Home rehabilitation system

Abstract

according to the difficulty index; obtaining a human posture image in a time period; analyzing the human posture image; identifying the human posture image to obtain a plurality of current node coordinates; comparing the plurality of current node coordinates with a plurality of standard node coordinates of the standard rehabilitation model to calculate an error value; and defining a current rehabilitation score in the difficulty index according to the error value.

Description

Technical Field

[0001] This invention relates to a rehabilitation system, and more particularly to a home rehabilitation system and method. Prior Technology

[0002] Commercially available rehabilitation devices typically require patients to visit medical institutions in person and undergo rehabilitation under the guidance of professional rehabilitation therapists. This not only increases time and financial costs but also causes inconvenience. Although existing smart rehabilitation devices can record patients' rehabilitation movements, their hardware is expensive and their operation is complex, making them unfriendly to elderly patients or first-time users. Furthermore, these devices lack real-time correction capabilities, failing to provide timely and effective advice during patient operation. Moreover, many rehabilitation devices rely on traditional cameras to monitor and identify patient movements, but their analysis of human skeletal posture is not precise enough to provide high-precision rehabilitation guidance.

[0003] Therefore, the inventor believed that the above-mentioned defects could be improved. So he devoted himself to research and applied scientific principles, and finally proposed an invention that is reasonably designed and effectively improves the above-mentioned defects. Summary of the Invention

[0004] The present invention provides a home rehabilitation system and rehabilitation method that can effectively improve the defects that may occur in existing rehabilitation devices.

[0005] This invention discloses a home rehabilitation system, comprising: a posture photography device for acquiring a human posture image over a time period; and a rehabilitation module electrically connected to the posture photography device, the rehabilitation module being able to receive the human posture image from the posture photography device, and the rehabilitation module comprising: a model database storing multiple standard rehabilitation models, each standard rehabilitation model including multiple standard node coordinates; a processing unit and a recognition unit; the processing unit being electrically connected to the model database and the recognition unit, and the processing unit being able to analyze the human posture image for the recognition unit to identify multiple current node coordinates; wherein, when the... When the identification unit identifies multiple current node coordinates and sends them back to the processing unit, the processing unit can compare the multiple current node coordinates with the multiple standard node coordinates of the standard rehabilitation model to calculate an error value; and an evaluation unit, electrically connected to the processing unit, establishes a difficulty index, and the difficulty index defines a difficulty score from 0 to 100; wherein, the evaluation unit can define the difficulty score of the standard rehabilitation model as 100 points, and the evaluation unit defines a current rehabilitation score in the difficulty index based on the error value of the multiple current node coordinates relative to the multiple standard node coordinates.

[0006] This invention also discloses a home rehabilitation method, comprising: having a difficulty index, defining a difficulty score from 0 to 100; defining the difficulty score of a standard rehabilitation model as 100 based on the difficulty index; acquiring a human posture image over a time period; analyzing the human posture image; identifying the human posture image to obtain multiple current node coordinates; comparing the multiple current node coordinates with multiple standard node coordinates of the standard rehabilitation model to calculate an error value; and defining a current rehabilitation score in the difficulty index based on the error value.

[0007] In summary, the home rehabilitation system and method disclosed in the embodiments of the present invention, through the technical solution of "a processing unit and a recognition unit; the processing unit is electrically connected to the model database and the recognition unit, and the processing unit can analyze the human posture image for the recognition unit to identify multiple current node coordinates", can provide a user with the ability to maintain the accuracy of rehabilitation movements while performing rehabilitation in different locations.

[0008] To further understand the features and technical content of this invention, please refer to the following detailed description and accompanying drawings. However, these descriptions and drawings are only for illustrating the invention and are not intended to limit the scope of protection of the invention in any way. Simple Explanation of the Diagram

[0009] Figure 1 is a block diagram of a home rehabilitation system according to the present invention.

[0010] Figure 2 is a schematic diagram of the use of the home rehabilitation system of the present invention.

[0011] Figure 3 is another schematic diagram of the use of the home rehabilitation system of the present invention.

[0012] Figure 4 is a flowchart illustrating a home rehabilitation method according to the present invention.

[0013] Figure 5 is a flowchart illustrating the process of issuing adjustment suggestions for the home rehabilitation method described in this paper. Implementation

[0014] The following specific embodiments illustrate the implementation of the "home rehabilitation system and rehabilitation method" disclosed in this invention. Those skilled in the art can understand the advantages and effects of this invention from the content disclosed in this specification. This invention can be implemented or applied through other different specific embodiments, and various details in this specification can also be modified and changed based on different viewpoints and applications without departing from the concept of this invention. Furthermore, the accompanying drawings of this invention are for simple illustrative purposes only and are not depictions of actual dimensions; this is stated in advance. The following embodiments will further describe the relevant technical content of this invention in detail, but the disclosed content is not intended to limit the scope of protection of this invention.

