JP6802467B2 - Processing device, judgment device and appropriate range identification method - Google Patents

Description

The present invention is a processing device that specifies an appropriate range when determining a matter related to swallowing using a determination parameter within an appropriate range, a determination device that is provided with the processing device and determines a matter related to swallowing, and within the appropriate range. It relates to an appropriate range specifying method for specifying an appropriate range when determining a matter related to swallowing using the determination parameters of.

As a device for measuring a measured amount related to muscle activity when swallowing food and drink, a swallowing activity measuring device disclosed by the applicant in Patent Document 1 below is known. This swallowing activity measuring device is configured to include a sensor sheet, a recorder unit, a measurement data analysis unit, a muscle activity analysis unit, and a recording / display unit. In this swallowing activity measuring device, electric signals output from a plurality of electromyographic electrodes and a plurality of vibration pickups arranged on a sensor sheet attached to the anterior neck of the subject are stored in a storage device, and the electric signals are stored in the storage device. The measurement data analysis unit performs the filtering process. Next, the muscle activity analysis unit determines the activity duration of the muscle group related to swallowing and the order of the start and end of the activity of the muscle group (determination of matters related to swallowing) based on the electrical signal filtered by the measurement data analysis unit. (Parameter used for) is calculated.

Japanese Unexamined Patent Publication No. 2014-168490 (Pages 7-8, Fig. 1)

However, the above-mentioned conventional swallowing activity measuring device has the following problems to be improved. That is, in this swallowing activity measuring device, the measurement data analysis unit filters the electric signals output from the EMG electrodes and the vibration pickup, and calculates the activity duration and the like based on the electric signals after the filtering. There is. In this case, the electrical signal after filtering also contains noise. In addition, it may contain signal components associated with movements different from swallowing such as mastication. Therefore, due to these effects, the value of the determination parameter such as the activity duration used for determining the swallowing-related matter may be an inappropriate value for the determination. Specifically, due to the influence of noise, a time shorter than the actual activity duration may be calculated as the activity duration. However, since the conventional swallowing activity measuring device does not have a means for grasping the appropriate range of the value of the determination parameter, the swallowing-related matter is determined by using the determination parameter of the inappropriate value. It may be difficult to make an accurate judgment result, and improvement in this respect is desired.

The present invention has been made in view of such a problem, and mainly provides a processing device, a determination device, and an appropriate range specifying method capable of specifying an appropriate range of values of a determination parameter used for determining a matter related to swallowing. The purpose.

The processing device according to claim 1 in order to achieve the above object is the determination parameter when the value of the determination parameter calculated based on the measurement parameter measured by attaching a sensor to the neck of a person is within an appropriate range. The processing unit is provided with a processing unit that specifies the appropriate range when determining a matter related to swallowing, and the processing unit statistically processes a plurality of the determination parameters calculated based on the measurement parameters measured a plurality of times. A processing device that specifies a first threshold value as the minimum value of the appropriate range based on the processing result obtained, and specifies a range in which the value of the determination parameter is equal to or greater than the first threshold value as the appropriate range . The processing unit executes a process for obtaining a mixed Gaussian distribution of each determination parameter as the statistical process, and specifies the first threshold value based on the result obtained from the mixed Gaussian distribution as the processing result.

Further, the processing apparatus according to claim 2 is the processing apparatus according to claim 1 , wherein the processing unit has a probability density distribution of the first distribution having the maximum mixing ratio in the mixed Gaussian distribution and an adjacent first distribution. The average value of the first distribution and the average of the second distribution among the values of the determination parameters that are equal to the probability density distribution of the second distribution whose average value is smaller than the average value of the first distribution. The value of the determination parameter located between the value and the value is specified as the first threshold value.

Further, the processing apparatus according to claim 3, in the processing apparatus according to claim 1 or 2, wherein the processing unit identifies a second threshold value to the maximum value of the proper range based on the processing result, the determination A range in which the value of the parameter is equal to or higher than the first threshold value and equal to or lower than the second threshold value is specified as the appropriate range.

The processing apparatus according to claim 4, wherein, in the processing apparatus according to claim 3, wherein the processing unit executes a process for obtaining a mixed Gaussian distribution of the respective determination parameter as said statistical process, as the process result The second threshold is specified based on the results obtained from the mixed Gaussian distribution.

