Gesture recognition upper limb rehabilitation evaluation training system combined with artificial intelligence

Abstract

The utility model provides an upper limb rehabilitation evaluation training system combined with artificial intelligence gesture recognition, and relates to the technical field of rehabilitation equipment. The system is installed on a medical sickbed and comprises a stand column; the first end of the middle-position rocker arm is vertically connected to the side wall of the stand column in a sliding mode; an angle sensor is mounted at the rotary joint of the front rocker arm and the middle rocker arm; the arm support is driven by the second end of the front rocker arm to extend or retract, and a pressure sensor and an infrared vision sensor are installed on the arm support; and the connection limiting stopper is installed between the front rocker arm and the middle rocker arm, and the rotation angle of the front rocker arm is limited through the connection limiting stopper. The training state of the patient can be obtained in combination with the sensor, accurate feedback data is provided for a therapist, and therefore a training scheme better meeting the individual requirements of the patient is formulated, and the training effect is effectively improved.

Description

Technical Field

[0001] The utility model relates to the technical field of rehabilitation equipment, in particular to an upper limb rehabilitation evaluation and training system combining artificial intelligence gesture recognition. Background Art

[0002] With the development and application of technology, a variety of upper limb rehabilitation training devices have emerged: upper limb suspension rehabilitation devices, upper limb weight-bearing rehabilitation devices, upper limb resistance training, upper limb robot devices, etc. Through timely rehabilitation training, the occurrence of these complications can be effectively prevented, and the pain and discomfort of patients can be alleviated. However, traditional rehabilitation suspension systems often rely on the experience and judgment of therapists to set training parameters, making it difficult to achieve personalized and precise treatment, and difficult to adjust parameters in real time during training, resulting in limited training effects.

[0003] Therefore, there is an urgent need for an upper limb rehabilitation evaluation and training system combining artificial intelligence gesture recognition that can achieve precise treatment and effectively improve training effects. Content of the Utility Model

[0004] The purpose of the utility model is to provide an upper limb rehabilitation evaluation and training system combining artificial intelligence gesture recognition, which solves the technical problems existing in the prior art, such as difficult to achieve personalized and precise treatment, difficult to adjust parameters in real time during training, resulting in limited training effects. The many technical effects that can be produced by the preferred technical solution among the many technical solutions provided by the utility model are described in detail below.

[0005] To achieve the above purpose, the utility model provides the following technical solutions:

[0006] An upper limb rehabilitation evaluation and training system combining artificial intelligence gesture recognition provided by the utility model is installed on a medical bed and includes:

[0007] A column, the bottom end of which is detachably installed on the medical bed through a gripper;

[0008] A middle swing arm, the first end of which is vertically slidably connected to the side wall of the column;

[0009] A front swing arm, the first end of which is rotatably connected to the second end of the middle swing arm in a horizontal plane, and an angle sensor is installed at the rotational connection of the front swing arm and the middle swing arm;

[0010] An arm support, which is rotatably connected to the second end of the front swing arm and is driven to extend or retract by the second end of the front swing arm, and a pressure sensor and an infrared vision sensor are installed on the arm support;

[0011] A connecting limiter is installed between the first end of the front swing arm and the second end of the middle swing arm, and the rotation angle of the front swing arm is limited by the connecting limiter.

[0012] Preferably, the connecting limiter includes:

[0013] An arc-shaped chute is opened at the second end of the middle swing arm and is coaxially arranged with the second rotating shaft between the middle swing arm and the front swing arm;

[0014] Limiting rods, two groups of the limiting rods are limited in the arc-shaped chute and are respectively located on both sides of the first end of the front swing arm.

[0015] Preferably, the limiting rod includes:

[0016] A short rod is limited at the end of the second rotating shaft;

[0017] A long rod is slidably connected in the arc-shaped chute and is vertically and fixedly connected to the short rod.

