. for measuring, processing, and recording a physical movement of a person and the receiving computer 1000 are usually in a sleep state. When the contraction of the muscle fibers exceeds a certain level and the microcomputer 110 determines that the threshold value has been exceeded, the acceleration sensor is woken up ((4) in FIG. 14) and the receiving computer 1000 is woken up (FIG. 14). (5)). And power consumption and use resources are reduced by making it sleep again as needed, such as when there is no signal for a fixed period.

  FIG. 15 is a flowchart for explaining the flow of processing when the wake-up processing as described above is performed.

  As shown in FIG. 15, first, a synchronization signal or a time signal is transmitted from the receiving computer 1000 to the sensor device main body 100 arranged in a wired or extremely short distance (S100). At this time, when a plurality of sensor device main bodies 100 are used for measurement, a plurality of synchronization signals or time signals are simultaneously transmitted. Thereafter, the receiving computer 1000 enters a sleep state until the start of measurement (S101).

  When the sensor device main body 100 receives the synchronization signal or the time signal, the microcomputer 110 is activated to reset the count operation of the clock generation unit 104 (S102).

  As described above, when measurement is started, the microcomputer 110 sets, for example, the variable n to 1 in order to specify the number of measurements (S106).

  The microcomputer 110 acquires T = clock (t) as the measurement time T (S107), receives the myoelectric signal (S108) amplified and A / D converted by the myoelectric amplifier 134, and the myoelectric potential is, for example, a predetermined value. When it is detected that the change has exceeded the threshold value (S111), a wakeup signal is transmitted to the receiving computer 1000 and the acceleration sensor 120 (S112), and the receiving computer 1000 and the acceleration sensor 120 are activated. The microcomputer 110 receives the myoelectric signal (S108) amplified and A / D converted by the myoelectric amplifier 134, and receives the acceleration signal (S109) amplified and A / D converted by the acceleration sensor 120, and receives sensor information. Is acquired (S110).

  The microcomputer 110 adds a time stamp to the measurement data corresponding to the myoelectric signal and acceleration signal acquired in a lump and transmits the measurement data (S122).

  The subsequent processing is the same as the processing in FIG.

  In FIG. 15, the receiving computer 1000 and the acceleration sensor 120 are configured to be woken up. However, only the acceleration sensor 120 may be woken up.

  With the above configuration, the power consumption of the sensor system can be reduced. In particular, the reduction in power consumption of the sensor device main body 100, which is difficult to mount a large-capacity battery due to the limitation in shape, results in an increase in measurable time.

  In the above description, each of the second to fourth embodiments has been described on the premise of the sensor system of the first embodiment. However, Embodiments 2 to 4 are not mutually exclusive, and any of these can be combined to form a sensor system.

  Embodiment disclosed this time is an illustration of the structure for implementing this invention concretely, Comprising: The technical scope of this invention is not restrict | limited. The technical scope of the present invention is shown not by the description of the embodiment but by the scope of the claims, and includes modifications within the wording and equivalent meanings of the scope of the claims. Is intended.

  DESCRIPTION OF SYMBOLS 100 Sensor apparatus main body, 102 Clock I / F, 104 Clock generation part, 106 Data input / output I / F, 110 Microcomputer, 120 Accelerometer, 130.1, 130.2 Electrode, 132 Sensor I / F, 134 Myoelectric amplifier , 1000 Receiving computer, 2010 Computer main body, 2020 Memory disk drive, 2030 Optical disk drive, 2040 CPU, 2050 Bus, 2060 ROM, 2070 RAM, 2080 Hard disk, 2100 Keyboard, 2110 Mouse, 2120 Display, 2210 Memory card, 2200 Optical disk.

