Hand neurosurgery postoperative rehabilitation training device

Abstract

The invention relates to the field of medical instruments, in particular to a hand neurosurgery postoperative rehabilitation training device which comprises a wearable mechanical glove and a controller. The wearable mechanical glove is used for driving hand joints of a patient to move and collecting finger action data of the patient; the controller is used for marking a first finger with limited neurological function and a second finger with healthy neurological function of the patient; motion relation data among the fingers under the actions are stored in the controller; and the controller is used for controlling the wearable mechanical glove to drive and pull the first finger according to the action data of the first finger corresponding to the second finger action completion degree percentage in the motion relation data. By the adoption of the technical scheme, under the preset action, according to the percentage of the action completed by the finger with the healthy nerve function, the finger with the limited nerve function can be dragged to move, and a patient can control the traction process of the finger with the limited nerve function at will under the coherent action.

Description

Technical Field

[0001] The present invention relates to the field of medical devices, and particularly to a rehabilitation training device for hand neurosurgery after operation. Background Art

[0002] Hand nerve injuries are usually caused by trauma, diseases or surgeries, which may lead to the loss or limitation of hand functions, seriously affecting the daily life and working ability of patients. Postoperative rehabilitation training is a key link in restoring hand functions, aiming to promote nerve regeneration, muscle strength recovery and joint range of motion improvement through systematic exercises. Traditional hand rehabilitation training mainly relies on the manual operations of physical therapists and simple auxiliary tools, such as rubber balls, handgrip strengtheners, etc. Although these methods are effective to a certain extent, there are many limitations. First of all, the traditional training methods lack precise control and feedback mechanisms, and it is difficult to make personalized adjustments according to the individual differences and rehabilitation progress of patients. Secondly, the intensity and frequency of manual training are difficult to quantify, which may lead to insufficient or excessive training and affect the rehabilitation effect. In addition, the long-term reliance on physical therapists has a high labor cost, and patients lack professional guidance when performing self-training at home, which is likely to result in incorrect training postures or methods, further affecting the rehabilitation process.

[0003] With the progress of technology, intelligent and automated technologies have gradually been applied to the field of rehabilitation medicine, and rehabilitation training devices for hand neurosurgery after operation have emerged as the times require. These devices usually combine sensors, microprocessors, mechanical structures and software algorithms to achieve precise force control, motion trajectory monitoring and real-time feedback. For example, some devices use a force feedback system and virtual reality technology to simulate hand movements in daily life to help patients perform functional training in a safe environment. In addition, intelligent rehabilitation devices can also record the training data of patients, generate personalized rehabilitation plans, and share them with doctors or therapists through a cloud platform to achieve remote monitoring and guidance.

[0004] In the prior art, most of them adopt active or passive training methods. Among them, active training is for patients to perform nerve control training by themselves, and passive training is to intervene with an external driving force to make patients move in accordance with the external driving force. Patent publication number CN119097525A discloses a hand function rehabilitation system controlled by a nerve interface, which identifies the motion intention of patients by collecting the myoelectric signals of the forearms and drives accordingly with a corresponding external driving force. Patent publication number CN114869691A analyzes the motion intention by detecting the force conditions of various parts of the patient's hand, and then drives with a corresponding external driving force. However, the area required to cover the patient's limb for collecting the myoelectric signals of the forearms is too large, making the device cumbersome to carry and use. Analyzing the motion intention by detecting the force conditions of various parts of the patient's hand can simplify the device volume. However, if the finger nerve function is limited, the reaction will be rigid and slow, and it is difficult to effectively convey instructions, and patients cannot train as they wish. Summary of the Invention

[0005] To solve the above problems, the present invention provides a rehabilitation training device for hand neurosurgery, which is used to, under a preset action, according to the percentage of the completion of the action by the fingers with healthy nerve function, traction the fingers with limited nerve function to move, so that the patient can, under a coherent action, freely control the traction process of the fingers with limited nerve function.

