A fuzzy expert control method for osseointegration

Abstract

The present invention discloses a fuzzy expert control method for bone integration, which uses the advantages of expert algorithms and fuzzy algorithms to achieve intelligent control of prostheses. The fuzzy algorithm meets the requirements for prosthetic control and has low processor requirements. It can significantly reduce the size and cost of the prosthesis, and can also reduce the running time with low latency. Expert solutions are then provided for each user, and the expert algorithm is used to adjust the kinematic control angle based on the common analysis of experts to achieve control voltage adjustment of the prosthetic motor, which helps the user's physical recovery.

Description

Technical Field

[0001] The present invention relates to the technical field of medical devices, and more particularly to a fuzzy expert control method for osseointegration. Background Art

[0002] Recent attempts at intelligent control technologies in intelligent prosthetics research include: in the control of powered knee joints, after comparing neural network control and fuzzy control in terms of handling uncertain data and the need for training, fuzzy PID control was ultimately chosen to control the knee joint's DC motor; intelligent above-the-knee prostheses based on cerebellar model neural network controllers; motion pattern recognition based on BP neural networks; and gait recognition based on probabilistic neural networks. Furthermore, research is ongoing on hybrid intelligent control technologies such as neural network expert control and neural network fuzzy logic control.

[0003] The development of intelligent prosthetic control has evolved from open-loop to closed-loop, and from single-variable control to single-variable feedback to multivariable control and then to multivariable feedback. Currently, most widely used prosthetics are passive, and the limited research on intelligent prosthetics also uses open-loop control. Therefore, existing prosthetic control technology lacks research on intelligent control methods.

[0004] Therefore, how to achieve intelligent control of prostheses is an urgent problem that needs to be solved by those skilled in the art. Summary of the Invention

[0005] In light of this, the present invention provides a fuzzy expert control method for osseointegration, leveraging the advantages of both expert and fuzzy algorithms to achieve intelligent prosthetic control. Considering speed, professionalism, convenience, and affordability, the present invention uses fuzzy expert control as a control method for intelligent prostheses during osseointegration surgery, addressing the shortcomings of current inexpensive prostheses, such as the lack of active control, and the unsmoothness of expensive intelligent prostheses. The systems in intelligent prosthetic designs are nonlinear and unpredictable. Fuzzy algorithms meet the requirements for prosthetic control, placing low demands on processors, significantly reducing the size and cost of prostheses, and also reducing runtime and latency. Furthermore, expert solutions are provided for each user, utilizing expert algorithms to adjust the kinematic control angle based on common expert analysis, thereby adjusting the control voltage of the prosthetic motor, aiding the user's recovery.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] A fuzzy expert control method for osseointegration comprises the following steps:

[0008] Step 1: Collect the residual leg thigh muscle signal, residual leg sciatic nerve signal and residual leg position signal, and output the residual leg movement status according to the set residual leg expert basic logic;

[0009] Step 2: Collect the good leg thigh muscle signal, good leg calf muscle signal and good leg position signal, and output the good leg movement status according to the set good leg expert basic logic;

[0010] Step 3: Obtain comprehensive movement status based on the movement status of the injured leg and the movement status of the good leg;

[0011] Step 4: Obtain motor control instructions based on the current comprehensive motion situation and the comprehensive motion situation at the previous moment, and transmit the motor control instructions to the prosthetic motor.

[0012] Preferably, the residual leg thigh muscle signal includes the residual leg thigh front muscle signal and the residual leg thigh back muscle signal; the good leg thigh muscle signal includes the good leg thigh front muscle signal and the residual leg thigh back muscle signal; the good leg calf muscle signal includes the good leg calf front muscle signal and the good leg calf back muscle signal.

[0013] Preferably, the fuzzy control logic specifically processes and calculates the residual leg thigh muscle signal, residual leg sciatic nerve signal and residual leg position signal to obtain the residual leg thigh front signal amplitude and time, residual leg thigh back muscle signal amplitude and time, residual leg sciatic nerve signal amplitude and time, residual leg position signal angle and amplitude; judges the amplitude strength and time length according to the set threshold value to obtain the judgment result.

[0014] Preferably, the basic logic of the disabled leg expert and the basic logic of the good leg expert are adopted to judge the movement situation and trend according to the judgment results.

[0015] Preferably, the basic logic of the residual leg expert is: if the sciatic nerve signal of the residual leg is a strong pulse signal, the muscle signal of the front thigh of the residual leg is long and high in amplitude, and the muscle signal of the back thigh of the residual leg has no obvious high amplitude, the movement trend is judged to be lifting the residual leg;

[0016] The sciatic nerve signal of the residual leg is a strong pulse signal, the muscle signal of the posterior thigh of the residual leg is long and high in amplitude, and the muscle signal of the anterior thigh of the residual leg has no obvious high amplitude, and it is judged that the movement trend is to retract the residual leg;

[0017] The sciatic nerve signal of the residual leg is a weak pulse signal, and the movement trend is judged to be that the residual leg is at rest;

[0018] The movement angle and the initial value of the residual leg position signal change, and it is determined that the state at the moment before the change is whether the residual leg is raised or lowered.

