JP2023133526A - Detection device and detection method - Google Patents

Abstract

To provide a minute evaluation result concerning periodic time-series data.SOLUTION: A periodic time-series data abnormal part detection system 1 divides whole data 44A being periodic time-series data so as to generate partial data 45A, calculates its partial data feature amount 45B, outputs a partial data abnormality detection result 45C by displaying, which is the detection result of an abnormality in the partial data 45A based on the partial data feature amount 45B. Thus the abnormality part detection system 1 detects an abnormality for each piece of partial data 45A obtained by dividing the whole data 44A, thereby to enable presentation of information that may identify which part of the periodic time-series data is abnormal.SELECTED DRAWING: Figure 2

Description

The present invention relates to information processing service technology. Further, the present invention relates to a technique for realizing a detection device and a detection method for detecting abnormal portions of periodic time-series data.

In fields such as healthcare, medical care, and nursing care, systems that measure data on people are increasing. These systems provide valuable information to users by calculating analysis results from the obtained data and providing feedback to the users. The above data is often periodic time series data.

An example of such a system is a system (finger tap measurement and analysis system) that simply evaluates cognitive functions and motor functions by measuring and analyzing a user's finger tap movements (for example, Patent Document 1).

Here, the finger tapping movement is a movement in which the thumb and index finger are repeatedly opened and closed. Periodic time series data can be obtained by measuring finger tapping movements. It is known that the performance of the finger tapping movement varies depending on the presence or absence and severity of brain dysfunction such as dementia or Parkinson's disease. It has been pointed out that the results of analysis of periodic time-series data measured by the above system have the potential to enable early detection of brain dysfunction in users and evaluation such as estimating the severity.

Japanese Patent Application Publication No. 2013-109540

If the evaluation results obtained by analyzing periodic time-series data of finger tapping movements are poor, it is desirable to obtain more detailed results regarding the evaluation results.

An object of the present invention is to provide more detailed evaluation results for periodic time series data.

The above and other objects and novel features of the present invention will become apparent from the description of this specification and the accompanying drawings.

As a means for solving the above problem, the technology described in the claims is used.

One example is a detection device that detects an abnormality using periodic information indicating the state of a living body, which includes a periodic information acquisition section that acquires periodic information, and a periodicity acquired by the periodic information acquisition section. A partial information generation unit that generates partial information based on the cycle from the partial information, a partial information feature calculation unit that calculates the feature amount of the partial information generated by the partial information generation unit, and a partial information feature calculation unit. a partial information anomaly detection section that detects an anomaly in the partial information generated by the partial information generation section based on the calculated feature amount; an output section that outputs information based on a detection result by the partial information anomaly detection section; Equipped with

By using the technology of the present invention, more detailed evaluation results can be provided for periodic time series data.

FIG. 1 is a configuration diagram of a human data measurement system including a periodic time series data abnormality detection system according to a first embodiment. FIG. 1 is a configuration diagram of a periodic time series data abnormal portion detection system according to a first embodiment. FIG. 1 is a configuration diagram of a measuring device according to a first embodiment. FIG. 1 is a configuration diagram of a terminal device according to a first embodiment. FIG. 2 is a diagram showing a state in which a magnetic sensor, which is a movement sensor, is attached to a user's finger. FIG. 3 is a diagram illustrating a detailed configuration example of a motion sensor control unit and the like of the measuring device. It is a flowchart which shows the procedure of the whole process in the human data measurement system of 1st Embodiment. FIG. 3 is a diagram showing an example of a waveform signal of a feature amount. It is a figure which shows the whole data feature-value list. It is a figure which shows the continuation of the whole data feature-value list. FIG. 3 is a diagram showing an example of definition of partial data. FIG. 3 is a diagram showing a partial data feature amount list. FIG. 7 is a diagram showing an example of partial data in which an abnormality has been detected. It is a figure showing a practice menu list. It is a figure showing a practice menu correspondence table. It is a figure showing an example of a menu screen which is an initial screen of a service. It is a figure showing a task measurement screen. It is a figure showing an evaluation result screen. It is a figure which shows an abnormal part detection result screen. It is a figure which shows the periodic time series data abnormal part detection system of 2nd Embodiment. It is a diagram showing the configuration of a server. FIG. 2 is a diagram illustrating an example data structure of user information managed by a server in a DB.

Examples of embodiments of the present invention will be described below with reference to the drawings. In addition, in all the figures for explaining the embodiment, the same parts are given the same reference numerals in principle, and repeated explanations thereof will be omitted.

Embodiments will be described in detail using the drawings. However, the present invention should not be construed as being limited to the contents described in the embodiments shown below. Those skilled in the art will readily understand that the specific configuration can be changed without departing from the spirit or spirit of the present invention.
When there are multiple elements having the same or similar functions, the same reference numerals may be given different subscripts for explanation. However, if there is no need to distinguish between multiple elements, the subscript may be omitted in the explanation.

In this specification, etc., expressions such as "first," "second," and "third" are used to identify constituent elements, and do not necessarily limit the number, order, or content thereof. isn't it. Further, numbers for identifying components are used for each context, and a number used in one context does not necessarily indicate the same configuration in another context. Furthermore, this does not preclude a component identified by a certain number from serving the function of a component identified by another number.

The position, size, shape, range, etc. of each component shown in the drawings etc. may not represent the actual position, size, shape, range, etc. in order to facilitate understanding of the invention. Therefore, the present invention is not necessarily limited to the position, size, shape, range, etc. disclosed in the drawings or the like.

(First embodiment)
A periodic time series data abnormality detection system (detection device) according to the first embodiment will be described using FIGS. 1 to 18. The periodic time series data abnormal part detection system of the first embodiment has a function of detecting an abnormal part in the periodic time series data (periodic information indicating the state of a living body) obtained by measuring a subject. . With this function, abnormal parts can be detected with high precision.

[Human data measurement system]
FIG. 1 is a configuration diagram of a human data measurement system including a periodic time series data abnormality detection system according to a first embodiment. In the first embodiment, a human data measurement system is provided in a facility such as a hospital or a facility for the elderly, a user's home, or the like. The human data measurement system includes a periodic time series data abnormality detection system 1 and a measurement system 2 which is a magnetic sensor type finger tap movement system, and these are connected through a communication line. The measurement system includes a measurement device 3 and a terminal device 4, which are connected through a communication line. A plurality of measurement systems 2 may be provided within the facility.

The measurement system 2 is a system that measures finger movements using a magnetic sensor type movement sensor. A motion sensor is connected to the measuring device 3. The motion sensor is attached to the user's finger. The measuring device 3 measures finger movements through a movement sensor and obtains measurement data including time-series waveform signals. The terminal device 4 displays various information including partial data abnormality detection results on a display screen, and accepts operation input from the user. In the first embodiment, the terminal device 4 is a PC.

The periodic time series data abnormal portion detection system 1 has a function of providing an abnormal portion detection service as a service based on information processing. The periodic time series data abnormal portion detection system 1 has a partial data abnormality detection function as its function. The partial data anomaly detection function is a function that detects an abnormal part in the periodic time series data measured by the measurement system 2.

The periodic time series data abnormal portion detection system 1 receives, for example, periodic time series data as input data from the measurement system 2 . The periodic time-series data abnormality detection system 1 outputs, for example, partial data abnormality detection results as output data to the measurement system 2. The partial data abnormality detection result includes a partial data abnormality degree and a partial data abnormality feature amount in addition to the partial data abnormality determination result.

The human data measurement system of the first embodiment is applicable not only to facilities such as hospitals and elderly facilities and their subjects, but also to a wide range of general facilities and people. The measuring device 3 and the terminal device 4 may be configured as an integrated measuring system. The measurement system 2 and the periodic time series data abnormal portion detection system 1 may be configured as an integrated device. The terminal device 4 and the periodic time series data abnormal portion detection system 1 may be configured as an integrated device. The measuring device 3 and the periodic time series data abnormal portion detection system 1 may be configured as an integrated device.

