JP6061300B2 - Training effect evaluation method, training effect evaluation calculation device, and program using hemiplegic motor function recovery training device - Google Patents

Description

  The present invention relates to a hemiplegic motor function recovery training device for training a paralyzed site such as a finger or a forearm to promote motor function recovery for a patient whose left or right half is paralyzed (hereinafter referred to as hemiplegia) The present invention relates to a training effect evaluation method, a training effect evaluation calculation device, and a program.

  Hemiplegia often remains after a stroke. This hemiplegia can restore a part of motor function by rehabilitation. As an effective rehabilitation that can discover the plasticity of the brain, there is conventionally a method called repeated push therapy. This method induces automatic extension (hereinafter, referred to as stretch reflex) of a part paralyzed in training, and increases voluntary movement by causing voluntary movement, thereby realizing voluntary automatic movement intended by the patient. The current rehabilitation is performed by skilled doctors and therapists, but since the training takes a long time, the physical burden on the doctors and occupational therapists is great.

  For this reason, for example, Patent Document 1 discloses a training apparatus that can automatically perform treatment training of the upper limbs by using a machine, and Patent Document 2 includes, for example, an extremity, a neck, and a waist. A training apparatus for generating a motor sensation in a patient who has paralyzed motor function such as a motion effector is disclosed in Patent Document 3, for example, in which the motor skill of a paralyzed part is improved and the motor skill is evaluated A skill evaluation device is disclosed.

WO / 131340 pamphlet JP 2009-82209 A JP 2008-79712 A

  However, the techniques disclosed in Patent Documents 1 to 3 described above all relate to a method for controlling the operation of the training apparatus and a technique for evaluating exercise ability, and evaluate the effect of restoring motor function by training on a paralyzed part. Not what you want. Applicant must excite specific nerve tracts to be strengthened to improve the paralysis of the stroke, finger, forearm, etc., ie, target extension with respect to the finger, forearm pronation (or pronation) Has developed a method and apparatus for inducing voluntary movements by frequently inducing stretch reflexes to the body, and has obtained good results in promoting the recovery of the paralyzed site, and has applied for a patent (for example, Japanese Patent Application Laid-Open No. 2011-56079, JP2012-50728). However, during and after training, a method for quantitatively evaluating the motor function recovery effect by training in a paralyzed part has not yet been established.

  The present invention was made to solve the above-described problems, and realized quantitative evaluation of results aimed at improving the paralysis of fingers and forearms using a hemiplegic motor function recovery training apparatus that realizes repeated repeated therapy. It is an object of the present invention to provide a training effect evaluation method, a training effect evaluation calculation device, and a program using a hemiplegic motor function recovery training device that can improve the training effect.

  The invention according to claim 1 is provided with a training mechanism for performing training for stimulating muscles of a hemiplegic site to induce stretch reflex, and repeating the training for a predetermined number of times to promote recovery of the motor function of the hemiplegic site. A training effect evaluation method using a hemiplegic motor function recovery training apparatus, wherein an automatic motion by stretch reflex of the hemiplegic region obtained from the detected motion information of the training mechanism (an exercise following stretch reflex and an optional subsequent step) A first step of multiply-accumulating the speed of the motion (including motion) and a predetermined sampling time.

  The invention according to claim 2 is the training effect evaluation method according to claim 1, further comprising a second step of obtaining a maximum speed of the automatic motion by the stretch reflection.

  According to a third aspect of the present invention, in the training effect evaluation method according to the first or second aspect, each of the results of the product-sum operations for the predetermined number of times is compared with an average value of the results of the product-sum operations. And a third step.

  According to a fourth aspect of the present invention, in the training effect evaluation method according to the third aspect dependent on the second aspect, at least one of the result of the product-sum operation, the maximum speed of the automatic motion by the stretch reflection, and the result of the comparison operation. The method further includes a fourth step of displaying one on a display device and prompting an evaluation of the degree of recovery of the paralyzed motor function.

  The invention according to claim 5 is the training effect evaluation method according to any one of claims 1 to 3, wherein the maximum value, the minimum value, the first quantile, the second quantile of the automatic movement by the stretch reflex, and A fifth step of urging evaluation of the degree of recovery of the hemiplegic motor function by generating output information expressing the box whisker by calculating the third quantile and outputting it to the output device; Features.

  The invention according to claim 6 is provided with a training mechanism for performing a training for stimulating the muscles of the hemiplegic site to induce stretch reflex, and repeating the training for a predetermined number of times to promote the recovery of the motor function of the hemiplegic site. A training effect evaluation computing device for evaluating the degree of motor function recovery of the hemiplegic site using a hemiplegic motor function recovery training device, wherein the extension of the hemiplegic site obtained from the detected motion information of the training mechanism An arithmetic processing unit that performs a product-sum operation on the speed of automatic movement by reflection and a predetermined sampling time, and an output unit that outputs a result of the product-sum operation calculated by the arithmetic processing unit.

  According to a seventh aspect of the present invention, in the training effect evaluation computing device according to the sixth aspect, the training mechanism reciprocates up and down with the apparatus base and the vicinity of the wrist placed on the apparatus base as a swing fulcrum. A hand rest, a finger bending / extension mechanism that does not hinder one of the fingers of the hand placed on the hand rest to bend down and bends, and that prevents the bent finger from returning. A first servomotor for driving; a first encoder for measuring a rotation angle of the first servomotor; and the hand for inducing the extension motion by hitting a root of the finger pushed down by the finger bending / extension mechanism. A finger tapping mechanism provided on a platform; a second servo motor that drives the finger tapping mechanism; a second encoder that measures a rotation angle of the second servo motor; and a mechanism control unit; The mechanism control unit includes the first Control for obtaining information on the rotation angle from the encoder and the second encoder, and stimulating the finger by a tapping operation by the finger tapping mechanism at the end of bending of the finger to induce stretch reflex of the finger; Continued contact with the finger during the reflective reflection of the finger gives further stimulation, and the finger striking mechanism moves to the base of the finger along with the extended reflection, enabling continuous automatic finger extension (self-powered) It is characterized in that the control to perform is repeated for a predetermined number of training iterations.

  The invention according to claim 8 is the training effect evaluation computing device according to claim 6, wherein the training mechanism is supported by the device base and the device base so as to be rotatable about a horizontal axis, and the upper surface is opened. A cylindrical body, a stick that is provided in the half cylinder and attaches a finger of a forearm or can be gripped by a finger, a wrist support that is provided in the half cylinder and supports a wrist, and one end of the half cylinder A drive shaft connected to the body and extending horizontally; a servo motor provided on the apparatus base for driving the drive shaft; an encoder provided on the servo motor for measuring a rotation angle of the motor shaft; and a mechanism control unit. The mechanism control unit obtains the rotation angle information from the encoder and repeats the forward rotation, stop, reverse rotation, and stop operations of the half cylinder. In muscle stretch In order to excite the nerve cells by causing reflexes, the first angular velocity, the rapid acceleration, and the second angular velocity higher than the first angular velocity are used to control the muscle stimulation speed. Control for enabling continuous forearm automatic (self-powered) rotation by an operation of applying a resistance force to maintain tension is repeatedly performed a predetermined number of training repetitions.

  The invention according to claim 9 is provided with a training mechanism that is controlled by a computer and performs a training for stimulating a muscle of a hemiplegic site to induce a stretch reflex, and the training is repeated a predetermined number of times to perform the training of the hemiplegic site. A program of a training effect evaluation computing device that evaluates the degree of motor function recovery of the hemiplegic site using a hemiplegic motor function recovery training device that promotes recovery of motor function, the computer including the detected motion of the training mechanism in the computer Performing a product-sum operation on the speed of automatic movement by stretch reflection of the hemiplegic region obtained from information and a predetermined sampling time, and a procedure for outputting the result of the product-sum operation It is characterized by.

  According to the invention according to claim 1, the value obtained by multiply-accumulating the speed of automatic movement by the stretch reflection of the hemiplegic site obtained from the detected movement information of the training mechanism and the predetermined sampling time, Since the degree of motor function recovery at the hemiplegic site can be quantitatively evaluated, the training effect can be improved.

  According to the second aspect of the present invention, the maximum speed of the automatic motion by the stretch reflex is obtained, and how much stimulation effect is obtained by the stretch reflex can be quantitatively evaluated by the value, so that the training effect can be improved.

  According to the invention of claim 3, each of the results of the product-sum operation and the average value of the results of the product-sum operation can be compared, and the degree of variation in the training effect can be quantitatively evaluated by the value. , Can improve the training effect.

  According to the invention of claim 4, by displaying on the display device at least one of the result of the product-sum operation, the maximum speed of the automatic movement by the stretch reflection, and the result of the comparison operation, a visual check of the doctor or occupational therapist is performed. It is possible to evaluate the degree of recovery of motor function at the hemiplegic site due to.

  According to the invention according to claim 5, the maximum value, the minimum value, the first quantile, the second quantile, and the third quantile are obtained by calculation, and the box whisker plot is displayed. It becomes possible to evaluate the degree of hemiplegic motor function recovery with less error compared to the average value evaluation.

  According to the invention of claim 6, the arithmetic processing unit performs a product-sum operation between the speed of the automatic movement due to the stretch reflection of the hemiplegic part obtained from the detected movement information of the training mechanism and the predetermined sampling time. The output unit can output the calculated product-sum operation result to quantitatively evaluate the degree of motor function recovery, and can provide a training effect evaluation arithmetic device that realizes improvement of the training effect. .

