Exercise rehabilitation data enhancement method and sensing equipment

Abstract

The invention provides an exercise rehabilitation data enhancement method, the method is applied to sensing equipment, the sensing equipment comprises an inertia measurement unit and a control unit, and the control unit is used for controlling the inertia measurement unit. Obtaining a random offset at a current moment, a relative motion offset between the inertial measurement unit and a human body part attached to the inertial measurement unit at a previous moment, and motion data at the current moment sensed by the inertial measurement unit; obtaining the relative motion offset at the current moment according to the relative motion offset at the previous moment and the random offset at the current moment; the motion data at the current moment are corrected according to the relative motion offset at the current moment to obtain enhanced motion rehabilitation data, the enhanced motion rehabilitation data are used for training a deep learning model, and a human body posture determination model is obtained. The exercise rehabilitation data accuracy, the robustness and generalization ability of the human body posture determination model and the determination accuracy of the human body posture can be improved, so that the application progress of the IMU in medical rehabilitation is promoted.

Description

Technical Field

[0001] This application relates to the technical field of motion rehabilitation data enhancement, and in particular, to a method for enhancing motion rehabilitation data and a sensing device. Background Art

[0002] Inertial Measurement Units (IMUs) have shown great potential in the field of medical rehabilitation, especially in the motion monitoring of patients and the assessment of rehabilitation progress. By combining deep learning techniques, IMUs can track the limb movements of patients and provide real-time motion rehabilitation data to help doctors adjust treatment plans more accurately. During human movement, due to factors such as soft tissue deformation, gravity, and inertial effects, the IMU will rotate and there will be a relative displacement between the IMU and the human body part to which it is attached, and it cannot always maintain a strict spatial position consistency with the attached human body part. Therefore, due to the occurrence of relative displacement and rotation, when a person repeats the same action, different measurement results will be generated, resulting in an inability to accurately determine the human body posture, which limits the popularization of the combination of deep learning techniques and IMUs in actual clinical practice. Summary of the Invention

[0003] This application provides a method for enhancing motion rehabilitation data and a sensing device, which can improve the accuracy of motion rehabilitation data, the robustness, generalization ability of the human body posture determination model, and the determination accuracy of the human body posture, so as to promote the application progress of IMUs in medical rehabilitation.

[0004] In the first aspect of this application, a method for enhancing motion rehabilitation data is provided. The method is applied to a sensing device, which includes an inertial measurement unit and a control unit. The control unit is configured to obtain a random offset at the current moment, a relative motion offset at the previous moment between the inertial measurement unit and the human body part to which the inertial measurement unit is attached, and motion data sensed by the inertial measurement unit at the current moment; obtain the relative motion offset at the current moment according to the relative motion offset at the previous moment and the random offset at the current moment; correct the motion data at the current moment according to the relative motion offset at the current moment to obtain enhanced motion rehabilitation data, and the enhanced motion rehabilitation data is used to train a deep learning model to obtain a human body posture determination model.

[0005] In some embodiments of the first aspect, the enhanced motion rehabilitation data includes enhanced angular velocity samples and enhanced acceleration samples.

[0006] In some embodiments of the first aspect, the random offset includes the random angular velocity offset generated by the inertial measurement unit due to rotation. Obtaining the random offset at the current moment includes: sampling from three independent standard Gaussian distributions respectively to obtain three random angular velocity offset components of the inertial measurement unit in the X-axis direction, Y-axis direction, and Z-axis direction at the current moment; multiplying the three random angular velocity offset components to obtain the random angular velocity offset at the current moment.

[0007] In some embodiments of the first aspect, the relative motion offset includes a relative angular velocity offset. Obtaining the relative motion offset at the current moment based on the relative motion offset at the previous moment and the random offset at the current moment includes: multiplying the relative angular velocity offset at the previous moment and the random angular velocity offset at the current moment to obtain the relative angular velocity offset at the current moment.