[Periodic time series data abnormality detection system]
FIG. 2 is a configuration diagram of the periodic time series data abnormal portion detection system 1 of the first embodiment. The periodic time series data abnormal portion detection system 1 includes a control section 101, a storage section 102, an input section 103, an output section 104, a communication section 105, etc., which are connected via a bus. The input unit 103 is a part for inputting operations by the administrator of the periodic time-series data abnormal portion detection system 1 or the like. The output unit 104 is a part that displays a screen for the administrator of the periodic time series data abnormal portion detection system 1 and the like. The communication unit 105 is a part that has a communication interface and performs communication processing with the measurement device 3 and the terminal device 4.

The control unit 101 controls the entire periodic time series data abnormality detection system 1, and is configured with a Central Processing Unit (CPU), a Read Only Memory (ROM), a Random Access Memory (RAM), etc., and is used for software program processing. Based on this, a data processing unit that performs partial data abnormality detection etc. is realized. The data processing section of the control section 101 includes a user information management section 11 , a task processing section 12 , a whole data evaluation section 13 , a partial data abnormality evaluation section 14 , a practice menu determination section 15 , and a result output section 16 . The control unit 101 has a function of inputting measurement data from the measurement device 3, a function of processing and analyzing the measurement data, a function of outputting control instructions to the measurement device 3 and the terminal device 4, and a function of outputting data for display to the terminal device 4. Achieve functions such as output.

The user information management unit 11 performs processes such as registering and managing user information input by a user in the user information 41 of the DB 40, and checking the user information 41 of the DB 40 when the user uses a service. The user information 41 includes attribute values for each individual user, usage history information, user setting information, and the like. Attribute values include gender, age, etc. Usage history information is information that manages the history of users using services provided by this system. The user setting information is setting information set by the user regarding the functions of this service.

The task processing unit 12 is a part that performs processing related to tasks for analysis and evaluation of motor functions and the like. In other words, a task is a predetermined finger movement. The task processing unit 12 outputs a task to the screen of the terminal device 4 based on the task data 42 in the DB 40. Further, the task processing unit 12 acquires the measurement data of the task measured by the measuring device 3 (periodic information indicating the state of the living body), and stores it in the DB 40 as the overall data 44A. Here, the whole data means the whole periodic time series data measured for a predetermined period of time. In this way, the task processing unit 12 (periodic information acquisition unit) acquires periodic information indicating the state of the living body.

The overall data evaluation unit 13 includes an overall data feature calculation unit 13A (periodic information feature calculation unit) and an overall data evaluation unit 13B (periodic information abnormality detection unit). The overall data feature calculation unit 13A calculates a feature representing the nature of the overall data 44A (periodic time series data) based on the user's overall data 44A, and stores it in the DB 40 as an overall data feature 44B. The whole data evaluation unit 13B evaluates the whole data based on the whole data feature amount 44B while referring to the whole data DB 43, and stores it in the DB 40 as the whole data evaluation result 44C.

The partial data anomaly evaluation section 14 includes a partial data generation section 14A (partial information generation section), a partial data feature amount calculation section 14B (partial information feature amount calculation section), and a partial data anomaly detection section 14C (partial information anomaly detection section). include. The partial data generation unit 14A divides the entire data 44A to generate partial data 45A, and stores the partial data 45A in the DB 40. The partial data feature amount calculation unit 14B calculates a feature amount for each piece of partial data 45A, and stores it in the DB 40 as a partial data feature amount 45B. The partial data abnormality detection unit 14C determines abnormality of the partial data based on the partial data feature amount 45B while referring to the partial data obtained from the whole data DB 43, and stores it in the DB 40 as a partial data abnormality detection result 45C. The partial data abnormality detection result 45C includes a partial data abnormality degree 45Ca, partial data abnormality presence/absence 45Cb, and partial data abnormality feature amount 45Cc.

In this way, the partial data abnormality detection unit 14C collects information indicating the degree of abnormality of the partial information generated by the partial data generation unit 14A and whether or not the partial information generated by the partial data generation unit 14A is abnormal. and information indicating an abnormal feature quantity, which is a feature quantity that is the basis for detecting that the partial information generated by the partial data generation unit 14A is abnormal. In this case, the periodic time series data abnormality part detection system 1 generates detailed information regarding the abnormality of the partial information, and therefore can provide more detailed information than information that can specify the part where the abnormality has occurred.

The practice menu determining unit 15 determines a practice menu 46 from the partial data abnormal feature amount 45Cc based on the practice menu list 50C and the practice menu correspondence table 50D, and stores it in the DB 40. In this way, the practice menu determining unit 15 determines a practice menu for improving the abnormal feature amount calculated by the partial data abnormality detecting unit 14C.

The result output unit 16 performs a process of outputting the overall data evaluation result 44C, the partial data abnormality detection result 45C, and the practice menu 46 to the screen of the terminal device 4. The overall data evaluation section 13 and the partial data abnormality evaluation section 14 perform screen output processing in cooperation with the practice menu determination section 15 and the result output section 16. In this way, the result output unit 16 further outputs the menu determined by the practice menu determination unit 15. In this case, the periodic time series data abnormal portion detection system 1 presents a practice menu regarding the abnormal portion of the partial data, and therefore can present information useful for eliminating the abnormal portion.

Further, since the result output unit 16 outputs the overall data evaluation result 44C, it also outputs the overall result, so it is possible to provide information from a plurality of viewpoints based on periodic time series data.

The data and information stored in the DB 40 of the storage unit 102 include user information 41, task data 42, whole data DB 43, whole data 44A, whole data feature quantity 44B, whole data evaluation result 44C, partial data 45A, partial data feature quantity. 45B, partial data abnormality detection result 45C, etc. The control unit 101 maintains and manages the management table 50 in the storage unit 102.

The administrator can set the contents of the management table 50. The management table 50 includes a whole data feature list 50A for setting features of the whole data, a partial data feature list 50B for setting features of partial data, a practice menu list 50C for setting practice menu candidates, and a list of partial data abnormalities. A practice menu correspondence table 50D, etc., which sets the correspondence between the feature amount 45Cc and the practice menu, is stored.

[Measuring device]
FIG. 3 is a configuration diagram of the measuring device 3 of the first embodiment. The measuring device 3 includes a motion sensor 20, a housing section 301, a measuring section 302, a communication section 303, and the like. The housing section 301 includes a motion sensor interface section 311 to which the motion sensor 20 is connected, and a motion sensor control section 312 that controls the motion sensor 20. The measurement unit 302 measures a waveform signal through the motion sensor 20 and the storage unit 301, and outputs it as measurement data. The measurement unit 302 includes a task measurement unit 321 that obtains measurement data. The communication unit 303 has a communication interface, communicates with the surrounding time-series data abnormal portion detection system 1, and transmits the measurement data to the surrounding time-series data abnormal portion detection system 1. The motion sensor interface section 311 includes an analog-to-digital conversion circuit, and converts the analog waveform signal detected by the motion sensor 20 into a digital waveform signal by sampling. The digital waveform signal is input to the motion sensor control section 312.

Note that the measuring device 3 may hold each measurement data in a storage means, or the measuring device 3 may not hold each measurement data and only the periodic time series data abnormality detection system 1 may hold it. good.