  According to the seventh aspect of the present invention, the information on the rotation angle is acquired from the encoder, and at the end of bending the finger, the finger is stimulated by tapping operation to induce the finger's stretching reflection, and the finger's stretching reflection Using a hemiplegic finger function recovery training device that gives further stimulation by continuing to touch the finger and automatically performs control that enables continuous extension by a stroking movement that moves to the base of the finger with stretching reflex Evaluation of extension motor function recovery of hemiplegic finger site. Therefore, it eliminates the need for repeated repeated therapy of the patient's hemiplegic finger by skilled doctors and therapists, reducing the physical burden on the doctors and occupational therapists, and enables quantitative evaluations using numbers. Therefore, the burden on the evaluation side is also reduced.

  According to the eighth aspect of the invention, the rotation angle information is acquired from the encoder, and the half cylinder is rotated forward, stopped, reversely rotated, and stopped. At that time, in normal rotation, muscle stimulation is applied at a first angular velocity, a rapid acceleration, and a second angular velocity faster than the first angular velocity in order to excite nerve cells by causing muscle stretching and reflex due to muscle tension for functional recovery. In the reversal of the hemiplegic forearm region using a hemiplegic forearm function recovery training device that controls the speed and in reverse, performs a series of controls that provide resistance to sustain muscle stimulation and maintain muscle tone Evaluate recovery of supination motor function. Therefore, in order to eliminate the need for repeated repeated therapy for hemiplegic forearms of patients by skilled doctors and therapists, the physical burden on the doctors and therapists can be reduced, and quantitative Since the degree of functional recovery can be known by simple evaluation, the burden on the evaluation side is also reduced.

  According to the invention of claim 9, the computer is caused to execute a procedure for calculating the product sum of the speed of the automatic movement by the stretch reflection of the hemiplegic region and the predetermined sampling time from the detected movement information of the training mechanism. By displaying the results obtained, for example, the degree of motor function recovery can be quantitatively evaluated, and a program for improving the training effect can be provided.

It is a block diagram which shows the structure of the hemiplegic motor function recovery | restoration training system containing the training effect evaluation calculating apparatus which concerns on embodiment of this invention. It is the figure quoted in order to demonstrate the mode of the training which recovers a hemiplegia finger extension function. It is a figure which shows the cross-section of the hemiplegia function recovery training apparatus used in Example 1. FIG. It is a perspective view of the finger bending extension mechanism of Example 1. It is a figure which shows the coordinate of the fingertip position in the finger bending extension mechanism of Example 1. FIG. It is sectional drawing of the finger tapping mechanism of Example 1. It is a block diagram which shows the structure of the training effect evaluation calculating apparatus of Example 1. FIG. 6 is a flowchart illustrating a processing procedure for training effect evaluation processing according to the first embodiment. 3 is a flowchart illustrating a detailed procedure of a calculation process of an automatic motion speed by extension reflection in the first embodiment. 3 is a flowchart illustrating a detailed procedure of product-sum operation processing according to the first exemplary embodiment. 6 is a graph cited for explaining tapping time adjustment in Example 1. FIG. It is a figure which shows an example of the data of the automatic motion by the expansion | extension reflection of the finger | toe of Example 1. FIG. In Example 1, it is a figure which shows an example of the data which measured the extension reflection of the fingertip with and without a tapping stimulus by myoelectric potential. In Example 1, it is the figure which showed the extension locus | trajectory of the fingertip when stimulated with the tapping pattern 1, and the resultant force and speed of the fingertip during extension. In Example 1, it is the figure which showed the extension locus | trajectory of the fingertip when stimulated with the tapping pattern 2, and the resultant force and speed of the fingertip during extension. It is a figure which shows the evaluation parameter | index (finger automatic extension degree evaluation) in Example 1. FIG. It is a figure which shows the evaluation parameter | index (finger automatic extension agility evaluation) in Example 1. FIG. It is a figure which shows the evaluation parameter | index (effect achievement degree at the time of repeated training) in Example 1. FIG. 2 is a diagram illustrating a specific example of Example 1. FIG. In Example 1, an example of the finger | toe automatic extension exercise | movement by the extending | stretching reflection of a finger | toe at the time of giving the combination of an electrical stimulus and a vibration stimulus other than a tapping stimulus is shown. The example of an automatic finger extension degree evaluation by the average value of Example 1 is shown. The finger | toe automatic extension degree evaluation example by the box-and-whisker diagram of Example 1 is shown. The finger automatic extension agility evaluation example by the box-and-whisker diagram of Example 1 is shown. An example of evaluating the degree of achievement of the same effect at the time of repetitive training using the box-and-whisker diagram of Example 1 is shown. It is sectional drawing of the hemiplegic forearm function recovery training apparatus used in Example 2. FIG. It is a perspective view of the hemiplegic forearm function recovery training apparatus used in Example 2. It is an effect | action figure of the half cylinder body of Example 2. FIG. It is a figure explaining the other dynamic rotational motion of Example 2. FIG. It is a front view of the torque detection mechanism of Example 2. It is a figure explaining the principle of the torque detection mechanism of Example 2. FIG. It is a figure explaining operation | movement of the mechanism control part of Example 2. FIG. It is a figure which shows the other dynamic pronation exercise | movement, automatic supination exercise | movement, etc. by the training of Example 2. FIG. In Example 2, it is the graph which measured the extension reflection to the supination by myoelectric potential. In Example 2, it is the graph which measured the stretch reflex with a myoelectric potential for every training pattern. FIG. 6 is a diagram showing angular velocities in Example 2. FIG. 6 is a diagram showing the maximum rotation angle of Example 2. It is a figure which shows the maximum rotation speed of Example 2. FIG. 10 is a flowchart illustrating a detailed procedure of calculation processing of automatic motion by stretch reflection in the second embodiment. 12 is a flowchart illustrating a detailed procedure of a product-sum operation process in the second embodiment. In Example 2, it is a figure which shows the verification result by each training pattern. It is an enlarged view of the partial time area | region of FIG. The forearm automatic rotation degree evaluation example by the box-and-whisker diagram of Example 2 is shown. The forearm automatic rotation agility evaluation example by the box-and-whisker diagram of Example 2 is shown.

  Hereinafter, a training effect evaluation method according to an embodiment of the present invention (hereinafter referred to as the present embodiment) will be described in detail.

  As shown in FIG. 1, the hemiplegic finger function recovery training system 1 includes a hemiplegic finger function recovery training device 10 and a training effect evaluation arithmetic device 30 according to the present embodiment. The hemiplegic finger function recovery training device 10 is provided with a training mechanism that is controlled by the mechanism control unit 20 and performs training for inducing stretch reflex by stimulating the muscles of the paralyzed site. Promotes motor function recovery. The partial exercise information (movement speed) of the hemiplegic finger function recovery training device 10 by the mechanism control unit 20 is output to the training effect evaluation calculation device 30.

  The training effect evaluation computing device 30 performs a product-sum operation between the speed of automatic movement by the stretched reflection of the obtained paralyzed part and a predetermined sampling time, and processes or displays the obtained value as a graph or the like Then, promote quantitative evaluation of the degree of motor function recovery in the paralyzed part, or evaluate by yourself. Details will be described later.

  Hereinafter, the hemiplegic finger function recovery training apparatus 10A for training the hemiplegic finger is set as Example 1, the hemiplegic forearm function recovery training apparatus 10B for training the hemiplegic forearm is set as Example 2, and the training effect evaluation calculation apparatus 30 is used. Each training effect evaluation method will be described.

  First, the content of training aimed at by the hemiplegic finger function recovery training device 10A (hereinafter simply referred to as the training device 10A) according to the first embodiment will be described with reference to FIG. In FIG. 2A, the assistant applies force F1 to the index finger 11 of the patient's hand Ha, and starts the finger bending operation as indicated by the arrow (1). At the same time, the caregiver bends the bent wrist 12 (arrow (2)).

  Then, as shown in (b), the finger 11 is continuously bent by the force F2 (arrow (3)). Further, as shown in (c), the finger 11 is continuously bent by the force F3 (arrow (4)), and the wrist 12 is bent (arrow (5)). In (a) to (c), bending the finger 11 with the force F of the assistant is called other dynamic bending.

  Next, at (d), the assistant taps the base 13 of the finger 11 as shown by the arrow (6). By tapping, the patient's motor nerve is stimulated, and the finger 11 is stretched and reflected as shown in (e) (arrow (7)), and returns to (a). At this time, the assistant's fingers 15 and 16 are attached to the upper surface of the patient's finger 11, and resistance is lightly applied while following the extension operation of the finger 11 to maintain the muscle tension of the finger 11. In (d) to (e) to (a), the finger 11 extends on its own without any external force. This is called automatic extension.

  The structure of the training apparatus 10A capable of performing the other dynamic bending and automatic extension described above will be briefly described with reference to FIGS. As shown in FIG. 3, the training apparatus 10 </ b> A includes a rectangular base plate 21, vertical wall portions 101 and 101 (only one on the back side is displayed) standing upright and parallel to the base plate 21, and these The slope part 25 that is passed to the oblique sides of the vertical wall parts 101 and 101 and supports the patient's arm, the hand platform 102 that is supported by the swing shaft 26 that is passed to the vertical wall parts 101 and 101 and supports the patient's hand, and this hand Side plates 103 and 103 (only one on the back side) mounted on the side of the platform 102 and swingable with the hand platform 102, and the tip of the support plate 104 spanned between these side plates 103 and 103 A finger bending / extending mechanism 105 that pushes down the finger supported by the finger, a finger striking mechanism 108 that is provided in the vicinity of the support plate 104 and moves the third arm 107 forward and backward, and a hand rest provided below the hand rest 102. That specifies the range of reciprocating motion And Tsu path mechanism 109, including. Further, a mechanism control unit 150 (mechanism control unit 20 in FIG. 1) that controls the finger bending / extension mechanism 105 and the finger tapping mechanism 108 is provided.

