A spinal surgery bypass robot based on electromagnetic wave reflection coefficient measurement

Abstract

This invention discloses a spinal surgery initiation robot based on electromagnetic wave reflection coefficient measurement. A radio frequency signal generator produces electromagnetic wave signals, which are emitted and received by a probe from biological tissue. A P-parameter measurement unit is designed to calculate the electromagnetic wave reflection coefficient, i.e., the P-parameter, based on the difference between the emitted and reflected signals. P-parameter data from different locations are collected to train a classification model, thereby identifying different bone tissues within the spine. An alarm device issues an alarm signal based on the processing results, guiding the probe's direction of travel. Compared with existing technologies, this invention overcomes the influence of body fluids on measurement accuracy by employing electromagnetic wave reflection coefficient measurement technology, enabling accurate measurement of anterior bone tissue and guiding precise positioning and safe operation of the spinal probe. This invention realizes a handheld, miniaturized spinal surgery initiation robot, reducing costs while ensuring reliability.

Description

Technical Field

[0001] This invention relates to the field of orthopedic surgical robots, and more particularly to a spinal surgery open-path robot based on electromagnetic wave reflection coefficient measurement. Background Technology

[0002] Traditional orthopedic surgery relies heavily on the surgeon's experience and control over the arm muscles, resulting in extremely high surgical risks. Compared to traditional orthopedic surgery, robot-assisted orthopedic surgery offers several advantages: precise access, smaller incisions, higher stability, and better postoperative outcomes. From an industry perspective, considering market size and technological development, orthopedic surgical robots have a promising future. However, my country's orthopedic surgical robot industry started relatively late and is still in its early stages of industrialization. Therefore, designing a spinal surgery incision robot at a lower cost while ensuring reliability is a pressing technical challenge. Summary of the Invention

[0003] The purpose of this invention is to address the shortcomings of existing technologies by providing a spinal surgery open-path robot based on electromagnetic wave reflection coefficient measurement.

[0004] The objective of this invention is achieved through the following technical solution: a spinal surgery open-path robot based on electromagnetic wave reflection coefficient measurement, comprising a probe and a handle;

[0005] The probe has a front end and a rear end, with the front end used for drilling holes in spinal surgery.

[0006] The handle is internally equipped with a looper, an RF signal generator, an RF signal receiver, a downconverter, an analog-to-digital converter, and a control processing module.

[0007] The looper is connected to the rear end of the probe; the radio frequency signal generator and the radio frequency signal receiver are both connected to the looper;

[0008] The radio frequency signal generator is used to generate electromagnetic wave signals of a specified frequency, which are transmitted to the probe via the looper, emitted into the biological tissue through the probe, and received by the radio frequency signal receiver.

[0009] Both the radio frequency signal generator and the radio frequency signal receiver are connected to the control processing module via a downconverter and an analog-to-digital converter, respectively.

[0010] The downconverter is used to downconvert high-frequency signals; the analog-to-digital converter is used to sample the frequency-converted signal, convert it into a digital signal, and then input it into the control processing module.

[0011] The control processing module is connected to the PC and the alarm device; the control processing module is used to process the sampling signal of the analog-to-digital conversion circuit, calculate the P parameter based on the difference between the transmitted signal and the reflected signal, and identify different bone tissues inside the spine.

[0012] The PC is used to train a classification model based on the P parameters of different tissues drilled by the probe, and to deploy the trained classification model to the control and processing module.

[0013] The alarm device issues an alarm signal based on the processing result of the control processing module, i.e., the real-time drilling area of ​​the probe, to guide the direction of the probe's movement.

[0014] Furthermore, the control processing module includes a P-parameter measurement unit, which is used to calculate the P-parameters at different tissue locations reached by the probe, specifically:

[0015] The electromagnetic wave signal generated by the radio frequency signal generator is transmitted to the biological tissue through a probe via a looper, and the other signal is transmitted to the control processing module after passing through a first down-converter and a first analog-to-digital converter to obtain a first digital signal.