[0006] To achieve the above object, the technical solution of the present invention is as follows: A rehabilitation training device for hand neurosurgery, comprising a wearable mechanical glove and a controller;

[0007] The wearable mechanical glove is used to drive the movement of the patient's hand joints and collect the finger movement data of the patient;

[0008] The controller is used to mark the first finger with limited nerve function and the second finger with healthy nerve function of the patient;

[0009] The controller stores the movement relationship data between the fingers under each action, and the movement relationship data includes the finger movement data of each finger, and the finger movement data of each finger is paired according to the percentage of the completion of the single finger movement;

[0010] The controller is used to, under a preset action, according to the second finger movement data collected by the wearable mechanical glove, and according to the movement data of the first finger corresponding to the percentage of the completion of the second finger movement in the movement relationship data, control the wearable mechanical glove to drive and traction the first finger.

[0011] The beneficial effects of adopting the above solution are as follows:

[0012] 1. In this solution, rehabilitation training is carried out for patients with limited nerve function in some fingers. The patient wears the wearable mechanical glove, which has the functions of finger movement driving and movement data collection of each finger respectively. It can drive the movement of the fingers with limited nerve function respectively, traction the fingers to carry out corresponding activities, and collect the finger movement data of the fingers with healthy nerve function.

[0013] 2. In this solution, the user can preset actions, and the patient can perform corresponding rehabilitation training by repeating the preset actions. During this process, the movement process of the second finger with healthy nerve function is collected in real time, and according to the completion degree of the second finger within the corresponding action, the wearable mechanical glove is controlled to drive the first finger with limited nerve function to perform the corresponding action. The patient can freely control the execution speed of the second finger, so that the first finger will also be driven related to the execution speed of the second finger. The patient can also limit the percentage of the final completion degree of the second finger, so as to control the maximum movement amplitude of the first finger. During the whole training process, the driving speed and amplitude of the restricted first finger are controlled by the patient himself, and the overall hand movement is coherent, still maintaining the cooperative linkage relationship between the fingers.

[0014] Furthermore, the number of the first finger and the second finger marked by the controller is at least 1.

[0015] Beneficial effect: Since the patient maintains the cooperative linkage relationship between the fingers during the whole training process, the number of the first finger and the second finger marked by the controller can be multiple. After being marked as multiple, the movement relationship data is still used for driving.

[0016] Furthermore, the movement relationship data is collected based on the hand actions of the preset sample population. The controller classifies the movement habit gradients of each action according to the different finger movement conditions of the sample population under the same action, based on the movement amplitude and direction angle gradient of each finger.

[0017] Beneficial effect: There are differences in the movement habits of patients. Since only the first finger with limited nerve function is driven in the coherent movement formed during the training process, if the preset action is different from the patient's habit, the patient may feel uncomfortable. Therefore, big data sample population collection is adopted, and by providing various movement habit options, appropriate action parameters are selected.

[0018] Furthermore, it also includes a trained BP neural network. The BP neural network is used to train according to the mapping data samples of the demographic characteristics and movement habit gradients of the sample population, and the BP neural network is used to input the demographic characteristics of the patient and output the movement habit gradient of the patient.

[0019] Beneficial effect: There is a relationship between the movement habit and the demographic characteristics of the patient. Therefore, it can be trained based on the demographic characteristics of the sample population, so as to construct a BP neural network that maps the demographic characteristics and movement habits. The demographic characteristics of the patient are input into the BP neural network, so as to obtain the movement habit paired with the patient.

[0020] Furthermore, the demographic characteristics include the patient's age, gender, height, weight, race, occupation and location.

[0021] Beneficial effects: Patients of different ages, genders, heights, weights, ethnicities, occupations, and locations may have different exercise habits. Therefore, BP neural network learning is performed based on the above parameters to establish a mapping relationship associated with the exercise habits of the patients.

[0022] Furthermore, the controller is also used to preset the training actions; the controller is also used to formulate a training guideline for the patient. The training actions are divided into gradients according to the movement amplitude intensity of each action, and the training guideline is to gradually increase the movement amplitude intensity of the training actions along the total training duration.

[0023] Beneficial effects: The controller can set the actions that the patient needs to perform, and as the total training time of the patient increases, action switching is performed according to the different movement amplitudes of each action.