  As shown in FIG. 4, the finger flexion / extension mechanism 105 includes a first arm 111 supported by a first shaft 41 via a first torque mechanism 90, and a first slider movably provided on the first arm 111. 112, a second arm 113 supported by the second shaft 45 via a second torque mechanism 95, a second slider 114 provided movably on the second arm 113, and upward from the first slider 112. A first extension 116 extending toward the first extension 116 and a finger pressing portion 118 provided at an intersection C with the second extension 117 extending from the second slider toward the first extension 116.

  Returning to FIG. 3, the training apparatus 10 </ b> A can be said as follows. In the training apparatus 10A for causing the patient's finger to perform passive flexion and automatic extension, a mechanism for moving the patient's finger (finger bending / extension mechanism 105) includes a first arm 111, a second arm 113, and the first of them. Intersection of the first servo motor 44 and the second servo motor 47 that drive the first arm 111 and the second arm 113 and the first and second extension portions 116 and 117 extended from the first slider 112 and the second slider 114 The fingertip pressing unit 118 arranged in C is configured. The first arm 111 is driven by the first servo motor 44, and the second arm 113 is driven by the second servo motor 47.

  By controlling the output of the first servo motor 44 and the output of the second servo motor 47 with the mechanism control unit 150 (reference numeral 20 in FIG. 1), the fingertip pressing unit 118 can be held at an arbitrary position. Can be moved on the movement trajectory. The first servo motor 44 and the second servo motor 47 provided on the hand platform 102 are respectively provided with a first encoder 52 and a second encoder 53 that monitor the rotation angle of the motor shaft. The mechanism control unit 150 comprehensively controls the first servo motor 44 and the second servo motor 45 by acquiring rotation angle information from the first encoder and the second encoder.

  The training apparatus 10A includes a special first torque measurement mechanism 90 that can measure the drive torque (Term1) of the first arm 111 at the base of the first arm 111, and the mechanism control in which torque information is not shown. Send to department. Similarly, a special second torque measuring mechanism 95 that can measure the driving torque (Term2) of the second arm 113 is provided at the base of the second arm 113, and torque information is not shown here (not shown). 1). Torque is obtained by measuring strain with a strain gauge and converting this strain. It can be considered that a strain gauge for this purpose is directly attached to the first arm 111. However, when the distortion generated in the first arm 111 is very small, the influence of the error becomes remarkable, and the reliability of measurement is lowered. As a countermeasure, in this embodiment, special torque measuring mechanisms 90 and 95 that can expand (amplify) the distortion are employed.

  In FIG. 3, when the center of the first axis 41 is A, the center of the second axis 45 is O, and the intersection 118 is C, the triangle OAC shown in FIG. 5 can be drawn. The side OA has a fixed length, and the other side AC and side OC have variable lengths.

  In the triangle OAC shown in FIG. 5, the axis passing through the side OA is the x axis, the axis passing through the point O (origin) and perpendicular to the x axis is the y axis, the length of the side OA is L (constant value), and the length of the side OC L1 (variation value), the angle O is θ1 (measured by the encoder 53 provided in the second servomotor 47 of the second arm 113), and the angle (outer angle) A is θ2 (first of the first arm 111). Measured by an encoder 52 provided in the servo motor 44). The angle O + angle C = angle A is transformed into angle C = angle A−angle O = θ 2 −θ 1, and the angle C becomes (θ 2 −θ 1). The inner angle A is (2π−θ2).

If the coordinates of the point C are (x, y), x = L1 · cos θ1 and y = L1 · sin θ1. From the sine theorem, L / sin (θ2−θ1) = L1 / sin (2π−θ2). Here, sin (2π−θ2) = sin θ2. When L / sin (θ2−θ1) = L1 / sinθ2 is arranged with respect to L1, L1 = L · sin θ2 / sin (θ2−θ1) is obtained. This L1 is substituted into x = L1 · cos θ1 and y = L1 · sin θ1.
x = L · cos θ1 · sin θ2 / sin (θ2−θ1)
y = L · sin θ1 · sin θ2 / sin (θ2−θ1)

  The point C, that is, the position of the fingertip pressing portion 118 is uniquely determined by θ1 and θ2 measured by the encoders 53 and 52 of the second arm 113 and the first arm 111, and the fingertip pressing portion 118 can be controlled on a two-dimensional plane. Become. For this reason, the relational expression of the speed (fingertip speed) and force (fingertip external force) concerning the point C can be expressed as follows.

  6 is arranged between a rail member 154 (FIG. 3) including a lower rail 152 and an upper rail 153 and upper and lower rails 152 and 153 and travels on the rail member 154. A movable body 155, a third arm 107 swingably supported via a shaft member 156 of the movable body 155, and a finger provided rotatably at the tip of the third arm 107 via a shaft member 157. And a finger tapping roller 158 for tapping the base of

  The upper and lower rails 152 and 153 are rectangular plate members. The third arm 107 has a notch (not shown) along the moving direction of the movable body 155 so that the third arm 107 can swing. The rail member 154, the movable body 155, and the third arm 107 can be manufactured using general-purpose products. The moving body 155 includes a shaft member 156, plate bodies 161 and 161 provided at both ends of the shaft member 156, and a plurality of movable bodies 155 that are rotatably supported by the plate bodies 161 and 161 (in this case, a total of eight movable bodies 155). ) Of the roller member 162.

  The third arm 107 has two third arm portions 163 and 163 whose base portions are swingably supported by the shaft member 156 of the moving body 155, and the distal ends of these third arm portions 163 and 163 are arranged. The shaft member 157 is provided so as to be connected and supports the finger tapping roller 158. The finger hitting roller 158 is preferably made of urethane resin or urethane rubber.

  The finger tapping mechanism 108 is provided with a third servo motor 55 on the upper surface of the upper rail 153. A third shaft 167 is rotatably provided on the third servo motor 55 via pulleys 164 and 165 and a belt 166. A transmission member 168 that transmits the force of the third servo motor 55 and advances and retracts the third arm 107 is provided on the third shaft 167.

  The transmission member 168 is mounted on the third shaft 167 and can be rotated together with the third shaft 167, and the third torque measuring mechanism 171 and the third shaft 167 are provided on the third shaft 167. Consists of a base 172 that circulates in the interior, a box 173 connected to the base 172, and a roller member 174 that is provided in the box 173 and is in contact with the upper and lower surfaces of the third arm 163. The basic configuration of the third torque mechanism 171 is the same as that of the first torque mechanism 90 and the second torque mechanism 95 shown in FIG.

  Here, a mechanism control unit (not shown in FIG. 1), which is not shown here, performs control to give a stimulus to the finger by a tapping operation by the finger tapping mechanism 108 at the end of bending of the finger and induce a finger stretch reflex. Further, the mechanism control unit 20 gives further stimulation by continuing to touch the finger when the finger is stretched and reflected, and continuously stretches by a tracing operation in which the finger tapping mechanism 108 moves to the base of the finger along with the finger stretching and reflection. Control to enable. These two types of control are repeatedly performed by a predetermined (about 10 times) number of training iterations under the control of the mechanism control unit 20.

  The structure of the training apparatus 10A is disclosed in detail as Example 2 in Japanese Patent Application Laid-Open No. 2012-50728 filed on September 2, 2010 by the present applicant and published on March 15, 2012. Therefore, further explanation is omitted in order to avoid duplication.

  FIG. 7 shows the configuration of the training effect evaluation calculation device 30 of FIG. According to FIG. 7, the training effect evaluation arithmetic device 30 includes an arithmetic processing unit 300, a storage unit 301, A / D (Analog to Digital) converters 302 and 303, a communication unit 304, and an output unit 305. Consists of.

  The arithmetic processing unit 300 is, for example, a microprocessor, and hemiplegic obtained from the detected motion information of the training apparatus 10A by sequentially reading and executing a program stored in the storage unit 301 that is built in or externally attached. A product-sum operation is performed on the speed of the automatic extension movement due to the stretch reflection of the finger part and a predetermined sampling time, and the result is output to the output unit 305 and displayed or printed.

  The arithmetic processing unit 300 may obtain the maximum speed of the automatic extension movement by extension reflection. Further, the arithmetic processing unit 300 may perform a comparison operation between each of the results of the product-sum operation for a predetermined number of times and an average value of the result of the product-sum operation. The arithmetic processing unit 300 outputs to the output unit 305 at least one of the result of the product-sum operation, the maximum speed of the automatic extension movement due to the stretch reflection, and the result of the comparison operation.

  The storage unit 301 is assigned a program area and a work area, and the program area according to the embodiment of the present invention relates to the program area, and the work area relates to the speed of the automatic extension movement by extension reflection calculated at a predetermined sampling interval. Data is stored in time series. The storage unit 301 is mounted with a semiconductor storage element such as a DRAM or an SRAM.

  The A / D converter 302 takes in sensing information of contact force at the time of automatic extension movement by extension reflection of the patient's finger detected by the force sensor 40 at a predetermined sampling interval, converts it into a digital signal, and converts it into an arithmetic processing unit. To 300. At this time, the sensing information, which is an analog quantity, is converted into respective digital values by dividing it into n according to the resolution. The A / D converter 303 takes in sensing information of the patient's myoelectric potential detected by the EMG sensor 50 at a predetermined sampling interval, converts it into a digital signal, and outputs it to the arithmetic processing unit 300.