[0016] The radio frequency signal receiver receives the electromagnetic wave signal reflected by biological tissue, which is then processed by the second down-converter and the second analog-to-digital converter to obtain a second digital signal, which is then transmitted to the control processing module.

[0017] In the P-parameter measurement unit, the result of Fourier transforming the first digital signal is used as the incident energy P. I The result of the Fourier transform of the second digital signal is taken as the reflected energy P. R According to the incident energy P I With reflected energy P R The formula for calculating the P parameter is as follows:

[0018] Furthermore, the PC terminal calculates the P parameters at different tissue locations by the probe based on the received P parameter measurement unit, and trains a classification model using the P parameters as features.

[0019] The classification model is set into three major categories: cortical bone, cancellous bone, and blood. The cortical bone category is further divided into two subcategories: cortical bone contaminated with blood and cortical bone uncontaminated with blood. The cancellous bone category is further divided into three subcategories: cancellous bone contaminated with blood, cancellous bone contaminated with cortical bone fragments, and cancellous bone uncontaminated with both blood and cortical bone fragments.

[0020] Furthermore, based on the differences in P parameters among different bone tissues, the classification model is set as a four-level support vector machine (SVM) structure, with the first level containing SVM1, the second level containing SVM2, the third level containing SVM3 and SVM4, and the fourth level containing SVM5.

[0021] The SVM1 is used to identify whether it belongs to the "blood" category, and samples identified as not belonging to the "blood" category are input into the SVM2.

[0022] The SVM2 is used to identify the "cortical bone" category and the "cancellous bone" category. Samples identified as "cortical bone" are input into SVM3, and samples identified as "cancellous bone" are input into SVM4.

[0023] The SVM3 is used to identify two subcategories: "cortical bone stained with blood" and "cortical bone not stained with blood"; the SVM4 is used to identify whether it is "cancellous bone stained with blood", and samples identified as not "cancellous bone stained with blood" are input into the SVM5.

[0024] The SVM5 is used to identify two subclasses: "cancellous bone contaminated with cortical bone fragments" and "cancellous bone that is neither contaminated with blood nor contaminated with cortical bone fragments".

[0025] Furthermore, the PC deploys the trained classification model to the control and processing module. The control and processing module collects the P parameters of the probe during real-time movement through the P parameter measurement unit and inputs them into the trained classification model to obtain the real-time drilling area of ​​the probe.

[0026] Furthermore, the probe is composed of a coaxial line, consisting of an inner conductor, an insulator, and a rigid outer conductor marked with graduations from the inside out; the probe tip adopts a pointed antenna structure.

[0027] Furthermore, the handle adopts an ellipsoidal transparent hollow handle, and various functional modules are installed inside the handle; the probe and handle are designed as an integrated structure or a detachable structure.

[0028] Furthermore, the microcontroller unit (MCU) of the control processing module reserves multiple traditional serial port channels as expansion interfaces to facilitate the addition of new functional modules in the future.

[0029] Furthermore, the warning device employs an audible and visual alert method, providing an alarm through light and / or buzzer when the probe enters a dangerous area; for different tissue types identified, it uses different colored lights, different frequency lights, different frequency buzzers, different volume buzzers, or any combination of the above methods to provide an alert.

[0030] A method for guiding and alarming a spinal puncture probe using the aforementioned spinal surgery incision robot, the method comprising the following steps:

[0031] Step 1: Deploy a P-parameter measurement unit with a measurement frequency range of 50MHz to 500MHz in the control and processing module. It includes two modes: point frequency and sweep frequency. The measurement principle is to calculate the P-parameter based on the incident energy and reflected energy. The measured P-parameters will vary for different tissues during the spinal incision process.

[0032] Step 2: Use the P-parameter measurement unit to calculate the P-parameters at different tissue locations reached by the probe, transmit the calculated P-parameters to the PC, train a classification model on the PC using the P-parameters as features, and deploy the trained classification model to the control and processing module on the PC.

