TWI791234B - Wearable inertial sensing system for upper limbs - Google Patents

Abstract

The proposed invention is a wearable inertial sensing system for upper limbs, which includes one or several wearable motion detection device(s), a network gateway, a database and a display device. The wearable motion detection devices are worn on the upper limbs of a user. The devices generate the corresponding motion data according to the upper limbs activities of the user when the power of the device turns on. The network gateway is electrically connected to the wearable motion detection devices. The network gateway is an edge device with storage and computing capabilities to collect, identify and analyze upper limb motion data to obtain analysis results. The database is electrically connected to the network gateway to receive and store the analysis results. The display device is electrically connected to the database and receives the analysis results and displays the visualized results. Thereby, the proposed invention can effectively detect and analyze the upper limb movements of the user and can be applied for rehabilitation field.

Description

Upper Limb Wearable Inertial Sensing System

The invention relates to an upper limb wearable inertial sensing system, which can effectively identify and analyze the movement of a user's upper limb, and can be used for rehabilitation purposes.

Limb paralysis or limb disability is a common clinical symptom of chronic diseases today. For example, stroke patients may experience temporary disability of upper body limbs, and such patients will not be able to move their hands normally like healthy people; it is also commonly known as "fifty shoulders" disease , refers to the inflammation of the shoulder joint capsule tissue of the patient, resulting in shoulder pain and the inability to raise his hands, which in turn affects his normal life. The current treatment methods for such patients usually require the patient to undergo rehabilitation treatment under the guidance and companionship of a rehabilitation therapist. Through long-term reciprocating dynamic exercises, as well as the adjustment and practice of the patient's limb movements, the patient's muscles can be gradually slowed down. In the case of sticking, the range of motion and orientation of the limbs can be improved, so that the function of the limbs of the patient can return to normal.

However, during the treatment period, the patient needs to commute between the hospital and home for a long time, which is quite inconvenient. If the patient can perform rehabilitation exercises at home outside the treatment time, it will help the recovery of limb function; however, whether the patient's movements are correct or not , It will also affect the effectiveness of rehabilitation exercises. Therefore, improving the accuracy of patients' self-rehabilitation exercises is a very important research and development direction.

Today, in view of the fact that the relevant systems applicable to home rehabilitation actions are still relatively insufficient, the inventor is a tireless spirit, assisted by his rich professional knowledge and years of practical experience, and improved it, and based on this Research and create the present invention.

The present invention relates to an upper limb wearable inertial sensing system, comprising at least one wearable motion detection device, a network gateway, a database and a display device; the wearable motion detection device is worn on the user's upper limb , especially worn on the user's wrist, to collect the user's upper limb movement data, the collected data includes the mean, median, quartile, skewness and kurtosis of each direction component; The gateway is electrically connected to the wearable motion detection device, and includes a storage unit and a computing unit, wherein the storage unit receives and stores the user's upper limb movement data, and the computing unit recognizes and analyzes the upper limb movement data , and mark it as various actions, after extracting the sensing data features of various actions, convert them into feature vectors to form a data set, divide the data set, and divide the data set into training data set and test data set The data set, the training data set is used to establish the action classification model, and the test data set is used to evaluate the action classification model, and then select the best model according to the performance of the test data set in each model to obtain an analysis result; the database system It is electrically connected to the network gateway to receive and store the obtained analysis results; the display device is electrically connected to the database to receive the obtained analysis results and display the analysis results.

In one embodiment of the present invention, the wearable motion detection device includes a microprocessor, a wireless communication transmission chip, an inertial sensor, a power circuit and a power supply unit.

In an embodiment of the invention, the inertial sensor includes at least one of an accelerometer, a gyroscope or a magnetometer.

In one embodiment of the present invention, the inertial sensor measures the acceleration, angular velocity and magnetic field of the user's upper limb movement.

In one embodiment of the present invention, the inertial sensor measures the X-axis, Y-axis and Z-axis direction components of the acceleration and angular velocity of the user's upper limb movement, and measures the X-axis, Y-axis and Y-axis of the magnetic field of the user's upper limb movement. Axis and the direction component of the Z axis.

In one embodiment of the present invention, the wearable motion detection device is electrically connected to the network gateway through a wireless network.

In one embodiment of the present invention, the database is a cloud database.

In an embodiment of the present invention, the display device is a smart glasses or a projection display device.

In one embodiment of the present invention, at least one wearable motion detection device includes an insulating pad and a wristband.

In this way, the upper limb wearing inertial sensing system of the present invention can accurately detect the user's upper limb movement, analyze and compare it, and present the analysis result to the user in real time, so that the user can know in real time what he has done Accuracy of movements, and instant adjustments.