[0015] It should be understood that while terms such as "first," "second," and "third" may be used in this document to describe various components or signals, these components or signals should not be limited by these terms. These terms are primarily used to distinguish one component from another, or one signal from another. Furthermore, the term "or" as used herein should, as appropriate, include any combination of one or more of the associated listed items.

[0016] [Example 1]

[0017] Please refer to Figures 1 to 3, which illustrate a home rehabilitation system 100 and a rehabilitation method disclosed in this invention. The home rehabilitation method can be applied to the home rehabilitation system 100 (but this invention is not limited thereto). Therefore, the following first describes the components of the home rehabilitation system 100 and their connection relationships, followed by the implementation steps of the home rehabilitation method.

[0018] The home rehabilitation system 100 enables a user P to perform home rehabilitation and provides the user P with adjustment suggestions as needed. In other words, the user P can perform rehabilitation at home or in any place where rehabilitation can be performed with the correct posture, without the need for additional caregivers (such as doctors or rehabilitation therapists) to accompany or guide them.

[0019] The home rehabilitation system 100 includes a posture photography device 1 and a rehabilitation module 2 electrically connected to the posture photography device 1. The posture photography device 1 is used to acquire a human posture image F over a period of time. It should be noted that, as shown in Figures 2 and 3, the posture photography device 1 in this embodiment is exemplified using a desktop computer's webcam, but the invention is not limited to this; that is, in other embodiments not illustrated in the figures, the posture photography device 1 can also be a camera, mobile phone, or other device capable of acquiring the human posture image F.

[0020] Please refer to Figure 1. The rehabilitation module 2 can receive the human posture image F from the posture photography device 1. The rehabilitation module 2 includes a model database 21, a processing unit 22 electrically connected to the model database 21, a recognition unit 23 electrically connected to the processing unit 22, an evaluation unit 24 electrically connected to the processing unit 22, a prompting unit 25 electrically connected to the processing unit 22, and a data storage unit 26 electrically connected to the processing unit 22.

[0021] In detail, the model database 21 stores multiple standard rehabilitation models, each containing multiple standard node coordinates. Specifically, the multiple standard rehabilitation models are different rehabilitation postures and are composed of multiple different standard node coordinates, so that the user P can refer to or imitate the corresponding rehabilitation movements.

[0022] The processing unit 22 can analyze the human posture image F so that the recognition unit 23 can identify multiple current node coordinates. In this embodiment, the processing unit 22 uses the image processing program of Open Source Computer Vision Library (hereinafter referred to as OpenCV) to analyze and process the human posture image F; the recognition unit 23 uses the image recognition program of Media Pipe to identify various parts of the body in the human posture image F, but the present invention is not limited thereto.

[0023] Specifically, OpenCV is a cross-platform computer vision library that can preprocess the human pose image F (e.g., image filtering, color conversion, geometric transformation, etc.), and output the preprocessed human pose image F for subsequent operations, thereby improving the quality of the human pose image F and making subsequent recognition and operations more accurate.

[0024] Next, Media Pipe is an open-source, cross-platform framework that provides efficient and accurate image processing technology. For example, Media Pipe uses 33 key joint nodes extracted from a human skeleton model to identify the human pose image F. Each node represents a key joint (e.g., head, shoulder, elbow, knee, ankle, etc.). Furthermore, Media Pipe can receive the human pose image F output from OpenCV and accurately identify various parts of the human body (e.g., head, shoulder, elbow, knee, ankle, etc.), and calculate the current node coordinates of each joint using the 33 key joint nodes, thereby achieving real-time and accurate pose analysis of the human pose image F.

[0025] In other words, the processing unit 22 can first preprocess the human posture image F and transmit the processed image to the recognition unit 23. Then, the recognition unit 23 identifies each human body part in the processed image (e.g., joints such as head, shoulders, elbows, knees, and ankles; that is, each joint), calculates the current node coordinates of each human body part, and sends them back to the processing unit 22. The processing unit 22 then directly marks the multiple current node coordinates on the human posture image F (as shown in Figures 2 and 3).

[0026] It is worth mentioning that the processing unit 22 can also calculate the position of each standard node coordinate based on the Euclidean distance, calculate an adjustment value based on the ratio between the standard node coordinate and the current node coordinate, and input the adjustment value into multiple standard node coordinates to generate multiple adjusted node coordinates.

