Control program, control method, and control device

Abstract

To provide a control program, a control method, and a control device capable of supplying electrical stimulation with a waveform pattern corresponding to the condition of a treatment object person.SOLUTION: A control program according to one aspect of the present invention causes a processor 11 of an information terminal 10 to execute processing of storing a learned model 60 that takes the joint movable amount of a user as input and takes waveform pattern information indicating a waveform pattern of electrical stimulation applied to the user's body by a low-frequency therapy device 20 as output, acquiring the user's pre-treatment joint movable amount, acquiring the waveform pattern information by inputting the acquired joint movable amount to the learned model 60, and controlling the waveform pattern of the electrical stimulation applied to the user's body by the low-frequency therapy device 20 on the basis of the acquired waveform pattern information.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a control program, a control method, and a control device.

Background Art

[0002] Conventionally, a low-frequency therapeutic apparatus is known that attaches a pad having a conductive layer to a user's body and supplies a low-frequency pulsed current to the body to perform treatment such as relieving the user's stiff shoulders.

[0003] Patent Document 1 describes a biological information processing apparatus in which a reference value of target electrical characteristics (which varies depending on the site and attributes) is input, and the parameters of the electrical output are changed using a learning model so that the output measured by the therapeutic apparatus matches the reference value. Patent Document 2 discloses a vibration device that uses a learning model to obtain vibrations that relax the user. Patent Document 3 discloses an electrical stimulation application device that controls the treatment time according to the usage environment and the treatment site. Patent Document 4 discloses an electrotherapeutic apparatus that allows a user to specify treatment content (treatment mode, treatment time, electrical stimulation intensity).

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Patent Document 3

Patent Document 4

Summary of the Invention

Problems to be Solved by the Invention

[0005] Treatment such as relieving stiff shoulders with low-frequency pulsed current usually has different treatment contents suitable for each patient receiving the treatment. Also, in each patient, the suitable treatment content varies depending on the state of the patient during the treatment.

[0006] However, in the conventional technology, although it is disclosed to change the treatment content according to the part and attributes of the patient to be treated, it does not assume providing appropriate treatment content according to the state of the patient's disorder during the treatment. Therefore, there is room for further improvement in providing treatment content suitable for the state of the patient during treatment.

[0007] In one aspect, the present invention has been made in view of such circumstances, and its object is to provide a control program, a control method, and a control device capable of supplying electrical stimulation of a waveform pattern according to the state of a subject.

Means for Solving the Problems

[0008] In order to solve the above problems, the present invention adopts the following configuration.

[0009] (1) Store in a computer a learned model that takes the user's joint mobility as input and outputs waveform pattern information indicating the waveform pattern of the electrical stimulation applied by the electrotherapeutic device to the user's body, acquire the user's joint mobility, acquire the waveform pattern information by inputting the acquired joint mobility into the learned model, and control the waveform pattern of the electrical stimulation applied by the electrotherapeutic device to the user's body based on the acquired waveform pattern information. A control program that causes the above processing to be executed.

[0010] ​(1) According to this, it is possible to obtain waveform pattern information from a learned model according to the state of the user's joint movement amount and generate a waveform pattern appropriate for the treatment of the user. Thereby, it is possible to provide treatment content suitable for the state of the user when receiving treatment.

[0011] (2) The control program according to (1), wherein the learned model is based on machine learning using, as teacher data, the improvement amount of the joint movement amount of the subject by treatment with electrical stimulation. Control program.

[0012] (2) According to this, it is possible to provide treatment content with a high improvement amount of joint movement amount suitable for the state of the subject.

[0013] (3) The control program according to (2), wherein the learned model is based on machine learning using, as teacher data, the joint movement amount of the user, the waveform pattern information, and the improvement amount. Control program.

[0014] (3) According to this, it is possible to provide treatment content with an even higher improvement amount of joint movement amount suitable for the state of the subject.

[0015] (4) The control program according to any one of (1) to (3), wherein the learned model takes as input the joint movement amount of the user and treatment policy information indicating the treatment policy desired by the user, and in the process of obtaining the waveform pattern information, the joint movement amount of the user and the treatment policy information are input to the learned model. Control program.

[0016] As in (4), in order to provide treatment with a high improvement amount of joint movement amount, it is preferable to include information on the treatment policy desired by the user as input information.

[0017] (5) (4) The control program according to claim (4), wherein the treatment policy includes an improvement period desired by the user, Control program.

[0018] As in (5), in order to provide treatment with a high improvement amount of joint movement amount, it is preferable that the treatment policy information includes an improvement period desired by the user.

[0019] (6) (4) or (5) The control program according to claim (4) or (5), wherein the treatment policy includes a treatment intensity desired by the user, Control program.

[0020] As in (6), in order to provide treatment with a high improvement amount of joint movement amount, it is preferable that the treatment policy information includes a treatment intensity desired by the user.

[0021] (7) (4) to (6) The control program according to any one of (4) to (6), wherein the treatment policy includes a treatment site desired by the user, Control program.

[0022] As in (7), in order to provide treatment with a high improvement amount of joint movement amount, it is preferable that the treatment policy information includes a treatment site desired by the user.

[0023] (8) (4) to (7) The control program according to any one of (4) to (7), wherein the treatment policy includes the time of one treatment desired by the user, Control program.

[0024] As in (8), in order to provide treatment with a high improvement amount of joint movement amount, it is preferable that the treatment policy information includes the time of one treatment desired by the user.

[0025] (9) A control program according to any one of (4) to (8), wherein the treatment policy includes a stimulation pattern desired by the user, Control program.

[0026] As in (9), in order to provide treatment with a high improvement amount of joint movement amount, it is preferable that the treatment policy information includes a stimulation pattern desired by the user.

[0027] (10) A control program according to any one of (1) to (9), wherein the learned model takes as inputs the joint movement amount of the user and attribute information indicating the attributes of the user, In the process of acquiring the waveform pattern information, the joint movement amount of the user and the attribute information are input to the learned model, Control program.