[0024] Furthermore, the finger movement data includes the activity angle of the metacarpophalangeal joint and the bending angle of the interphalangeal joint. The combinations of the activity angle of the metacarpophalangeal joint and the bending angle of the interphalangeal joint of the training actions preset by the controller are different under the same movement amplitude intensity, and the number of training actions preset by the controller under the same movement amplitude intensity is at least 1.

[0025] Beneficial effects: The joint movements of the fingers include the opening and closing between the fingers and the bending of the fingers. Therefore, different action types can be arranged and combined with different opening and closing and finger bending. Therefore, under the same movement amplitude intensity, multiple actions are set, and the patient can perform different action trainings by changing the combination of the activity angle of the metacarpophalangeal joint of the second finger and the bending angle of the interphalangeal joint.

[0026] Furthermore, the BP neural network is also used to train based on the mapping data samples of the patient's demographic characteristics, training guidelines, training actions assigned under each same movement amplitude intensity, and rehabilitation effects. The BP neural network is used to input the patient's demographic characteristics and output the training guidelines and training action types that meet the preset rehabilitation effects under the same movement amplitude intensity.

[0027] Beneficial effects: Different training modes can achieve different training effects. Through the BP neural network, based on the patient's demographic characteristics, the training guidelines and the action combinations under the same movement amplitude intensity under the training guidelines are trained. Thus, a mapping of the rehabilitation effects of patients with different demographic characteristics under each training guideline and action combination under the same movement amplitude intensity is established. By outputting the training guidelines and action combinations under the preset rehabilitation effects, targeted training is performed on the patient.

[0028] Furthermore, the wearable mechanical gloves include several mechanical joints, the position and number of which correspond to the joints of the patient's hand. The mechanical joints at the patient's metacarpophalangeal joints are used to drive the patient's fingers to bend, straighten, open and close, and the mechanical joints at the patient's phalanges are used to drive the patient's fingers to bend and straighten. The mechanical joints are all equipped with angle sensors, and the driving mode of the mechanical joints is one of liquid drive, gas drive and electric drive. There is a casing between the mechanical joints that fits the patient's hand.

[0029] Beneficial Effects: The mechanical joints can drive the patient's hand joints. The housing connects the mechanical joints, attaching them to the housing to form a glove-like structure. The patient wears the housing and mechanical joints, pairing the mechanical joints with the patient's hand joints.

[0030] Furthermore, the controller is also used to obtain the load of the mechanical joint when driving the patient's finger, and when the mechanical joint load is greater than a preset value, the mechanical joint driving is stopped.

[0031] Beneficial Effects: Patients may make mistakes, such as moving their second finger too dramatically, which can cause the first finger's movement to change too dramatically. Therefore, the controller monitors the load on the mechanical joints as they drive the patient's fingers, determining the resistance generated by the current drive when pulling the patient's first finger. If the resistance exceeds a preset value, indicating that the patient's first finger is moving too stiffly, the controller immediately stops the mechanical joint drive to avoid causing excessive damage to the patient.

[0032] Additional aspects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description which follows, or may be learned by practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is a schematic diagram of a wearable mechanical glove according to an embodiment of the hand neurosurgery postoperative rehabilitation training device of the present invention;

[0034] Figure 2 This is a control logic diagram of an embodiment of the hand neurosurgery postoperative rehabilitation training device of the present invention;

[0035] Figure 3 A schematic diagram of the movement habit gradient logic of an embodiment of the hand neurosurgery postoperative rehabilitation training device of the present invention;

[0036] Figure 4 This is a logical diagram of the training policy and training action planning of an embodiment of the hand neurosurgery postoperative rehabilitation training device of the present invention.

[0037] The reference numerals in the drawings of the specification include: 1. mechanical joint at the metacarpophalangeal joint; 2. mechanical joint at the phalanges; 3. housing; 4. angle sensor. Detailed implementation manners

[0038] The technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative work belong to the scope of protection of the present invention.

[0039] In the description of the present invention, it should be noted that the orientation or positional relationship indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. is based on the orientation or positional relationship shown in the accompanying drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention. In addition, the terms "first", "second", and "third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0040] In the description of the present invention, it should be noted that unless otherwise clearly specified and defined, the terms "installed", "connected", and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0041] The following is a further detailed description through specific implementation manners:

[0042] Embodiment 1:

[0043] As shown in the attached... Figures 1-4 A postoperative rehabilitation training device for hand neurosurgery, including a wearable mechanical glove and a controller. The controller is signal-connected to the wearable mechanical glove by wired or wireless connection.