[Terminal device]
FIG. 4 is a configuration diagram of the terminal device 4 of the first embodiment. The terminal device 4 includes a control section 401, a storage section 402, a communication section 403, an input device 404, and a display device 405. The control unit 401 performs display of overall data evaluation results, partial data abnormality detection results, etc. as control processing based on software program processing. The storage unit 402 stores user information, task data, whole data (periodic time series data), whole data evaluation results, partial data abnormality detection results, etc. obtained from the periodic time series data abnormality detection system 1. The communication unit 403 has a communication interface, communicates with the periodic time series data abnormal part detection system 1, receives various data from the periodic time series data abnormal part detection system 1, and detects the periodic time series data abnormal part. Send user instruction input information, etc. to the system 1. The input device 404 includes a keyboard, a mouse, and the like. Display device 405 displays various information on display screen 406. Note that the display device 405 may be a touch panel.

[Finger, motion sensor, finger tap measurement]
FIG. 5 is a diagram showing a state in which a magnetic sensor, which is the movement sensor 20, is attached to a user's finger. The motion sensor 20 has a transmitting coil section 21 and a receiving coil section 22, which are paired coil sections through a signal line 23 connected to the measuring device 3. The transmitting coil section 21 generates a magnetic field, and the receiving coil section 22 detects the magnetic field. In the example of FIG. 5, on the user's right hand, the transmitting coil section 21 is attached near the nail of the thumb, and the receiving coil section 22 is attached near the nail of the index finger. The finger to be worn can be changed to another finger. It can be attached not only near the nails.

As shown in FIG. 5, the movement sensor 20 is attached to the user's target fingers, for example, the thumb and index finger of the left hand. In this state, the user performs a finger tap, which is a repeated movement of opening and closing two fingers. In finger tapping, a movement is performed that transitions between a state in which the two fingers are closed, that is, a state in which the fingertips of the two fingers are in contact, and a state in which the two fingers are spread, that is, a state in which the fingertips of the two fingers are spread. Along with this movement, the distance between the transmitting coil section 21 and the receiving coil section 22, which corresponds to the distance between the fingertips of the two fingers, changes. The measuring device 3 measures a waveform signal corresponding to a change in the magnetic field between the transmitting coil section 21 and the receiving coil section 22 of the motion sensor 20.

Note that the motion sensor 20 may be any sensor other than a magnetic sensor as long as it can measure the distance between two fingers. For example, the distance waveform of the two fingers may be obtained by attaching the two fingers to a tablet terminal or a touch-panel PC and repeatedly opening and closing the fingers. Alternatively, the shape of the hand and the position of the fingertips may be detected using an infrared sensor to obtain the distance waveform of the two fingers.

More specifically, finger taps include the following types of tasks: Examples of the exercise include one-handed free run, one-handed metronome, two-handed free run, two-handed alternate free run, two-handed metronome simultaneously, and two-handed alternate metronome. One-handed free running refers to tapping as many times as quickly as possible with the two fingers of one hand. One-handed metronome refers to tapping the two fingers of one hand in time with a stimulus at a constant pace. Simultaneous free running with both hands refers to finger tapping with two fingers of the left hand and two fingers of the right hand at the same timing. Two-handed alternating free run refers to performing finger taps with two fingers of the left hand and two fingers of the right hand at alternate timings. There are also finger taps that follow markers.

[Movement sensor control unit and finger tap measurement]
FIG. 6 is a diagram showing a detailed configuration example of the motion sensor control section 312 and the like of the measuring device 3. As shown in FIG. In the movement sensor 20, the distance D between the transmitting coil section 21 and the receiving coil section 22 is shown. The motion sensor control unit 312 includes an AC generation circuit 312a, a current generation amplifier circuit 312b, a preamplifier circuit 312c, a detection circuit 312d, an LPF circuit 312e, a phase adjustment circuit 312f, an amplifier circuit 312g, and an output signal terminal 312h.

A current generation amplifier circuit 312b and a phase adjustment circuit 312f are connected to the AC generation circuit 312a. The transmitting coil section 21 is connected to the current generating amplifier circuit 312b through the signal line 23. The receiving coil section 22 is connected to the preamplifier circuit 312c through the signal line 23. A detection circuit 312d, an LPF circuit 312e, an amplifier circuit 312g, and an output signal terminal 312h are connected in this order after the preamplifier circuit 312c. A detection circuit 312d is connected to the phase adjustment circuit 312f.

The AC generating circuit 312a generates an AC voltage signal of a predetermined frequency. The current generation amplifier circuit 312b converts the AC voltage signal into an AC current of a predetermined frequency and outputs it to the transmitting coil section 21. The transmitting coil section 21 generates a magnetic field using alternating current. The magnetic field causes the receiving coil section 22 to generate an induced electromotive force. The receiving coil section 22 outputs an alternating current generated by induced electromotive force. The alternating current has the same frequency as the predetermined frequency of the alternating current voltage signal generated by the alternating current generating circuit 312a.

The preamplifier circuit 312c amplifies the detected alternating current. The detection circuit 312d detects the amplified signal based on the reference signal 312i from the phase adjustment circuit 312f. The phase adjustment circuit 312f adjusts the phase of the AC voltage signal of a predetermined frequency or double frequency from the AC generation circuit 312a, and outputs it as a reference signal 312i. The LPF circuit 312e band-limits and outputs the detected signal, and the amplifier circuit 312g amplifies the signal to a predetermined voltage. Then, an output signal corresponding to the measured waveform signal is output from the output signal terminal 312h.

The waveform signal that is the output signal has a voltage value representing the distance D between the two fingers. The distance D and the voltage value can be converted based on a predetermined calculation formula. The calculation formula can also be obtained through calibration. In the calibration, measurements are performed, for example, with the user holding a block of a predetermined length with two fingers of the target hand. A predetermined calculation formula is obtained as an approximate curve that minimizes the error from a data set of voltage values and distance values in the measured values. Furthermore, the size of the user's hand may be determined through calibration and used for normalizing feature amounts, etc. In the first embodiment, the above magnetic sensor is used as the motion sensor 20, and a measuring means corresponding to the magnetic sensor is used. The present invention is not limited to this, and other detection means and measurement means such as an acceleration sensor, a strain gauge, and a high-speed camera may be applied.

[Processing flow]
FIG. 7 is a flowchart showing the overall procedure of processing mainly performed by the periodic time series data abnormality detection system 1 in the human data measurement system of the first embodiment. FIG. 7 includes steps S1 to S10. Below, the steps will be explained in order.

(Step S1) First, the user operates the measurement system 2. Specifically, the terminal device 4 displays an initial screen on the display screen in response to the user's operation. The user selects a desired operation item on the initial screen. For example, an operation item for detecting and processing abnormal data is selected. The terminal device 4 transmits instruction input information corresponding to the selection to the periodic time series data abnormal portion detection system 1. The user can also input and register user information such as gender and age by operating the initial screen. In that case, the terminal device 4 transmits the input user information to the periodic time series data abnormal portion detection system 1. The user information management unit 11 of the periodic time series data abnormal portion detection system 1 registers the user information in the user information 41.

(Step S2) The task processing unit 12 of the periodic time series data abnormal portion detection system 1 transmits task data for the user to the terminal device 4 based on the instruction input information and the finger tap task data 42 of S1. The task data includes information on one or more types of tasks related to hand and finger movements, such as one-handed free running, two-handed simultaneous free running, and two-handed alternate free running. The terminal device 4 displays hand and finger exercise task information on the display screen based on the received task data. The user performs a hand and finger movement task according to the task information on the display screen. The measuring device 3 measures the task and transmits the measured data to the periodic time series data abnormal portion detection system 1. The periodic time series data abnormal portion detection system 1 stores the measurement data in the measurement data 42B.

(Step S3) The whole data feature calculation unit 13A of the periodic time series data abnormal portion detection system 1 calculates the whole data feature 44B from the whole data 44A based on the whole data feature list 50A. Then, the overall data evaluation unit 13B applies statistical methods such as multivariate analysis and machine learning to the overall data feature amount 44B to obtain an overall data evaluation result 44C. In this way, the overall data evaluation unit 13B detects an abnormality in the periodic information based on the feature amount calculated by the overall data feature amount calculation unit 13A.