[0028] As in (10), in order to provide treatment with a high improvement amount of joint movement amount, it is preferable to include the attribute information of the user as input information.

[0029] (11) A control program according to any one of (1) to (10), wherein the learned model takes as inputs the joint movement amount of the user and environment information indicating the environment during treatment, In the process of acquiring the waveform pattern information, the joint movement amount of the user and the environment information are input to the learned model, Control program.

[0030] As in (11), in order to provide treatment with a high improvement amount of joint movement amount, it is preferable to include the environment information during treatment as input information.

[0031] (12) A computer, Store a learned model that takes the user's joint range of motion as input and outputs waveform pattern information indicating the waveform pattern of the electrical stimulation applied by the electrotherapeutic device to the user's body. Obtain the user's joint range of motion. Obtain the waveform pattern information by inputting the obtained joint range of motion into the learned model. Based on the obtained waveform pattern information, control the waveform pattern of the electrical stimulation applied by the electrotherapeutic device to the user's body. Control method.

[0032] (12) According to this, it is possible to obtain waveform pattern information from a learned model according to the state of the user's joint range of motion and generate a waveform pattern appropriate for the user's treatment. As a result, it is possible to provide treatment content suitable for the state of the user when receiving treatment.

[0033] (13) A storage unit that stores a learned model that takes the user's joint range of motion as input and outputs waveform pattern information indicating the waveform pattern of the electrical stimulation applied by the electrotherapeutic device to the user's body; A joint range of motion acquisition unit that acquires the user's joint range of motion; A waveform pattern information acquisition unit that acquires the waveform pattern information by inputting the acquired joint range of motion into the learned model; A control unit that controls the waveform pattern of the electrical stimulation applied by the electrotherapeutic device to the user's body based on the acquired waveform pattern information; A control device comprising:

[0034] (13) According to this, it is possible to obtain waveform pattern information from a learned model according to the state of the user's joint range of motion and generate a waveform pattern appropriate for the user's treatment. As a result, it is possible to provide treatment content suitable for the state of the user when receiving treatment.

Effect of the Invention

[0035] According to the present invention, it is possible to provide a control program, a control method, and a control device capable of supplying electrical stimulation of a waveform pattern according to the state of a subject to be treated.

Brief Description of the Drawings

[0036]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Embodiments for Carrying Out the Invention

[0037] Hereinafter, embodiments according to one aspect of the present invention will be described with reference to the drawings.

[0038] <Configuration of Information Terminal 10 and Low-Frequency Therapeutic Apparatus 20> FIG. 1 is a diagram showing an example of an information terminal 10 of the present invention and a low-frequency therapeutic apparatus 20 capable of communicating with the information terminal 10. FIG. 2 is a diagram showing a state in which the pad unit 2 and the main body unit 3 of the low-frequency therapeutic apparatus 20 are separated. The information terminal 10 is an example of the "control device" of the present invention. The low-frequency therapeutic apparatus 20 is an example of the "electrical therapeutic apparatus" of the present invention.

[0039] The low-frequency therapeutic apparatus 20 is a device for treating a user by supplying a low-frequency pulsed current to the user to give an electrical stimulus. The frequency of the low-frequency pulsed current is, for example, about 1 to 1200 [Hz]. The low-frequency therapeutic apparatus 20 is a cordless type low-frequency therapeutic apparatus and includes a pad unit 2 and a main body unit 3.

[0040] The pad unit 2 is composed of a pad 21 and a holder 22. The pad 21 and the holder 22 are integrated in this example. "Integrated" means a state in which the pad 21 and the holder 22 are combined inseparably from each other in a normal use state. However, the pad 21 and the holder 22 may be separable from each other.

[0041] The pad 21 is a part attached to the user's body. The pad 21 includes a conductive layer 21a that supplies a low-frequency pulsed current to the user. The conductive layer 21a is exposed at least partially on each of the front and back surfaces of the pad 21. In this example, the conductive layer 21a is exposed on the entire back surface 211 of the pad 21 facing the user's body side and a part of the surface opposite to the back surface 211 of the pad 21.

[0042] The attachment of the pad 21 to the user's body is performed by attaching the back surface 211 of the pad 21 to the user's skin via a conductive gel attached to the treatment portion 21Y on the back surface 211.

[0043] The pad 21 is formed by laminating a carbon layer, which is a conductor, on the surface of a base material made of a soft synthetic resin by printing, and this carbon layer becomes the conductive layer 21a. The pad 21 has flexibility. The conductive layer 21a is provided separately for each polarity (+ pole and - pole) during energization. Note that since the polarity of the pad 21 may be alternately switched for energization, the conductive layer 21a dedicated to the + pole and the conductive layer 21a dedicated to the - pole do not exist fixedly, and the polarity is variable.

[0044] For example, as shown in FIG. 2, the pad 21 includes an attachment portion 21X attached on the holder 22, and a treatment portion 21Y extending from at least one of the attachment portions 21X with the conductive layer 21a exposed.

[0045] The conductive layer 21a is also exposed on the surface, which is the surface facing the main body portion 3 in the attachment portion 21X of the pad 21, and this exposed portion becomes the pad-side electrode portion 212. This pad-side electrode portion 212 is formed for electrical connection with the electrode (not shown) of the main body portion 3.

[0046] In this example, the conductive layer 21a corresponding to one pole (for example, the + pole) is exposed at one end in the width direction (the direction of arrow W21 in FIG. 2) of the attachment portion 21X, and the conductive layer 21a corresponding to the other pole (for example, the - pole) is exposed at the other end.

[0047] The holder 22 is a portion that holds the pad 21. In this example, the holder 22 is made of a hard resin and holds the attachment portion 21X of the pad 21 with a double-sided adhesive tape. Thereby, the pad 21 and the holder 22 are integrated.

[0048] This holder 22 includes a pad holding portion 221 that holds the attachment portion 21X of the pad 21, and wall portions 222 located at both ends of the pad holding portion 221. Note that the holding of the pad 21 is not limited to the method using a double-sided adhesive tape, and for example, a method using heat welding or a method using paste or an adhesive may also be used.