Further, the processing apparatus according to claim 5 is the processing apparatus according to claim 4 , wherein the processing unit has a value of a probability density distribution of a first distribution having a maximum mixing ratio in the mixed Gaussian distribution and the first distribution. The average value of the first distribution and the mean value of the first distribution among the values of the determination parameters that are adjacent to and have the same mean value as the value of the probability density distribution of the third distribution that is larger than the average value of the first distribution. The value of the determination parameter located between the average value of the three distributions is specified as the second threshold value.

Further, the determination device according to claim 6 uses the processing device according to any one of claims 1 to 5 and the determination parameter within the appropriate range specified by the processing device to perform matters related to swallowing. It is provided with a determination unit for determination.

Further, the method for specifying the appropriate range according to claim 7 uses the determination parameter when the value of the determination parameter calculated based on the measurement parameter measured by attaching the sensor to the neck of a person is within the appropriate range. a appropriate range specification method Ru is specific to the processing unit in the processing apparatus according to the appropriate range at the time of determining the matters relating swallowing to any one of claims 1 to 5 Te, to the processing unit, a plurality Based on the processing results obtained by statistically processing a plurality of the determination parameters calculated based on the measurement parameters measured once, the first threshold value to be the minimum value of the appropriate range is specified, and the determination parameters A range having a value equal to or higher than the first threshold value is specified as the appropriate range , and a process for obtaining a mixed Gaussian distribution of each determination parameter is executed as the statistical process, and is obtained from the mixed Gaussian distribution as a result of the process. The first threshold is specified based on the result .

According to the processing device according to claim 1, the determination device according to claim 6 , and the appropriate range specifying method according to claim 7, a plurality of determination parameters calculated based on measurement parameters measured a plurality of times are statistically processed. By specifying the first threshold value which is the minimum value of the appropriate range based on the processing result obtained in the above procedure, the appropriate range of the determination parameter used for determining the matter related to swallowing can be surely specified. Therefore, according to this processing device, the determination device, and the appropriate range identification method, it is difficult to determine whether or not it is appropriate to use the calculated determination parameter for determining a matter related to swallowing (for example,). Even in a situation where there is no expert who can determine whether or not it is appropriate), is it appropriate to use the determination parameter for determining matters related to swallowing by comparing the appropriate range with the determination parameter? As a result of being able to reliably and easily determine whether or not it is possible, it is possible to accurately determine matters related to swallowing.

Further, this processing apparatus, a result according to the determination device and the appropriate range specification method this, which executes the processing for obtaining a mixed Gaussian distribution of each determination parameter as statistical processing, were obtained from a mixed Gaussian distribution as the processing result By specifying the first threshold value based on the above, a highly reliable first threshold value can be specified as the minimum value in the appropriate range.

Further, according to the processing apparatus according to claim 2 and the determination apparatus according to claim 6 , the probability density distribution of the first distribution having the maximum mixing ratio in the mixed Gaussian distribution and the distribution adjacent to the first distribution. Judgment that the mean value is smaller than the mean value of the first distribution and the probability density distribution of the second distribution is equal. Judgment located between the mean value of the first distribution and the mean value of the second distribution among the values of the judgment parameters. By specifying the value of the parameter as the first threshold value, a more reliable first threshold value can be specified as the minimum value in the appropriate range.

The processing apparatus according to claim 3, and according to the determination apparatus of claim 6, the second threshold value to the maximum value of the proper range based on the processing result obtained by statistically processing a plurality of determination parameter By specifying, it is possible to specify an appropriate range in which both extremely short judgment parameters and extremely long judgment parameters can be excluded from the judgment parameters used for judgment, so that the judgment parameters are swallowed. It is possible to more reliably determine whether or not it is appropriate to use it for determining matters related to.

Further, according to the processing apparatus according to claim 4 and the determination apparatus according to claim 6 , a process for obtaining a mixed Gaussian distribution of each determination parameter is executed as statistical processing, and the mixed Gaussian distribution as a processing result is obtained. By specifying the second threshold value based on the result, it is possible to specify a highly reliable second threshold value as the maximum value in the appropriate range.