[0018] Preferably, it further includes:

[0019] A screw handle passes through two groups of the short rods and is also threadedly connected to the axis center of the second rotating shaft.

[0020] Preferably, the armrest includes:

[0021] A first fixed support and a second fixed support, and the first fixed support and the second fixed support are synchronously and rotatably connected to the second end of the front swing arm.

[0022] Preferably, it further includes:

[0023] A first rotating shaft, the top end of the first rotating shaft is rotatably connected to the second end of the front swing arm, and the bottom end of the first rotating shaft is rotatably connected between the first fixed support and the second fixed support.

[0024] Preferably, it further includes:

[0025] A linear motor is installed between the first end of the middle swing arm and the column, and the middle swing arm is driven by the linear motor.

[0026] In the technical solution provided by the present utility model, the height of the middle swing arm on the column is adjusted according to the actual use situation of the patient. The arm is placed in the armrest at the end of the front swing arm. The activity angle information of the patient's arm can be obtained through the angle sensor to timely understand the activity range of the patient. The maximum activity range of the patient is restricted by connecting a limiter, and the activity range can also be adjusted according to the patient's recovery situation. The real-time pressure of the arm is detected by a pressure sensor, and the texture features and palm shape of the patient's palm are collected by an infrared vision sensor. Twenty-one key points of the human hand are detected through a gesture recognition network to restore the arm skeleton features. According to the captured key point information of the hand, the real-time state and pose of the hand are detected, and corresponding evaluation data are fed back to provide accurate data information for doctors to evaluate the hand. Overall, this application can combine sensors to obtain the training status of the patient, provide accurate feedback data for the therapist, and thus formulate a training plan that better meets the individual needs of the patient, effectively improving the training effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions in the embodiments of the present utility model or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present utility model. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0028] Figure 1 is a schematic diagram of the overall structure of the present utility model;

[0029] Figure 2 is a schematic diagram of the structure of the armrest of the present utility model;

[0030] Figure 3 is a schematic diagram of the structure of the connection limiter of the present utility model.

[0031] In the figure, 1 - clamp; 2 - main shaft; 3 - support rotator; 4 - column; 5 - middle swing arm; 6 - connection limiter; 7 - front swing arm; 8 - telescopic structure; 9 - first rotating shaft; 10 - armrest; 11 - first fixed support; 12 - pressure sensor; 13 - second fixed support; 14 - infrared vision sensor; 15 - arm fixing belt; 16 - screw handle; 17 - short rod; 18 - long rod; 19 - arc-shaped chute. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0032] To make the objectives, technical solutions and advantages of the present utility model clearer, the technical solutions of the present utility model will be described in detail below. Obviously, the described embodiments are only a part of the embodiments of the present utility model, rather than all the embodiments. Based on the embodiments of the present utility model, all other implementation manners obtained by those of ordinary skill in the art without creative efforts belong to the scope protected by the present utility model.

[0033] Referring to Figures 1-3 , a specific embodiment of the present utility model provides an upper limb rehabilitation assessment and training system combined with artificial intelligence gesture recognition, which is installed on a medical bed and includes:

[0034] A column 4, the bottom end of the column 4 is detachably installed on the medical bed through a clamp 1;

[0035] A middle-position swing arm 5, the first end of the middle-position swing arm 5 is vertically slidably connected to the side wall of the column 4;

[0036] A front-position swing arm 7, the first end of the front-position swing arm 7 is rotatably connected to the second end of the middle-position swing arm 5 in a horizontal plane, and an angle sensor is installed at the rotation connection of the front-position swing arm 7 and the middle-position swing arm 5;

[0037] An arm support 10, the arm support 10 is rotatably connected to the second end of the front-position swing arm 7 and is driven to extend or retract by the second end of the front-position swing arm 7. A pressure sensor 12 and an infrared vision sensor 14 are installed on the arm support 10;

[0038] A connection limiter 6, the connection limiter 6 is installed between the first end of the front-position swing arm 7 and the second end of the middle-position swing arm 5, and the rotation angle of the front-position swing arm 7 is limited by the connection limiter 6.