[0049] Since the holder 22 is made of a hard resin, it is a non-conductor. Therefore, when the pad 21 is placed across the backbone on the user's back, the non-conductor holder 22 can be aligned with the backbone so that the treatment part 21Y of the pad 21 does not overlap the backbone.

[0050] Thereby, it is possible to suppress the flow of the low-frequency pulsed current through the user's backbone and spinal cord. Thus, it is possible to suppress damage to the backbone and spinal cord caused by the current, and the low-frequency treatment device 20 can be used safely. Also, in the attachment part 21X of the pad 21, since it is not necessary to separately cover the part overlapping the backbone with an insulating member, the configuration of the pad part 2 can be simplified.

[0051] The pad 21 is a consumable and is detachable from the main body part 3 when being replaced or the like. In this example, the holder 22 is integrated with the pad 21 to form the pad part 2, and the main body part 3 is configured to be detachable from the holder 22. The replacement of the pad 21 is performed, for example, together with the holder 22.

[0052] The main body part 3 is a part for applying a low-frequency pulsed current to the conductive layer 21a of the pad 21 by being attached to the holder 22. Inside the main body part 3, a power supply part such as a battery and an electric circuit (substrate) for forming a desired low-frequency pulsed current are arranged, and switches and a display part may be provided on the outside.

[0053] Although not shown, from the lower surface of the main body part 3 facing the holder 22, electrodes that are electrically connected to the pad-side electrode part 212 of the pad part 2 protrude. The electrodes of the main body part 3 are provided according to the polarity.

[0054] In a state where the holder 22 is locked to the main body 3, the width dimension W22 of the holder 22 is formed to be smaller than the width dimension W3 of the main body 3. Since the holder 22 is made of a hard resin, it has poor flexibility. On the other hand, since the pad 21 has flexibility, by making the holder 22 narrower in width than the main body 3, the flexibility of the pad 21 is less likely to be inhibited by the main body 3. For this reason, it is easy to align the pad 21 along the curved surface of the user's body, so the pad portion 2 has good fitness to the user's body.

[0055] The information terminal 10 is an electronic device such as a smartphone or a tablet terminal possessed by the user. The information terminal 10 can perform wireless communication with the low-frequency therapeutic apparatus 20. The information terminal 10 receives treatment policy information indicating the treatment policy desired by the user, attribute information indicating the attributes of the user, environmental information indicating the environment during treatment, etc. based on the input operation from the user (the person to be treated) of the low-frequency therapeutic apparatus 20. In addition, the information terminal 10 receives information regarding the current state of the site where the user wants to receive treatment, for example, information regarding the joint movement amount indicating the range within which the joint can move. Based on the received user information, the information terminal 10 acquires waveform pattern information indicating the waveform pattern of the electrical stimulation applied by the low-frequency therapeutic apparatus 20 to the user's body from the learned model provided in the information terminal 10. The information terminal 10 transmits the waveform pattern information acquired from the learned model to the low-frequency therapeutic apparatus 20.

[0056] <Hardware configuration of the main body 3> FIG. 3 is a block diagram showing an example of the hardware configuration of the main body 3. The main body 3 includes, for example, as shown in FIG. 3, a processor 31, a memory 32, a communication interface 33, a power supply unit 34, and a pad driving unit 35.

[0057] The processor 31 is a processor such as a CPU (Central Processing Unit) or an MPU (Micro Processing Unit), for example. By reading and executing the program stored in the memory 32, the processor 31 serves as a control unit that controls the operations of each part of the low-frequency therapeutic apparatus 20. Note that the processor 31 may be a combination of a plurality of processors.

[0058] The memory 32 is realized by a RAM (Random Access Memory), a ROM (Read Only Memory), a flash memory, or the like. The memory 32 stores programs executed by the processor 31, data used by the processor 31, and the like.

[0059] The communication interface 33 may be a wireless communication interface or a wired communication interface. By providing the communication interface 33, the main body 3 can, for example, receive commands from other communication devices such as the information terminal 10 to control the main body 3, or transmit information of the main body 3 to other communication devices such as the information terminal 10. The communication interface 33 is controlled by the processor 31.

[0060] In addition to the communication interface 33, the main body 3 may include a user interface. The user interface includes, for example, an input device that receives operation inputs from the user and an output device that outputs information to the user. The input device can be realized by, for example, keys or a remote control. The output device can be realized by, for example, a display or a speaker. Also, both the input device and the output device may be realized by a touch panel or the like. The main body 3 may receive, for example, via the input device, treatment policy information indicating the treatment policy desired by the user, attribute information indicating the attributes of the user, environment information indicating the environment during treatment, and the like.

[0061] The power supply unit 34 supplies power to each component of the low-frequency therapeutic apparatus 20. As the power supply, for example, an alkaline dry battery or a secondary battery such as a lithium-ion battery or a nickel-metal hydride battery is used, and a driving voltage that stabilizes the battery voltage and supplies it to each component is generated. Further, the power supply unit 34 may supply power to each component of the low-frequency therapeutic apparatus 20 using power supplied from a household power supply or the like, not limited to a battery.

[0062] The pad driving unit 35 controls the supply of a low-frequency pulse current to the user by the pad 21 by applying a pulse voltage to the pad 21. The pad driving unit 35 is controlled by the processor 31. The pulse voltage applied to the pad 21 by the pad driving unit 35 of the main body unit 3 will be described later with reference to FIG. 5.

[0063] <Hardware Configuration of Information Terminal 10> FIG. 4 is a block diagram showing an example of the hardware configuration of the information terminal 10. The information terminal 10 includes, for example, a processor 11, a memory 12, a communication interface 13, and a user interface 14 as shown in FIG. 4. The processor 11 is an example of the "computer" of the present invention. Further, the processor 11 is an example of the "control unit" and "acquisition unit" of the present invention. The memory 12 is an example of the "storage unit" of the present invention.

[0064] The processor 11 is, for example, a processor such as a CPU or an MPU. The processor 11 serves as a control unit that controls the operations of each part of the information terminal 10 by reading and executing a program stored in the memory 12. Note that the processor 11 may be a combination of a plurality of processors.