Further, according to the processing apparatus according to claim 5 and the determination apparatus according to claim 6 , the value of the probability density distribution of the first distribution having the maximum mixing ratio in the mixed Gaussian distribution and the distribution adjacent to the first distribution. The mean value is larger than the mean value of the first distribution. The value of the probability density distribution of the third distribution is equal. Among the values of the judgment parameters, between the mean value of the first distribution and the mean value of the third distribution. By specifying the value of the determination parameter located in, as the second threshold value, a more reliable second threshold value can be specified as the maximum value in the appropriate range.

It is a block diagram which shows the structure of the determination device 1. It is a top view which shows the structure of the sensor sheet 11. It is explanatory drawing which shows the wearing state of the determination target person 100 (subject 300) wearing a sensor sheet 11. It is 1st explanatory drawing explaining the calculation method of activity time Ta. It is a 2nd explanatory diagram explaining the calculation method of the activity time Ta. It is explanatory drawing explaining the method of specifying an appropriate range Tf.

Hereinafter, the forms of the processing device, the determination device, and the appropriate range specifying method will be described with reference to the attached drawings.

First, the configuration of the determination device 1 shown in FIG. 1 as an example of the determination device will be described. The determination device 1 is configured to be able to determine matters related to swallowing. Specifically, in the determination device 1, as a matter related to swallowing, whether the person to be determined (hereinafter, also referred to as “determination target person 100”) is a person with reduced swallowing function (a person with reduced swallowing function) or swallowing. It is possible to determine whether a person has normal function (a person whose swallowing function has not deteriorated) as a matter related to swallowing (hereinafter, also referred to as "swallowing function determination").

Further, as shown in FIG. 1, the determination device 1 includes a sensor sheet 11, a microphone 12, an operation unit 13, a storage unit 14, a display unit 15, and a processing unit 16. In this case, the processing device is composed of the storage unit 14 and the processing unit 16.

As shown in FIG. 2, the sensor sheet 11 includes an adhesive sheet 21, and a plurality of (eight as an example) electromyographic electrodes 22a to 22h (an example of a sensor, hereinafter, when not distinguished, “electromyographic electrodes 22”. (Also referred to as), and a mounting index 23. One side of the adhesive sheet 21 is an adhesive surface. In the figure, the sensor sheet 11 is shown as viewed from the surface opposite to the adhesive surface of the adhesive sheet 21.

The electromyogram electrode 22 detects a voltage generated by muscle activity when swallowing food and drink, and outputs an electric signal S1 indicating a change with time (measurement parameter) of the voltage. In this sensor sheet 11, as shown in FIG. 2, as an example, four sets of two EMG electrodes 22 as one set (a total of eight: a set of EMG electrodes 22a and 22b, an EMG electrode 22c). , 22d, EMG electrodes 22e, 22f, EMG electrodes 22g, 22h) EMG electrodes 22 are provided. Further, each EMG electrode 22 is fixed to the adhesive surface of the adhesive sheet 21.

As shown in FIG. 3, the index 23 is an index (mark) for aligning the sensor sheet 11 when the sensor sheet 11 is attached (attached) to the neck Ne of a person (determination target person 100 or subject 300 described later). Therefore, as shown in FIG. 2, it is composed of notches formed at the edges of the adhesive sheet 21.

In this case, in this sensor sheet 11, as shown in FIG. 3, the adhesive sheet 21 is attached to the neck Ne so that the index 23 is located near (outside) the Adam's apple Lp (Adam's apple) in the human neck Ne. (Hereinafter, this state is also referred to as a "specified state"), so that each EMG electrode 22 faces the measurement target muscle (muscle involved in swallowing), each EMG electrode 22 and the index 23 The position is specified. Specifically, in the specified state, each electromyogram electrode 22 is located in the vicinity of the sternothyroid muscle, the sternohyoid muscle, the thyrohyoid muscle, and the mentalis muscle (all not shown) as examples of the muscles to be measured. Are configured to face each other. More specifically, in the specified state, the EMG electrodes 22a and 22b face the vicinity of the sternothyroid muscle, the EMG electrodes 22c and 22d face the vicinity of the sternohyoid muscle, and the EMG electrodes 22e , 22f are configured to face the vicinity of the sternohyoid muscle, and the electromyogram electrodes 22g, 22h are configured to face the vicinity of the mentalis muscle.