  The communication unit 304 controls inter-processor communication performed between the arithmetic processing unit 300 and the mechanism control unit 20. For example, the communication unit 304 indicates a tapping pattern to be described later from the arithmetic processing unit 300 to the mechanism control unit 20. The parameter is transmitted, and sensing information acquired as a result of driving the hemiplegic finger function recovery device 10A according to the parameter is transmitted from the mechanism control unit 20 to the arithmetic processing unit 300.

  The output unit 305 is composed of a display device such as an LCD (Liquid Crystal Device) or an OLED (Organic Electro-Luminescence), for example. As a result of the product-sum operation generated by the arithmetic processing unit 300, the maximum of automatic extension movement due to stretch reflection is obtained. Display representing box-and-whisker plots based on at least one of the results of speed and comparison operation, or the maximum value, minimum value, first quantile, second quantile, and third quantile of automatic extension movement by stretch reflection It has a role as a user interface that generates information and displays it on the display device described above to promote evaluation of the degree of recovery of the paralyzed motor function. Note that the output unit 305 may be a plotter instead of a display device, or a combination thereof.

  Hereinafter, the operation of the training effect evaluation computing device 30 shown in FIGS. 1 and 7 will be described in detail with reference to the flowcharts of FIGS. 8 to 10 and the graphs shown in FIGS.

  By the way, it is important for facilitating repetitive therapy to cause stretch reflex by tapping stimulation and to promote automatic extension. Since the stimulus for causing the stretch reflex may vary among individuals, it is necessary to control the training apparatus 10A so as to flexibly cope with the patient. For this reason, a plurality of tapping stimulation patterns using the tapping hand speed and the contact time as parameters were prepared.

  The training effect evaluation computing device 30 (arithmetic processing unit 300) first selects an optimal pattern for a patient from a plurality of muscle stimulation patterns prepared in advance by cooperative control with the mechanism control unit 20, and the muscle stimulation. The paralyzed site is stimulated according to the pattern (step S10). Specifically, the mechanism control unit 20 adjusts the speed at which the tapping hand (finger tapping roller 158) strikes by the finger tapping mechanism 108 (hereinafter referred to as tapping hand speed). As the tapping hand speed increases, the amount of stimulation given to the patient increases. If this amount of stimulation is excessive, it will be a passive movement without the automatic extension being promoted well. It is important to give an appropriate amount of tapping stimulation, as automatic movement, not passive movement, is essential for recovery from hemiplegia.

  Next, when tapping (finger bending), the mechanism control unit 20 adjusts the time from when the tapping hand touches the second finger joint until transition to tapping (extension). This time is called contact time. A large difference in force stimulation is controlled by adjusting this time according to the degree of paralysis. However, since there is no means for directly sensing that the tapping hand touches the base of the paralyzed finger, the contact is determined here by paying attention to the angular velocity of the motor. Specifically, the time at which the current angular velocity is decelerated by the resistance at the base of the finger relative to the target angular velocity is measured as the contact start time (tapping start). When the contact time becomes excessive, the extension of the finger is hindered. Therefore, an example in which an upper limit of the contact time is provided and the contact time is set at 0.2 s is shown in a graph in FIG. In FIG. 11, the horizontal direction is the time axis, and the target and the current angular velocity [rad / s] are graduated in the vertical direction, and the change in angular velocity for each elapsed time is shown.

  In Example 1, the magnitude of the tapping stimulus was controlled by setting the tapping hand speed and the contact time. There are infinite combinations of these two parameters, but four patterns were created here and applied to the subject. Here, the mechanism control unit 20 uses the one pattern selected by controlling the training apparatus 10A, and repeats the bending and extension operations of the subject's finger 10 times (step S10). Then, the training effect evaluation calculation device 30 (calculation processing unit 300) takes in information on the rotation angle from the encoders 52 and 53 at a predetermined sampling interval, and measures the trajectory, fingertip force, and fingertip speed during finger extension each time. And it preserve | saves to the memory | storage part 301 in time series (step S20). FIG. 12 shows the data extracted from the 3rd to the 7th time when statistically relatively accurate data can be obtained in 10 trainings. The data obtained here may be output to the output unit 305 for display or printing. In FIG. 12, the horizontal direction is a time axis, and the fingertip force (Fourse [N]) and the fingertip speed (Speed [m / s]) are scaled in the vertical direction, and the time series changes of the fingertip force and the fingertip speed are made. Is shown.

  In the graph of FIG. 12, it is unclear which part of the speed change of the fingertip speed is due to the stretch reflection. Therefore, it was decided to confirm using EMG here. That is, the training effect evaluation calculation device 30 (calculation processing unit 300) performs training using the training device 10A by cooperative control with the mechanism control unit 20, and measures the myoelectric potential of the patient at that time. Here, the subject is a healthy person, and a comparison is made between the case where tapping is performed and the case where tapping is not performed. However, in both cases, voluntary movement was not performed, and thereby, it was confirmed whether or not an unconscious stretch reflex was generated by a tapping stimulus. FIG. 13 (a) shows a graph for confirming the effect of stretching reflection by the extension speed of the fingertip when there is no tapping stimulus and FIG. 13 (b) shows a tapping stimulus.

  As is apparent from FIG. 13A, EMG is constant when there is no tapping stimulus. As shown in FIG. 13B, when there is a tapping stimulus, an increase in EMG can be confirmed about 100 [ms] after the start of tapping. Here, since the voluntary movement is not performed, it is proved that the increase in EMG is due to stretch reflection. The stretch reflex includes a stretch reflex up to the spinal cord and a stretch reflex reaching the brain. Since it is known that the former takes several tens [ms] from the stimulus and the latter takes several hundreds [ms], the fact that EMG increased after about 100 [ms] from the start of tapping indicates that the latter brain It can be seen that the stretch reflection reaches up to. Further, since the change in the fingertip speed corresponds to the change in the EMG, the training effect is evaluated based on the fingertip speed.

  FIG. 14 shows an automatic finger extension movement (extension movement by extension reflection and subsequent movement) when tapping is performed in accordance with stimulus pattern 1 (tapping speed is 110 [mm / s], contact time is 0.15 [s]). The extension trajectory of the fingertip (including the optional extension movement) (FIG. 14A), the resultant force of the fingertip during the extension (Force [N]) and the speed (Speed [m / s]) are shown (FIG. 14 (b)). FIG. 15 shows the extension trajectory of the fingertip (FIG. 15A) when tapping is performed according to the stimulation pattern 2 (tapping speed is 110 [mm / s], contact time is 0.2 [s]). The resultant force (Force [N]) and speed (Speed [m / s]) at the time of extension are shown (FIG. 15B). In any case, the test subject is a hemiplegic patient, and the data of the third to seventh times are extracted from the ten times of training.

  Here, the description returns to the flowchart. FIG. 9 shows a detailed procedure for the automatic extension motion speed detection process (step S20) in the basic operation flow shown in FIG. Prior to the calculation of the automatic extension movement speed by extension reflection, the arithmetic processing unit 300 stores the minimum value “Vmini” of the fingertip speed in one training and the minimum fingertip speed in a predetermined number of trainings in a built-in register (not shown). The value “Vminj” and the maximum value “Vmaxj” of the fingertip speed at a predetermined number of exercises are assigned. However, j is the number of training (for example, 10 times).

  In FIG. 9, the arithmetic processing unit 300 always monitors the contact time of the tapping hand with a built-in timer (step S301). When the set contact time of 0.2 seconds is reached (step S301 “YES”), the arithmetic processing unit 300 sets Vi at that time in the register Vmini (step S302). Then, Vminj and Vi are compared (step S303). If Vminj is equal to or higher than the fingertip speed Vi at that time (step S303 “YES”), the arithmetic processing unit 300 sets the fingertip speed Vi to Vminj (step S304). If Vminj is less than the fingertip speed Vi (step S303 “NO”), the process of step S303 is repeated.

  Next, the arithmetic processing unit 300 compares Vmaxj and Vi (step S305). Here, if Vmaxj is less than Vi (step S305 “YES”), the arithmetic processing unit 300 sets Vi at that time to Vmaxj (step S306), and if Vmaxj is not less than Vi (step S305 “NO”). "), It is determined whether the fingertip speed Vi is positive or negative (step S307). Subsequently, if Vi is positive (step S307 “NO”), the process of step S303 is repeatedly executed. If Vi is negative (step S307 “YES”), j is updated by +1 (step S308). . The above processing is repeatedly executed until j counts, for example, “10” (step S309 “YES”).

  Returning to FIG. The training effect can be evaluated based on the fingertip automatic extension speed calculated as described above. First, the first important evaluation is the finger automatic extension degree evaluation. The finger automatic extension degree evaluation is an index representing how much automatic finger extension has been performed by extension reflection. For this reason, if the change of the fingertip speed in a set of training results is calculated, it is possible to evaluate the degree of automatic finger extension. Specifically, in the automatic extension movement by extension reflection, the area Si is calculated by multiplying the fingertip speed at each sampling time by the sampling time 8 [ms], and the area Si is calculated by the area Si in this time domain. The degree can be evaluated. Details will be described later.

  FIG. 10 shows a detailed procedure for the product-sum operation process (step S30) of the automatic extension movement speed and the sampling time in FIG. According to FIG. 10, the arithmetic processing unit 300 constantly monitors the contact time with the fingertip by the built-in timer (step S401). When the set contact time of 0.2 seconds is reached (step S401 “YES”), the arithmetic processing unit 300 sets Vi at that time in the register Vmini (step S402). Then, Vminj and Vi are compared (step S403). If Vminj is greater than or equal to Vi (step S403 “YES”), arithmetic processing unit 300 sets Vi at that time to Vminj (step S404). If Vminj is less than Vi (step S403 “NO”), the process of step S403 is repeated.