(Step S4) The result output unit 16 of the periodic time series data abnormal portion detection system 1 sends the overall data evaluation result 44C to the terminal device 4 and displays it on the screen. In this way, the result output unit 16 outputs information based on the detection results by the overall data evaluation unit 13B. Users can check the evaluation results of their own periodic time series data on the screen.

(Step S5) The partial data generation unit 14A of the periodic time series data abnormality detection system 1 generates partial data 45A from the entire data 44A. The partial data generation unit 14A generates partial information based on a cycle (for example, information for one cycle) from the entire data 44A. Then, the partial data feature amount calculation unit 14B calculates the partial data feature amount 45B from the partial data 45A based on the partial data feature amount list 50B. Then, the partial data anomaly detection unit 14C applies a statistical method such as multivariate analysis or machine learning to the partial data feature amount 45B to obtain a partial data anomaly detection result 45C. As described above, the partial data anomaly detection unit 14C detects an abnormality in the partial data 45A based on the feature amount calculated by the partial data feature amount calculation unit 14B.

(Step S6) The result output unit 16 of the periodic time series data abnormal portion detection system 1 sends the partial data abnormality detection result 45C to the terminal device 4 and displays it on the screen. Users can check abnormal parts of their own periodic time series data on the screen.

(Step S7) The practice menu determining unit 15 of the periodic time series data abnormal portion detection system 1 selects the practice menu 46 based on the partial data abnormal feature amount 45Cc while referring to the practice menu list 50C and the practice menu correspondence table 50D. generate.

(Step S8) The result output unit 16 of the periodic time series data abnormality detection system 1 sends the practice menu 46 to the terminal device 4 and displays it on the screen. In this way, the result output unit 16 outputs information based on the detection results. As a result, the user can check the exercise menu that he or she should perform on the screen.

[Overall data feature calculation]
FIG. 8 shows an example of a waveform signal of a feature amount. 8(a) shows a waveform signal of the distance D between the two fingers, FIG. 8(b) shows a waveform signal of the velocity of the two fingers, and FIG. 8(c) shows a waveform signal of the acceleration of the two fingers. show. The velocity shown in FIG. 8(b) is obtained by time differentiation of the distance waveform signal shown in FIG. 8(a). The acceleration shown in FIG. 8(c) is obtained by time differentiation of the velocity waveform signal shown in FIG. 8(b). The overall data feature amount calculation unit 13A obtains a waveform signal of a predetermined feature amount as in this example from the waveform signal of the overall data 44A based on calculations such as differentiation and integration. Further, the overall data feature amount calculation unit 13A obtains a value based on a predetermined calculation from the feature amount.

FIG. 8(d) is an enlarged view of FIG. 8(a) and shows an example of the feature amount. The maximum value Dmax of the finger tap distance D, the tap interval TI, etc. are shown. The horizontal broken line indicates the average value Dav of the distance D over the entire measurement time. The maximum value Dmax indicates the maximum value of the distance D during the entire measurement time. The tap interval TI is the time corresponding to the cycle TC of one finger tap, and particularly indicates the time from the minimum point Pmin to the next minimum point Pmin. In addition, the maximum point Pmax and minimum point Pmin within one cycle of the distance D, the opening operation time T1 and the closing operation time T2, which will be described later, are shown.

Below, further detailed examples of feature amounts will be shown. In the first embodiment, a plurality of feature quantities obtained from the distance, velocity, and acceleration waveforms are used. Note that in other embodiments, only some of the plurality of features may be used, other features may be used, and the details of the definition of the feature may be described. Not limited.

FIG. 9 is a diagram showing the entire data feature list 50A. This association setting is just an example and can be changed. In the entire data feature list 50A in FIG. 9, columns include feature classification, identification number, and feature parameter. The feature amount classification includes [distance], [velocity], [acceleration], [tap interval], [phase difference], and [marker tracking].

For example, the feature [distance] has a plurality of feature parameters identified by identification numbers (A1) to (A11). The unit in the parentheses [ ] of the feature parameter is indicated. (A1) "Maximum amplitude of distance" [mm] is the difference between the maximum value and minimum value of amplitude in the distance waveform ((a) of FIG. 8). (A2) "Total movement distance" [mm] is the sum of absolute values of distance changes over the entire measurement time of one measurement.

(A3) "Average of local maximum values of distance" [mm] is the average of local maximum values of amplitude in each period. (A4) "Standard deviation of local maximum value of distance" [mm] is the standard deviation regarding the above value. (A5) "Inclination of the approximated curve (attenuation rate) of the maximum point of distance" [mm/sec] is the slope of the curve that approximates the maximum point of amplitude. This parameter mainly represents the amplitude change due to fatigue during the measurement time. (A6) "Variation coefficient of maximum value of distance" is a coefficient of variation of maximum value of amplitude, and the unit is a dimensionless quantity (indicated by [-]). This parameter is a value obtained by normalizing the standard deviation by the average, and therefore, individual differences in finger length can be eliminated. (A7) "Standard deviation of local maximum values of distance" [mm] is the standard deviation of local maximum values of amplitude at three adjacent locations.

This parameter is a parameter for evaluating the degree of amplitude variation in a local short time period. (A8) "Average of minimum values of distance" [mm] is the average of minimum values of amplitude in each cycle. (A9) "Standard deviation of minimum value of distance" [mm] is the standard deviation regarding the above value. (A10) "Variation coefficient of minimum value of distance" is a coefficient of variation of minimum value of amplitude, and the unit is a dimensionless quantity (indicated by [-]). This parameter is a value obtained by normalizing the standard deviation by the average, and therefore, individual differences in finger length can be eliminated. (A11) "Standard deviation of local minimum values of distance" [mm] is the standard deviation of local minimum values of amplitude at three adjacent locations. This parameter is a parameter for evaluating the degree of amplitude variation in a local short period of time.

The feature [speed] has feature parameters indicated by the following identification numbers (A12) to (A26). (A12) "Maximum amplitude of velocity" [m/sec] is the difference between the maximum value and minimum value of velocity in the velocity waveform ((b) of FIG. 8). (A13) "Average of local maximum values of opening speed" [m/sec] is the average of maximum values of speeds during the opening operation of each finger tap waveform. The opening motion is a motion of changing the two fingers from the closed state to the maximum open state ((d) in FIG. 8). (A14) "Average of minimum values of closing speed" [m/sec] is the average of minimum values of speed during the closing operation. The closing motion is a motion in which the two fingers are brought from the maximum open state to the closed state. (A15) "Standard deviation of maximum value of opening speed" [m/sec] is the standard deviation of maximum value of speed during the opening operation.

(A16) "Average of minimum points of closing speed" [m/sec] is the standard deviation of the minimum value of the speed during the closing operation. (A17) "Energy balance" [-] is the ratio between the sum of squares of the speed during the opening motion and the sum of the squares of the speed during the closing motion. (A18) "Total energy" [m2/sec2] is the sum of squares of velocities during the entire measurement time. (A19) "Variation coefficient of maximum value of opening speed" [-] is a coefficient of variation of the maximum value of speed during the opening operation, and is a value obtained by normalizing the standard deviation by the average. (A20) "Average of minimum values of closing speed" [m/sec] is the coefficient of variation of the minimum value of speed during the closing operation. (A21) "Number of tremors" [-] is the number obtained by subtracting the number of large opening/closing finger taps from the number of reciprocating times when the velocity waveform changes from positive to negative. (A22) "Average distance ratio at peak opening speed" [-] is the average value of the ratio of the distance when the speed is at its maximum value during the opening movement, when the finger tap amplitude is 1.0. be. (A23) "Average of distance ratios at the time of peak closing speed" [-] is an average of similar ratios regarding distances at the minimum value of speed during the closing operation. (A24) "Ratio of distance ratios at peak speed" [-] is the ratio between the value of (A22) and the value of (A23). (A25) "Standard deviation of distance ratio at peak opening speed" [-] is the standard deviation of the ratio when the finger tap amplitude is 1.0 regarding the distance at the maximum speed during opening movement. It is. (A26) "Standard deviation of distance ratio at peak closing speed" [-] is the standard deviation of a similar ratio regarding the distance at the minimum value of speed during the closing operation.