[0008] In some embodiments of the first aspect, the motion data at the current moment includes the angular velocity at the current moment. Correcting the motion data at the current moment based on the relative motion offset at the current moment to obtain the enhanced motion rehabilitation data includes: multiplying the inverse matrix of the relative angular velocity offset at the current moment, the matrix of the angular velocity at the current moment, and the transformation matrix from the preset global inertial coordinate system to the sensor coordinate system to obtain the enhanced angular velocity sample.

[0009] In some embodiments of the first aspect, the random offset further includes the random acceleration offset generated by the inertial measurement unit due to relative displacement. Obtaining the random offset at the current moment includes: sampling from three independent standard Gaussian distributions respectively to obtain two sets of random displacement offset components of the inertial measurement unit at the previous moment and the current moment, and both sets of random displacement offset components include three random displacement offset components in the X-axis direction, Y-axis direction, and Z-axis direction; obtaining the velocity offset at the current moment based on the two sets of random displacement offset components at the previous moment and the current moment; subtracting the calculated velocity offset at the previous moment from the velocity offset at the current moment to obtain the random acceleration offset at the current moment.

[0010] In some embodiments of the first aspect, the relative motion offset includes a relative acceleration offset. Obtaining the relative motion offset at the current moment based on the relative motion offset at the previous moment and the random offset at the current moment includes: adding the relative acceleration offset at the previous moment and the random acceleration offset at the current moment to obtain the relative acceleration offset at the current moment.

[0011] In some embodiments of the first aspect, the motion data at the current moment includes the acceleration at the current moment. The motion rehabilitation data after enhancement is obtained by correcting the motion data at the current moment according to the relative motion offset at the current moment, including: multiplying the sum of the random acceleration offset at the current moment and the acceleration at the current moment by the inverse transformation matrix from the preset global inertial coordinate system to the sensor coordinate system and the inverse matrix of the enhanced angular velocity sample that has been calculated to obtain the enhanced acceleration sample.

[0012] In some embodiments of the first aspect, the random offset at the current moment is adjusted according to the empirical value range generated by the inertial measurement unit due to self-rotation and relative displacement.

[0013] The second aspect of the present application provides a sensing device, including: a processor and a memory; the memory is coupled to the processor, and the memory is used to store computer program code, and the processor calls the computer program code to enable the sensing device to execute the method as in the first aspect.

[0014] It can be understood that the present application provides a method for enhancing motion rehabilitation data and a sensing device. By modeling the relative motion offset between the inertial measurement unit and the human body part to which the inertial measurement unit is attached as a variable that can change with time (that is, the relative motion offset is obtained according to the relative motion offset at the previous moment and the random offset at the current moment), the relative motion offset that changes with time is more in line with the actual human motion process. Then, the accuracy of the enhanced motion rehabilitation data obtained by correcting the motion data at the current moment according to the relative motion offset at the current moment is improved, and the accuracy index of posture feature extraction and human body part positioning is enhanced. The robustness, generalization ability, and the determination accuracy of the human body posture of the human body posture determination model trained by the enhanced motion rehabilitation data are improved, so as to promote the application progress of IMU in medical rehabilitation. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The accompanying drawings herein are incorporated into the specification and constitute a part of this specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.

[0016] Figure 1 It is a schematic flowchart of a method for enhancing motion rehabilitation data provided by an embodiment of the present application; Figure 2 It is a schematic diagram of an application scenario of a method for enhancing motion rehabilitation data provided by an embodiment of the present application; Figure 3 It is a schematic structural diagram of a sensing device provided by an embodiment of the present application.

[0017] Through the above-mentioned drawings, specific embodiments of the present application have been shown, and will be described in more detail hereinafter. These drawings and the written description are not intended to limit the scope of the concept of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by reference to specific embodiments. Detailed Description of the Embodiments

[0018] Here, the exemplary embodiments will be described in detail, and the examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0019] The terms "first", "second", etc. involved in the present application are only for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features.