[0019] Preferably, the basic logic of the good leg expert is: if the good leg front thigh muscle signal and the good leg calf muscle signal have a long duration and a high amplitude, and the good leg calf muscle signal and the good leg posterior thigh muscle signal have no obvious high amplitude, the movement trend is judged to be lifting the good leg;

[0020] The good leg posterior thigh muscle signal and the good leg calf muscle signal have a long duration and a high amplitude, and the good leg posterior thigh muscle signal and the good leg calf muscle signal do not have an obviously high amplitude, and the movement trend is judged to be retracting the good leg;

[0021] The good leg thigh muscle signal and the good leg calf muscle signal do not have an obviously high amplitude, and it is judged that the movement trend is that the good leg is at rest;

[0022] The movement angle and initial value of the good leg position signal change, and it is determined that the state before the change is whether the good leg is raised or lowered.

[0023] Preferably, the residual leg thigh muscle signal and the good leg thigh muscle signal are spindle-shaped, and the residual leg sciatic nerve signal is pulse-shaped; the residual leg position signal and the good leg position signal use a gyroscope to collect three-axis angles.

[0024] Preferably, in step 4, if the motion result generated by the current motor control instruction is the same as the comprehensive motion situation at the previous moment, the motor runs directly based on the current motor control scheme; otherwise, the motor runs slowly based on the current motor control scheme.

[0025] Preferably, the motor control instruction outputs a corresponding instruction according to the judgment rules specified by the integrated motion logic library.

[0026] It can be seen from the above technical solution that compared with the existing technology, the present invention discloses a fuzzy expert control method for bone integration, which uses fuzzy control as the control method of intelligent prostheses in bone integration surgery, and adopts expert basic logic for motion judgment, and comprehensively controls the movement of intelligent prostheses, which can achieve the requirements of speed, professionalism, convenience and low cost, and can provide expert solutions for each user, which helps the user's physical recovery. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0028] Figure 1 The accompanying drawing is a flow chart of the fuzzy expert control method for osseointegration provided by the present invention. DETAILED DESCRIPTION

[0029] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0030] The embodiment of the present invention discloses a fuzzy expert control method for osseointegration, comprising the following steps:

[0031] S1: Collect the residual leg thigh muscle signal, residual leg sciatic nerve signal and residual leg position signal, and output the residual leg movement status in the residual leg movement selector according to the set residual leg expert basic logic and fuzzy control logic;

[0032] S2: Collect the good leg thigh muscle signal, good leg calf muscle signal and good leg position signal, and output the good leg movement status in the good leg movement selector according to the set good leg expert basic logic and fuzzy control logic;

[0033] S3: in the comprehensive movement selector, the comprehensive movement condition is obtained by using fuzzy control logic according to the movement condition of the disabled leg and the movement condition of the good leg;

[0034] S4: Obtaining a motor control instruction in the integrated motion controller according to the current integrated motion condition and the integrated motion condition at the previous moment, and transmitting the motor control instruction to the prosthetic motor.

[0035] In order to further optimize the above technical solution, the thigh muscle signal of the disabled leg includes the front thigh muscle signal of the disabled leg and the back thigh muscle signal of the disabled leg; the thigh muscle signal of the good leg includes the front thigh muscle signal of the good leg and the back thigh muscle signal of the disabled leg; the calf muscle signal of the good leg includes the front calf muscle signal of the good leg and the back calf muscle signal of the good leg.

[0036] In order to further optimize the above technical solution, the fuzzy control logic is specifically to process the residual leg thigh muscle signal, the residual leg sciatic nerve signal and the residual leg position signal to calculate the residual leg thigh front signal amplitude and time, the residual leg thigh back muscle signal amplitude and time, the residual leg sciatic nerve signal amplitude and time, the residual leg position signal angle and amplitude;

[0037] The amplitude strength and duration are judged according to the set threshold to obtain the judgment result.

[0038] In order to further optimize the above technical solution, a fuzzy control knowledge base is established according to the basic logic of the disabled leg expert and the good leg expert, and the movement situation and trend are judged based on the judgment results combined with the fuzzy control knowledge base.