[0033] Step 3: During the drilling process, the radio frequency signal generator emits electromagnetic wave signals to the human tissue and then reflects the echo signals. The P parameter measurement unit calculates the P parameters of the probe during real-time movement and inputs them into the trained classification model to obtain the classification result, which is the real-time drilling area of ​​the probe. Then, the classification result guides the precise positioning and safe operation of the probe. When the probe is identified as entering a dangerous area, an alarm is triggered.

[0034] The beneficial effects of this invention are as follows: This invention proposes a spinal surgery initiation robot based on electromagnetic wave reflection coefficient measurement. A radio frequency signal generator produces electromagnetic wave signals, which are emitted and received by a probe from biological tissue. A P-parameter measurement unit is designed to calculate the electromagnetic wave reflection coefficient, i.e., the P-parameter, based on the difference between the emitted and reflected signals. P-parameter data from different locations are collected to train a classification model, thereby identifying different bone tissues within the spine. An alarm device issues an alarm signal based on the processing results, guiding the probe's direction of travel. Compared with existing technologies, this invention overcomes the influence of body fluids on measurement accuracy by employing electromagnetic wave reflection coefficient measurement technology, enabling accurate measurement of the bone tissue in front, guiding the precise positioning and safe operation of the spinal probe. Furthermore, a control processing module and other functional modules are installed inside the handle, providing corresponding prompts to the user based on the probe head's position. This invention realizes a handheld, miniaturized spinal surgery initiation robot, ensuring reliability while reducing costs, filling the gap in domestic handheld spinal robot technology. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 A schematic diagram illustrating the connections of various parts of a spinal surgery open-path robot, as shown in an exemplary embodiment;

[0037] Figure 2 A schematic diagram of the overall structure of a spinal surgery bypass robot, as shown in an exemplary embodiment;

[0038] Figure 3 A schematic diagram of the internal structure of a probe shown in an exemplary embodiment;

[0039] Figure 4 A partially enlarged view of the probe tip shown in an exemplary embodiment;

[0040] Figure 5 A diagram of tissues penetrated during spinal surgery (from left to right) illustrating an exemplary embodiment;

[0041] Figure 6 A schematic diagram of a classification model shown in an exemplary embodiment;

[0042] Figure 7 A flowchart illustrating the workflow of a spinal surgery incision robot, as shown in an exemplary embodiment;

[0043] Figure 8 This is a flowchart illustrating a signal processing example. Detailed Implementation

[0044] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0045] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0046] like Figure 1 , Figure 2 As shown, this embodiment of the invention provides a spinal surgery open-path robot based on electromagnetic wave reflection coefficient measurement, including a probe 1 and a handle 12, wherein the probe 1 and the handle 12 are designed as an integral structure or a detachable structure.

[0047] The probe 1 is composed of a coaxial line, and its internal structure is as follows: Figure 3 As shown, from the inside out, the components are: inner conductor 1-1, insulator 1-2, and rigid outer conductor 1-3 with graduations; a magnified view of the probe tip is shown below. Figure 4 As shown, it employs a pointed antenna structure, which facilitates penetration of bone;

[0048] The handle 12 is internally equipped with a looper 2, an RF signal generator 3, an RF signal receiver 4, a downconverter, an analog-to-digital conversion circuit, and a control processing module 9; the looper 2 is connected to the rear end of the probe 1; the RF signal generator 3 and the RF signal receiver 4 are both connected to the looper 2;

[0049] The radio frequency signal generator 3 is used to generate electromagnetic wave signals of a specified frequency, which are transmitted to the probe 1 via the looper 2, emitted into the biological tissue by the probe 1, and received by the radio frequency signal receiver 4.

[0050] The frequency range of the electromagnetic wave signal generated by the radio frequency signal generator 3 is between 50MHz and 500MHz, and includes two transmission modes: frequency sweep and frequency spot.