[0039] Traditional rehabilitation suspension systems often rely on the experience and judgment of therapists to set training parameters, making it difficult to achieve personalized and precise treatment, and difficult to adjust parameters in real time during training, resulting in limited training effects. In this application, the height of the middle swing arm 5 on the column 4 is adjusted according to the actual usage situation of the patient (lying or sitting). The arm is placed in the arm support 10 at the end of the front swing arm 7. The angular sensor can obtain the angular information of the patient's arm, timely understand the patient's range of motion, limit the patient's maximum range of motion by connecting the limiter 6, and can also adjust the range of motion according to the patient's recovery situation. The real-time pressure of the arm is detected by the pressure sensor 12, and the texture features and palm shape of the patient's palm are collected by the infrared vision sensor 14. 21 key points of the human hand are detected by the gesture recognition network to restore the arm skeleton features. According to the captured hand key point information, the real-time state and pose of the hand are detected, and the corresponding evaluation data are fed back, providing accurate data information for doctors to evaluate the hand. Overall, this application can combine sensors to obtain the patient's training status, provide accurate feedback data for therapists, and thus formulate a training plan that better meets the individual needs of patients, effectively improving the training effect.

[0040] For a further optimized solution, the connecting limiter 6 includes:

[0041] An arc-shaped chute 19 is opened at the second end of the middle swing arm 5 and is coaxially arranged with the second rotating shaft between the middle swing arm 5 and the front swing arm 7;

[0042] Limiting rods, two groups of limiting rods are limited in the arc-shaped chute 19 and are respectively located on both sides of the first end of the front swing arm 7.

[0043] When the two groups of limiting rods are respectively located at the two ends of the arc-shaped chute 19, the distance between the two groups of limiting rods along the direction of the arc-shaped chute 19 is the farthest, and the rotation angle range of the front swing arm 7 is the largest; to reduce the rotation angle, the distance between the two groups of limiting rods is reduced, that is, the rotation angle range is limited.

[0044] For a further optimized solution, the limiting rod includes:

[0045] A short rod 17 is limited at the end of the second rotating shaft;

[0046] A long rod 18 is slidably connected in the arc-shaped chute 19 and is perpendicularly and fixedly connected to the short rod 17.

[0047] For a further optimized solution, it further includes:

[0048] A screw handle 16 passes through the two groups of short rods 17 and is also threadedly connected to the axis center of the second rotating shaft.

[0049] When the screw handle 16 is in a relaxed state, the long rod 18 can slide freely by hand within the arc-shaped chute 19. By tightening the screw handle 16, the short rod 17 is limited on the second rotating shaft, and at the same time, the long rod 18 is limited within the arc-shaped chute 19. With this setting, the rotation angle range of the front swing arm 7 is adjusted.

[0050] For a further optimized solution, the armrest 10 includes:

[0051] The first fixed support 11 and the second fixed support 13, and the first fixed support 11 and the second fixed support 13 are synchronously rotatably connected to the second end of the front swing arm 7.

[0052] For a further optimized solution, it further includes:

[0053] The first rotating shaft 9, the top end of the first rotating shaft 9 is rotatably connected to the second end of the front swing arm 7, and the bottom end of the first rotating shaft 9 is rotatably connected between the first fixed support 11 and the second fixed support 13.

[0054] The first fixed support 11 and the second fixed support 13 are synchronously rotatably connected to the second end of the front swing arm 7 through the first rotating shaft 9; when the arm is placed on the armrest 10, it can meet the requirement of moving the arm within a certain range.

[0055] For a further optimized solution, it further includes:

[0056] A linear motor, which is installed between the first end of the middle swing arm 5 and the column 4, and the middle swing arm 5 is driven by the linear motor.