[0020] Hereinafter, the technical solution of the present application and how the technical solution of the present application solves the technical problems will be described in detail with specific embodiments. The following several specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

[0021] Please refer to Figure 1 and Figure 2 , Figure 2 which is a schematic flowchart of a motion rehabilitation data enhancement method provided by the present application. The execution subject of the motion rehabilitation data enhancement method can be a sensing device, specifically the control unit of the sensing device. As Figure 2 shown, the motion rehabilitation data enhancement method may include the following steps: Step S110: Obtain the random offset at the current moment, the relative motion offset at the previous moment between the inertial measurement unit and the human body part to which the inertial measurement unit is attached, and the motion data sensed by the inertial measurement unit at the current moment.

[0022] Specifically, as Figure 1As shown, the sensing device includes an inertial measurement unit and a control unit. The inertial measurement unit and the control unit can be integrated or physically separated. The inertial measurement unit can sense motion data at different times (including the current time and the previous time), and the motion data includes acceleration and angular velocity. The random offset can be randomly generated. In one implementation, the random offset can be adjusted according to the empirical value range generated by the inertial measurement unit due to self-rotation and relative displacement. The relative motion offset of the previous time is obtained from the storage unit (not shown in the figure) of the sensing device. It can be understood that the relative motion offset is the offset between the motion data sensed by the IMU and the actual motion data of the human body part to which it is attached.

[0023] Step S120: Obtain the relative motion offset at the current time according to the relative motion offset at the previous time and the random offset at the current time.

[0024] It can be understood that during the human body movement, the human body will generate actual motion data. Due to factors such as soft tissue deformation, gravity, and inertial effects, the IMU will generate self-rotation or relative displacement, resulting in a deviation between the motion data sensed by the IMU and the actual motion data of the human body. This deviation can be solved by the random offset.

[0025] Exemplarily, the random offset includes the random angular velocity offset generated by the inertial measurement unit due to self-rotation. The relative motion offset includes the relative angular velocity offset. The determination formula of the relative angular velocity offset is as follows: (1) Where, represents the relative angular velocity offset, represents the relative angular velocity offset at the previous time (it can be understood that the calculation method of the relative angular velocity offset at the previous time refers to formula (1)), represents the random angular velocity offset at the current time. Formula (1) means multiplying the relative angular velocity offset at the previous time and the random angular velocity offset at the current time to obtain the relative angular velocity offset at the current time.

[0026] It can be understood that there is a deviation between the angular velocity sensed by the IMU and the actual angular velocity of the human body part. The random angular velocity offset can simulate this deviation to solve the calculation error problem of the angular velocity.

[0027] Exemplarily, the random offset includes the random acceleration offset generated by the inertial measurement unit due to the relative displacement with the human body part to which it is attached. The relative motion offset includes the relative acceleration offset. The determination formula of the relative acceleration offset is as follows: (2) Wherein, represents the relative acceleration offset, represents the relative acceleration offset at the previous moment (it can be understood that the relative acceleration offset at the previous moment refers to formula (2)), represents the random acceleration offset at the current moment. Formula (2) means adding the relative acceleration offset at the previous moment and the random acceleration offset at the current moment to obtain the relative acceleration offset at the current moment.

[0028] It can be understood that there is a deviation between the acceleration sensed by the IMU and the actual acceleration of the human body part. The random acceleration offset can simulate this deviation to solve the problem of calculation error of acceleration.

[0029] It can be understood that in step S120, based on the relative angular velocity offset under the ideal state, the relative angular velocity offset under the non-ideal state is modeled. Compared with the relative angular velocity offset which is a fixed value under the ideal state, the relative angular velocity offset can vary with time and is more in line with the actual motion situation. Similarly, compared with the relative acceleration offset which is a fixed value under the ideal state, the relative acceleration offset can vary with time and is more in line with the actual motion situation.

[0030] Step S130: Correct the motion data at the current moment according to the relative motion offset at the current moment to obtain enhanced motion rehabilitation data, and the enhanced motion rehabilitation data is used to train a deep learning model to obtain a human posture determination model.