[0039] To further optimize the above technical solution, the basic logic of the residual leg expert is as follows: if the sciatic nerve signal of the residual leg is a strong pulse signal, the muscle signal of the front thigh of the residual leg is long and high in amplitude, and the muscle signal of the back thigh of the residual leg does not have a significant high amplitude, the movement trend is judged to be lifting the residual leg;

[0040] The sciatic nerve signal of the residual leg is a strong pulse signal, the muscle signal of the posterior thigh of the residual leg is long and high in amplitude, and the muscle signal of the anterior thigh of the residual leg has no obvious high amplitude. It is judged that the movement trend is to retract the residual leg;

[0041] The sciatic nerve signal of the residual leg is a weak pulse signal, and the movement trend is judged to be that the residual leg is at rest;

[0042] The movement angle and the initial value of the residual leg position signal change, and the state before the change is determined to be whether the residual leg is raised or lowered.

[0043] To further optimize the above technical solution, the basic logic of the Good Leg Expert is as follows: if the good leg front thigh muscle signal and the good leg back calf muscle signal have a long duration and high amplitude, and the good leg back thigh muscle signal and the good leg front calf muscle signal do not have a significantly high amplitude, the movement trend is judged to be lifting the good leg;

[0044] The good leg's posterior thigh muscle signal and the good leg's calf muscle signal have long durations and high amplitudes, while the good leg's posterior thigh muscle signal and the good leg's calf muscle signal do not have significantly high amplitudes, indicating that the movement trend is to retract the good leg.

[0045] The good leg thigh muscle signal and the good leg calf muscle signal have no obvious high amplitude, and the movement trend is judged to be that the good leg is at rest;

[0046] The movement angle and initial value of the good leg position signal change, and the state before the change is determined to be whether the good leg is raised or lowered.

[0047] In order to further optimize the above technical solution, the thigh muscle signals of the residual leg and the good leg are spindle-shaped, and the sciatic nerve signals of the residual leg are pulse-shaped; the three-axis angles of the residual leg position signal and the good leg position signal are collected by a gyroscope.

[0048] In order to further optimize the above technical solution, in S4, if the motion result generated by the current motor control instruction is the same as the comprehensive motion situation at the previous moment, the motor runs directly based on the current motor control scheme; if not, the motor runs slowly based on the current motor control scheme.

[0049] In order to further optimize the above technical solution, the motor control instructions output corresponding instructions according to the judgment rules specified by the comprehensive motion logic library.

[0050] The motor voltage of the prosthetic motor ensures that its own parameters are set within a safe range while the motor cannot be overloaded, ensuring that the patient is in a relatively safe and stable state, and minimizing secondary damage to the patient's residual limb or walking imbalance.

[0051] Example

[0052] Fuzzy logic control (also known as fuzzy control) is a computer digital control technology based on fuzzy set theory, fuzzy language variables, and fuzzy logic reasoning. The basic methods and main steps of fuzzy controller design generally include:

[0053] (1) Fuzzy control defines variables, that is, determines the observed conditions and the actions to be controlled. In the present invention, the first step is the residual leg thigh muscle signal, the residual leg sciatic nerve signal and the residual leg position signal; the second step is the good leg thigh muscle signal, the good leg calf muscle signal and the good leg position signal; the third step is the output results of the first and second steps: the movement of the residual leg and the movement of the good leg; the fourth step is the output result of the third step: the current comprehensive movement situation and the output result of the third step at the previous moment: the comprehensive movement situation at the previous moment.

[0054] (2) Fuzzy control fuzzification, collects electromyographic signals, neural signals and position signals, and judges the current motion state and motion trend according to the set segmentation threshold, as shown in Table 1 below:

[0055] Table 1 Segmentation threshold judgment table

[0056]

[0057] (3) Fuzzy control knowledge base

[0058] It consists of two parts: a database and a rule base. The database provides the relevant definitions for processing fuzzy data; while the rule base is a group of language control rules that describe control objectives and strategies. The fuzzy knowledge base is expressed in the above language.

[0059] (4) Fuzzy control logic judgment

[0060] By simulating the fuzzy concepts of human judgment, fuzzy logic and fuzzy inference are used to generate fuzzy control signals. This is the essence of fuzzy controllers. Fuzzy logic judgment involves the process of calculating the actual situation at the time of use.

[0061] (5) Fuzzy control defuzzification

[0062] Defuzzification: Convert the fuzzy value obtained by inference into a clear control signal as the input value of the system. The present invention adopts the centroid method as the defuzzification method, taking the centroid of the area enclosed by the membership function curve and the horizontal coordinate as the final output value of the fuzzy reasoning, that is,

[0063] Output = Σ(input × weight) / Σ weight

[0064] The center of gravity method has a smoother output inference control, and the output changes even in response to small changes in the input signal.

[0065] (6) The judgment rules specified by the comprehensive motion logic library are shown in Table 2.