Claims (7)

A sensor device for measuring changes in physical quantity and myoelectric potential associated with movement of a measurement object,
A clock generation means that is reset in response to an external instruction;
First measurement means for making a myoelectric potential data a signal from an electrode that is mounted on the measurement object and measures the myoelectric potential;
A second measuring means for making a change in the physical quantity generated in the sensor device mounted on the measurement object into measurement data;
The myoelectric potential data from the first measuring means and the measurement data from the second measuring means are acquired in a lump, and a time stamp from the clock generating means is added to the outside by wireless communication. Control means for transmitting ,
Said second measuring means, in response to the measurement signal from the first measuring means, Ru is activated by the wake-up signal given from the control unit, the sensor device. A measurement system,
A sensor device for measuring a change in physical quantity and myoelectric potential accompanying movement of a measurement object,
The sensor device includes:
A clock generation means that is reset in response to an external instruction;
First measurement means for making a myoelectric potential data a signal from an electrode that is mounted on the measurement object and measures the myoelectric potential;
A second measuring means for making a change in the physical quantity generated in the sensor device mounted on the measurement object into measurement data;
The myoelectric potential data from the first measuring means and the measurement data from the second measuring means are acquired in a lump, and a time stamp from the clock generating means is added to the outside by wireless communication. Control means for transmitting,
A data management device including a storage device for giving an instruction to the clock generation unit to the sensor device and storing the myoelectric potential data, the measurement data, and the time stamp transmitted from the sensor device in association with each other Further comprising
The storage device measures, for each muscle to be measured, an electrode for measuring the myoelectric potential and an image showing a mounting position of the sensor device, and the electrode to be attached to the muscle to be measured. And discriminating data for discriminating the measurement pattern when
The data management device includes:
Presenting means for presenting the image to the user before starting measurement,
For the measurement data at the electrode position that is currently mounted, further includes a determination means for determining the mounting state based on the determination data,
It said presenting means, you come discrimination result of the discriminating means, the measuring system. The discrimination means discriminates the movement of the muscle as a transition of a plurality of time-series states based on the measured myoelectric potential data and the measurement data,
The measurement system according to claim 2 , wherein the presenting unit displays the transition of the plurality of time-series states. Said second measuring means, the first in response to the measurement signal from the measuring means, is activated by a wakeup signal supplied from said control means, according to claim 2 or 3, wherein the measurement system. The measurement system according to claim 4 , wherein the data management device enters a sleep state after giving an instruction to the clock generation unit, and is activated by the wake-up signal. A measurement program that is executed by an arithmetic unit of a data management device that controls a measurement operation of a measurement system having a sensor device for measuring a change in physical quantity and a myoelectric potential associated with movement of a measurement object,
The sensor device includes a clock generation unit that is reset in response to an instruction based on the measurement program, and a first signal for using myoelectric potential data as a signal from an electrode that is attached to the measurement target and measures the myoelectric potential. Measuring means, second measuring means for using the change in the physical quantity generated in the sensor device mounted on the measuring object as measurement data, the myoelectric potential data from the first measuring means, and the first Control means for acquiring the measurement data from the two measurement means in a lump, adding a time stamp from the clock generation means, and transmitting to the outside by wireless communication,
Transmitting a signal instructing the clock generation means to reset to the sensor device;
Receiving the myoelectric potential data, the measurement data, and the time stamp transmitted from the sensor device, and storing them in a storage device in association with each other .
The storage device measures, for each muscle to be measured, an electrode for measuring the myoelectric potential and an image showing a mounting position of the sensor device, and the electrode to be attached to the muscle to be measured. And discriminating data for discriminating the measurement pattern when
The measurement program is
Presenting the image to the user before starting the measurement;
A step of determining a mounting state based on the determination data with respect to measurement data at a currently mounted electrode position;
The step of presenting the determination result to the user, further, Ru is executed by the arithmetic unit, measuring program. Based on the measured myoelectric potential data and the measured data, determining the movement of the muscle as a transition of a plurality of time-series states,
The measurement program according to claim 6 , further causing the arithmetic device to execute a step of presenting the transition of the plurality of time-series states to the user.