[0057] The main function of the linear motor is to drive the middle swing arm 5 to rise or fall, so that the height of the armrest 10 can be adapted to patients with different postures.

[0058] For a further optimized solution, it further includes:

[0059] A telescopic structure 8, the first end of the telescopic structure 8 is fixedly connected to the second end of the front swing arm 7, and the top end of the first rotating shaft 9 is rotatably connected to the second end of the telescopic structure 8.

[0060] The telescopic structure 8 can be a hydraulic cylinder, which is used to adjust the horizontal position of the armrest 10.

[0061] For a further optimized solution, it further includes:

[0062] The controller is electrically connected to the angle sensor, pressure sensor 12, infrared vision sensor 14, linear motor, and telescopic structure 8; it can also use the acquired hand state information as a signal to control the overall equipment action, such as lifting or telescopic action, to increase the interest of the training process. Among them, the infrared vision sensor 14 first identifies whether there is a palm in the image. If there is a palm, it immediately collects the characteristic information of the person's palm, sends the data back to the host computer for processing, and after processing, obtains 21 key points of the hand through the key point recognition network, thereby restoring the hand skeleton information, realizing the recognition of the hand's angle information, displacement information, hand posture, etc., and can collect accurate hand information during the day or at night to adapt to different light environments and improve the accuracy of recognition. At the same time, the hand rest of this design can realize the up and down movement and left and right rotation of the front end, which improves the comfort of the user.

[0063] Control process: The hand movement is acquired through the infrared vision sensor 14, and the movement is sent to the controller as a corresponding action instruction, and the linear motor or telescopic structure 8 is driven by the controller to control the adjustment of the height or horizontal position of the arm rest 10; the above control principle adopts the intelligent gesture interaction control principle in the publication number CN107807662A or CN104460991A, and the interactive arm rest 10 is controlled by gestures, that is, the position adjustment is completed through human-computer interaction.

[0064] Further optimization plans also include:

[0065] The arm fixing belts 15 are connected to the first fixing bracket 11 and the second fixing bracket 13 respectively.

[0066] The arm is fixed on the first fixing support 11 and the second fixing support 13 by the arm fixing belt 15, so as to ensure the safety and stability of the arm movement.

[0067] Further optimization plans also include:

[0068] The main shaft 2 has a bottom end fixedly connected to the clamp 1, and a top end of the main shaft 2 is rotatably connected to the column 4 via a support rotator 3; wherein the support rotator 3 may be a bearing.

[0069] The clamp 1 of the present application may be a common clamping structure used for hospital beds, and is not specifically limited here.

[0070] The infrared vision sensor 14 is installed on the first fixed support 11 and the second fixed support 13 through a rotating motor, and the horizontal angle can be adjusted by the rotating motor. An electric push rod is also installed between the infrared vision sensor 14 and the first fixed support 11 and the second fixed support 13. The infrared vision sensor 14 is driven to rise and fall by the electric push rod, which is convenient for identifying hand or facial information.

[0071] For a further optimized solution, pressure sensors are also installed on the front swing arm 7 and the telescopic structure 8. A rotary motor (not shown in the figure) is installed at the rotational connection between the front swing arm 7 and the middle swing arm 5. The front swing arm 7 is driven to rotate by the rotary motor, which can drive the patient's arm for rehabilitation training; the middle swing arm 5, the front swing arm 7 and the telescopic structure 8 form a robotic arm structure.

[0072] When the swing arm, the telescopic structure 8 collide with an external object or are affected by an external force, the force output of the robotic arm, i.e., the rotation angle of the front swing arm 7, can be adjusted according to the data fed back by the sensors, so that the robotic arm can better adapt to the external environment. According to the force information obtained by the pressure sensors and combined with the dynamic model of the robotic arm, precise control of the position of the end of the robotic arm can be achieved; by continuously adjusting the control input, the robotic arm can maintain a stable position when interacting with the external environment. Using the force information obtained by the pressure sensors, tracking control of an external object can be achieved; by comparing the difference between the target force and the actual force, the motion trajectory of the robotic arm is adjusted to achieve precise tracking of the external object.