[0065] The memory 12 is realized by a RAM, a ROM, a flash memory, etc. The memory 12 stores a program executed by the processor 11, data used by the processor 11, etc. For example, the memory 12 takes as input the joint movement amount of the user (the subject), treatment policy information indicating the treatment policy desired by the user, attribute information indicating the attributes of the user, environmental information indicating the environment during treatment, etc., and outputs waveform pattern information indicating the waveform pattern of the electrical stimulation applied by the low-frequency treatment device 20 to the user's body, and stores a learned model.

[0066] The "joint movement amount" refers to, for example, the movable range angles of the upper limbs, fingers, lower limbs, trunk, etc. The upper limbs include the shoulder (including the scapula), elbow, forearm, hand, etc. The fingers include the thumb, fingers, etc. The lower limbs include the thigh, knee, ankle and foot, the first toe, the big toe, the toes, etc. The trunk includes the neck, the thoracic and lumbar regions, etc. The "joint movement amount of the user" refers to the current joint movement amount of the user.

[0067] The "treatment policy" includes the improvement period desired by the user, the treatment intensity desired by the user, the treatment site desired by the user, the time of one treatment desired by the user, the stimulation pattern desired by the user, etc. The "desired improvement period" refers to how long of a treatment period is desired for improvement. For the improvement period, it is possible to select, for example, "short-term improvement" and "long-term improvement". When "short-term improvement" is selected, a waveform pattern is generated that is estimated to result in a large improvement amount in the short term (e.g., with one treatment). Also, when "long-term improvement" is selected, a waveform pattern is generated that is estimated to result in a large improvement amount in the long term (e.g., one month later when treating daily). For the improvement amount to be large in the long term, the average value or the minimum value or the maximum value of the improvement amount during that period may be increased, or the improvement amount after the end of that period may be increased. For the minimum value of the improvement amount to be large means increasing the value indicating that the joint should be moved at least this much, and for the maximum value of the improvement amount to be large means increasing the value indicating that the joint should be moved up to this much at most.

[0068] "Treatment intensity" refers to the strength of electrical stimulation, which can be switched, for example, between strong, normal, and weak. "Treatment site" refers to the part of the user where the low-frequency therapeutic device 20 is attached, such as the shoulder, arm, hand, waist, knee, ankle, etc. "Treatment time per session" refers to the duration for which electrical stimulation treatment continues and can be switched, for example, between 10 minutes, 20 minutes, etc. The treatment time per session may also be switched, for example, between longer and shorter. The "stimulation pattern" can be switched to patterns such as "kneading", "tapping", "pushing", "sweeping", etc.

[0069] "Attribute information" refers to, for example, the user's gender, age, height, weight, health status, etc. "Environmental information" refers to, for example, climate (weather, temperature, humidity, etc.), season, etc.

[0070] "Waveform pattern information" refers to information that can reproduce a waveform pattern and may be parameters such as waveform intensity, waveform frequency, waveform pulse width, waveform output pattern, etc., or may be time-series data of waveform intensity.

[0071] The "trained model" is a model of rules and patterns of data obtained by machine learning that prepares a large amount of combined information of the joint mobility (before treatment), treatment policy information, attribute information, environmental information, waveform pattern information, and the improvement amount of the user's joint mobility when treatment is performed under these conditions (joint mobility + treatment policy information + attribute information + environmental information + waveform pattern information), and uses these combined information as teacher data. The "improvement amount of joint mobility" refers to the improvement amount of joint mobility in the treatment by the electrical stimulation of the low-frequency therapeutic device 20 and can be calculated as (joint mobility after treatment) - (joint mobility before treatment). The method of machine learning is not particularly limited, and for example, any method such as logistic regression, decision tree, random forest, gradient boosting decision tree, neural network, etc. can be used.

[0072] The communication interface 13 may be a wireless communication interface or a wired communication interface. By providing the communication interface 13, the information terminal 10 can, for example, receive information from an electrotherapy device such as a low-frequency therapy device 20, or transmit a command to an electrotherapy device such as the low-frequency therapy device 20 to control the electrotherapy device. Also, the information terminal 10 can communicate with a server 40 (described later with reference to FIGS. 11 and 12) having a learned model via the communication interface 13.

[0073] The user interface 14 includes, for example, an input device that receives an operation input from the user, an output device that outputs information to the user, and the like. The input device can be realized by, for example, a touch panel of a display in the information terminal 10. The output device can be realized by, for example, a display or a speaker. The user interface 14 is controlled by the processor 11.

[0074] For example, the processor 11 obtains waveform pattern information by inputting the user's joint movement amount, treatment policy information, attribute information, and environmental information into a learned model, and controls the waveform pattern of the electrical stimulation applied by the low-frequency therapy device 20 to the user's body based on the obtained waveform pattern information. The "user's joint movement amount" is obtained, for example, by detecting the movement amount while the user actually moves the joint. The detection may be performed by taking a photograph with a smartphone and performing image recognition, or by attaching a motion sensor to the user's body. Also, the user may manually input (for example, input a numerical value) to the display of the information terminal 10. "Controlling the waveform pattern" means controlling the waveform pattern of the electrical stimulation by transmitting the waveform pattern information to the low-frequency therapy device 20.

[0075] <The pulsed voltage applied by the main body 3 to the pad 21> FIG. 5 is a diagram showing an example of a pulse voltage applied by the main body unit 3 to the pad 21. The pulse voltage waveform 50 shown in FIG. 5 is an example of the waveform of the pulse voltage applied by the pad driving unit 35 of the main body unit 3 to the pad 21.

[0076] The parameters of the pulse voltage waveform 50 include an amplitude (voltage) V, a pulse width W, and a pulse period T (pulse frequency F = 1 / T). The main body unit 3 can change the treatment content for the user by changing at least one of these parameters.