[0044] The wearable mechanical glove is used to drive the movement of the patient's hand joints and collect the finger movement data of the patient. The wearable mechanical glove includes a number of mechanical joints, and the positions and numbers of the mechanical joints correspond to the patient's hand joints. The mechanical joint 1 at the patient's metacarpophalangeal joint is used to drive the patient's fingers to bend, straighten, open, and close. The mechanical joint 1 at the patient's metacarpophalangeal joint is driven by an electric universal joint. The mechanical joint 2 at the patient's phalanx is used to drive the patient's fingers to bend and straighten, and the mechanical joint 2 at the patient's phalanx is driven by a micro motor. Angle sensors 4 are installed on all mechanical joints, and a housing 3 that fits the patient's hand is provided between all mechanical joints.

[0045] The controller is used to mark the first finger with limited nerve function of the patient and the second finger with healthy nerve function.

[0046] The controller stores the motion relationship data between each finger under each action. The motion relationship data includes the action data of each finger. Among them, the action data of each finger is paired according to the completion percentage of a single finger action.

[0047] The controller is also used to preset the actions of the training; the controller is also used to formulate the training guidelines for the patient. The actions of the training are gradient-divided according to the motion amplitude intensity of each action. The training guideline is to gradually increase the motion amplitude intensity of the training action along the total training duration. The controller is used to drive and traction the first finger by the wearable mechanical glove according to the action data of the second finger collected by the wearable mechanical glove under the preset action and the action data of the first finger corresponding to the completion percentage of the second finger action in the motion relationship data.

[0048] This embodiment is applicable to the rehabilitation training of patients with limited nerve function in some fingers. The patient wears a wearable mechanical glove, which has the functions of finger motion driving and action data collection for each finger, can drive the fingers with limited nerve function respectively, traction the fingers to perform corresponding activities, and collect the action data of the fingers with healthy nerve function.

[0049] The mechanical joint 1 at the metacarpophalangeal joint of the wearable mechanical glove is driven by an electric universal joint, which can adapt to the degrees of freedom of the metacarpophalangeal joint and can drive the metacarpophalangeal joint to open and close and the phalanges to bend respectively. The mechanical joint 2 at the phalanx is driven by a micro motor for phalanx bending. The angle sensor 4 is used to collect the rotation angles of each hand joint, so as to collect the finger action data.

[0050] The training guideline covers the action gradients of multiple motion amplitude intensities and is switched according to the training duration of the patient. When the patient is training, he can practice actions while wearing the wearable mechanical glove. At this time, during the process of performing actions, the second finger with healthy nerve function can smoothly complete the action practice, while for the first finger with limited nerve function, due to nerve damage, its actions are difficult to execute or there are situations such as stiffness, jamming, and tremors. During this process, the motion process of the second finger with healthy nerve function will be collected in real time, and according to the completion degree of the second finger in the corresponding action, the wearable mechanical glove will be controlled to drive the first finger with limited nerve function to perform the corresponding action. The patient can freely control the execution speed of the second finger, so that the first finger will also be driven related to the execution speed of the second finger. The patient can also limit the completion percentage of the second finger finally, so as to control the maximum motion amplitude of the first finger. During the whole training process, the driving speed and amplitude of the restricted first finger are controlled by the patient himself, and the overall hand movement is coherent, still maintaining the cooperative linkage relationship between the fingers.

[0051] For example, the motion relationship data of the second finger and the first finger are both performing flexion. When the second finger performs 0% to 10% of the flexion amplitude process, the first finger will also synchronously perform 0% to 10% of the flexion under the traction of the wearable mechanical glove. The patient can adjust the speed of the second finger's flexion through their own feelings, such as pain, touch, etc. feedback. Keep the traction of the wearable mechanical glove at a moderate level, not too aggressive nor too gentle. In addition, for example, when the first finger bends to 30%, the patient feels severe pain. At this time, the patient can reduce the flexion degree of the second finger, so that the first finger also reduces the flexion degree, thereby relieving the symptoms in time and controlling the flexion degree of the second finger within 30% at the current stage to avoid excessive training amplitude causing damage to the patient.