The microphone 12 is another example of the sensor, and as shown in FIG. 3, when the food or drink passes through the pharynx when swallowing, it is attached by an adhesive tape near the laryngeal ridge Lp in the human neck Ne. (Swallowing sound) is detected, and an electric signal S2 indicating a time-dependent change (measurement parameter) of the voltage according to the loudness of the swallowing sound is output.

The operation unit 13 is configured to enable various operations and outputs an operation signal according to the operation.

The storage unit 14 stores the activity time Ta (corresponding to the determination parameter) calculated by the calculation process described later executed by the processing unit 16. Further, the storage unit 14 stores the threshold value Tt1 (corresponding to the first threshold value) specified by the specific processing described later executed by the processing unit 16.

The display unit 15 displays images of the waveforms of the electric signals S1 and S2 output from the sensor sheet 11 and the result of the determination process executed by the processing unit 16 under the control of the processing unit 16.

The processing unit 16 performs various processes according to the operation signal output from the operation unit 13. Specifically, the processing unit 16 measures the time-dependent change (measurement parameter) of the voltage specified by the electric signals S1 and S2 output from the electromyographic electrode 22 and the microphone 12 of the sensor sheet 11, respectively. Further, the processing unit 16 executes a calculation process for calculating the activity time Ta (determination parameter) of the muscle activity associated with swallowing based on the time-dependent change of the measured voltage. Further, the processing unit 16 executes a specific process for specifying the threshold value Tt1 as the minimum value of the appropriate range Tf of the activity time Ta. In addition, the processing unit 16 functions as a determination unit, and determines the swallowing function (determination of matters related to swallowing) using the activity time Ta within the appropriate range Tf.

As described above, the determination device 1 is based on the time-dependent change (measurement parameter) of the voltage specified by the electric signals S1 and S2 output from the electromyographic electrode 22 and the microphone 12 attached to the human neck Ne. It is possible to determine the swallowing function (determination of matters related to swallowing) of the determination target person 100 by using the activity time Ta (determination parameter) of the muscle activity calculated as described above. Further, in this determination device 1, as shown in FIG. 4, the start time Ts and the end time (rise time and fall time) when the voltage value Vm of the voltage specified by the electric signal S1 crosses the reference value Vs are set. The activity time Ta is calculated as the time Te.

Here, since the electric signal S1 generally contains noise, as shown in FIG. 5, the voltage value Vm crosses the reference value Vs in a short time due to the noise, and the muscle occurs during swallowing. The time shorter than the actual activity time during which is active may be calculated as the activity time Ta. Therefore, in order to make a correct judgment, when the activity time Ta is extremely short (when the activity time Ta is outside the appropriate range Tf), the activity time Ta is excluded from the activity time Ta used for the judgment (filtering). There is a need to do. In this determination device 1, it is possible to specify the appropriate range Tf of the activity time Ta used for the determination. Hereinafter, a method for specifying the appropriate range Tf of the activity time Ta will be described.

A plurality of subjects 300 (19 in this example) are selected to specify the appropriate range Tf. Next, the measurement parameters when each subject 300 swallows food and drink are measured and stored. Specifically, as shown in FIG. 3, the adhesive sheet of the sensor sheet 11 so that the target position of the subject 300 (in this example, the position on the left side of the adam's apple Lp) and the attachment index 23 of the sensor sheet 11 coincide with each other. 21 is attached to the anterior neck An in the neck Ne of the subject 300. At this time, the electromyogram electrode 22 of the sensor sheet 11 faces the measurement target muscle of the subject 300. In addition, the microphone 12 is attached to the vicinity of the laryngeal ridge Lp in the neck Ne of the subject 300 using an adhesive tape.

Next, the operation unit 13 is operated to instruct the processing unit 16 to start the processing, and then instruct the subject 300 to swallow food and drink. When the subject 300 swallows food or drink according to the instruction, the electromyographic electrode 22 of the sensor sheet 11 detects the voltage generated by the muscle activity during swallowing, and indicates the time-dependent change of the voltage as a measurement parameter. The signal S1 is output. Further, the microphone 12 detects the swallowing sound during swallowing and outputs an electric signal S2 indicating a time-dependent change in voltage according to the loudness of the swallowing sound as a measurement parameter.

Next, the processing unit 16 executes a calculation process for calculating the activity time Ta as a determination parameter based on the electric signals S1 and S2 output from the EMG electrode 22 and the microphone 12, respectively. In this calculation process, the processing unit 16 first measures the time when the swallowing sound is maximized (the peak time of the swallowing sound) based on the electric signal S2.