[0031] Exemplarily, the enhanced motion rehabilitation data includes enhanced angular velocity samples, and the determination formula of the enhanced angular velocity samples is as follows: (3) Wherein, represents the enhanced angular velocity sample, represents the transformation matrix from the global inertial coordinate system to the sensor coordinate system, which can be preset, represents the angular velocity at the current moment, which can be represented in matrix form, represents the inverse matrix of the relative angular velocity offset. Formula (3) means multiplying the inverse matrix of the relative angular velocity offset at the current moment, the matrix of the angular velocity at the current moment, and the preset transformation matrix from the global inertial coordinate system to the sensor coordinate system to obtain the enhanced angular velocity sample.

[0032] Exemplarily, the enhanced motion rehabilitation data further includes enhanced acceleration samples, and the determination formula for the enhanced acceleration samples is as follows: (4) Wherein, represents the enhanced acceleration sample, represents the inverse matrix of the enhanced angular velocity sample that has been calculated (calculated according to reference formula (3)), represents the inverse transformation matrix from the global inertial coordinate system to the sensor coordinate system, represents the matrix of the acceleration at the current moment, represents the random acceleration offset, which can appear in the form of a matrix. Formula (4) means multiplying the sum of the random acceleration offset and the matrix of the acceleration at the current moment, the preset inverse transformation matrix from the global inertial coordinate system to the sensor coordinate system, and the inverse matrix of the enhanced angular velocity sample that has been calculated to obtain the enhanced acceleration sample.

[0033] After obtaining the enhanced samples, using the enhanced acceleration samples and angular velocity samples as input data and the human body posture as the output target to train the deep learning model, a human body posture determination model can be obtained.

[0034] It can be understood that in the related art, directly using the motion data sensed by the inertial measurement unit as the motion rehabilitation data cannot eliminate the errors generated by the human body motion in the motion data, resulting in a decrease in the accuracy of the human body posture determination of the human body posture determination model.

[0035] In the above technical solution of the present application, by modeling the relative motion offset between the inertial measurement unit and the human body part to which the inertial measurement unit is attached as a variable that can change with time (that is, the relative motion offset is obtained according to the relative motion offset at the previous moment and the random offset at the current moment), the relative motion offset that changes with time is more in line with the actual human body motion process. Then, the accuracy of the enhanced motion rehabilitation data obtained by correcting the motion data at the current moment according to the relative motion offset at the current moment is improved, and the accuracy indexes of posture feature extraction and human body part positioning are enhanced. The robustness, generalization ability, and the accuracy of determining the human body posture of the human body posture determination model trained by the enhanced motion rehabilitation data are improved, so as to promote the application progress of the IMU in medical rehabilitation.

[0036] That is, the above technical solution uses the characteristic that different relative displacements and rotations of the IMU during human movement generate different measurement results to perform data augmentation. This method directly mines data diversity from the actual application scenario and can efficiently generate a large number of different "IMU reading motion data - human posture" samples without a complex hardware simulation process. In the field of medical rehabilitation, in the scenario where the patient's movement conditions are complex and variable, this motion rehabilitation data augmentation method based on the IMU displacement during actual movement can more accurately reflect the diversity of real data, provide richer and more practical motion rehabilitation data for the deep learning model, improve the model's learning ability for complex motion patterns, and has obvious advantages in terms of data utilization efficiency and diversity improvement.

[0037] In some embodiments, the value range of the random offset F is , that is, the random offset F follows a standard Gaussian distribution, and its mean is , and the variance is , and it includes random offset components in three directions of the XYZ axes.

[0038] In one implementation manner, the random offset F includes a random angular velocity offset. Obtaining the random offset at the current moment in step S110 includes the following steps: Step S111: Sample from three independent standard Gaussian distributions to obtain three random angular velocity offset components of the inertial measurement unit in the X-axis direction, Y-axis direction, and Z-axis direction at the current moment.