[0077] Specifically, the main body unit 3 boosts the power supply voltage to a predetermined voltage and adjusts the boosted voltage to a voltage corresponding to the set amplitude. For example, the main body unit 3 can adjust the amplitude V of the pulse voltage at a predetermined number of levels (10 levels) according to an instruction from the user. When the main body unit 3 receives a setting input of a certain level from the user, it generates a treatment waveform (pulse waveform) corresponding to the treatment mode based on the amplitude V corresponding to that level, and outputs the generated treatment waveform to the pad-side electrode portion 212 of the pad 21. The level of the amplitude V of the pulse voltage is a parameter corresponding to the treatment intensity of the treatment policy.

[0078] The low-frequency therapeutic apparatus 20 is pre-provided with a plurality of treatment modes. For example, treatment modes include "kneading", "tapping", "pushing", "sweeping" modes, etc. The main body unit 3 gives an electrical stimulation corresponding to various modes from the pad 21 to the user by changing the waveform of the pulse voltage applied to the pad 21. The treatment mode is a parameter corresponding to the stimulation pattern of the treatment policy.

[0079] (Method for generating a machine-learned model) FIG. 6 is a flowchart for explaining an example of a method for generating a machine-learned model. In this example, the case of generating a learned model with the joint mobility before treatment and the treatment policy information as inputs will be described.

[0080] For each combination of a subject who receives treatment with the low-frequency therapeutic apparatus 20 and the waveform pattern of the electrical stimulation applied to the subject, the generator of the learned model calculates the amount of improvement in the joint mobility of the subject who has received the treatment. It is preferable to have a large number of subjects receive treatment for each of the prepared waveform patterns and collect a large amount of combined information on the amount of improvement in joint mobility after the treatment.

[0081] The generator of the learned model receives an input of the joint mobility of the subject before the treatment and the treatment policy information of the subject from the subject to be treated (step S1). As described above, the joint mobility before the treatment is the range of motion angle before the treatment of the joint (such as the upper limb, fingers, lower limb, trunk, etc.) that the subject is about to receive treatment for. The treatment policy information is information such as the desired improvement period, the intensity of the electrical stimulation, the treatment site, the time for one treatment, and the stimulation patterns such as "kneading" and "tapping". The joint mobility may be measured by the subject actually moving the joint and the generator side that generates the learned model measuring the mobility, or may be detected by the subject taking a picture with a smartphone and recognizing the image taken. The treatment policy information may be set by the generator side that generates the learned model and notified to the subject who is to receive the treatment before the treatment. Note that the desired improvement period of the subject in this example will be described as being short (one treatment).

[0082] The generator of the learned model treats the subject to be treated with the target waveform pattern based on the treatment policy information received from the subject (step S2).

[0083] The generator of the learned model receives an input of the joint mobility after the treatment from the subject to be treated, for example, after the treatment of the subject is completed (step S3). The joint mobility after the treatment may be measured by the subject actually moving the joint and the generator side that generates the learned model measuring the mobility, or may be detected by the subject taking a picture with a smartphone and recognizing the image taken, in the same way as before the treatment.

[0084] The generator of the learned model calculates the improvement amount of the joint range of motion based on the pre-treatment joint range of motion received in step S1 and the post-treatment joint range of motion received in step S3 (step S4).

[0085] Next, the generator of the learned model generates a learned model for each combination of the subject and the waveform pattern based on the information collected in the above steps S1 to S4.

[0086] The generator of the learned model creates a data set composed of a combination of the pre-treatment joint range of motion and treatment policy information received in step S1, the waveform pattern for treating the subject in step S2, and the improvement amount of the joint range of motion calculated in step S4 (step S5).

[0087] The generator of the learned model generates a learned model by machine learning using the data set created in step S5 as teacher data (step S6). The learned model outputs a waveform pattern estimated to maximize the improvement amount of the joint range of motion for the input of the pre-treatment joint range of motion and treatment policy information. In this example, the pre-treatment joint range of motion and treatment policy information are used as inputs, but it is preferable to further add the subject's attribute information and treatment environment information as inputs.

[0088] Note that the machine learning in step S5 is performed by an arbitrary computer. For example, the machine learning in step S5 is performed on a server, and the learned model generated by the server is transmitted to the information terminal 10.

[0089] In addition, in this example, the case of calculating the amount of improvement in joint mobility in the short term (one treatment) of the subject to generate a learned model has been described, but it is not limited to this. For example, the amount of improvement in joint mobility in the long term (for example, one-month treatment) for the same subject may be calculated to generate a learned model. Specifically, the long-term amount of improvement in joint mobility for the same subject is obtained, and a plurality of "joint mobility before treatment" + "treatment policy information" + "waveform pattern" and "long-term improvement amount" of the subject during that long-term period are included in the dataset. Thereby, a learned model capable of outputting a waveform pattern with the maximum long-term improvement amount can be generated.

[0090] <Operation Example of Information Terminal 10 and Low-Frequency Therapeutic Apparatus 20> FIG. 7 is a sequence diagram showing an example of the operations of the information terminal 10 and the low-frequency therapeutic apparatus 20. In the example shown in FIG. 7, the case where the information terminal 10 has a learned model 60 will be described.

[0091] First, the user U1 operates the information terminal 10 to input the joint mobility before treatment of the user U1 and the desired treatment policy information (step S11). As described above, the joint mobility before treatment may be detected by the user U1 actually moving the joint and taking a picture of it with a smartphone.

[0092] Next, based on the joint mobility before treatment and the treatment policy information input from the user U1, the information terminal 10 obtains waveform pattern information indicating the waveform pattern of the electrical stimulation to be applied to the body of the user U1 for the treatment of the user U1 from the learned model 60 (step S12). The information terminal 10 transmits the obtained waveform pattern information to the low-frequency therapeutic apparatus 20 (step S13).

[0093] Next, the low-frequency therapeutic apparatus 20 applies an electrical stimulation corresponding to the waveform pattern to the user U1 based on the waveform pattern information transmitted from the information terminal 10 (step S14).