Subsequently, the processing unit 16 measures the voltage value Vm (voltage associated with muscle activity) based on the electric signal S1 and specifies the rise time at which the voltage value Vm rises above a predetermined reference value Vs. In this case, when there are a plurality of rise times, a rise time in which the voltage value Vm is less than the reference value Vs is specified for a predetermined time Tr1 or more consecutively before each rise time, and the swallowing sound peak is the most among them. The rise time close to the time is defined as the start time Ts of the activity time Ta.

Further, after the start time Ts, the time when the voltage value Vm falls from the reference value Vs or more to less than the reference value Vs and becomes less than the reference value Vs is specified as a continuous fall time of Tr2 or more for a predetermined time. , The fall time closest to the start time Ts is set as the end time Te of the activity time Ta. As an example, the reference value Vs can be twice the average value of the voltage value Vm in the time zone when there is no reaction related to swallowing.

Next, the processing unit 16 calculates the activity time Ta from the start time Ts and the end time Te. Subsequently, the processing unit 16 stores the activity time Ta in the storage unit 14. As described above, one calculation process for calculating one activity time Ta for one subject 300 is completed. Next, the second calculation process is executed for the same subject 300. In this example, for one subject 300, the calculation process of swallowing two kinds of foods and drinks having different hardness (difficulty of swallowing) is performed 6 times for each food and drink (12 for one subject 300). (Times) It shall be executed. Therefore, in this example, it is assumed that the calculation process is executed a total of 228 times for 19 subjects 300.

Subsequently, when the calculation processing for all the subjects 300 is completed, the operation unit 13 is operated to specify the threshold value Tt1 which is the minimum value of the appropriate range Tf of the activity time Ta suitable for use in determining the swallowing function. Instructs the processing unit 16 to execute the specific processing. In this specific processing, the processing unit 16 specifies the threshold value Tt1 based on the processing result obtained by statistically processing the plurality of activity time Tas calculated for the plurality of subjects 300. In this case, the processing unit 16 executes a process of obtaining a mixed Gaussian distribution of each activity time Ta as a statistical process.

Here, the following three are assumed when the threshold value Tt1 is specified by the process of obtaining the mixed Gaussian distribution. First, it is assumed that the distribution of activity time Ta in one cluster follows the probability density distribution of one parameter, and that different clusters derive from the probability density distribution of different parameters. Second, it is assumed that the overall probability density distribution for all the calculated activity time Ta follows a mixture distribution represented by a weighted linear sum of the plurality of probability density distributions. Third, it is assumed that the individual distributions that make up the overall distribution are all normal distributions.

Assuming the above three assumptions, the probability density distribution p (x) of the activity time Ta based on the mixture distribution is given by the following equation (1).
p (x) = Σ [i = 1, n] ξ _i N _i (x: μ _i σ _i ² ) ... (1)
Here, x is a random variable corresponding to the calculated activity time Ta, and n is the assumed number of distributions (number of clusters). Further, Ni (x: μ _i σ _i ² ) is the i-th distribution (normal distribution) having an average value of μ _i and a variance of σ _i ² . Further, ξ _i is a mixing ratio in the i-th distribution, and corresponds to the degree of contribution of the i-th distribution in the probability density distribution p (x).
Further, the total of the mixing ratios ξi in the n distributions is 1 as shown in the following equation (2).
Σ [n, i = 1] ξ _i = 1 ... (2)

In this determination device 1, the processing unit 16 expresses the overall probability density distribution p (x) of all the calculated activity time Tas by a mixed Gaussian distribution, that is, each parameter of the above equation (1), that is, each normal distribution. The mean value μ _i , the variance σ _i ² , and the mixing ratio ξ _i are specified by the EM algorithm. In this case, as shown in FIG. 6, the probability density distribution p (x) of each activity time Ta calculated by the above calculation process is represented by each curve shown in the figure by each parameter of the equation (1).

The processing unit 16, the appropriate range Tf outside the active time Ta derived from the distribution N _k than the average value of all active time Ta calculated with a sufficiently small average value _{μ i (1 ≦ k ≦ n} th) of the active time Ta, determines activity time Ta derived other distribution _{n k (n th + 1 ≦} k ≦ n) and activity time Ta in the proper range Tf. That is, the processing unit 16 sets any activity time Ta in the distribution _Nk as the threshold value Tt1.