[0051] The radio frequency signal generator 3 and the radio frequency signal receiver 4 are both connected to the control processing module 9 via a downconverter and an analog-to-digital converter circuit, respectively. Specifically, the radio frequency signal generator 3 is connected to the control processing module 9 via a first downconverter 6 and a first analog-to-digital converter circuit 8 in sequence; the radio frequency signal receiver 4 is connected to the control processing module 9 via a second downconverter 5 and a second analog-to-digital converter circuit 7 in sequence.

[0052] The downconverter is used to downconvert high-frequency signals; the analog-to-digital converter is used to sample the frequency-converted signal, convert it into a digital signal, and then input it into the control processing module 9.

[0053] The control processing module 9 is connected to a PC, and also to an alarm device 10 and a battery 11. The control processing module 9 processes the digital signal sampled by the analog-to-digital conversion circuit, calculates the electromagnetic wave reflection coefficient (P parameter) based on the difference between the transmitted and reflected signals, and identifies different bone tissues within the spine, including cortical bone, cancellous bone, and blood vessels. Figure 5 As shown; the PC terminal is used to train a lightweight classification model based on the P parameters of the probe drilling into different tissues, and deploys the trained classification model inside the control processing module; the alarm device 10 issues an alarm signal based on the processing result of the control processing module, i.e., the real-time drilling area of ​​the probe, to guide the direction of the probe's movement.

[0054] Furthermore, the control processing module includes a P-parameter measurement unit, which is used to calculate the electromagnetic wave reflection coefficient, i.e., the P-parameter, at different tissue locations reached by the probe. The specific process is as follows:

[0055] The electromagnetic wave signal generated by the radio frequency signal generator is transmitted to the biological tissue through a probe via a looper, and the other signal is converted into a digital signal 1 by the first down-converter and the first analog-to-digital conversion circuit and then transmitted to the control processing module.

[0056] The electromagnetic wave signal reflected by biological tissue is received by the radio frequency signal receiver, and after passing through the second down-converter and the second analog-to-digital conversion circuit, a digital signal 2 is obtained and transmitted to the control processing module.

[0057] In the P-parameter measurement unit of the control processing module, the result of Fourier transforming the digital signal 1 is taken as the incident energy P. I The result of the Fourier transform of digital signal 2 is taken as the reflected energy P. R According to the incident energy P I With reflected energy P R The formula for calculating the P parameter is as follows: In this embodiment, a 256-point Fast Fourier Transform is performed using 32-bit floating-point numbers, but it is not limited to this.

[0058] Furthermore, the PC terminal calculates the P parameters at different tissue locations obtained by the received P parameter measurement unit, and trains a classification model using the P parameters as features; the classification model can be implemented using a support vector machine; at the initial moment when the probe penetrates from the superficial layer (cortical bone) to the deep layer (cancellous bone), the probe head will be mixed with blood or debris from the superficial layer, so data collection and classification are also required for cases with impurities; in this embodiment, the classification model is set to three major categories: cortical bone, cancellous bone, and blood; the cortical bone category is further divided into two subcategories: cortical bone contaminated with blood and cortical bone uncontaminated with blood; the cancellous bone category is further divided into three subcategories: cancellous bone contaminated with blood, cancellous bone contaminated with cortical bone debris, and cancellous bone uncontaminated with both blood and cortical bone debris; specifically, as Figure 6 As shown, based on the differences in P-parameters among different bone tissues, the classification model is set up as a 4-level Support Vector Machine (SVM) structure. The first level contains SVM1, the second level contains SVM2, the third level contains SVM3 and SVM4, and the fourth level contains SVM5. SVM1 is used to identify whether a sample belongs to the "blood" category. Samples identified as not belonging to the "blood" category are input into SVM2. SVM2 is used to identify the "cortical bone" and "cancellous bone" categories, and samples identified as belonging to the "cortical bone" category are input into SVM5. Samples of the class are input into SVM3. Samples identified as belonging to the "cancellous bone" class are input into SVM4. SVM3 is used to identify two subclasses: "cortical bone contaminated with blood" and "cortical bone not contaminated with blood". SVM4 is used to identify whether a sample is "cancellous bone contaminated with blood". Samples identified as not being "cancellous bone contaminated with blood" are input into SVM5. SVM5 is used to identify two subclasses: "cancellous bone contaminated with cortical bone fragments" and "cancellous bone neither contaminated with blood nor contaminated with cortical bone fragments".