  Next, the arithmetic processing unit 300 calculates the area Si by multiplying the sampling time Δt by Vi at each sampling time and calculating the area Si (step S405). FIG. 16 shows the relationship between the fingertip speed on the time axis and the area Si calculated at that time. Subsequently, the arithmetic processing unit 300 determines whether Vi is positive or negative (step S406). If Vi is positive (step S406 “NO”), the processes in and after step S403 are repeatedly executed, and if Vi is negative (step S406 “YES”), j is updated by +1 (step S407). The above processing is repeatedly executed until j counts, for example, “10” (step S408 “YES”).

  The description returns to the basic operation flow of FIG. The calculation process of the fingertip automatic extension speed (step S20) and the product-sum calculation process (step S30) of the fingertip automatic extension speed and the sampling time are repeatedly executed for the number of training iterations (step S40). Further calculates the average value (step S50). Subsequently, the arithmetic processing unit 300 generates a display information by graphing the result, and outputs the display information to the output unit 305 to prompt evaluation (step S70).

  The most important evaluation is the degree of automatic finger extension (degree of automatic finger extension). The patient with a higher degree of paralysis has a weaker ability to exercise the fingers automatically. Specifically, as shown in FIG. 16, the degree of automatic extension of the finger is evaluated based on the area Si of the extension reflection area in the extension reflection time area indicated by hatching. In addition to the evaluation of the degree of automatic extension of the finger, it is also possible to evaluate the agility of the automatic extension of the finger indicating how much voluntary movement is caused by the stretch reflex using the maximum fingertip speed during the automatic extension. Specifically, as shown in FIG. 17, the automatic finger extension agility is evaluated based on the maximum fingertip velocity Vmax in the automatic extension movement time region by extension reflection expressed by hatching.

  It is also possible to evaluate the degree of variation in the training effect (degree of achievement of the effect) from the fingertip speed change curve (speed change) and the median curve of the speed change at each automatic extension. Specifically, as shown in FIG. 18, the area Si of the extended reflection region in the extended reflection time region indicated by hatching is obtained each time, and the same effect achievement level at the time of repeated training is evaluated by the following arithmetic expression. As H is closer to 1, the variation of each training is smaller.

  That is, a difference (Sm-Si) between the area Si and its median value Sm in each extended reflection is obtained, a ratio (Sm-Si / Sm) between the difference and the median value is obtained, and the ratio is subtracted from 1. FIGS. 19A and 19B show specific examples. In this case, the degree of achievement of the effect is Ha> Hb. It can be said that if the above-described evaluation of the automatic finger extension and the automatic finger extension agility are good and the effect achievement degree is high, a constant stretch reflection is promoted every time, and the same effect is achieved.

  20 (a) to 20 (f) show the training results of the subject A when not only tapping stimulation but also electrical stimulation and vibration stimulation are supplementarily used for the finger extensor muscles depending on the degree of paralysis of the patient. ing. Electrical stimulation is not a muscle by passing a pulse current through an electrode attached to the body surface, but a motor nerve with a much smaller threshold that controls the muscle is stimulated, and the excitement of the nerve is transmitted to the muscle and the muscle contracts To do. With the assistance of such electrical stimulation, the joint movement threshold value is lowered, that is, the sensitivity of the stretch reflex is increased, so that the stretch reflex caused by the tapping stimulus is easily caused effectively.

  In addition, the vibration stimulus is a stimulus that can directly increase the sensitivity of the stretch reflex by directly acting on the deep sense part so that the stimulus potential can be weighted to the stimulus potential by the tapping stimulus through the vibration sense receptor. This method is highly safe, and tends to effectively cause stretch reflection due to tapping stimulation. 20 (a) shows no tapping stimulus, FIG. 20 (b) shows only a tapping stimulus, FIG. 20 (c) shows a tapping stimulus and an electrical stimulus, FIG. 20 (d) shows a tapping stimulus and a vibration stimulus, and FIG. Tapping stimulation, electrical stimulation, and vibration stimulation are used in combination, and FIG. 20 (f) shows the verification result of subject A by training when there is no tapping stimulation after training.

  FIG. 21 shows the evaluation result of the finger automatic extension degree using the average value. This evaluation with the average value is effective in the case of online control of the training apparatus 10A. That is, the training effect evaluation calculation device 30 (the calculation processing unit 300) uses the training device 10A to perform the tapping stimulus and the vibration stimulus when there is no tapping stimulus, when there is, and when the tapping stimulus and the electrical stimulus are used together. When used together, when tapping stimulation, electrical stimulation, and vibration stimulation are used together, the average value of the training results for 10 sets in each case for only tapping stimulation after training is obtained and output to the output unit 305 (for example, indicate.

  According to the evaluation result of FIG. 21, a better result is obtained when the tapping stimulus is not performed. This shows the usefulness of the tapping mechanism of the training apparatus 10A. In addition, regarding the evaluation of the automatic finger extension agility, the results obtained when the tapping stimulus and the electrical stimulus were used in combination were better than the other cases. In the absence of tapping stimulation, the result was clearly inferior agility in automatic finger extension. In addition, it can be understood that the effect is stably obtained when the tapping stimulus, the electrical stimulus, and the vibration stimulus are all used because the same effect achievement evaluation degree at the time of repeated training is high.

  Next, evaluation based on a box-and-whisker diagram will be described with reference to FIGS. 22, 23, and 24. Again, the evaluation was made using the combinations shown in FIGS. FIG. 22 shows the automatic finger extension degree evaluation, FIG. 23 shows the automatic finger extension agility evaluation, and FIG. 24 shows the same effect achievement degree evaluation during repeated training. As in the case of the average value, the boxplot shown here also shows that the subject A using the training apparatus 10A has no tapping stimulus, if there is, and if tapping stimulus and electrical stimulation are used together, the tapping stimulus and vibration stimulus are used. When used together, this is realized by obtaining an average value of 10 sets of training results for each set when tapping stimulation, electrical stimulation, and vibration stimulation are used together, and outputting (for example, displaying) to the output unit 305.

  As shown in FIG. 22 and FIG. 23, the subject A obtained better results in the training including the tapping stimulus in the automatic finger extension degree evaluation and the automatic finger extension agility evaluation than in the case without the tapping stimulus. Can understand. In particular, it can be understood that, when tapping stimulation, electrical stimulation, and vibration stimulation are used in combination, there is little variation and training can be performed stably. This can also be confirmed from the same effect achievement evaluation at the time of repeated training shown in FIG. Further, by performing training without tapping stimulation before and after training and comparing the results, it is possible to confirm the improvement in the effect by the automatic finger extension degree evaluation and the automatic finger extension agility evaluation.

  According to the training effect evaluation method of the first embodiment described above, instead of the selective exercise training that requires a lot of time and effort, the training device 10A is controlled to sense the motion information and force information. In this way, the training effect when performing the facilitating technique is quantitatively evaluated, and the finger paralysis is improved by facilitating voluntary finger extension movement and repetition. The program by this training effect evaluation method is transplanted to the training effect evaluation calculation device 30, and this training effect evaluation calculation device 30 adjusts and adapts the training method by the training effect evaluation, and maximizes the training effect. Thus, hemiplegic finger function recovery is promoted instead of a doctor or occupational therapist, and labor can be reduced and medical effects can be improved. In addition, since it is possible to quantitatively analyze and evaluate changes in exercise and force when performing a push-through procedure using the motion information and force information of the training device 10A, the evaluation results are sent to the training device 10A. Further improvement of the training effect can be realized by changing the training pattern by feedback.

  In addition, according to the training effect evaluation calculation device 30 according to the first embodiment, the calculation processing unit 300 performs automatic extension exercise by stretch reflection of a hemiplegic region obtained from the detected exercise information of the training mechanism (training device 10A). The product-sum operation of the speed and a predetermined sampling time is performed and output to the output unit 305 to prompt a doctor or occupational therapist to perform quantitative evaluation. For example, training of a subject using the training device 10A The results of quantitative evaluation based on the results are compiled into a large database, and the training effect evaluation computing device 30 also includes visualizing the training effect of the subject by displaying the result of comparison and displaying the progress of the improvement result. It can also be realized.

  Next, a training effect evaluation method according to this embodiment in which the forearm is a training target instead of the finger of the first embodiment will be described as a second embodiment.

  FIG. 25 shows the structure of a hemiplegic forearm function recovery training device (hereinafter simply referred to as training device 10B). As shown in FIG. 25, the training apparatus 10 </ b> B includes a moving table 212 including casters (universal wheels) 211, a forearm mounting table 213 provided at one end on the moving table 212, and a device base provided on the moving table 212. 214, a half cylinder 215 supported on the apparatus base 214 so as to be rotatable about a horizontal axis and having an open upper surface, a stick 216 provided in the half cylinder 215, and a half cylinder 215 A wrist support portion 217 provided at the end, a drive shaft 218 having one end connected to the half cylinder 215 and extending along the horizontal axis, and bearing bases 219 and 219 provided on the apparatus base 214 and rotatably supporting the drive shaft 218. A servo motor 221 provided on the apparatus base 214 for driving the drive shaft 218 and an encoder for measuring the rotation angle of the motor shaft 222 provided on the servo motor 221. The rotation angle information is acquired from the encoder 223 and the encoder 223, and the half cylinder 215 is rotated forward in one direction, stopped, reversed, and stopped repeatedly. In order to excite the nerve cells by causing the muscle to reflex by the tension of the muscle to be planned, the first angular velocity, the rapid acceleration, and the second angular velocity faster than the first angular velocity are controlled, and the muscle stimulation speed is reversed. A control unit 224 that performs a series of controls to apply resistance to maintain the muscle tension and maintain muscle tone, and a torque conversion unit 225 that calculates the torque applied to the stick 216 and sends it to the mechanism control unit 224 Become.