1: Wearable motion detection device

2: Network Gateway

21: storage unit

22: Operation unit

3: Database

4: Display device

Figure 1: A schematic diagram of the composition of the upper limb wearable inertial sensing system of the present invention.

The second picture: the acceleration signal diagram of the upper limb movement measured by the present invention.

The third picture: the angular velocity signal diagram of the upper limb movement measured by the present invention.

In order to enable the technical means of the present invention to achieve a more complete and clear disclosure, the following specific examples are used to describe the scope of practical application of the present invention in detail, but it is not intended to limit the scope of the present invention in any form. For the scope of the invention, please also refer to the disclosed drawings.

The invention relates to an upper limb wearable inertial sensing system, which is used for detecting and identifying and analyzing a user's upper limb.

Please refer to the first figure, the upper limb wearable inertial sensing system of the present invention mainly includes at least one wearable motion detection device (1), network gateway (2), database (3) and display device (4).

The wearable motion detection device (1) is worn on the user's upper limbs, preferably on the user's wrist, to collect the user's upper limb motion data; the wearable motion detection device (1) includes a microprocessor, A wireless communication transmission chip, an inertial sensor, a power circuit and a power supply unit, wherein the inertial sensor includes at least one of an accelerometer, a gyroscope, or a magnetometer. Wearable motion devices may include insulating pads or wristbands. The microprocessor is connected to the inertial sensor and receives the output sensing signal; the function of the microprocessor is mainly to sample the analog signal output by the inertial sensor, convert it into a digital signal form, and transmit it through wireless The communication is transmitted to the network gateway (2).

The network gateway (2) is electrically connected to the wearable motion detection device (1). In this embodiment, it is connected to the wearable motion detection device (1) through wireless network (Wi-Fi) communication. ; The network gateway (2) includes a storage unit (21) and a computing unit (22), the storage unit (21) is used to receive and store the upper limb movement data of the user, and the computing unit (22) is to identify And analyze the received upper limb movement data to obtain an analysis result.

A database (3) is electrically connected to the network gateway (2) to receive and store analysis results. In the embodiment of this case, the database is a cloud database.

The display device (4) is electrically connected to the database (3), and receives the analysis result, and displays the analysis result; the display device (4) can be a pair of smart glasses for the user to wear, so as to know the analysis result in real time ; The display device (4) can also be a projection display device, which is also used to present the analysis results, so as to let the user know the analysis results immediately.

In actual use, first wear the wearable motion detection device (1) on the user's upper limbs, preferably on the wrist, and make the user start to move. At this time, the wearable motion detection device (1) ) is used to measure the X-axis, Y-axis and Z-axis direction components of the acceleration of the user's upper limbs, and the X-axis, Y-axis and Z-axis direction components of the angular velocity of the user's upper limbs. When the user performs different types of actions, the measured X-axis, Y-axis, and Z-axis direction components are different. Therefore, feature extraction can be performed according to the appearance of the collected data, and through the simplified The features are classified to obtain the actions performed by the user; the collected data includes the mean, median, quartile, skewness, and kurtosis of each direction component.

In the following embodiment, the user first makes the arm hang down naturally, and then performs the single pendulum movement of clockwise rotation, and uses the inertial sensor to detect the acceleration and angular velocity of the user's arm movement; the second picture shows the detected The acceleration signal diagram of the user's arm movement, the third diagram is the angular velocity signal diagram of the detected user's arm movement, where X represents the measured acceleration under the X-axis direction component, Y represents the acceleration under the Y-axis direction component, and Z represents the acceleration in the Z-axis direction component.

The second picture (A) is the original signal waveform diagram of the acceleration of the user's arm motion detected by the inertial sensor. The wearable motion detection device (1) will transmit this data to the storage of the network gateway (2) unit (21), and then transmitted to the computing unit (22) of the network gateway (2), and the computing unit will analyze and identify the received data, the second figure (B) is that the computing unit (22) will The second picture (A) is the original signal diagram and the denoised signal diagram obtained after denoising.

The third picture (A) is the waveform diagram of the raw signal of the angular velocity of the user’s arm motion detected by the inertial sensor. The wearable motion detection device (1) will transmit this data to the storage of the network gateway (2) unit (21), and then transmitted to the computing unit (22) of the network gateway (2), and the computing unit will analyze and identify the access The received data, the third picture (B) is the denoised signal picture obtained by the computing unit after removing the noise from the original signal picture in the third picture (A).