[0027] For example, the processing unit 22 can compare the current node coordinates and height ratio of the user P with the standard node coordinates and height ratios of multiple standard rehabilitation models, calculate the adjustment value using Euclidean distance, and adjust the relative positions between the multiple standard nodes of the multiple standard rehabilitation models using the adjustment value, thereby adjusting the multiple standard node coordinates to multiple adjusted node coordinates suitable for the user P to perform (that is, individualized standard rehabilitation movements).

[0028] Furthermore, in this embodiment, the multiple adjustment node coordinates can also be generated through other calculation methods. For example, the processing unit 22 can also compare the current node coordinates and height ratio of the user P with the standard node coordinates and height ratios of multiple standard rehabilitation models, and calculate the angular differences between the multiple standard nodes based on vector and trigonometric function relationships, so as to adjust the standard node coordinates of the standard rehabilitation model to conform to the height ratio of the user P, and generate multiple adjustment node coordinates.

[0029] It is worth mentioning that when the identification unit 23 identifies multiple current node coordinates and sends them back to the processing unit 22, the processing unit 22 can compare the multiple current node coordinates with the multiple standard node coordinates (or multiple adjustment node coordinates) of the standard rehabilitation model to calculate an error value.

[0030] Furthermore, the evaluation unit 24 establishes a difficulty index, and the difficulty index defines a difficulty score from 0 to 100. Specifically, the evaluation unit 24 can define the difficulty score of the standard rehabilitation model as 100 points, and the evaluation unit 24 defines a current rehabilitation score in the difficulty index based on the error values ​​of multiple current node coordinates relative to multiple standard node coordinates.

[0031] Next, the processing unit 22 determines whether the coordinates of the current nodes match the coordinates of the standard nodes (or the coordinates of the adjustment nodes) based on the current rehabilitation score; if the current rehabilitation score is less than 90 points, it means that the coordinates of the current nodes do not match the coordinates of the standard nodes, and the prompting unit 25 issues an adjustment suggestion.

[0032] In other words, when the user P performs rehabilitation, the processing unit 22 and the recognition unit 23 analyze the human posture image F to obtain multiple current node coordinates; then, the processing unit 22 compares the multiple current node coordinates with the multiple standard node coordinates (or multiple adjusted node coordinates) in the model database and calculates the error value, and the evaluation unit 24 defines the current rehabilitation score.

[0033] At this time, when the current rehabilitation score is greater than 90 points, it indicates that the coordinate positions of multiple current node coordinates of user P are extremely close to the coordinate positions of multiple standard node coordinates (or multiple adjustment node coordinates), and the prompting unit 25 will issue a rehabilitation success notification to user P. Conversely, when the current rehabilitation score is less than 90 points, it indicates that the coordinate positions of multiple current node coordinates of user P are significantly different from the coordinate positions of multiple standard node coordinates (or multiple adjustment node coordinates), and the processing unit 22 will calculate the adjustment suggestion and issue the adjustment suggestion to user P through the prompting unit 25.

[0034] In this embodiment, the adjustment suggestion may be given to the user P through animation (e.g., moving the human posture image F of the user P from multiple current node coordinates to multiple standard node coordinates) or through text description.

[0035] Furthermore, in this embodiment, the data storage unit 26 can be used to store the current node coordinates and multiple current rehabilitation scores. Specifically, the data storage unit 26 can store and integrate the current rehabilitation record of the user P, enabling the user P to continuously track their rehabilitation progress.

[0036] The above describes the components and their connections in the home rehabilitation system 100. The home rehabilitation method is described below. This home rehabilitation method applies the aforementioned home rehabilitation system 100, but the invention is not limited thereto. Referring to Figure 4, the home rehabilitation method includes steps S101 to S107, and any one of these steps can be omitted or replaced with a reasonable variation depending on the designer's needs.

[0037] Step S101 includes: having a difficulty index, which defines a difficulty score from 0 to 100.

[0038] Step S102 includes: defining the difficulty score of a standard rehabilitation model as 100 points based on the difficulty index.

[0039] Step S103 includes: acquiring a human posture image within a time period.

[0040] Step S104 includes: analyzing the human posture image.

[0041] Step S105 includes: identifying the human posture image to obtain multiple current node coordinates.

[0042] Step S106 includes: comparing multiple current node coordinates with multiple standard node coordinates of the standard rehabilitation model to calculate an error value.