  Note that a part of the caster 211 may be a normal wheel (non-universal wheel). Further, the caster 211 may be omitted. In this case, the movable table 212 is a simple table. As long as it is a table, an appropriate table or desk can be used instead of the table. Therefore, whether or not the movable table 212 is arranged below the apparatus base 214 is arbitrary. However, since the servo motor 221 and the encoder 223 are heavy, it is recommended that the apparatus base 214 and the movable table 212 be integrated as in this embodiment in order to reduce the human burden.

  Further, the half cylinder 215 only needs to have a part of the cylinder, and the notch ratio is not limited to 50%. Furthermore, the half cylinder 215 is basically a half cylinder, but may be a polygonal cylinder.

  A torque detection mechanism 260 is provided between the half cylinder 215 and the drive shaft 218. The configuration and operation of this torque detection mechanism 260 will be described later.

  FIG. 26 is a perspective view of the training apparatus 10B, FIGS. 27A, 27B, and 27C are operational views of the half cylinder 215, and FIG. 28 is a diagram showing other dynamic rotational motion. For example, as shown in FIG. 26, the patient needs to train the left hand, for example, and places the wrist 256 on the wrist support portion 217 and the forearm portion 258 on the forearm platform 213. Turning the left wrist 256 viewed from the patient to the center of the body as indicated by the arrow (a) is called “pronation”, and turning outward as indicated by the arrow (b) is called “extraversion”. Call.

  The movement that is turned to the direction of the arrow (a) and the arrow (b) is called “automatic rotation movement”. In order to cause this automatic rotational motion to a greater extent, a motion in which a force is applied from the outside and rotated to the other is referred to as “other dynamic rotational motion”. For example, by performing “other dynamic rotational motion” in the direction of arrow (a), “automatic rotational motion” can be caused in the direction of arrow (b).

  In FIG. 27A, the wrist is not shown because the drawing becomes complicated, but the wrist is in a neutral state, and training can be started from this state. As shown in FIG. 27B, the servo motor 221 performs other dynamic rotational motion. Specifically, as shown in FIG. 28, first, the other dynamic rotational motion is performed at the first angular velocity, the rapid acceleration is performed, and the other dynamic rotational motion is performed at the second angular velocity. For example, the first angular velocity is 2.5 radians / second, and the second angular velocity is 11.0 radians / second. That is, the second angular velocity is about 2 to 5 times the first angular velocity.

  The sudden acceleration is set to the maximum acceleration that the servo motor 221 can produce. For example, the transition is made from the first angular velocity to the second angular velocity in 0.01 seconds. At this time, the angular acceleration is set to (11.0−2.5) /0.01=about 1000 radians / second. That is, after constant rotation at the first angular velocity, the muscle tension is increased by rapidly rotating at the rapid acceleration and the second angular velocity to restore the function. Then, stretch reflex is excited by muscle tone. The stretch reflex promotes the contraction of the muscle, and as shown in FIG. 27 (c), automatic rotation is performed at the patient's will. However, in order to maintain the muscle tone by maintaining the muscle stimulation with the servo motor 221, a light resistance force is generated.

  In order to generate a light resistance force by the servo motor 221, it is necessary for the patient to know the torque generated during automatic rotation. That is, if the torque is large, the resistance force (resistance torque) is increased, and if the torque is small, the resistance force is decreased, thereby avoiding a burden on the patient. For this purpose, a torque detection mechanism is provided. The shaft torque can be detected by attaching a strain gauge to the drive shaft 218. However, when the outer diameter of the drive shaft 218 is small, the amount of twisting of the shaft (see FIG. 30, reference sign θa) is small, so that measurement is difficult and measurement error increases. Therefore, as described below, a torque detection mechanism with good measurement accuracy is adopted.

  As shown in FIG. 29, the torque detection mechanism 260 includes a pin 261 erected on the outer periphery of the disc 242 so as to extend to the drawing table, and a rod-shaped member 238 that has one end fitted to the pin 261 and extends to the center of rotation of the disc 242. And a constricted portion 262 provided in the middle of the rod-shaped member 238, a strain cage 263 attached to the constricted portion 262, and a torque converting portion 225 for converting strain information obtained by the strain cage 263 into torque. .

  One end of the rod-shaped member 238 is fitted into the pin 261 through a long hole 264 extending in the longitudinal direction of the rod-shaped member 238, thereby allowing the rod-shaped member 238 to move with respect to the pin 261. In addition, the other end of the rod-shaped member 238 is sandwiched between other angular shaft portions 267 of the drive shaft 218 by flanges 265 and 266. The rod-shaped member 238 may be a round bar or a square bar in addition to the band plate, and the shape is arbitrary. However, since the width of both ends is large in the case of a band plate, one end can be easily attached to the disc 242 by utilizing this width, and the other end can be easily attached to the drive shaft 218.

  The operation of the torque detection mechanism 260 having such a configuration will be described next. As shown in FIG. 30A, torque Tm is applied to the disc 242 through the stick 216 by automatic rotation. This torque Tm is directly transmitted to the drive shaft 218 and indirectly transmitted to the drive shaft 218 via the rod-shaped member 238. If the directly transmitted torque is Ta and the indirectly transmitted torque is Ts, then Tm = Ta + Ts. The strain cage 263 is provided only on the rod-shaped member 238.

  As shown in FIG. 30 (b), the rod-shaped member 238 bends in a dogleg shape with the constricted portion 262 as a bending point due to the torque Ts. As the torque Ts increases, the bending increases. Due to this regularity, the torque Ts can be calculated structurally. Here, the length from the bending point to the drive shaft 218 is L1, the length from the bending point to the pin 261 is L2, the length of the constricted portion 262 is Lx, and the width of the constricted portion 262 is b. Although not shown in the figure, the thickness of the constricted portion 262 (the front and back dimensions of the drawing) is h. Also, let E be the longitudinal elastic modulus (Young's modulus) of the constricted portion 262.

  As shown in FIG. 30C, when the torque Ta acts on the drive shaft 218, the drive shaft 218 is slightly twisted. The angle θa in the figure increases as the torque Ta increases. Due to this regularity, the torque Ta can also be calculated structurally. Here, the drive shaft 218 has a length LL and a diameter R. Further, the lateral elastic modulus of the drive shaft 218 is G. Although detailed description is omitted, the torque Ta is a function of Ts as shown below.

  Since the torque Ts can be obtained by the strain gauge 263, the torque conversion unit 225 obtains the torque Tm based on the strain information obtained by the strain gauge 263. Resistive torque is generated by the servo motor 221 in accordance with the torque Tm. Details of this resistance torque will be described later. If an appropriate amount of resistance is applied during the automatic rotational movement, the patient's forearm can sustain stimulation and maintain muscle tone, and can rotate more greatly.

  In FIG. 30B, as the width b of the constricted portion 262 is smaller, the rod-shaped member 238 is bent more greatly, and a large strain is generated in the constricted portion 262. Since the strain gauge 263 is attached to the constricted portion 262 in this manner, the torque Tm can be obtained with high accuracy. That is, it is possible to detect torque with much higher accuracy than directly attaching a strain gauge to the drive shaft 218 shown in FIG.

  Next, how to obtain the resistance torque will be described. When the current angle of the servo motor 221 is θ, the virtual inertia is I, the virtual viscosity is C, and the virtual elasticity is K, the impedance equation for the sensed torque Tm is as follows.

  At the constant speed, the magnitude of the resistance force is determined by adjusting the value of C. During acceleration / deceleration, the resistance torque can be determined by adjusting K1 and K2. In particular, when it is desired to reduce the time delay, the value of I may be set to approach K2 = 0 after C is determined. Thus, the resistance torque Tn is set to an appropriate value.

  As shown in FIG. 28, the first angular velocity, the rapid acceleration, and the second angular velocity are set, the resistance torque is set, and then training is performed in the following cycle. In FIG. 31, the other dynamic rotation motion is a minus (−) region. As illustrated in FIG. 31, the mechanism control unit 224 controls the training device 10 </ b> B to perform a passive movement that includes the first angular velocity, the rapid acceleration, and the second angular velocity. That is, the patient's forearm is rotated by the servo motor 221. At this time, muscle tension increases due to the rapid acceleration and the second angular velocity.

  At the point P1, the drive by the servo motor 221 is stopped. Then, an extensional reflex that turns the arm in the reverse direction is induced, and the trainee then tries to voluntarily rotate the forearm in the reverse direction on his / her own will (corresponding to an automatic rotational movement by the extension reflex). ). However, light resistance is applied by a servo motor after the point P1. At the point P1, the angular velocity is − (minus) due to the inertia so far, but the angular velocity rapidly decreases due to the rotation in the reverse direction and turns to + (plus). At the end of rotation in the reverse direction, the rotation slows down and finally the angular velocity becomes zero. The mechanism control unit (reference numeral 224 in FIG. 25) repeats this cycle several tens of times, with the other dynamic rotational motion during normal rotation and the automatic rotational motion during reverse rotation as one cycle.

  Below, the myoelectric potential measurement for healthy subjects is performed using several training patterns, and a highly effective training pattern is established. Then, a verification experiment for a hemiplegic patient is performed to verify its effectiveness. The purpose of the experiment is to measure the myoelectric potential measurement that a stretch reflex can be induced more when the target speed is reached at infinite acceleration than when the target speed is reached at some acceleration by attaching a myoelectric sensor to the forearm to be trained. This is to confirm. In addition, this is for comparing the myoelectric potential responses in each pattern using the determined training pattern and confirming whether stretch reflex is effectively induced.