[0073] It should be noted that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", etc. described in this article indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings. They are only for the convenience of describing this application 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 construed as a limitation to this application. In addition, the terms "first", "second", "third", etc. are only used for descriptive purposes and cannot be construed as indicating or implying relative importance.

[0074] In the description of this article, it should also be noted that unless otherwise clearly specified and defined, the terms "installed", "connected", "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. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0075] The above is only the specific implementation manner of the present utility model, but the protection scope of the present utility model is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present utility model can easily think of changes or substitutions, which should all be covered within the protection scope of the present utility model. Therefore, the protection scope of the present utility model should be subject to the protection scope of the claims.

Claims

1. An upper limb rehabilitation assessment and training system combining artificial intelligence gesture recognition, installed on a medical hospital bed, characterized in that, Comprising: A vertical column (4), the bottom end of the vertical column (4) is detachably installed on the medical bed through a gripper (1); A middle swing arm (5), the first end of the middle swing arm (5) is vertically slidably connected to the side wall of the vertical column (4); A front swing arm (7), the first end of the front swing arm (7) is rotatably connected to the second end of the middle swing arm (5) in a horizontal plane, and an angle sensor is installed at the rotational connection between the front swing arm (7) and the middle swing arm (5); An arm support (10), the arm support (10) is rotatably connected to the second end of the front swing arm (7), and is driven to extend or retract by the second end of the front swing arm (7), and a pressure sensor (12) and an infrared vision sensor (14) are installed on the arm support (10); A connection limiter (6), the connection limiter (6) is installed between the first end of the front swing arm (7) and the second end of the middle swing arm (5), and the rotational angle of the front swing arm (7) is limited by the connection limiter (6).

2. The upper limb rehabilitation evaluation and training system combining artificial intelligence gesture recognition according to claim 1, characterized in that The connection limiter (6) includes: An arc-shaped chute (19), the arc-shaped chute (19) is opened at the second end of the middle swing arm (5), and is coaxially arranged with the second rotation axis between the middle swing arm (5) and the front swing arm (7); Limiting rods, two groups of the limiting rods are limited in the arc-shaped chute (19), and are respectively located on both sides of the first end of the front swing arm (7).

3. The upper limb rehabilitation assessment and training system combining artificial intelligence gesture recognition according to claim 2, characterized in that, The limiting rod includes: A short rod (17), the short rod (17) is limited at the end of the second rotation axis; A long rod (18), the long rod (18) is slidably connected in the arc-shaped chute (19), and is perpendicularly and fixedly connected to the short rod (17).

4. The upper limb rehabilitation assessment and training system combining artificial intelligence gesture recognition according to claim 3, characterized in that, Also included is: A screw handle (16), the screw handle (16) passes through two groups of the short rods (17), and is also threadedly connected to the axis center of the second rotation axis.

5. The upper limb rehabilitation assessment and training system combined with artificial intelligence gesture recognition according to claim 1, characterized in that The arm support (10) includes: A first fixed support (11) and a second fixed support (13), the first fixed support (11) and the second fixed support (13) are synchronously rotatably connected to the second end of the front swing arm (7).

6. The upper limb rehabilitation assessment and training system combined with artificial intelligence gesture recognition according to claim 5, characterized in that Also included is: A first rotation axis (9), the top end of the first rotation axis (9) is rotatably connected to the second end of the front swing arm (7), and the bottom end of the first rotation axis (9) is rotatably connected between the first fixed support (11) and the second fixed support (13).

7. The upper limb rehabilitation assessment and training system combining artificial intelligence gesture recognition according to claim 1, characterized in that, Also included is: A linear motor, the linear motor is installed between the first end of the middle swing arm (5) and the vertical column (4), and the middle swing arm (5) is driven by the linear motor.

Images