[0059] Furthermore, the PC deploys the trained classification model to the control and processing module. The control and processing module collects the P parameters of the probe during real-time movement through the P parameter measurement unit and inputs them into the trained classification model to obtain the real-time drilling area of ​​the probe.

[0060] Furthermore, the handle of this spinal surgery bypass robot is an ellipsoidal transparent hollow handle, in which various functional modules are installed.

[0061] Furthermore, the probe and handle of this spinal surgery open-path robot are designed to be detachable. During use, the contamination of bodily fluids is limited to the probe part. When used again, only the probe needs to be replaced. The handle and its internal functional modules are reusable.

[0062] Furthermore, the microcontroller unit (MCU) of the control and processing module of the spinal surgery open-path robot reserves multiple traditional serial port channels as expansion interfaces to facilitate the addition of new functional modules in the future. The expansion interfaces can be connected to a serial port screen for graphic display, or connected to an antenna tuning device to automatically tune the antenna.

[0063] Furthermore, the warning device of this spinal surgery bypass robot uses audio-visual prompts. When the probe enters a dangerous area, it provides an alarm through lights and / or a high-frequency buzzer. For example, different colored lights, lights of different frequencies, buzzers of different frequencies, buzzers of different volume, or any combination of the above methods can be used to provide prompts for different types of identified tissues.

[0064] In another embodiment, a method for guiding and alarming a spinal puncture probe using the aforementioned spinal surgery open-path robot is provided, such as... Figure 7 As shown, the method includes the following steps:

[0065] Step 1: Deploy a P-parameter measurement unit with a measurement frequency range of 50MHz to 500MHz in the control processing module. This unit includes both spot frequency and sweep frequency modes. The measurement principle is based on the incident energy P. I With reflected energy P R Calculate the P parameter:

[0066]

[0067] Among them, the incident energy P I With reflected energy P RThe calculation method is as follows: An electromagnetic wave signal generated by a radio frequency signal generator is transmitted through a looper to the inside of the biological tissue via a probe. Another signal is converted to digital signal 1 by a first down-converter and a first analog-to-digital converter, and then transmitted to the control processing module. The electromagnetic wave signal reflected from the biological tissue is received by a radio frequency signal receiver, and converted to digital signal 2 by a second down-converter and a second analog-to-digital converter, and then transmitted to the control processing module. The Fourier transform result of digital signal 1 is used as the incident energy P. I The result of the Fourier transform of digital signal 2 is taken as the reflected energy P. R .

[0068] For example Figure 5 The P-parameters measured during the spinal bypass procedure vary depending on the different tissue structures (cortical bone, cancellous bone, blood, etc.).

[0069] Step 2: The P-parameter measurement unit is used to calculate the P-parameters at different tissue locations reached by the probe. The control and processing module transmits the calculated P-parameters to the PC. On the PC, a classification model is trained using the P-parameters as features. At the initial moment when the probe penetrates from the shallow layer to the deep layer, the probe tip may be mixed with blood or debris from the shallow layer. Therefore, it is also necessary to collect and classify data in cases where impurities are present. The trained classification model is then deployed to the control and processing module on the PC.