  As an experimental method, in order to confirm the reaction of the muscle in the supination movement, a change is measured by attaching a myoelectric potential sensor to the supination muscle. However, no conscious voluntary movements will be allowed for any supination movement. The number of training is 20 times, and the maximum training time is 60 seconds.

  FIG. 32 shows other dynamic rotational motion, automatic rotational motion, and the like in training. In the other dynamic rotation, a low speed operation is performed in the speed section 1 and a high speed rotation is performed in the speed section 2. Thereafter, an automatic rotational movement is performed. In the other dynamic rotational motion, since the training device 10B operates the patient's forearm, the myoelectric potential is low in the speed section 1, and when the speed section 2 is reached, the muscle tone is increased by the speed change of the training device 10B. Increases so that myoelectric potential increases rapidly. This reaction becomes stretch reflection.

  FIG. 33A shows a training result at an infinite acceleration, and FIG. 33B shows a training result at a finite acceleration. Comparing FIG. 33A and FIG. 33B, the change from the minimum point to the maximum point of the myoelectric potential from the passive movement to the automatic movement (the magnitude of the stretch reflection) is shown in FIG. Is bigger. Here, as an effective training, it is required to induce a stretch reflex of the forearm more. Therefore, the training method for accelerating to the target speed with infinite acceleration is selected.

  FIGS. 34 (a), (b), and (c) show comparisons of myoelectric potential responses in patterns determined from the above results. FIG. 34 (a) shows the experimental results of pattern 1, FIG. 34 (b) shows the pattern 2, and FIG. According to FIGS. 34 (a), (b), and (c), the acceleration change from the speed section 1 to the speed section 2 from the pattern 1 to the pattern 3 increases, and the change in the myoelectric potential also increases accordingly. Therefore, comparing the rotational speed of the automatic rotational motion and the area of the time domain between the horizontal axes in the three patterns, it can be seen that pattern 3 is the largest and pattern 1 is the smallest. From the above, it was confirmed that pattern 1 has the smallest training effect, pattern 3 has the greatest effect, and pattern 2 has an intermediate effect between patterns 1 and 3.

○ First experiment:
Experimental device: Training device 10B
Experimental conditions: 1st angular velocity, sudden acceleration, 2nd angular velocity Result: The graph shown to Fig.35 (a).

  According to the result of the first experiment, as shown in FIG. 34 (a), after the other dynamic rotational motion composed of the first angular velocity, the rapid acceleration, and the second angular velocity, the automatic rotational motion by the stretched reflection (by the stretched reflection). Rotational motion followed by optional rotational motion). The maximum angular velocity θa · max in the automatic rotational motion was about 6 radians / second. Further, since the product of the angular velocity and time becomes an angle, an integral value, that is, an area Sa surrounded by the curve and the time axis becomes a rotation angle.

○ Second experiment (comparative example):
Experimental device: Training device 10B
Experimental conditions: First angular velocity only (rapid acceleration, no second angular velocity)
Results: Graph shown in FIG. 35 (b). For comparison, a part of the curve in FIG.

  In the comparative example, as shown in FIG. 35 (b), after the other dynamic rotational motion consisting only of the first angular velocity, an automatic rotational motion by stretch reflection is caused. The maximum angular velocity θb · max in the automatic rotational motion was about 3 radians / second. An area Sb surrounded by the curve and the time axis is a rotation angle.

  The above two experiments were repeated a predetermined number of times (20 times). Then, as shown in FIG. 36, the maximum angular velocity θa · max in the example is 4.5 to 7.0 radians / second, and the maximum angular velocity θb · max in the comparative example is 2.5 to 5. 2 radians / second. As shown in FIG. 36, the maximum rotation angle Sa in the example is 1.9 to 2.0 radians, and the maximum rotation angle Sb in the comparative example is 1.65 to 1.75 radians. It was.

  In the automatic rotational movement by stretch reflex, the greater the maximum rotation angle, the greater the muscle contraction. That is, as shown in FIG. 34 (a), by performing other dynamic rotation including sudden acceleration, as shown in FIG. 37, it was confirmed that the rotation angle was increased and a large stretch reflection was obtained. .

  Here, the training effect evaluation calculation device 30 according to the second embodiment will be described. Except for the structure of the training device 10B, the training effect evaluation calculation device 30 of the second embodiment is the same as the first embodiment shown in FIG. 7, and the basic operation is the same as that of the first embodiment shown in FIG. is there. FIG. 38 shows the detailed procedure for calculating the automatic rotational speed by the forearm stretch reflection (step S20 in FIG. 8). FIG. 39 shows the product-sum operation of the automatic rotational movement speed by the forearm rotational speed and the sampling time. A detailed procedure for the process (step S30 in FIG. 8) is shown.

  As in the first embodiment, prior to the calculation processing of the automatic rotational motion speed (rotational angular velocity) by the stretch reflection, the arithmetic processing unit 300 sets the current automatic rotational motion speed “Vmini” by the stretch reflection to a predetermined register (not shown). The minimum speed (rotational angular speed) “Vminj” at the number of exercises is assigned to the maximum speed (rotational angular speed) “Vmaxj” at the predetermined number of exercises. However, j is the number of trainings.

  In FIG. 38, first, the arithmetic processing unit 300 determines whether Vi (rotational angular velocity) is positive or negative (step S311). If Vi is positive (step S311 “YES”), the arithmetic processing unit 300 further compares Vmaxj and Vi (step S312). If Vmaxj <Vi (step S312 “YES”), the current value is added to Vmaxj. Vi is set (step S313), and if Vmaxj <Vi is not satisfied (step S312 “NO”), the process returns to step S311 to obtain the maximum speed of the automatic rotational motion by extension reflection.

  On the other hand, if it is determined in step S311 that Vi is 0 or negative (step S311 “NO”), the arithmetic processing unit 300 determines whether Vi is positive or negative, and if it is still negative (step S314 ”). NO "), the process after step S311 is repeatedly executed to obtain the maximum automatic rotation speed of the forearm. If 0 or positive (step S314" YES "), j indicating the number of exercises is updated by 1 (step S315). . The above processing is repeatedly executed, for example, until the number of exercises j counts “20” (step S316 “YES”).

  FIG. 39 shows a detailed procedure for a product-sum operation process (step S40 in FIG. 8) of the rotation speed of the forearm and the sampling time. According to FIG. 39, the arithmetic processing unit 300 determines whether Vi (rotational angular velocity) is positive or negative (step S411). If Vi is positive (step S411 “YES”), the arithmetic processing unit 300 calculates the area Si of FIG. 34 by multiplying the sampling time Δt by the forearm rotational angular velocity Vi during the automatic rotational motion at each sampling time. (Step S412). FIG. 34 shows the relationship between the angular velocity on the time axis and the area Si calculated at that time.

  The arithmetic processing unit 300 repeats the product-sum operation described above only when Vi is positive, and when the forearm velocity V (i−N) in the automatic rotational motion is equal to 0 or negative (step S411 “NO”). Whether positive or negative is determined (step S413). If it is not positive (step S413 “NO”), the processing after step S411 is repeatedly executed. If it becomes 0 or positive (step S413 “YES”), j indicating the number of exercises is updated by +1 (step S414). The above processing is repeatedly executed until, for example, the number of exercises j counts “20” (step S415 “YES”).

  Returning to the basic operation flow of FIG. 8, the forearm automatic rotation speed calculation process (step S20) and the product-sum calculation process (step S30) of the automatic rotation speed and sampling time are repeatedly executed for the number of training iterations. (Step S40), the arithmetic processing unit 300 further calculates the average value or each value necessary for the boxplot (Step S50). Subsequently, the arithmetic processing unit 300 generates a display information by graphing the result, and displays it on the output unit 305 to prompt evaluation (step S60).

  The experiment described above is for evaluating the difference in the occurrence of stretch reflexes related to the training effect in which training is performed on a hemiplegic patient using three patterns whose effects have been confirmed by myoelectric potential measurement. The target rotational motion of the paralyzed patient targeted for training is supination motion. However, the supination motion includes a motion due to the stretch reflex and a voluntary motion caused by the stretch reflex. For this reason, the number of exercises was set to 20, and the maximum exercise time was set to 60 seconds.

  Since the patient who performed this experiment was not able to return to the initial point, external assistance was added. In the results shown below, the force and speed were set to zero. FIG. 40 shows other dynamic rotational motion, automatic rotational motion, and the like of the experimental results. Through a verification experiment using myoelectric potential, it is possible to determine whether the patient's stretch reflex is more induced by comparing the number of rotations in the automatic rotational movement and the area of the time axis. Therefore, it was decided to compare the area of the rotational speed and time axis in the automatic rotational motion.

  40 (a) to 40 (c) are graphs showing a part of the results of the demonstration experiment using the above-described pattern 1, pattern 2, and pattern 3, respectively. In both cases, the horizontal axis indicates the time axis and the vertical axis indicates the rotational speed. Comparing the areas of the rotational speed and the time axis in the automatic rotational motion in pattern 1 and pattern 2, it can be seen that the area of pattern 2 is larger. Similarly, comparing the area of the rotational speed and the time axis in pattern 2 and pattern 3, it can be seen that pattern 3 is larger. This can be more effectively induced than the stretch reflex of a hemiplegic patient by changing the training pattern.

  FIG. 41 shows an enlarged view of the partial time region of FIG. In the second embodiment, as described above, the degree of automatic rotation of the forearm is evaluated based on the area Si of the automatic rotational movement time region by the stretched reflection indicated by hatching in FIG. In addition to the evaluation of the degree of automatic rotation of the forearm, it is also possible to evaluate the automatic rotation agility indicating how much voluntary movement is caused by the stretch reflection using the maximum speed Vmax during automatic rotation. That is, as shown in FIG. 41, the automatic rotational agility is evaluated based on the maximum speed Vmax in the automatic rotational movement time region due to the stretched reflection indicated by hatching.