The computing unit (22) will further compare the signal obtained after denoising with a data set in the storage unit (21), compare the user's action with the signal of the standard action in the data set, and obtain the used The angle accuracy, signal similarity and movement stability of the operator's movements.

The establishment of the above data set is to first collect the sensing data of various types of actions to be identified, and mark them as various actions; after extracting the features of the sensing data of various actions, and converting them into feature vectors, to form a data set set. Next, divide the data set and divide the data set into training data set and test data set. The training data set is used to establish the action classification model, while the test data set is used to evaluate the action classification model. Finally, according to the test The data set is performed on each model to select the best model. Also, according to the type of predicted motion, it can be used for analysis of corresponding motions, such as angle analysis on specific motions, calculation of movement times, calculation of displacement, etc.

The action data obtained after the user performs an action, such as the above-mentioned signal diagram, is processed by the computing unit (22), and then compared with the data set to compare the correctness of the (a) angle of the user action data, that is The difference between the user's motion and the standard motion at an angle of 90 degrees, (b) signal similarity, which is the difference between the signal generated by comparing the user's motion and the standard signal collected in the dataset, and (c) motion stability Sexuality, that is to calculate the size of the angle standard deviation between the movement data obtained by the user in several repetitions, so as to judge the stability of the movement. After the user's action data is processed and compared by the computing unit (22), an analysis result will be obtained, and the analysis result will be sent to the database (3) of the present invention for storage. In a preferred embodiment of this case Among them, the database (3) is a cloud database.

Then, the database (3) transmits the analysis results and presents them on the display device (4), so that users can watch the analysis results of their actions in real time, which is very convenient to use. In addition, according to the analysis results, Make action corrections.

To sum up, the upper limb wearing inertial sensing system of the present invention can indeed achieve the expected use effect through the above disclosed embodiments, and the present invention has not been disclosed before the application, and it has fully complied with the provisions of the patent law with requirements. ￠It is really convenient to file an application for a patent for invention according to the law, and ask for the review and approval of the patent.

However, the above-mentioned descriptions are only preferred embodiments of the present invention, and are not intended to limit the scope of protection of the present invention; those who are familiar with the art generally do other things based on the characteristics and scope of the present invention. Any effect change or modification should be regarded as not departing from the design scope of the present invention.

1: Wearable Motion Detection Device 2: Network Gateway 21: Storage unit 22: Computing unit 3:Database 4:Display device

Claims (9)

An upper limb wearable inertial sensing system includes: at least one wearable motion detection device, which is worn on a user's upper limb to collect movement data of the user's upper limb, and the collected data includes the average value of components in each direction , median, quartile, skewness and kurtosis; a network gateway, which is electrically connected to the wearable motion detection device and includes a storage unit and a computing unit, wherein the storage unit is Receive and store the upper limb movement data of the user, and the computing unit identifies and analyzes the upper limb movement data, and marks them as various movements, extracts the sensing data features of various movements, and converts them into feature vectors , to form a data set, divide the data set, and divide the data set into a training data set and a test data set, the training data set is used to establish an action classification model, and the test data set is used to evaluate the action classification model, Then select the best model according to the performance of the test data set in each model to obtain an analysis result; and a database electrically connected to the network gateway to receive and store the analysis result; and a display device , is electrically connected to the database and receives the analysis result, and displays the analysis result. The upper limb wearable inertial sensing system as described in Claim 1, wherein the wearable motion detection device includes a microprocessor, a wireless communication transmission chip, an inertial sensor, a power circuit and a power supply unit. According to claim 2, the upper limb wearable inertial sensing system, wherein the inertial sensing device includes at least one of an accelerometer, a gyroscope, or a magnetometer. The upper limb wearable inertial sensing system according to claim 2, wherein the inertial sensor measures the acceleration, angular velocity and magnetic field of the user's upper limb movement. The upper limb wearable inertial sensing system as described in Claim 4, wherein the inertial sensor measures the acceleration and angular velocity components of the X-axis, Y-axis and Z-axis direction of the user's upper limb movement, and measures the user's upper limb movement The direction components of the X-axis, Y-axis and Z-axis of the magnetic field. The upper limb wearable inertial sensing system as described in Claim 1, wherein the at least one wearable motion detection device is electrically connected to the network gateway through a wireless network. The upper limb wearable inertial sensing system as described in Claim 1, wherein the database is a cloud database. According to claim 1, the upper limb wearable inertial sensing system, wherein the display device is a smart glasses or a projection display device. The upper limb wearable inertial sensing system according to claim 1, wherein the at least one wearable motion detection device comprises an insulating spacer and a wristband.

Images