[0070] The drilling process of the handheld spinal surgery opener robot is as follows: (1) epidermis, (2) dermis, (3) cortical bone, (4) cancellous bone. When drilling in the dermis, the probe surface is easily contaminated with blood and water, and when drilling in the cortical bone layer, it will also be contaminated with cortical bone fragments. The specific classification categories of the classification model are set as three major categories: cortical bone, cancellous bone, and blood. The cortical bone category is further divided into two subcategories: cortical bone contaminated with blood and cortical bone not contaminated with blood and water. The cancellous bone category is further divided into three subcategories: cancellous bone contaminated with blood and water, cancellous bone contaminated with cortical bone fragments, and cancellous bone not contaminated with blood and water and not contaminated with cortical bone fragments.

[0071] Pig bones were selected as the experimental subject due to their smaller vertebrae. Data were collected sequentially from different tissues within the spine, including blood, cortical bone, and cancellous bone, using both single-frequency measurements and frequency sweeps. The frequency sweep range was set to 50MHz–500MHz with 2MHz steps. The P-parameters of different tissues were measured. Next, bovine bones were selected as the experimental subject due to their larger vertebrae. Data were collected sequentially from different tissues within the spine, including blood, cortical bone, and cancellous bone, using both single-frequency measurements and frequency sweeps. The frequency sweep range was set to 50MHz–500MHz with 2MHz steps. The P-coefficients of different tissues were measured. Finally, clinical testing was conducted to obtain clinical data. [The text then abruptly shifts to a seemingly unrelated topic:] ...using... Figure 6 The five support vector machines (SVMs) shown (SVM1-5) complete the classification task.

[0072] Step 3, the signal processing flowchart is as follows: Figure 8 As shown, during the drilling process, the radio frequency signal generator emits electromagnetic wave signals to the human tissue and then reflects the echo signals. The P-parameter measurement unit of the control processing module calculates the P-parameters of the probe during real-time movement and inputs them into the trained classification model to obtain the classification result, which is the real-time drilling area of ​​the probe. Then, based on the classification result, the probe is guided to accurately position and operate safely. When the probe is identified to enter a dangerous area, an alarm is triggered through the alarm device to ensure the safety of the operation.

[0073] The above description is merely a preferred embodiment of one or more embodiments of this specification and is not intended to limit the scope of one or more embodiments of this specification. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of one or more embodiments of this specification should be included within the protection scope of one or more embodiments of this specification.

Claims

1. A spinal surgery incision robot based on electromagnetic wave reflection coefficient measurement, characterized in that, Includes probe and handle; The probe has a front end and a rear end. The front end adopts a pointed antenna structure for drilling holes in spinal surgery. The handle is internally equipped with a looper, an RF signal generator, an RF signal receiver, a downconverter, an analog-to-digital converter, and a control processing module. The looper is connected to the rear end of the probe; the radio frequency signal generator and the radio frequency signal receiver are both connected to the looper; The radio frequency signal generator is used to generate electromagnetic wave signals of a specified frequency, which are transmitted to the probe via the looper, emitted into the biological tissue through the probe, and received by the radio frequency signal receiver. The frequency range of the electromagnetic wave signal generated by the radio frequency signal generator is between 50MHz and 500MHz, and includes two transmission modes: frequency sweep and frequency spot. Both the radio frequency signal generator and the radio frequency signal receiver are connected to the control processing module via a downconverter and an analog-to-digital converter, respectively. The downconverter is used for downconverting high-frequency signals; The analog-to-digital converter circuit is used to sample the frequency-converted signal, convert it into a digital signal, and then input it into the control processing module. The control processing module is connected to the PC and the alarm device; the control processing module is used to process the sampling signal of the analog-to-digital conversion circuit, calculate the P parameter based on the difference between the transmitted signal and the reflected signal, and identify different bone tissues inside the spine. The control processing module includes a P-parameter measurement unit, which is used to calculate the P-parameters at different tissue locations reached by the probe. Specifically: The electromagnetic wave signal generated by the radio frequency signal generator is transmitted to the biological tissue through a probe via a looper, and the other signal is transmitted to the control processing module after passing through a first down-converter and a first analog-to-digital converter to obtain a first digital signal. The radio frequency signal receiver receives the electromagnetic wave signal reflected by biological tissue, which is then processed by the second down-converter and the second analog-to-digital converter to obtain a second digital signal, which is then transmitted to the control processing module. In the P-parameter measurement unit, the result of Fourier transforming the first digital signal is used as the incident energy. P I The result of the Fourier transform of the second digital signal is used as the reflected energy. P R According to the incident energy P I With reflected energy P R The formula for calculating the P parameter is as follows: P=P R / P I ; The PC is used to train a classification model based on the P parameters of the probe drilling into different tissues, and to deploy the trained classification model in the control and processing module. The control and processing module collects the P parameters of the probe during real-time movement through the P parameter measurement unit and inputs them into the trained classification model to obtain the real-time drilling area of ​​the probe. The alarm device issues an alarm signal based on the processing result of the control processing module, i.e., the real-time drilling area of ​​the probe, to guide the direction of the probe's movement.