The feature quantity [acceleration] has feature quantity parameters indicated by the following identification numbers (A27) to (A36). (A27) "Maximum amplitude of acceleration" [m/sec2] is the difference between the maximum value and minimum value of acceleration in the acceleration waveform ((c) of FIG. 8). (A28) "Average of local maximum values of opening acceleration" [m/s2] is the average of local maximum values of acceleration during the opening motion, and is the average of the maximum values of acceleration during the opening movement, and is the average of the maximum values of acceleration that appear during one cycle of finger tapping. It has a value of 1. (A29) "Average of minimum values of opening acceleration" [m/sec2] is the average of minimum values of acceleration during the opening operation, and is the second value among four types of extreme values.

(A30) "Average of local maximum values of closing acceleration" [m/sec2] is the average of local maximum values of acceleration during the closing operation, and is the third value among four types of extreme values. (A31) "Average of minimum values of closing acceleration" [m/sec2] is the average of minimum values of acceleration during the closing operation, and is the fourth value among four types of extreme values. (A32) "Average contact time" [seconds] is the average contact time in the closed state of the two fingers. (A33) "Standard deviation of contact time" [seconds] is the standard deviation of the contact time. (A34) "Contact time variation coefficient" [-] is the variation coefficient of the contact time. (A35) "Number of zero crossings of acceleration" [-] is the average number of times that acceleration changes from positive to negative during one cycle of finger taps. Ideally, this value would be twice. (A36) "Number of freezes" [-] is the value obtained by subtracting the number of large opening/closing finger taps from the number of reciprocating times in which the positive/negative acceleration changes during one cycle of finger taps.

Next, FIG. 10 is a diagram showing the continuation of the entire data feature list 50A. [Tap interval] has feature parameters indicated by the following identification numbers (A37) to (A45). (A37) "Number of taps" [-] is the number of finger taps during the total measurement time of one measurement. (A38) "Tap interval average" [seconds] is the average of the above-mentioned tap intervals ((d) in FIG. 8) in the distance waveform. (A39) "Tap frequency" [Hz] is the frequency at which the spectrum becomes maximum when the distance waveform is Fourier transformed. (A40) “Tap interval standard deviation” [seconds] is the standard deviation regarding the tap interval. (A41) "Tap interval variation coefficient" [-] is a variation coefficient regarding the tap interval, and is a value obtained by normalizing the standard deviation with the average value. (A42) "Tap interval variation" [mm2] is an integrated value of frequencies of 0.2 to 2.0 Hz when the tap interval is subjected to spectrum analysis.

(A43) "Skewness of tap interval distribution" [-] is the skewness of the frequency distribution of tap intervals, and represents the degree to which the frequency distribution is distorted compared to the normal distribution. (A44) "Standard deviation of local tap intervals" [seconds] is the standard deviation of tap intervals at three adjacent locations. (A45) “Slope of approximate curve of tap interval (attenuation rate)” [-] is the slope of the curve that approximates the tap interval. This slope mainly represents the change in tap interval due to fatigue during the measurement time.

The feature quantity [phase difference] has feature quantity parameters shown by the following identification numbers (A46) to (A49). (A46) "Average phase difference" [degrees] is the average phase difference in the waveforms of both hands. The phase difference is an index value representing the deviation of the finger taps of the left hand with respect to the right hand as an angle, assuming that one cycle of finger taps of the right hand is 360 degrees. The case where there is no deviation is defined as 0 degrees. (A47) “Standard deviation of phase difference” [degrees] is the standard deviation regarding the phase difference. The larger the value of (A46) or (A47), the greater the misalignment of both hands and the greater the instability. (A48) "Similarity of both hands" [-] is a value representing the correlation when the time difference is 0 when a cross-correlation function is applied to the waveforms of the left and right hands. (A49) "Time lag at which the similarity of both hands is maximum" [seconds] is a value representing the time lag at which the correlation in (A48) is maximum.

Regarding the feature quantity [marker tracking], it has feature quantity parameters indicated by the following identification numbers (A50) to (A51). (A50) "Average delay time from marker" [seconds] is the average delay time of finger taps relative to the time indicated by the periodic marker. The marker corresponds to a stimulus such as a visual stimulus, an auditory stimulus, or a tactile stimulus. This parameter value is based on the point in time when the two fingers are in the closed state. (A51) "Standard deviation of delay time from marker" [seconds] is the standard deviation regarding the delay time.

[Overall data evaluation]
The overall data evaluation unit 13B obtains an overall data evaluation result 44C representing the quality of the overall data based on the overall data feature amount 44B calculated by the overall data feature amount calculation unit 13A. For example, using the overall data DB 43, multiple regression analysis is applied using a plurality of feature amounts in the overall data feature amount 44B as explanatory variables and the degree of abnormality as an objective variable to obtain an estimation formula for estimating the degree of abnormality. The degree of abnormality is defined as an index that is smaller as it is normal and larger as it is abnormal. Examples of the degree of abnormality include the severity score of brain dysfunction, such as the Mini Mental State Examination (MMSE), which indicates the severity of dementia, and the Unified Parkinson's Disease Rating Scale (UPDRS), which indicates the severity of Parkinson's disease. Can be mentioned. However, these severity values have a property that the more normal the value is, the greater the value is, and the more abnormal the value is, the smaller the value. For example, in the MMSE, a score of 30 indicates the highest cognitive function, and as the score approaches 0, the cognitive function declines. Therefore, as a preprocess, the positive and negative values of MMSE and UPDRS are inverted and then used as the degree of abnormality. Here, in order to estimate the severity score, a similar method may be used instead of multiple regression analysis. For example, a discriminant regression analysis that simultaneously performs discrimination and regression using a linear model may be used.

Other regression methods such as support vector machine regression and neural networks may also be used. Alternatively, as a simpler method, one of the feature amounts of the entire data feature amount 44B may be selected and the quality of the entire data may be determined based on its size.

[Partial data generation]
The partial data generation unit 14A cuts out the finger tap waveform every cycle to obtain partial data 45A. In order to cut out the partial data 45A, as shown in FIG. 11, one period of finger tapping is defined as the time when the average value of the entire data 44A is crossed from top to bottom to the next time when it is crossed from top to bottom. In this way, by defining one cycle based on the average value of the entire data 44A, it is possible to adjust the distance value when the two fingers are fully opened (maximum value) is too small, or when the distance value when the two fingers are closed. If (minimum value) is too large, half-hearted up and down movements that cannot be called finger tapping movements can be eliminated. One period may be defined in other ways, such as from one minimum point to the next minimum point, or from one maximum point to the next maximum point. As for how to cut out the partial data 45A, it may be cut out not every period but every plural periods.

Note that the partial data feature amount calculation unit 14B, which will be described later, also calculates feature amounts (P19) to (P20) using the waveforms of both hands. A waveform is required. To do this, it is sufficient to extract one period from the right hand waveform, and then extract a waveform in the same time period from the left hand waveform. You can also find it by reversing the right and left hands.