[0094] Note that the waveform pattern information obtained from the learned model 60 is, for example, according to the desired improvement period in the treatment policy information of user U1. When "short-term improvement" is selected as described above, information on the waveform pattern is obtained, which is estimated to result in a large improvement amount in the short term (e.g., with one treatment). When "long-term improvement" is selected, information on the waveform pattern is obtained, which is estimated to result in a large improvement amount in the long term (e.g., one month later when treating daily).

[0095] <Example of a screen for receiving pre-treatment information> FIG. 8 is a diagram showing an example of a reception screen for pre-treatment information displayed on the information terminal 10.

[0096] For example, as the reception screen for pre-treatment user information in step S11 described with reference to FIG. 7, an input operation screen as shown in FIG. 8 is displayed on the touch panel 14a of the information terminal 10. Specifically, in order to receive the pre-treatment (current) joint mobility amount, for example, "Please tell me how much the joint can move currently." and its options "Take a picture with the camera" or "Enter a numerical value" are displayed. The option "Take a picture with the camera" means that, as described above, the user actually moves the joint and takes a picture of it with the camera of the information terminal 10. The method of taking the picture will be described with reference to FIG. 9. The option "Enter a numerical value" means that the user actually moves the joint and, if they recognize that they can move it this much, they input the approximate numerical value by operating the touch panel 14a themselves.

[0097] In addition, on the touch panel 14a, for example, "What is the desired improvement period?" and its options "Short-term improvement" or "Long-term improvement" are displayed in order to allow the user to input the desired treatment policy. Also, a "Treatment start" button is displayed, which is touched after each piece of information is input to start the treatment. As items to be displayed in order to allow the user to input the desired treatment policy, in addition to the improvement period, for example, the intensity of the stimulus, the treatment site, the time for one treatment, the stimulus pattern, etc. may be displayed.

[0098] Regarding the user's attribute information and environmental information, it may be accepted together with the treatment policy information, or may be registered in advance in the application of the information terminal 10. Also, the treatment policy information may be registered in advance in the application of the information terminal 10.

[0099] <Example of detecting joint range of motion> FIG. 9 is a diagram showing an example of detecting the joint range of motion by photographing with the information terminal 10. In this example, it is assumed that the treatment site of the user U1 is the "neck".

[0100] For example, on the touch panel 14a of the information terminal 10 shown in FIG. 8, when the user selects "Take a picture with the camera" from the options, a guide such as "In front of the camera, move your neck widely from side to side" is displayed on the touch panel 14a. The user U1 follows the guide and actually moves the neck from side to side while facing the camera of the information terminal 10 and takes a picture of the state as shown in FIG. 9. The information terminal 10 detects the range of motion of the user U1's neck by image recognition.

[0101] <Example of a screen for accepting the amount of improvement in joint range of motion due to treatment> FIG. 10 is a diagram showing an example of a screen for accepting the amount of improvement in joint range of motion after treatment. In steps S3 and S4 in the method for generating the learned model of FIG. 6, the amount of improvement in joint range of motion after treatment was described, but the acceptance of the amount of improvement in joint range of motion is not limited to the time of generating the learned model. For example, after the user has received treatment with the low-frequency therapeutic device 20 as shown in FIG. 7, the amount of improvement in joint range of motion due to the treatment may be accepted (collected). In that case, as a screen for accepting the amount of improvement after treatment, an input operation screen as shown in FIG. 10 is displayed on the touch panel 14a of the information terminal 10.

[0102] Specifically, in order to receive the amount of improvement after treatment, for example, "Please tell me how much the joint can move currently." and its options "Take a picture with a camera" or "Enter a numerical value" are displayed. Regarding "Take a picture with a camera" and "Enter a numerical value", they are the same as those described in FIG. 8. Note that the reception result of the improvement amount of the joint mobility may be used as information for updating the learned model by providing feedback and further machine learning.

[0103] As described above, the information terminal 10 receives, as inputs, the current (before treatment) joint mobility of the user, treatment policy information, attribute information, and environmental information, and inputs the received joint mobility, treatment policy information, attribute information, and environmental information into the learned model to obtain waveform pattern information indicating the waveform pattern of the electrical stimulation applied to the user's body from the learned model, and transmits the obtained waveform pattern information to the low-frequency therapeutic apparatus 20. According to this configuration, it is possible to obtain waveform pattern information suitable for the user's treatment from the learned model according to the joint mobility state of the user before treatment. Thereby, it is possible to provide the user with treatment having a high degree of improvement suitable for the state of the user's joint.

[0104] In addition, according to the information terminal 10, according to whether the user hopes for short-term improvement or long-term improvement as a treatment policy, waveform pattern information estimated to have a large improvement amount in each period can be obtained from the learned model. Thereby, it is possible to provide the user with treatment having a high degree of improvement suitable for the state of the user's joint according to the improvement period desired by the user.

[0105] In addition, every time the user receives treatment with the low-frequency therapeutic apparatus 20, the improvement amount of the joint mobility due to the treatment is collected and the learned model is updated, so that it is further possible to provide the user with treatment having a high degree of improvement suitable for the state of the user's joint.

[0106] <Operation examples of the information terminal 10, the server 40, and the low-frequency therapeutic apparatus 20> FIG. 11 is a sequence diagram showing an example of the operations of the information terminal 10, the server 40, and the low-frequency therapeutic apparatus 20. In the example of FIG. 7 described above, the information terminal 10 has the learned model 60, but the present invention is not limited to this. For example, as shown in FIG. 11, the server 40 capable of communicating with the information terminal 10 may have the learned model 60. The server 40 may be, for example, a physical server or a virtual server (cloud server).

[0107] Also in this case, first, the user U1 operates the information terminal 10 to input the joint mobility before treatment of the user U1 and the desired treatment policy information (step S21). The joint mobility before treatment may be detected from an image taken by the user U1 actually moving the joint and photographing it with a smartphone as described above.

[0108] Next, the information terminal 10 transmits the input joint mobility before treatment and treatment policy information to the server 40 (step S22).