[0039] Exemplarily, sample the random angular velocity offset component of the inertial measurement unit in the X-axis direction from the standard Gaussian distribution A . Sample the random angular velocity offset component of the inertial measurement unit in the Y-axis direction from the standard Gaussian distribution B , and sample the random angular velocity offset component of the inertial measurement unit in the Z-axis direction from the standard Gaussian distribution C .

[0040] Step S112: Multiply the three random angular velocity offset components to obtain the random angular velocity offset.

[0041] Specifically, the determination formula for the random angular velocity offset at the current moment is . That is, the random angular velocity offset has a value range of .

[0042] In another implementation manner, the random offset F further includes a random acceleration offset. Obtaining the random offset at the current moment in step S110 includes the following steps: Step S113: Sample two sets of random displacement offset components for the previous moment and the current moment of the inertial measurement unit from three independent standard Gaussian distributions. Both sets of random displacement offset components include three random displacement offset components in the X-axis direction, Y-axis direction, and Z-axis direction.

[0043] Exemplarily, sample a set of random displacement offset components for the previous moment , and this set of random displacement offset components includes three random displacement offset components in the X-axis direction, Y-axis direction, and Z-axis direction for the previous moment. Similarly, sample a set of random displacement offset components for the current moment , and this set of random displacement offset components includes three random displacement offset components in the X-axis direction, Y-axis direction, and Z-axis direction for the current moment.

[0044] Step S114: Obtain the velocity offset at the current moment based on the two sets of random displacement offset components for the previous moment and the current moment.

[0045] Exemplarily, the determination formula for the velocity offset at the current moment is .

[0046] Step S115: Subtract the calculated velocity offset at the previous moment from the velocity offset at the current moment to obtain the random acceleration offset at the current moment.

[0047] Exemplarily, the determination formula for the random velocity offset at the current moment is , where the velocity offset at the previous moment is calculated with reference to Steps S113 to S114.

[0048] In some embodiments, the step of adjusting the random offset at the current moment according to the empirical value range generated by the rotation of the inertial measurement unit includes: If the random angular velocity offset is not within the empirical value range generated by rotation, re-execute Steps S111 and S112 until the random angular velocity offset is within the empirical value range generated by rotation, and then obtain this random angular velocity offset.

[0049] Alternatively, if the random acceleration offset is not within the empirical value range generated by relative displacement, re-execute Steps S113 and S115 until the random acceleration offset is within the empirical value range generated by relative displacement, and then obtain this random acceleration offset.

[0050] It is understandable that the range of empirical values can be obtained through experimental tests. By restricting the random angular velocity offset and the random acceleration offset within the range of empirical values, the random angular velocity offset is made more consistent with the deviation between the angular velocity sensed by the IMU and the angular velocity of the human body, and the random acceleration offset is made more consistent with the deviation between the acceleration sensed by the IMU and the acceleration of the human body.

[0051] Figure 3 The following is a schematic structural diagram of the sensing device provided by this application. As Figure 3 shown, the sensing device 10 includes: a processor 11, a memory 12, and a bus 13; The memory 12 is used to store the computer program code of the processor 11; Among them, the processor 11 is configured to execute the technical solution of the motion rehabilitation data enhancement method in any of the foregoing method embodiments by executing the computer program code.

[0052] Optionally, the memory 12 can be either independent or integrated with the processor 11.

[0053] The memory 12 is connected to the processor 11 through the bus 13 and completes mutual communication.

[0054] Optionally, the memory 12 may include a random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory.

[0055] The bus 13 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of simplicity of representation, only a thick line is shown in the figure, but it does not mean that there is only one bus or one type of bus.

[0056] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0057] The sensing device 10 is used to execute the technical solutions provided in any of the foregoing method embodiments. The implementation principles and technical effects are similar and will not be elaborated herein.

[0058] This application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the motion rehabilitation data enhancement method as described above.

[0059] Those of ordinary skill in the art can understand that all or part of the steps of implementing the foregoing method embodiments can be completed by hardware related to program instructions. The foregoing program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps including the foregoing method embodiments; and the foregoing storage medium includes various media that can store program codes, such as ROM, RAM, magnetic disks, or optical discs.