[Partial data feature calculation]
FIG. 12 is a diagram showing the partial data feature amount list 50B. Based on this list, the partial data feature amount calculation unit 14B calculates the partial data feature amount 45B. The columns include feature classification, identification number, and feature parameter. The feature amount classification includes [distance], [velocity], [acceleration], [tap interval], [phase difference], and [marker tracking]. The partial data feature amount calculation unit 14B may calculate all the feature amounts in the partial data feature amount list 50B, or may select and calculate some feature amounts.

For example, the feature quantity [distance] has a plurality of feature quantity parameters identified by identification numbers (P1) to (P3). The unit in the parentheses [ ] of the feature parameter is indicated. (P1) "Minimum value of distance" [mm] is the minimum value of amplitude of partial data. (P2) "Maximum value of distance" [mm] is the maximum value of the amplitude of the partial data. (P3) "Total movement distance" [mm] is the sum of the absolute values of the distance changes during the entire measurement time of the partial data.

The feature [speed] has feature parameters indicated by the following identification numbers (P4) to (P11). (P4) "Maximum value of opening speed" [m/sec] is the maximum value of the speed during the opening operation of the partial data. The opening motion is a motion in which the two fingers are brought from a closed state to a maximum open state. (P5) "Minimum value of closing speed" [m/sec] is the minimum value of speed during the closing operation. The closing motion is a motion in which the two fingers are brought from the maximum open state to the closed state. (P6) "Energy balance" [-] is the ratio between the sum of squares of the speed during the opening motion and the sum of the squares of the speed during the closing motion. (P7) "Total energy" [m2/sec2] is the sum of squares of velocities during the entire measurement time of the partial data. (P8) "Number of tremors" [-] is the number obtained by subtracting 1, which is the number of finger taps, from the number of times the speed waveform changes sign or negative. (P9) "Distance ratio at peak opening speed" [-] is the distance when the speed is at its maximum value during the opening motion, assuming that the amplitude of the finger tap is 1. (P10) "Distance ratio at peak closing speed" [-] is the distance at the minimum value of the speed during the closing operation when the amplitude of the finger tap is 1. (P11) "Ratio of distance ratio at peak speed" [-] is the ratio between the value of (9) and the value of (10).

The feature quantity [acceleration] has feature quantity parameters indicated by the following identification numbers (P12) to (P17). (P12) "Maximum value of opening acceleration" [m/sec2] is the maximum value of acceleration during the opening movement, and is the first value of four types of extreme values that appear during one cycle of finger tapping. . (P13) "Minimum value of opening acceleration" [m/sec2] is the minimum value of acceleration during the opening operation, and is the second value among four types of extreme values. (P14) "Maximum value of closing acceleration" [m/sec2] is the local maximum value of acceleration during the closing operation, and is the third value among four types of extreme values. (P15) "Minimum value of closing acceleration" [m/sec2] is the minimum value of acceleration during the closing operation, and is the fourth value among four types of extreme values. (P16) "Contact time" [seconds] is the contact time in the closed state of the two fingers. (P17) "Number of freezes" [-] is the value obtained by subtracting 1, which is the number of large opening/closing finger taps, from the number of reciprocating times in which the positive/negative acceleration changes during one cycle of finger taps.

The feature [tap interval] has a feature parameter indicated by the following identification number (P18). (P18) "Tap interval" [seconds] is the time of one cycle of finger taps.

The feature quantity [phase difference] has feature quantity parameters indicated by the following identification numbers (P19) to (P20). (P19) "Phase difference" [degrees] is the phase difference between the waveforms of both hands. The phase difference is an index value representing the deviation of the finger taps of the left hand with respect to the right hand as an angle, assuming that one cycle of finger taps of the right hand is 360 degrees. The case where there is no deviation is defined as 0 degrees. (P20) "Similarity of both hands" [-] is a value representing the correlation when the time difference is 0 when a cross-correlation function is applied to the waveforms of the left and right hands.

The feature quantity [marker tracking] has a feature quantity parameter indicated by the following identification number (P21). This feature amount is calculated in the task of moving by following the marker. (P21) "Delay time from marker" [seconds] is the delay time of the finger tap with respect to the time indicated by the periodic marker. The marker corresponds to a stimulus such as a visual stimulus, an auditory stimulus, or a tactile stimulus. This parameter value is based on the point in time when the two fingers are in the closed state.

[Partial data abnormality detection]
In order to detect an anomaly in the partial data 45A by the partial data anomaly detection unit 14C, a 1-class Support Vector Machine (SVM), which is a type of machine learning, is employed. SVM, which is the premise of this method, is a method of defining classification boundaries in two-class classification so as to maximize the margin between the classification boundary (hyperplane expressed by a linear equation) and the data of each class. However, if the classification boundary is a hyperplane, it will not be possible to separate the two groups if the classification boundary has a complex shape. Therefore, in SVM, a kernel function is introduced to be able to handle classification boundaries with complex shapes. ing. 1-class SVM has the same concept as the two-class classification problem of SVM, but is a method of classifying a certain percentage of abnormal data and other normal data within one class.

Before applying 1-class SVM to the feature data in the previous section, the feature values are standardized to have an average of 0 and a standard deviation of 1 as preprocessing. By standardizing, it is possible to prevent uneven weighting of feature quantities due to different ranges for each feature quantity in the 1-class SVM calculation. The commonly used RBF kernel was used as the kernel function. The margin optimization algorithm uses sequential minimal optimization (SMO). The proportion of outliers is assumed to be 8%.

Note that methods other than 1-class SVM may be used to detect abnormalities in partial data. For example, assuming a normal distribution centered on the average of the feature quantity distribution of the partial data 46A, data having a large distance from the center of the normal distribution may be detected as abnormal.

[Partial data abnormality detection result]
In the 1-class SVM, a classification score y is calculated, and if the classification score y is a negative value, it is determined to be abnormal. The result detected by this determination is defined as partial data abnormality presence/absence 45Cb.

It is considered that the degree of abnormality increases as the classification score y becomes smaller and further away from 0. Therefore, this classification score y is converted by a function that asymptotically approaches z=0% at y=0 and z=100% at y=-∞, and z is set to a partial data abnormality degree of 45Ca.

In addition, in order to investigate which feature among all the features contributed to the abnormality determination for the periodic finger tap waveform that was determined to be abnormal by 1-class SVM, the standard deviation SD = The feature quantity that deviates by 2.0 or more is defined as the partial data abnormal feature quantity 45Cc.

[Execution example of partial data anomaly evaluation section]
When the above-described partial data abnormality generation unit was executed on the entire data DB 43 that stored the entire right-hand finger tap data of 228 healthy people, 12,898 pieces of partial data 45A were obtained. Then, the partial data feature amount 45B was obtained for each of the partial data 45A by the partial data feature amount calculation unit 14B. Here, 15 feature quantities P1 to P8 and P12 to P18 were selected and calculated from the partial data feature quantity list 50B. Then, when the partial data anomaly detection unit 14C was executed using the obtained partial data feature quantity 45B, a partial data anomaly detection result 45C was obtained. As a result, the number of partial data 45A in which an abnormality was detected was 1032 out of 12898.

FIG. 13 is a diagram showing an example of partial data 45A in which an abnormality has been detected. The topmost waveform is the entire data 44A in which no abnormality was detected in the partial data 45A. The lower four waveforms are the entire data 44A in which there was one or more pieces of partial data 45A in which an abnormality was detected. Partial data 45A in which an abnormality has been detected is overwritten with a thick line on the distance waveform of the finger tapping motion. Above it, the partial data abnormality feature amount is shown.

[Practice menu decided]
FIG. 14 is a practice menu list 50C showing index items representing the nature of the finger tapping movement and practice menus for improving the index items. The index items include [momentum], [endurance], [rhythm], [bilateral coordination], [marker following], [movement magnitude], [waveform balance], and [amplitude control]. The settings of the index items and practice menu are just examples, and can be changed.