[0052] Embodiment 2:

[0053] The difference from the above embodiment is that the number of the first finger and the second finger marked by the controller is at least 1.

[0054] Since during the entire training process, the patient maintains the cooperative linkage relationship between the fingers, the number of the first finger and the second finger marked by the controller can be multiple. After being marked as multiple, the motion relationship data is still used for driving.

[0055] Embodiment 3:

[0056] The difference from the above embodiment is that the motion relationship data is collected based on the hand movements of a preset sample population. The controller classifies the motion habits of each action according to the different finger motion conditions of the sample population under the same action, and according to the motion amplitude and direction angle gradient of each finger.

[0057] It also includes a trained BP neural network. The BP neural network is used to train according to the mapping data samples of the demographic characteristics and motion habit gradients of the sample population. The demographic characteristics at least include the patient's age, gender, height, weight, race, occupation, and location. The number of the sample population is at least greater than 100%, and the learning rate of the BP neural network is between 0.2 - 0.5. The BP neural network is used to input the patient's demographic characteristics and output the patient's motion habit gradient.

[0058] Patients have different exercise habits. Since only the first finger with limited nerve function is driven in the continuous movements formed during training, if the preset movements are different from the patients' habits, the patients may feel awkward. Therefore, a large sample of the population is collected, and by providing various exercise habit options, appropriate movement parameters are selected. There is a relationship between exercise habits and the demographic characteristics of patients. Therefore, training can be carried out based on the demographic characteristics of the sample population, and a BP neural network that maps demographic characteristics and exercise habits can be constructed. The demographic characteristics of the patient are input into the BP neural network to obtain the exercise habits paired with the patient.

[0059] Patients of different ages, genders, heights, weights, ethnicities, occupations, and locations may have different exercise habits. Therefore, BP neural network learning is performed based on the above parameters to establish a mapping relationship associated with the exercise habits of patients.

[0060] Example 4:

[0061] The difference from the above embodiment is that the finger movement data includes the joint movement angle of the metacarpophalangeal joint and the bending angle of the phalangeal joint. The combinations of the joint movement angle of the metacarpophalangeal joint and the bending angle of the phalangeal joint of the training movements under the same exercise amplitude intensity preset by the controller are all different, and the number of training movements under the same exercise amplitude intensity preset by the controller is at least 1.

[0062] The BP neural network is also used for training according to the mapping data samples of the patient's demographic characteristics, training guidelines, training movements allocated under each same exercise amplitude intensity, and rehabilitation effects. The number of patient samples for training is greater than 300. The BP neural network is used to input the patient's demographic characteristics and output the training guidelines and the types of training movements under the same exercise amplitude intensity that meet the preset rehabilitation effects.

[0063] The joint movements of the fingers include the opening and closing between the fingers and the finger bending. Therefore, different action types can be arranged and combined with different opening and closing and finger bending. Therefore, under the same exercise amplitude intensity, multiple actions are set, and the patient can perform different action trainings by changing the combination of the joint movement angle of the second finger metacarpophalangeal joint and the bending angle of the phalangeal joint.

[0064] Different training modes can achieve different training effects. Based on the patient's demographic characteristics, the BP neural network trains the training guidelines and the action combinations under the same exercise amplitude intensity under the training guidelines. Thus, a rehabilitation effect mapping of the actions of patients with different demographic characteristics under each training guideline and the same exercise amplitude intensity is established. By outputting the training guidelines and action combinations under the preset rehabilitation effects, targeted training is carried out on the patients.

[0065] Example 5:

[0066] The difference from the above embodiment is that the controller is further configured to obtain the load when the mechanical joint drives the patient's finger, and stop the mechanical joint drive when the mechanical joint load is greater than a preset value.