[0060] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit them. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for enhancing motion rehabilitation data, the method being applied to a sensing device, characterized in that, The sensing device includes an inertial measurement unit and a control unit, and the control unit is configured to, obtain a random offset at the current moment, a relative motion offset at the previous moment between the inertial measurement unit and the human body part to which the inertial measurement unit is attached, and motion data at the current moment sensed by the inertial measurement unit; obtain the relative motion offset at the current moment based on the relative motion offset at the previous moment and the random offset at the current moment; correct the motion data at the current moment according to the relative motion offset at the current moment to obtain enhanced motion rehabilitation data, and the enhanced motion rehabilitation data is used to train a deep learning model to obtain a human posture determination model.

2. The method according to claim 1, wherein The enhanced motion rehabilitation data includes enhanced angular velocity samples and enhanced acceleration samples.

3. The method according to claim 2, wherein The random offset includes a random angular velocity offset generated by the inertial measurement unit due to self-rotation. The obtaining of the random offset at the current moment includes: respectively sampling from three independent standard Gaussian distributions to obtain three random angular velocity offset components at the current moment of the inertial measurement unit in the X-axis direction, Y-axis direction, and Z-axis direction; multiplying the three random angular velocity offset components to obtain the random angular velocity offset at the current moment.

4. The method according to claim 3, wherein The relative motion offset includes a relative angular velocity offset. The obtaining of the relative motion offset at the current moment based on the relative motion offset at the previous moment and the random offset at the current moment includes: multiplying the relative angular velocity offset at the previous moment and the random angular velocity offset at the current moment to obtain the relative angular velocity offset at the current moment.

5. The method according to claim 4, wherein The motion data at the current moment includes the angular velocity at the current moment. The correcting of the motion data at the current moment according to the relative motion offset at the current moment to obtain enhanced motion rehabilitation data includes: multiplying the inverse matrix of the relative angular velocity offset at the current moment, the matrix of the angular velocity at the current moment, and a preset transformation matrix from the global inertial coordinate system to the sensor coordinate system to obtain enhanced angular velocity samples.

6. The method according to claim 2, wherein The random offset further includes a random acceleration offset generated by the inertial measurement unit due to relative displacement. The obtaining of the random offset at the current moment includes: respectively sampling from three independent standard Gaussian distributions to obtain two sets of random displacement offset components of the inertial measurement unit at the previous moment and the current moment, and both sets of random displacement offset components include three random displacement offset components in the X-axis direction, Y-axis direction, and Z-axis direction; obtaining the velocity offset at the current moment according to the two sets of random displacement offset components at the previous moment and the current moment; subtracting the velocity offset at the previous moment that has been calculated from the velocity offset at the current moment to obtain the random acceleration offset at the current moment.

7. The method according to claim 6, wherein The relative motion offset includes a relative acceleration offset. The obtaining of the relative motion offset at the current moment based on the relative motion offset at the previous moment and the random offset at the current moment includes: adding the relative acceleration offset at the previous moment and the random acceleration offset at the current moment to obtain the relative acceleration offset at the current moment.

8. The method according to claim 7, characterized in that, The motion data at the current moment includes the acceleration at the current moment. The enhanced motion rehabilitation data obtained by correcting the motion data at the current moment according to the relative motion offset at the current moment includes: Multiply the sum of the random acceleration offset at the current moment and the matrix of the acceleration at the current moment, the inverse transformation matrix from the preset global inertial coordinate system to the sensor coordinate system, and the inverse matrix of the calculated enhanced angular velocity samples to obtain the enhanced acceleration samples.

9. The method according to claim 1, wherein Adjust the random offset at the current moment according to the empirical value range generated by the inertial measurement unit due to rotation and relative displacement.

10. A sensing device, characterized in that, Including: A processor and a memory; The memory is coupled to the processor. The memory is used to store computer program code, and the processor calls the computer program code to enable the sensing device to execute the method according to any one of claims 1 to 9.

Images