2. The spinal surgery bypass robot based on electromagnetic wave reflection coefficient measurement according to claim 1, characterized in that, The PC terminal calculates the P parameters at different tissue locations obtained by the probe drilling based on the received P parameter measurement unit, and trains a classification model using the P parameters as features. The classification model is set into three major categories: cortical bone, cancellous bone, and blood. The cortical bone category is further divided into two subcategories: cortical bone contaminated with blood and cortical bone uncontaminated with blood. The cancellous bone category is further divided into three subcategories: cancellous bone contaminated with blood, cancellous bone contaminated with cortical bone fragments, and cancellous bone uncontaminated with both blood and cortical bone fragments.

3. The spinal surgery bypass robot based on electromagnetic wave reflection coefficient measurement according to claim 2, characterized in that, Based on the differences in P parameters of different bone tissues, the classification model is set as a four-level support vector machine (SVM) structure, with the first level containing SVM1, the second level containing SVM2, the third level containing SVM3 and SVM4, and the fourth level containing SVM5. The SVM1 is used to identify whether a sample belongs to the "blood" category, and samples identified as not belonging to the "blood" category are input into the SVM2. The SVM2 is used to identify the "cortical bone" category and the "cancellous bone" category. Samples identified as "cortical bone" are input into SVM3, and samples identified as "cancellous bone" are input into SVM4. The SVM3 is used to identify two subcategories: "cortical bone stained with blood" and "cortical bone not stained with blood"; the SVM4 is used to identify whether it is "cancellous bone stained with blood", and samples identified as not "cancellous bone stained with blood" are input into the SVM5. The SVM5 is used to identify two subclasses: "cancellous bone contaminated with cortical bone debris" and "cancellous bone that is neither contaminated with blood nor contaminated with cortical bone debris".

4. The spinal surgery bypass robot based on electromagnetic wave reflection coefficient measurement according to claim 1, characterized in that, The probe is composed of a coaxial line, consisting of an inner conductor, an insulator, and a rigid outer conductor with graduations, arranged from the inside out.

5. The spinal surgery bypass robot based on electromagnetic wave reflection coefficient measurement according to claim 1, characterized in that, The handle is an ellipsoidal transparent hollow handle, and various functional modules are installed inside the handle; the probe and handle are designed as an integrated structure or a detachable structure.

6. The spinal surgery bypass robot based on electromagnetic wave reflection coefficient measurement according to claim 1, characterized in that, The microcontroller unit (MCU) of the control processing module reserves multiple traditional serial port channels as expansion interfaces, facilitating the addition of new functional modules in the future.

7. The spinal surgery bypass robot based on electromagnetic wave reflection coefficient measurement according to claim 1, characterized in that, The warning device uses an audible and visual alert method. When the probe enters a dangerous area, it will provide an alarm through light and / or buzzer. For different types of tissues identified, it will use different colored lights, different frequency lights, different frequency buzzers, different volume buzzers, or any combination of the above methods to provide an alert.

Images