FIG. 15 is a diagram showing a practice menu correspondence table 50D regarding setting information of association between feature amounts and practice menu items. This association setting is just an example and can be changed. This table has columns such as feature classification, identification number, feature parameter, and index item. The feature amount classification includes [distance], [velocity], [acceleration], [tap interval], [phase difference], and [marker tracking]. The feature amounts in this list match the partial data feature amount list 50B and are associated with at least one or more of the index items set in the practice menu list 50C.

[Display screen (1) - Menu]
FIG. 16 is a diagram showing an example of a menu screen, which is an initial screen of the service, as an example of the display screen of the terminal device 4. This menu screen includes a user information field 1501, an operation menu field 1502, a setting field 1503, and the like.

In the user information column 1501, the user can input and register user information. Note that if there is user information that has already been input into an electronic medical record or the like, it may be configured to link with that user information. Examples of user information that can be input include user ID, name, date of birth or age, gender, dominant hand, disease/symptom, memo, and the like. The dominant hand can be selected from right hand, left hand, both hands, unknown, etc. The disease/symptom may be input by selecting it from options in a list box, or by inputting arbitrary text. When this system is used in a hospital or the like, a doctor or the like may input information in place of the user instead of the user. The ambient time series data abnormality detection system 1 can be applied even when there is no user information registered.

The operation menu column 1502 displays operation items for functions provided by the service. The operation items include "calibration", "measurement of finger movement", "abnormal data detection/processing", "termination", etc. In the case of selecting "Calibration", the above-mentioned calibration, that is, processing related to adjustment of the motion sensor 20 and the like with respect to the user's fingers is performed. The status of whether the adjustment has been completed or not is also displayed. In the case of selecting "Measure hand movement", the screen transitions to a task measurement screen for measuring hand movement tasks such as finger tapping. In the case of selecting "abnormal data detection/processing", an abnormality is detected in the measured data, the abnormal data detection result is displayed, and a transition is made to a screen for processing the detected abnormal data. If you select "Terminate", this service will be terminated.

In the settings field 1503, user settings can be made. For example, if there is a type of abnormality detection item that the user, measurer, or administrator desires to detect, that abnormality detection item can be selected from the options and set. Further, it is possible to select a process corresponding to each abnormality detection item. Additionally, threshold values for detecting abnormal data can also be set. These setting contents are sent to the periodic time series data abnormal part detection system 1 through the communication unit 105, and the periodic time series data abnormal part detection system 1 detects and processes abnormal data by referring to the settings specified here. do.

[Display screen (2) - Task measurement]
FIG. 17 is a diagram showing a task measurement screen as another example. This screen displays task information. For example, a graph 1600 with time on the horizontal axis and distance between two fingers on the vertical axis is displayed for each of the left and right hands. Other teaching information for explaining the task contents may be output on the screen. For example, a video area may be provided in which the task content is explained using video and audio. The screen includes operation buttons such as "start measurement,""redomeasurement,""endmeasurement," and "save (registration)," which the user can select. The user selects "Start measurement" according to the task information on the screen and performs the task exercise. The measuring device 3 measures the movement of the task and obtains a waveform signal. The terminal device 4 displays a measured waveform 1602 corresponding to the waveform signal being measured on a graph 1600 in real time. After exercising, the user selects "end measurement" and selects "save (registration)" to confirm. The measuring device 3 transmits the measured data to the surrounding time-series data abnormal portion detection system 1.

[Display screen (3) - Evaluation results]
FIG. 18 is a diagram showing an evaluation result screen as another example. This screen displays task analysis and evaluation result information. After analyzing and evaluating the task, this screen will be displayed automatically. In this example, a case is shown in which the feature quantities of five finger tapping movements A to E are displayed in a graph in the form of a radar chart. A solid frame line 1701 indicates the analysis evaluation result after the current task measurement. The estimated severity score of the overall data evaluation result 44C calculated by the overall data evaluation unit 13B is displayed. Also, multiple feature quantities are displayed in a radar chart. In addition, evaluation comments regarding the analysis and evaluation results may be displayed. The overall data evaluation unit 13 creates evaluation comments. For example, a message such as "(B) and (E) are good" is displayed. The screen has operation buttons such as "Check abnormal part of finger tap waveform" and "End". The periodic time series data abnormal part detection system 1 transitions to the abnormal part detection result screen when "Check the abnormal part of finger tap waveform" is selected, and returns to the initial screen when "End" is selected. Transition.

[Display screen (4) - Abnormal part detection results]
FIG. 19 is a diagram showing an abnormal portion detection result screen as another example. On this screen, the partial data abnormality detection result 45C calculated by the partial data abnormality detection unit 14C is presented to the user. The waveform of the entire data 44A is displayed as a thin line. Then, among the partial data abnormality presence/absence 45Cb, the partial data 45A having an abnormality is displayed as a thick line on the waveform. Above it, a partial data abnormality feature quantity 45Cc and a partial data abnormality degree 45Ca are displayed. For the partial data abnormal feature amount 45Cc, an upward arrow is attached if the feature value is too large, and a downward arrow is attached if the feature value is too small. The partial data abnormality degree of 45Ca is displayed as the abnormality degree. Evaluation comments regarding the partial data abnormal feature amount 45Cc are also displayed. Furthermore, a practice menu 46 for improving it is presented.

The screen display of the partial data evaluation results shown in FIG. 19 is not limited to a graph of time and distance, but may also be a graph of time and velocity, time and acceleration, etc. Further, the display is not limited to a graph display, and may be a display of numerical data or a moving image of a finger tapping motion. In the case of video display, a warning sound may be generated at the abnormal portion or the background of the video of the abnormal portion may be changed so that the abnormal portion can be recognized.

Further, it is more preferable to display the overall data evaluation result screen shown in FIG. 18 and the abnormal portion detection result screen shown in FIG. 19 in parallel on one screen. This allows the cause of the overall data evaluation result score to be inferred from the abnormal portion detection result screen, and has the effect of making it easier for the subject to understand or accept the score and the cause.

The overall data evaluation result in the parallel display of the content of the overall data evaluation result and the content of the abnormal part detection result may be a score only, a radar chart only, both a score and a radar chart, or a separate display. A method may also be used. Similarly, the display of abnormal portion detection results is not limited to the graph display shown in FIG. 19, but other display methods may be used as long as the abnormal portion is visually recognized in the overall data. can be used.

[Effects etc.]
The periodic time series data abnormal part detection system 1 of the first embodiment divides the whole data 44A, which is periodic time series data, to generate partial data 45A, calculates the partial data feature quantity 45B, and calculates the Based on the partial data feature amount 45B, a partial data abnormality detection result 45C, which is a result of detecting an abnormality in the partial data 45A, is displayed and output. In this way, the abnormal part detection system 1 detects an abnormality in each piece of partial data 45A obtained by dividing the entire data 44A, and therefore presents information that can specify which part of the periodic time series data is abnormal. I can do it. That is, the periodic time series data abnormal portion detection system 1 can provide more detailed evaluation results.

As a result, if the overall data evaluation result 44C is bad, the user can know specifically which part has the problem. Further, by presenting the practice menu 46 obtained by the practice menu determination unit 15, the user can learn a practice method for improving the problem.

In this embodiment, abnormal portion detection has been described using time-series data of finger tapping motions, but other data may be used as long as it is periodic time-series data. Examples include time-series data obtained by measuring electrocardiographic signals, magnetocardial signals, pulse waves, breathing, brain waves, walking, eye blinking, chewing, and the like.

(Second embodiment)
A periodic time series data abnormality detection system according to the second embodiment will be described using FIGS. 20 to 22. The basic configuration of the second embodiment is the same as that of the first embodiment, and parts of the configuration of the second embodiment that are different from the configuration of Embodiment 1 will be described below.