[0109] Next, based on the received joint mobility before treatment and treatment policy information, the server 40 acquires waveform pattern information indicating the waveform pattern of the electrical stimulation to be applied to the body of the user U1 for the treatment of the user U1 from the learned model 60 (step S23). The server 40 transmits the acquired waveform pattern information to the information terminal 10 (step S24).

[0110] Next, the information terminal 10 transmits the received waveform pattern information to the low-frequency therapeutic apparatus 20 (step S25).

[0111] Next, based on the received waveform pattern information, the low-frequency therapeutic apparatus 20 applies an electrical stimulation corresponding to the waveform pattern to the user U1 (step S26).

[0112] In this example, the case where the server 40 transmits waveform pattern information to the information terminal 10 and the information terminal 10 transmits the waveform pattern information to the low-frequency therapeutic apparatus 20 has been described. However, the present invention is not limited to this. For example, the server 40 and the low-frequency therapeutic apparatus 20 may be configured to be communicable, and the server 40 may transmit the waveform pattern information to the low-frequency therapeutic apparatus 20 without going through the information terminal 10.

[0113] <Hardware Configuration of Server 40> FIG. 12 is a block diagram showing an example of the hardware configuration of the server 40. The server 40 includes, for example, a processor 41, a memory 42, and a communication interface 43 as shown in FIG. 11. The processor 41 is an example of the "computer" of the present invention. Further, the processor 41 is an example of the "control unit" and "acquisition unit" of the present invention. The memory 42 is an example of the "storage unit" of the present invention.

[0114] The processor 41 is, for example, a processor such as a CPU or an MPU. The processor 41 reads and executes a program stored in the memory 42 to become a control unit that controls the operations of each part of the server 40. Note that the processor 41 may be a combination of a plurality of processors.

[0115] The memory 42 is realized by a RAM, a ROM, a flash memory, or the like. The memory 42 stores a program executed by the processor 41 or data used by the processor 41. For example, the memory 42 takes as input the joint movement amount of the user (the patient), the treatment policy information indicating the treatment policy desired by the user, the attribute information indicating the attributes of the user, the environmental information indicating the environment during treatment, etc., and outputs the waveform pattern information indicating the waveform pattern of the electrical stimulation applied by the low-frequency therapeutic apparatus 20 to the user's body, and stores the learned model 60.

[0116] The communication interface 43 may be a wireless communication interface or a wired communication interface. By providing the communication interface 43, the server 40 can communicate with, for example, the information terminal 10, the low-frequency therapeutic apparatus 20 (electrotherapeutic apparatus), and the like.

[0117] For example, the processor 41 obtains waveform pattern information by inputting the user's joint movement amount, treatment policy information, attribute information, and environmental information into the learned model, and controls the waveform pattern of the electrical stimulation applied to the user's body by the low-frequency therapeutic apparatus 20 based on the obtained waveform pattern information. The "user's joint movement amount" is obtained, for example, by detecting the movement amount while the user actually moves the joint. "Controlling the waveform pattern" means controlling the waveform pattern of the electrical stimulation by transmitting the waveform pattern information to the low-frequency therapeutic apparatus 20.

[0118] Even when the server 40 has a learned model as in this example, it is possible to obtain waveform pattern information suitable for the user's treatment from the learned model according to the user's joint movement state before treatment. Thereby, it is possible to provide the user with a treatment having a high degree of improvement suitable for the state of the user's joint. Similarly, it is also possible to provide the user with a treatment having a high degree of improvement according to the improvement period desired by the user.

[0119] <Operation example of the low-frequency therapeutic apparatus 20> FIG. 13 is a sequence diagram showing an example of the operation of the low-frequency therapeutic apparatus 20. As shown in FIG. 13, the learned model 60 may be provided in the low-frequency therapeutic apparatus 20. In this configuration, the processor 31 of the low-frequency therapeutic apparatus 20 serves as an example of the "computer" of the present invention, or an example of the "control unit" and "acquisition unit" of the present invention. Further, the memory 32 of the low-frequency therapeutic apparatus 20 serves as an example of the "storage unit" of the present invention.

[0120] First, user U1 operates the low-frequency treatment device 20 to input the pre-treatment joint mobility of user U1 and the desired treatment policy information (step S31). The pre-treatment joint mobility may be detected, for example, by user U1 actually moving the joint, photographing it with, for example, a smartphone, and detecting it from the photographed image.

[0121] Next, based on the pre-treatment joint mobility and treatment policy information input from user U1, the low-frequency treatment device 20 acquires waveform pattern information indicating the waveform pattern of the electrical stimulation to be applied to the body of user U1 for the treatment of user U1 from the learned model 60 (step S32). The low-frequency treatment device 20 applies an electrical stimulation corresponding to the acquired waveform pattern to user U1 based on the acquired waveform pattern information (step S33).

[0122] In this way, the processor of the low-frequency treatment device 20 inputs the joint mobility of user U1 received from user U1 and the treatment policy information (which may include attribute information and environmental information) into the learned model to acquire waveform pattern information, and controls the waveform pattern of the electrical stimulation applied to the body of user U1 based on the acquired waveform pattern information. "Controlling the waveform pattern" means applying an electrical stimulation based on the waveform pattern information to user U1.

[0123] As in this example, even when the low-frequency treatment device 20 has a learned model, it is possible to acquire waveform pattern information suitable for the treatment of the user from the learned model according to the pre-treatment joint movement state of the user. Thereby, it is possible to provide the user with treatment having a high degree of improvement suitable for the state of the user's joints. Similarly, it is also possible to provide the user with treatment having a high degree of improvement according to the improvement period desired by the user.

[0124] <Another configuration example of the low-frequency treatment device> FIG. 14 is a diagram showing another configuration example of the low-frequency treatment device. Although the configuration of the low-frequency treatment device described above has been described for the cordless type low-frequency treatment device 20, the low-frequency treatment device may be, for example, a wired type low-frequency treatment device 200 as shown in FIG. 13.