Specifically, the processing unit 16 has a probability density distribution p (x) and a distribution k of a distribution having the maximum mixing ratio ξ _i (corresponding to the first distribution, hereinafter this distribution is also referred to as “distribution k”). Probability density of a distribution adjacent to and whose mean value μ _k-1 is smaller than the mean value μ _{k of the} distribution k (corresponding to the second distribution, hereinafter this distribution is also referred to as “distribution k-1”). Of the activity time Ta (that is, x satisfying the following equation (3)) in which the distribution p (x) is equal, the mean value μ _k-1 of the distribution k _-1 and the mean value μ _{k of the} distribution k ₎ Let x (activity time Ta corresponding to x) between them be the threshold Tt1.
ξ _k-1 N _k-1 (x: μ _k-1 , σ ² _k-1 ) = ξ _k N _k (x: μ _k , σ ² _k ) ... (3)
As shown in FIG. 6, in this example, about 0.25 seconds is specified as the threshold Tt1. In this case, in this example, the threshold value Tt1 as the minimum value of the appropriate range Tf is specified, and the maximum value of the appropriate range Tf is not specified. Therefore, in this example, the range of the threshold value Tt1 or more is the appropriate range Tf.

Next, the processing unit 16 stores the specified threshold value Tt1 in the storage unit 14, and ends the specific processing.

Next, a method of determining the swallowing function of the determination target person 100 using the determination device 1 will be described.

First, the measurement parameter for the determination target person 100 is measured by the same procedure as the measurement of the measurement parameter for the subject 300. Next, the processing unit 16 is made to execute the calculation process. Subsequently, the operation unit 13 is operated to cause the processing unit 16 to execute the swallowing function determination. In this swallowing function determination, the processing unit 16 reads the threshold value Tt1 from the storage unit 14, and then whether or not the activity time Ta calculated by the calculation process is equal to or greater than the threshold value Tt1 (whether or not it is within the appropriate range Tf). ) Is determined. In this case, when the processing unit 16 determines that the activity time Ta is less than the threshold value Tt1 (not within the appropriate range Tf), the processing unit 16 does not determine the swallowing function and determines that the activity time Ta is less than the threshold value Tt1 (activity time Ta). (It is inappropriate to use for discrimination) is displayed on the display unit 15.

On the other hand, when the processing unit 16 determines that the activity time Ta is equal to or greater than the threshold value Tt1 (within the appropriate range Tf), the processing unit 16 compares the activity time Ta with a predetermined determination reference time to determine the person to be determined. It is determined whether 100 is a person with decreased swallowing function (a person with decreased swallowing function) or a person with normal swallowing function (a person with not decreased swallowing function). Next, the processing unit 16 displays the determination result on the display unit 15 and ends the swallowing function determination.

As described above, according to the processing device, the determination device 1, and the appropriate range specifying method, a plurality of activity time Tas (determination parameters) calculated based on the time-dependent changes (measurement parameters) of the voltage measured a plurality of times are statistically measured. By specifying the threshold value Tt1 which is the minimum value of the appropriate range Tf based on the processing result obtained by the processing, the appropriate range Tf of the activity time Ta used for determining the matter related to swallowing can be surely specified. Therefore, according to this processing device, the determination device 1, and the appropriate range specifying method, it is difficult to determine whether or not it is appropriate to use the calculated activity time Ta for determining items related to swallowing. For example, even in a situation where there is no expert who can determine whether or not it is appropriate), it is appropriate to use the activity time Ta for determining matters related to swallowing by comparing the appropriate range Tf and the activity time Ta. As a result of being able to reliably and easily determine whether or not there is, it is possible to accurately determine matters related to swallowing.

Further, according to this processing device, the determination device 1, and the appropriate range specifying method, a process of obtaining a mixed Gaussian distribution of each activity time Ta is executed as a statistical process, and based on the result obtained from the mixed Gaussian distribution as the processing result. By specifying the threshold value Tt1, it is possible to specify the highly reliable threshold value Tt1 as the minimum value of the appropriate range Tf.