[0066] Table 2 Comprehensive motion logic library judgment rules

[0067]

[0068]

[0069] After receiving a standard prosthesis in the hospital, amputees first undergo 10 walking tests, during which their electromyographic (EMG) threshold and gait data are recorded. A doctor then provides a walking recovery plan. Electrodes, sensors, and an osseointegrated prosthesis are then installed. These data (including EMG signals, sciatic nerve signals, position signals, movement of the residual leg, movement of the intact leg, combined movement, and motor control command values) are input into the sensors as segmented thresholds for the fuzzy algorithm. Once the patient returns home, using this algorithm in their daily lives can effectively improve their quality of life and recovery outcomes (recovery refers to muscle maintenance in the intact leg and the injured thigh, as well as upper body balance movements).

[0070] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.

[0071] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A fuzzy expert control method for osseointegration, characterized in that: The following steps are involved: Step 1: Collect the residual leg thigh muscle signal, residual leg sciatic nerve signal and residual leg position signal, and output the residual leg movement situation according to the set residual leg expert basic logic and fuzzy control logic; the residual leg thigh muscle signal includes the residual leg front thigh muscle signal and the residual leg back thigh muscle signal; the residual leg thigh muscle signal is spindle-shaped, the residual leg sciatic nerve signal is pulse-shaped, and the residual leg position signal and the three-axis angle are collected by a gyroscope; The fuzzy control logic specifically processes the residual leg thigh muscle signal, residual leg sciatic nerve signal and residual leg position signal to calculate the residual leg thigh front signal amplitude and time, residual leg thigh back muscle signal amplitude and time, residual leg sciatic nerve signal amplitude and time, residual leg position signal angle and amplitude; and determines the amplitude strength and duration based on the set threshold to obtain the judgment result; The basic logic of the residual leg expert is used to judge the movement situation and trend according to the judgment result; the basic logic of the residual leg expert is: the sciatic nerve signal of the residual leg is a strong pulse signal, the signal of the muscle on the front of the residual leg thigh is long and high in amplitude, and the signal of the muscle on the back of the residual leg thigh has no obvious high amplitude, and the movement trend is judged to be lifting the residual leg; the sciatic nerve signal of the residual leg is a strong pulse signal, the signal of the muscle on the back of the residual leg thigh is long and high in amplitude, and the signal of the muscle on the front of the residual leg thigh has no obvious high amplitude, and the movement trend is judged to be retracting the residual leg; the sciatic nerve signal of the residual leg is a weak pulse signal, and the movement trend is judged to be resting the residual leg; the movement angle and the initial value of the residual leg position signal change, and the state before the change is judged to be whether the residual leg is lifted or lowered; Step 2: Collect the good leg thigh muscle signal, the good leg calf muscle signal and the good leg position signal, and output the good leg movement status according to the set good leg expert basic logic and fuzzy control logic; the good leg thigh muscle signal includes the good leg front thigh muscle signal and the residual leg back thigh muscle signal; the good leg calf muscle signal includes the good leg front calf muscle signal and the good leg back calf muscle signal; the good leg thigh muscle signal is spindle-shaped, and the good leg position signal uses a gyroscope to collect three-axis angles; the fuzzy control logic performs fuzzy control based on the good leg thigh muscle signal, the good leg calf muscle signal and the good leg position signal to obtain a judgment result; The good leg expert basic logic is used to determine the movement status and trend based on the judgment results; the good leg expert basic logic is: if the good leg front thigh muscle signal and the good leg back calf muscle signal have a long duration and high amplitude, and the good leg back thigh muscle signal and the good leg front calf muscle signal do not have a significantly high amplitude, the movement trend is determined to be lifting the good leg; The good leg posterior thigh muscle signal and the good leg calf muscle signal have a long duration and a high amplitude, and the good leg posterior thigh muscle signal and the good leg calf muscle signal have no obvious high amplitude, and the movement trend is judged to be that the good leg is retracted; the good leg thigh muscle signal and the good leg calf muscle signal have no obvious high amplitude, and the movement trend is judged to be that the good leg is at rest; the movement angle and the initial value of the good leg position signal change, and the state before the change is judged to be that the good leg is raised or lowered; Step 3: Based on the movement of the disabled leg and the movement of the good leg, the fuzzy control logic is used to obtain the comprehensive movement situation; Step 4: Obtain motor control instructions based on the current comprehensive motion situation and the comprehensive motion situation at the previous moment, and transmit the motor control instructions to the prosthetic motor.

2. A fuzzy expert control method for osseointegration according to claim 1, characterized in that: In step 4, if the motion result generated by the current motor control instruction is the same as the comprehensive motion situation at the previous moment, the motor runs directly based on the current motor control scheme; otherwise, the motor runs slowly based on the current motor control scheme.

Images