[0043] Step S107 includes: defining a current rehabilitation score in the difficulty index based on the error value.

[0044] Please refer to Figure 5. In this embodiment, the home rehabilitation method can also issue an adjustment suggestion based on the difficulty score. Therefore, the home rehabilitation method also includes steps S201 to S203, and any one of the above steps can be omitted or replaced with a reasonable variation depending on the designer's needs.

[0045] Step S201 includes: determining whether the current rehabilitation score is greater than 90 points.

[0046] Specifically, based on the current rehabilitation score, it is determined whether the coordinates of multiple current nodes match the coordinates of multiple standard nodes. Specifically, if the current rehabilitation score is greater than 90, it is determined to be yes; if the current rehabilitation score is less than 90, it is determined to be no.

[0047] In other words, if the current rehabilitation score is greater than 90 points, it is determined to match the standard node coordinates, and step S202 is continued; if the current rehabilitation score is less than 90 points, it is determined to not match the standard node coordinates, and step S203 is continued.

[0048] Step S202 includes: if the determination is yes, issuing a rehabilitation success notification.

[0049] Step S203 includes: if the determination is no, calculating an adjustment suggestion and issuing the adjustment suggestion.

[0050] [Technical Effects of the Embodiments of the Invention]

[0051] In summary, the home rehabilitation system and method disclosed in the embodiments of the present invention, through the technical solution of "a processing unit and a recognition unit; the processing unit is electrically connected to the model database and the recognition unit, and the processing unit can analyze the human posture image so that the recognition unit can identify multiple current node coordinates", can provide the user with the ability to maintain the accuracy of rehabilitation movements while performing rehabilitation in different locations.

[0052] The content disclosed above is only a preferred and feasible embodiment of the present invention, and is not intended to limit the patent scope of the present invention. Therefore, all equivalent technical changes made using the contents of the present invention specification and drawings are included within the patent scope of the present invention.

[0053] 100: Home Rehabilitation System 1: Pose Photography Device 2: Rehabilitation Module 21: Model Database 22: Processing Unit 23: Identification Unit 24: Evaluation Unit 25: Prompt Unit 26: Data storage unit P: User F: Human posture image

Claims

1. A home rehabilitation system comprising: A pose photography device for acquiring an image of a human pose over a period of time; The system also includes: a rehabilitation module electrically connected to the posture imaging device, capable of receiving human posture images from the posture imaging device; the rehabilitation module comprising: a model database containing multiple standard rehabilitation models, each standard rehabilitation model containing multiple standard node coordinates; a processing unit and a recognition unit; the processing unit electrically connected to the model database and the recognition unit, and capable of analyzing the human posture images for the recognition unit to identify multiple current node coordinates; wherein, when the recognition unit identifies multiple current node coordinates and transmits them back to the processing unit, the processing unit can compare the multiple current node coordinates with the multiple standard node coordinates of the standard rehabilitation model to calculate an error value; and an evaluation unit electrically connected to the processing unit, the evaluation unit establishing a difficulty index, and the difficulty index defining a difficulty score from 0 to 100. The evaluation unit can define the difficulty score of the standard rehabilitation model as 100 points, and the evaluation unit defines a current rehabilitation score in the difficulty index based on the error values ​​of the current node coordinates relative to the standard node coordinates.

2. The home rehabilitation system as described in claim 1, wherein, The rehabilitation module also includes a prompting unit electrically connected to the processing unit. The processing unit can determine whether the coordinates of the current nodes match the coordinates of the standard nodes based on the current rehabilitation score. If the current rehabilitation score is less than 90 points, it means that the coordinates of the current nodes do not match the coordinates of the standard nodes, and the prompting unit issues an adjustment suggestion.

3. The home rehabilitation system as described in claim 1, wherein, The processing unit can calculate the position of each standard node coordinate based on the Euclidean distance, calculate an adjustment value based on the proportional relationship between multiple standard node coordinates and multiple current node coordinates, and input the adjustment value into the standard node coordinates to generate an adjusted node coordinate.

4. The home rehabilitation system as described in claim 1, wherein, The processing unit can calculate the angular differences between multiple standard nodes based on vector and trigonometric function relationships to generate an adjusted node coordinate.

5. The home rehabilitation system as described in claim 1, wherein, The rehabilitation module also includes a data storage unit electrically connected to the processing unit, the data storage unit being able to store the current node coordinates and multiple current rehabilitation scores.

6. The home rehabilitation system as described in claim 1, wherein, The processing unit is an image processing program using OpenCV; the recognition unit is an image recognition program using MediaPipe.

Images