Further, according to this processing device, the determination device 1, and the appropriate range specifying method, the probability density distribution p (x) having the maximum mixing ratio ξ _i in the mixed Gaussian distribution and the distribution adjacent to the distribution k. The mean value μ _k-1 of the distribution k-1 is equal to the probability density distribution p (x) of the distribution k-1 whose mean value μ k-1 is smaller than the mean value μ _k of the distribution _k. By specifying the activity time Ta located between ₁ and the mean value μ _k of the distribution k as the threshold value Tt1, a more reliable threshold value Tt1 can be specified as the minimum value of the appropriate range Tf.

The determination device and determination method are not limited to the above configurations and methods, and can be changed as appropriate. For example, the probability density distribution p (x) having the maximum mixing ratio ξ _i in the mixed Gaussian distribution and the distribution adjacent to the distribution k whose mean value μ _k-1 is larger than the mean value μ _{k of the} distribution k. Of the activity time Ta at which the probability density distribution p (x) of the small distribution k-1 is equal, the activity time Ta between the mean value μ _k-1 of the distribution k _-1 and the mean value μ _k of the distribution k Was described above as an example of specifying as the threshold value Tt1, but the distribution in which the mixing ratio ξ _i in the mixed Gaussian distribution is adjacent to the maximum distribution k and the mean value μ _k-1 is smaller than the mean value μ _{k of the} distribution k. It is also possible to adopt a configuration and method in which the mean value μ _k-1 of _k-1 is specified as the threshold value Tt1.

Further, a configuration and a method for specifying the threshold value Tt1 by statistical processing other than the processing for obtaining a mixed Gaussian distribution can also be adopted. For example, a configuration and method in which the threshold value Tt1 is set to μ−σ calculated from the average value μ of all the calculated activity time Tas and the standard deviation σ can be adopted. Further, it is also possible to adopt a configuration and a method of specifying the threshold value Tt1 by the Smirnov-Grubbs (SG) Test, which is a method of determining whether or not a certain value in the data group is an outlier.

Further, in the above configuration and method, only the threshold value Tt1 as the minimum value of the appropriate range Tf is specified, but the maximum value of the appropriate range Tf is based on the processing result obtained by statistically processing a plurality of activity time Tas. A configuration and method can be adopted in which the threshold value Tt2 (corresponding to the second threshold value) is specified, and the range in which the value of the activity time Ta is equal to or greater than the threshold value Tt1 and equal to or less than the threshold value Tt2 is set as the appropriate range Tf. According to this configuration and method, an appropriate range Tf that can exclude (filter) both extremely short activity time Ta and extremely long activity time Ta from the activity time Ta used for determination is specified. (The appropriate range Tf can be strictly defined), so that it is possible to more reliably determine whether or not it is appropriate to use the activity time Ta for determining matters related to swallowing.

Further, when specifying the threshold value Tt2 described above, a process of obtaining a mixed Gaussian distribution of each activity time Ta is executed as a statistical process, and the threshold value Tt2 is specified based on the result obtained from the mixed Gaussian distribution as the processing result. Can be done. According to this configuration and method, a highly reliable threshold value Tt2 can be specified as the maximum value of the appropriate range Tf.

Also, when specifying the threshold Tt2 based on the results obtained from the Gaussian mixture, a probability density distribution p of the largest distribution k mixing ratio xi] _i is the Gaussian mixture (x), a distributed adjacent to distribution k The mean of the distribution k of the activity time Ta at which the mean value μ _{k + 1} is equal to the probability density distribution p (x) of the distribution k + 1 (corresponding to the third distribution) that is larger than the mean value μ _k of the distribution k. The activity time Ta between the value μ _k and the mean value μ _{k + 1} of the distribution k + 1 can be specified as the threshold Tt2. According to this configuration and method, a more reliable threshold value Tt2 can be specified as the maximum value of the appropriate range Tf.

Further, when specifying the threshold value Tt2, a configuration and a method of specifying the average value μ _{k + 1} of the distribution _{k + 1} as the threshold value Tt2 can also be adopted. Further, similarly to the above-mentioned specification of the threshold value Tt1, a configuration and a method for specifying the threshold value Tt2 by statistical processing other than the processing for obtaining the mixed Gaussian distribution can also be adopted. For example, it is possible to adopt a configuration and method in which the threshold value Tt2 is set to μ + σ calculated from the average value μ of all the calculated activity time Tas and the standard deviation σ. It is also possible to adopt a configuration and method for specifying the threshold value Tt2 by the Smirnov-Grubbs (SG) Test.