[system]
FIG. 20 is a diagram showing a periodic time series data abnormal portion detection system according to the second embodiment. The periodic time series data abnormality detection system includes a server 6 of a service provider and systems 7 of a plurality of facilities, which are connected via a communication network 8. The communication network 8 and the server 6 may include a cloud computing system.

The facility can be of various types, such as a hospital, a health checkup center, a public facility, an entertainment facility, or the user's home. The facility is equipped with System 7. Examples of the facility systems 7 include a system 7A for a hospital H1, a system 7B for a hospital H2, and the like. For example, each hospital's system 7A and system 7B includes a measuring device 3 and a terminal device 4 that constitute a measuring system 2 similar to that of the first embodiment. The configuration of each system 7 may be the same or different. The facility system 7 may include a hospital electronic medical record management system or the like. The measuring device of the system 7 may be a dedicated terminal.

The server 6 is a device under the jurisdiction of a service provider. The server 6 has a function of providing a partial data anomaly detection service similar to the periodic time series data anomaly detection system 1 of the first embodiment to facilities and users as a service based on information processing. The server 6 provides service processing to the measurement system in a client-server manner. In addition to such functions, the server 6 has a user management function and the like. The user management function is a function that registers, accumulates, and manages user information, measurement data, analysis evaluation data, etc. of a user group obtained through the systems 7 of a plurality of facilities in a DB.

[server]
FIG. 21 is a diagram showing the configuration of the server 6. As shown in FIG. The server 6 includes a control section 601, a storage section 602, an input section 603, an output section 604, and a communication section 605, which are connected via a bus. The input unit 603 is a part for inputting operations by an administrator of the server 6 or the like. The output unit 604 is a part that displays a screen for the administrator of the server 6 and the like. The communication unit 605 is a part that has a communication interface and performs communication processing with the communication network 8. A DB 640 is stored in the storage unit 602. The DB 640 may be managed by a DB server or the like different from the server 6.

The control unit 601 controls the entire server 6, is configured with a CPU, ROM, RAM, etc., and realizes a data processing unit 600 that detects abnormal data, determines abnormal data processing, etc. based on software program processing. The data processing section 600 includes a user information management section 11 , a task processing section 12 , a whole data evaluation section 13 , a partial data abnormality evaluation section 14 , a practice menu determination section 15 , and a result output section 16 .

The user information management unit 11 registers and manages user information regarding user groups of the system 7 of a plurality of facilities in the DB 640 as user information 41. The user information 41 includes attribute values for each individual user, usage history information, user setting information, and the like. The usage history information includes track record information on how each user used the abnormal portion detection service in the past.

[Server management information]
FIG. 22 is a diagram showing an example of the data structure of the user information 41 managed by the server 6 in the DB 640. This table of user information 41 includes user ID, facility ID, in-facility user ID, gender, age, disease, severity score, symptoms, history information, and the like. The user ID is the user's unique identification information in this system. The facility ID is identification information of the facility where the system 7 is installed. Note that the communication addresses of the measuring devices of each system 7 are also managed separately. The in-facility user ID is user identification information managed within the facility or system 7, if such information exists. That is, the user ID and the in-facility user ID are managed in association with each other. The disease item or symptom item stores a value representing a disease or symptom selected and input by the user, or a value diagnosed by a doctor or the like at a hospital. The severity score is a value representing the severity of the disease.

The history information item is information that manages the user's track record of using the abnormal part detection service, and information such as the date and time of each use is stored in chronological order. In addition, the history information item stores each data when the practice was performed at that time, that is, the aforementioned measurement data, analysis evaluation data, abnormal data detection results, abnormal data processing contents, etc. The history information item may store information on addresses where each piece of data is stored.

[Effects etc.]
According to the surrounding time series data abnormal part detection system 1 of the second embodiment, similarly to the first embodiment, the whole data 44A, which is periodic time series data, is divided to generate partial data 45A, and the By calculating the data feature amount 45B and obtaining the partial data abnormality detection result 45C, an abnormal part in the entire data 44A can be detected and presented to the user. This allows the user to know specifically which part has the problem when the overall data evaluation result 44C is bad. Further, by presenting the practice menu 46 obtained by the practice menu determination unit 15, the user can learn a practice method for improving the problem.

Although the present invention has been specifically described above based on the embodiments, the present invention is not limited to the embodiments described above, and can be modified in various ways without departing from the gist thereof.

The present invention is not limited to the embodiments described above, and includes various modifications. For example, it is possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. Furthermore, it is possible to add, delete, or replace a part of the configuration of each embodiment with the configuration of other embodiments.

DESCRIPTION OF SYMBOLS 1... Periodic time series data abnormal part detection system, 2... Measurement system, 3... Measuring device, 4... Terminal device.

Claims (7)

A detection device that detects an abnormality using periodic information indicating the state of a living body,
a periodic information acquisition unit that acquires the periodic information;
a partial information generation unit that generates a plurality of time-series partial information based on the period from the periodic information acquired by the periodic information acquisition unit;
a partial information feature amount calculation unit that calculates a feature amount for each of the plurality of generated time-series partial information;
a partial information anomaly detection unit that performs a process of detecting an anomaly for each of the generated plurality of time-series partial information based on the feature amount of the partial information;
an output unit that outputs information based on a detection result by the partial information abnormality detection unit;
Equipped with
The partial information anomaly detection unit detects, with respect to the partial information detected to be abnormal among the plurality of generated time-series partial information, the characteristic that caused the partial information to be detected to be abnormal. Generate information indicating the abnormal feature quantity,
The output unit outputs the acquired periodic information, information indicating the partial information where the abnormality was detected, and the abnormal feature amount corresponding to the partial information where the abnormality was detected.
Detection device. The detection device according to claim 1,
The partial information abnormality detection unit generates a degree of abnormality of the partial information in which the abnormality has been detected,
The output unit outputs the degree of the generated abnormality.
Detection device. The detection device according to claim 1,
The partial information anomaly detection unit detects an anomaly in the partial information using a 1-class Support Vector Machine (SVM), which is a type of machine learning.
Detection device. The detection device according to claim 1,
The partial information anomaly detection unit assumes a normal distribution centered on the average of the feature quantity distribution of the partial information, and detects the partial information as abnormal if the distance from the center of the normal distribution is greater than a certain level. do,
Detection device. The detection device according to claim 1,
The periodic information represents a finger movement based on a finger tapping action that repeats a closing action and an opening action a plurality of times.
Detection device. The detection device according to claim 5,
The feature amounts include the distance between the fingers, the total distance traveled by the fingers, the phase difference between the finger tapping movements of both hands, the delay time of the finger tapping movement from the marker, the time of one cycle of the finger tapping movement, the speed of the fingers, and finger acceleration.
Detection device. A detection method executed by a detection device that detects an abnormality using periodic information indicating the state of a living body,
a periodic information acquisition step of acquiring the periodic information;
a partial information generation step of generating a plurality of time-series partial information based on the period from the periodic information acquired in the periodic information acquisition step;
a partial information feature amount calculation step of calculating a feature amount for each of the plurality of time-series partial information generated in the partial information generation step;
Detecting an abnormality in each of the plurality of time-series partial information generated in the partial information generation step based on the feature amount of each of the plurality of time-series partial information calculated in the partial information feature calculation step. a partial information anomaly detection step that performs processing to
an output step of outputting information based on the detection result obtained in the partial information abnormality detection step;
including;
In the partial information abnormality detection step, for the partial information detected to be abnormal among the plurality of generated time-series partial information, the feature that caused the partial information to be detected to be abnormal is detected. Generate information indicating the abnormal feature quantity,
The output step outputs the acquired periodic information, information indicating the partial information where the abnormality was detected, and the abnormality feature amount corresponding to the partial information where the abnormality was detected.
Detection method.

Images