  FIG. 42 shows an example of automatic forearm rotation evaluation using a box-and-whisker diagram, and FIG. 43 shows an example of automatic forearm rotation evaluation using a box-and-whisker diagram. 42 and 43 for each training (with facilitation (practice), without facilitation (before training), with both facilitation and electrical stimulation, with facilitation and vibration stimulation, with facilitation, vibration stimulation and electrical stimulation. The combination angle, no prompting (before training), only the prompting) shows the rotation angle [rad]. 42 and 43, it can be understood that, in the forearm automatic rotation degree evaluation and the forearm automatic rotation agility evaluation of the test subject, a better result was obtained by training including prompting than without prompting. . In particular, it can be understood that when electrical stimulation and vibration stimulation are used in combination, training can be performed stably with little variation. In addition, by performing training without prompting before and after training and comparing them, it is possible to confirm the improvement in effectiveness by evaluating the forearm automatic rotation degree and the forearm automatic rotation agility evaluation.

  In the above evaluation, as in the first embodiment, the arithmetic processing unit 300 obtains the average value of the automatic rotational motion speed due to the stretch reflection, or the maximum value, the minimum value, the first quantile of the automatic rotational motion due to the stretch reflection, For example, the second quantile and the third quantile are obtained by calculation, and for example, output information necessary for expressing a box plot is generated and output to the output device 305.

  As described above, according to the training effect evaluation method of the second embodiment, instead of the selective exercise training that requires a lot of time and labor, the training device 10B is controlled to control the motion information and force sense. The effect of training when sensing information is performed and the training technique is quantitatively evaluated, and the forearm paralysis is improved by the promotion of voluntary forearm extension exercise and repetition. The program by this training effect evaluation method is transplanted to the training effect evaluation calculation device 30, and this training effect evaluation calculation device 30 adjusts and adapts the training method according to the evaluation result of the training effect, and maximizes the training effect. Thus, hemiplegic forearm function recovery can be promoted instead of a doctor or occupational therapist, and labor can be reduced and medical effects can be improved.

  In addition, according to the training effect evaluation method of the second embodiment, quantitatively analyzing and evaluating changes in exercise and force when performing the push technique using the motion information and force information of the training device 10B. Therefore, further improvement of the training effect can be realized by changing the training pattern by feeding back the evaluation result to the training device 10B.

  Note that the program according to this embodiment is executed by a computer (training effect evaluation computing device 30) in FIG. 1, for example, and includes a training mechanism that performs training to stimulate stretch muscles and stimulate stretch muscles. A training effect evaluation calculation for evaluating the degree of motor function recovery of the hemiplegic site using the hemiplegic motor function recovery training device 10 that promotes recovery of the motor function of the hemiplegic site by repeating the training a predetermined number of times This is a program of the device 30. Then, for example, as shown in FIG. 8, the program stores the product sum of the speed of the automatic movement by the stretch reflection of the hemiplegic part obtained from the detected movement information of the training mechanism and a predetermined sampling time. The procedure for performing the calculation (steps S10 to S50) and the procedure for outputting the result of the product-sum operation (step S60) are executed.

  According to the program according to the present embodiment, the training effect evaluation computing device 30 calculates the sum of products of the speed of the automatic motion by the stretch reflection of the hemiplegic region obtained from the detected motion information of the training mechanism and the predetermined sampling time. By executing the process, the doctor or occupational therapist quantitatively determines the degree of motor function recovery of the paralyzed part of the hemiplegic patient by displaying the obtained result on the output device, for example, and prompting evaluation. As a result, the training effect can be improved.

  INDUSTRIAL APPLICABILITY The present invention can be advantageously used for evaluation of recovery of motor function in training using a training device that trains the finger or forearm of a patient whose left or right half is paralyzed to promote motor function recovery.

  DESCRIPTION OF SYMBOLS 1 ... Hemiplegic motor function recovery training system, 10 ... Hemiplegic motor function recovery training apparatus (10A, 10B: Training apparatus), 20, 150, 224 ... Mechanism control part, 21,214 ... Apparatus base, 30 ... Training effect evaluation Arithmetic unit, 41, 47, 221 ... Servo motor, 52, 53, 223 ... Encoder, 105 ... Finger flexion / extension mechanism, 108 ... Finger tapping mechanism, 215 ... Half cylinder, 216 ... Stick, 300 ... Arithmetic processing unit, 301 ... Storage unit, 302, 303 ... A / D converter, 304 ... Communication unit, 305 ... Output unit

Claims (9)

A training apparatus for recovering a hemiplegic motor function, comprising a training mechanism for performing a training for inducing stretch reflex by stimulating muscles at a hemiplegic site, and repeating the training a predetermined number of times to promote recovery of the motor function of the hemiplegic site An evaluation method using the training effect evaluation calculation device used,
A first step of multiply-accumulating a speed of automatic movement by a stretch reflection of the hemiplegic site obtained from the detected movement information of the training mechanism and a predetermined sampling time;
A training effect evaluation method characterized by comprising: A second step of determining a maximum speed of automatic movement by the stretch reflection;
The training effect evaluation method according to claim 1, further comprising: A third step of performing a comparison operation between each of the results of the product-sum operations for the predetermined number of times and an average value of the results of the product-sum operations;
The training effect evaluation method according to claim 1, further comprising: A fourth step of prompting an evaluation of the degree of recovery of the paralyzed motor function by displaying at least one of the result of the product-sum operation, the maximum speed of the automatic motion by the stretched reflection, and the result of the comparison operation on a display device;
The training effect evaluation method according to claim 3, which is dependent on claim 2. The maximum value, the minimum value, the first quantile, the second quantile, and the third quantile of the automatic motion due to the stretched reflection are obtained by calculation to generate output information that represents a box-and-whisker plot, and output it to the output device. A fifth step for prompting an evaluation of the degree of hemiplegic motor function recovery by
The training effect evaluation method according to claim 1, further comprising: A training apparatus for recovering a hemiplegic motor function, comprising a training mechanism for performing a training for inducing stretch reflex by stimulating muscles at a hemiplegic site, and repeating the training a predetermined number of times to promote recovery of the motor function of the hemiplegic site Use, a training effect evaluation computing device that evaluates the degree of motor function recovery of the hemiplegic site,
An arithmetic processing unit that performs a product-sum operation on the speed of the automatic movement by the stretch reflection of the hemiplegic site obtained from the detected movement information of the training mechanism and a predetermined sampling time;
An output unit that outputs the result of the product-sum operation calculated by the arithmetic processing unit;
A training effect evaluation calculation device characterized by comprising: The training mechanism is:
A device base,
A hand platform that reciprocates up and down with the vicinity of the wrist of the hand placed on the device base as a swing fulcrum;
A finger bending / extension mechanism that does not hinder one of the fingers of the hand placed on the table to bend and bends, and does not hinder the stretching operation of the bent finger to return;
A first servo motor for driving the finger bending and extending mechanism;
A first encoder for measuring a rotation angle of the first servomotor;
A finger striking mechanism provided in the hand platform for inducing the stretching motion by striking the root of the finger pressed down by the finger bending / stretching mechanism;
A second servo motor for driving the finger tapping mechanism;
A second encoder for measuring a rotation angle of the second servomotor;
A mechanism control unit,
The mechanism controller is
Information on the rotation angle is acquired from the first encoder and the second encoder, and when the finger is bent, the finger is stimulated by a tapping operation by the finger tapping mechanism to induce stretch reflex of the finger. Control and continuous contact with the finger when the finger is stretched and reflected, and further continuous finger automatic extension is provided by a tracing operation in which the finger hitting mechanism moves to the base of the finger along with the stretched reflection. 7. The training effect evaluation calculation apparatus according to claim 6, wherein the control to enable is repeatedly performed by a predetermined number of training repetitions. The training mechanism is:
A device base,
A half cylinder that is supported by the device base so as to be rotatable around a horizontal axis and whose upper surface is open;
A stick which is provided in the half cylinder and attaches a finger of the forearm or can be grasped with a finger;
A wrist support provided in the half cylinder and supporting the wrist;
A drive shaft having one end connected to the half cylinder and extending horizontally;
A servo motor that is provided in the apparatus base and drives the drive shaft;
An encoder provided in the servo motor for measuring the rotation angle of the motor shaft, and a mechanism control unit;
The mechanism controller is
The rotation angle information is acquired from the encoder, and the half cylinder is rotated forward, stopped, reversely rotated, and stopped repeatedly. In order to excite the nerve cells, the first angular velocity, the rapid acceleration, and the second angular velocity higher than the first angular velocity are used to control the muscle stimulation speed. In the reverse rotation, the muscle stimulation is sustained and the muscle tone is maintained. Therefore, the training effect evaluation computing device according to claim 6, wherein the control for enabling the continuous forearm automatic rotation by the operation of applying the resistance force is repeatedly performed a predetermined number of times of training. A hemiplegic exercise that is controlled by a computer and has a training mechanism for stimulating muscles at a hemiplegic site to induce stretch reflex, and promoting the recovery of the motor function of the hemiplegic site by repeating the training a predetermined number of times A program for a training effect evaluation computing device that evaluates the degree of motor function recovery of the hemiplegic site using a function recovery training device,
In the computer,
A procedure for performing a product-sum operation between the speed of automatic movement by stretch reflection of the hemiplegic site obtained from the detected movement information of the training mechanism and a predetermined sampling time;
A procedure for outputting the result of the product-sum operation;
A program that executes

Images