Further, the number of electromyographic electrodes 22 (sensors) arranged on the sensor sheet 11 is not limited to the eight illustrated in the above configuration, and can be arbitrarily specified. Further, the example of calculating the activity time Ta as a judgment parameter based on the time course of the voltage measured as a measurement parameter has been described above, but the judgment parameter is based on the vibration as a measurement parameter using a vibration pickup as a sensor. It is also possible to adopt a configuration and a method for calculating the amplitude and acceleration of the above.

Further, the example of determining the swallowing function as a matter related to swallowing using the determination device 1 has been described above, but as a matter related to swallowing, the food or drink to be determined is a food or drink that is easy to swallow (food or drink that is easy to swallow). It is also possible to use the determination device 1 to determine whether the food or drink is difficult to swallow (food or drink that is difficult to swallow).

1 Judgment device 12 Microphone 14 Storage unit 16 Processing unit 22 EMG electrode 100 Judgment target person 300 Subject k distribution k-1 distribution k + 1 distribution Ne Neck p (x) Probability density distribution Ta Activity time Tf Appropriate range Tt1 Threshold Tt2 Threshold μ _i mean value ξ _i mixing ratio

Claims (7)

When the value of the judgment parameter calculated based on the measurement parameter measured by attaching the sensor to the human neck is within the proper range, the appropriate range when judging matters related to swallowing using the judgment parameter Equipped with a processing unit to identify
The processing unit specifies a first threshold value as the minimum value of the appropriate range based on the processing result obtained by statistically processing a plurality of the determination parameters calculated based on the measurement parameters measured a plurality of times. A processing device that specifies a range in which the value of the determination parameter is equal to or greater than the first threshold value as the appropriate range .
The processing unit executes a process of obtaining a mixed Gaussian distribution of each determination parameter as the statistical process, and specifies the first threshold value based on the result obtained from the mixed Gaussian distribution as the processing result. .. In the processing unit, the probability density distribution of the first distribution having the maximum mixing ratio in the mixed Gaussian distribution and the second distribution which is adjacent to the first distribution and whose mean value is smaller than the mean value of the first distribution. Of the values of the determination parameters having the same probability density distribution of, the value of the determination parameter located between the average value of the first distribution and the average value of the second distribution is specified as the first threshold value. The processing apparatus according to claim 1 . The processing unit specifies a second threshold value as the maximum value of the appropriate range based on the processing result, and the appropriate range is a range in which the value of the determination parameter is equal to or greater than the first threshold value and equal to or less than the second threshold value. The processing apparatus according to claim 1 or 2, which is specified as. A claim in which the processing unit executes a process of obtaining a mixed Gaussian distribution of each determination parameter as the statistical process, and specifies the second threshold value based on the result obtained from the mixed Gaussian distribution as the processing result. 3. The processing apparatus according to 3 . In the processing unit, the value of the probability density distribution of the first distribution having the maximum mixing ratio in the mixed Gaussian distribution and the second distribution adjacent to the first distribution whose average value is larger than the average value of the first distribution. Probability of 3 distributions The value of the judgment parameter located between the mean value of the first distribution and the mean value of the third distribution among the values of the judgment parameters that are equal to the value of the density distribution is the first. 2. The processing apparatus according to claim 4, which is specified as a threshold value. A determination device including the processing device according to any one of claims 1 to 5 and a determination unit that determines a matter related to swallowing using the determination parameter within the appropriate range specified by the processing device. .. When the value of the judgment parameter calculated based on the measurement parameter measured by attaching the sensor to the human neck is within the proper range, the appropriate range when judging the matter related to swallowing using the judgment parameter a appropriate range specification method Ru is specific to the processing unit in the processing apparatus according to any one of claims 1 to 5,
A first threshold value, which is the minimum value in the appropriate range, is set for the processing unit based on the processing results obtained by statistically processing a plurality of the determination parameters calculated based on the measurement parameters measured a plurality of times. A process of specifying and specifying a range in which the value of the determination parameter is equal to or greater than the first threshold value as the appropriate range and obtaining a mixed Gaussian distribution of each determination parameter is executed as the statistical process and as the processing result. An appropriate range specifying method for specifying the first threshold value based on the result obtained from the mixed Gaussian distribution .

Images