[0125] The low-frequency therapeutic apparatus 200 includes a main body 205 of the therapeutic apparatus, a pair of pads 270 for attaching to a treatment site, and a cord 280 for electrically connecting the main body 205 and the pads 270. Similar to the low-frequency therapeutic apparatus 20, the low-frequency therapeutic apparatus 200 is an electrotherapeutic apparatus that performs treatments such as relieving the user's stiff shoulders by supplying a low-frequency pulsed current.

[0126] The pad 270 has a sheet-like shape and is attached to the user's body. A plug corresponding to an electrode (not shown) formed on the other surface (the surface that contacts the body) is provided on one surface of the pad 270 (the surface that does not contact the body). The electrode is formed of, for example, a conductive gel-like material or the like.

[0127] By connecting the plug 282 of the cord 280 and the plug on the pad 270 side and inserting the cord 280 into the jack of the main body 205, the main body 205 and the pads 270 are connected. When the polarity of the electrode formed on one pad 270 is positive, the polarity of the electrode formed on the other pad 270 is negative.

[0128] The main body 205 is provided with an operation interface 230 composed of various buttons and a display 260. The operation interface 230 includes a power button 232 for switching on / off the power, a mode selection button 234 for selecting a treatment mode, a treatment start button 236, and an adjustment button 238 for adjusting the strength (stimulation intensity) of the electrical stimulation.

[0129] Note that the operation interface 230 is not limited to the above configuration, and any configuration that can realize various operations by the user is acceptable. The operation interface 230 may be composed of, for example, other buttons, dials, switches, etc.

[0130] On the display 260, the electrical stimulation intensity, the remaining treatment time, the treatment mode, the wearing state of the pad 270, etc. are displayed, or various messages are displayed.

[0131] In the low-frequency therapeutic apparatus 200, the main body unit 205 and the pad 270 have functions corresponding to the main body unit 3 and the pad 21 in the low-frequency therapeutic apparatus 20. That is, the main body unit 205 controls the supply of a low-frequency pulsed current that gives an electrical stimulation for treatment from the pad 270.

[0132] <Control program> Note that the control methods of the information terminal 10, the server 40, and the low-frequency therapeutic apparatuses 20 and 200 described in the above-described embodiments can be realized by a computer executing a pre-prepared control program. This control program is recorded on a computer-readable storage medium and is executed by being read from the storage medium. Further, this control program may be provided in a form stored in a non-transitory storage medium such as a flash memory, or may be provided via a network such as the Internet.

Explanation of reference numerals

[0133] 2 Pad part 3, 205 Main body unit 10 Information terminal 11, 31, 41 Processor 12, 32, 42 Memory 13, 33, 43 Communication interface 14 User interface 14a Touch panel 20, 200 Low-frequency therapeutic apparatus 21, 270 Pad 21X Mounting part 21Y Treatment part 21a Conductive layer 22 Holder 34 Power supply unit 35 Pad drive unit 40 Server 50 Pulse voltage waveform 60 Learned model 211 Inside 212 Pad-side electrode part 221 Pad holding part 222 Wall part 230 Operation interface 232 Power button 234 Mode selection button 236 Treatment start button 238 Adjustment button 260 Display 280 Cord 282 Plug U1 User

Claims

1. Cause a computer to store a learned model that takes the user's joint movement amount as input and outputs waveform pattern information indicating the waveform pattern of the electrical stimulation applied by the electrotherapeutic device to the user's body, acquire the user's joint movement amount, acquire the waveform pattern information by inputting the acquired joint movement amount into the learned model, control the waveform pattern of the electrical stimulation applied by the electrotherapeutic device to the user's body based on the acquired waveform pattern information, A control program for executing the process.

2. The control program according to claim 1, wherein the learned model is based on machine learning using, as teacher data, the amount of improvement in the joint movement amount of the subject being treated by treatment with electrical stimulation, Control program.

3. The control program according to claim 2, wherein the learned model is based on machine learning using, as teacher data, the user's joint movement amount, the waveform pattern information, and the amount of improvement, Control program.

4. The control program according to claim 1, wherein the learned model takes as input the user's joint movement amount and treatment policy information indicating the treatment policy desired by the user, In the process of acquiring the waveform pattern information, the user's joint movement amount and the treatment policy information are input into the learned model, Control program.

5. The control program according to claim 4, wherein the treatment policy includes the improvement period desired by the user, Control program.

6. The control program according to claim 4, wherein the treatment policy includes the treatment intensity desired by the user, Control program.

7. The control program according to claim 4, wherein the treatment policy includes the treatment site desired by the user, Control program.

8. The control program according to claim 4, wherein the treatment policy includes the time of one treatment desired by the user, Control program.

9. The control program according to claim 4, wherein the treatment policy includes the stimulation pattern desired by the user, Control program.

10. The control program according to claim 1, wherein the learned model takes as input the user's joint movement amount and attribute information indicating the user's attributes, In the process of acquiring the waveform pattern information, the joint movement amount of the user and the attribute information are input to the learned model. Control program.

11. The control program according to any one of claims 1 to 10, wherein the learned model takes as inputs the joint movement amount of the user and environment information indicating the environment during treatment, and in the process of acquiring the waveform pattern information, the joint movement amount of the user and the environment information are input to the learned model. Control program.

12. A computer stores a learned model that takes as input the joint movement amount of a user and outputs waveform pattern information indicating the waveform pattern of the electrical stimulation applied by an electrotherapeutic device to the user's body, acquires the joint movement amount of the user, acquires the waveform pattern information by inputting the acquired joint movement amount into the learned model, and controls the waveform pattern of the electrical stimulation applied by the electrotherapeutic device to the user's body based on the acquired waveform pattern information. Control method.

13. A storage unit that stores a learned model that takes as input the joint movement amount of a user and outputs waveform pattern information indicating the waveform pattern of the electrical stimulation applied by an electrotherapeutic device to the user's body, a joint movement amount acquisition unit that acquires the joint movement amount of the user, a waveform pattern information acquisition unit that acquires the waveform pattern information by inputting the acquired joint movement amount into the learned model, and a control unit that controls the waveform pattern of the electrical stimulation applied by the electrotherapeutic device to the user's body based on the acquired waveform pattern information. A control device comprising the above.

Images