[0067] The patient may make mistakes. For example, the movement amplitude of the second finger is too large, resulting in too large a change in the movement amplitude of the first finger. Therefore, the controller monitors the load when the mechanical joint drives the patient's finger, judges the resistance generated when pulling the patient's first finger during the current driving process. When the resistance is greater than the preset value, the movement of the patient's first finger is too rigid at present, and the mechanical joint drive is stopped in time to avoid damaging the patient.

[0068] Obviously, the above embodiments are only examples given for clear illustration, and are not limitations on the implementation manners. For those of ordinary skill in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation manners here. And the obvious changes or modifications derived therefrom still fall within the protection scope of the present invention.

Claims

1. A rehabilitation training device for hand neurosurgery, comprising a wearable mechanical glove and a controller; characterized in that, The wearable mechanical glove is used to drive the movement of the patient's hand joints and collect the finger movement data of the patient; The controller is used to mark the first finger with limited nerve function and the second finger with healthy nerve function of the patient; The movement relationship data between each finger under each action is stored in the controller. The movement relationship data includes the action data of each finger, and the action data of each finger is paired according to the completion percentage of a single finger action; The controller is used to drive and traction the first finger by the wearable mechanical glove according to the action data of the first finger corresponding to the completion percentage of the second finger action in the movement relationship data based on the action data of the second finger collected by the wearable mechanical glove under the preset action.

2. The hand neurosurgery postoperative rehabilitation training device according to claim 1, wherein, The number of the first fingers and the second fingers marked by the controller is at least 1.

3. The hand neurosurgery postoperative rehabilitation training device according to claim 2, wherein, The movement relationship data is collected based on the hand actions of the preset sample population. The controller classifies the movement habit gradients of each action according to the finger movement conditions of the sample population under the same action and the movement amplitude and direction angle gradients of each finger.

4. The hand neurosurgery postoperative rehabilitation training device according to claim 3, characterized in that, It also includes a trained BP neural network. The BP neural network is used to train according to the mapping data samples of the demographic characteristics and movement habit gradients of the sample population. The BP neural network is used to input the demographic characteristics of the patient and output the movement habit gradient of the patient.

5. The hand neurosurgery postoperative rehabilitation training device according to claim 4, characterized in that, The demographic characteristics include the patient's age, gender, height, weight, race, occupation and location.

6. The hand neurosurgery postoperative rehabilitation training device according to claim 5, wherein, The controller is also used to preset the training actions; the controller is also used to formulate the patient training policy. The training actions are gradient-divided according to the movement amplitude intensity of each action, and the training policy is to gradually increase the movement amplitude intensity of the training actions along the total training duration.

7. The hand neurosurgery postoperative rehabilitation training device according to claim 6, characterized in that, The finger action data includes the metacarpophalangeal joint activity angle and the phalangeal joint bending angle. The combinations of the metacarpophalangeal joint activity angle and the phalangeal joint bending angle of the training actions under the same movement amplitude intensity preset by the controller are different, and the number of training actions under the same movement amplitude intensity preset by the controller is at least 1.

8. The hand neurosurgery postoperative rehabilitation training device according to claim 7, characterized in that, The BP neural network is also used to train according to the mapping data samples of the patient's demographic characteristics, training policy, training actions assigned under the same movement amplitude intensity and rehabilitation effect. The BP neural network is used to input the patient's demographic characteristics and output the training policy and the type of training actions under the same movement amplitude intensity that meet the preset rehabilitation effect.

9. The hand neurosurgery postoperative rehabilitation training device according to claim 8, wherein, The wearable mechanical glove includes several mechanical joints. The positions and numbers of the mechanical joints correspond to the hand joints of the patient. The mechanical joint (1) at the patient's metacarpophalangeal joint is used to drive the patient's fingers to bend, straighten, open and close. The mechanical joint (2) at the patient's phalanx is used to drive the patient's fingers to bend and straighten. Angle sensors (4) are provided on all mechanical joints. The driving mode of the mechanical joints is one of liquid driving, gas driving and electric driving. A casing (3) that fits the patient's hand is provided between the mechanical joints.

10. The hand neurosurgery postoperative rehabilitation training device according to claim 9, wherein The controller is also used to obtain the load when the mechanical joint drives the patient's finger. When the load of the mechanical joint is greater than the preset value, the driving of the mechanical joint is stopped.

Images