Construction method and application of a cutting state classification model for a medical bone oscillating saw cutting human bones

Abstract

This invention discloses a method for constructing and applying a classification model of cutting states when a medical bone oscillating saw cuts human bone. The method includes: S1. Preparing artificial cancellous bone as experimental material; S2. Classifying the cutting states into five categories: cutting preparation state, pre-cutting state, real-time cutting state, critical state, and penetration state; defining the acceleration signal characteristics for each state, and determining the cutting progress based on vibration and frequency changes; S3. Setting the acceleration signal sampling frequency to 4000Hz, with each sampling frame lasting 64ms, and extracting fifteen-dimensional features, including one-dimensional frequency domain features, five-dimensional wavelet features, and nine-dimensional time domain features; S4. Based on the fifteen-dimensional features obtained from feature extraction, manually classifying the acceleration signals during the cutting process into the five cutting states defined in step S2, and marking the start and end points of each state according to different feature changes; S5. Establishing a classification model using a Support Vector Machine (SVM) classification model with a multinomial kernel function.

Description

Technical Field

[0001] This application relates to the field of assisted orthopedic surgery, and in particular to the design of a method for constructing a classification model of the cutting state when a medical bone oscillating saw cuts human bone, and its application. Background Technology

[0002] In orthopedic surgery, the medical oscillating saw is a commonly used surgical tool, primarily for osteotomy. The saw blade oscillates strongly around its axis after being powered on. However, during surgery, the coolant and cutting debris can obstruct the surgeon's view, making it difficult to clearly observe the cutting area. Therefore, surgeons rely mainly on the sound of the blade contacting bone tissue and the vibrations felt by their hands to judge the cutting progress. For example, when cutting smoothly, the frequency, intensity, and rhythm of the vibrations change relatively little; however, when cutting bone tissues of different densities, such as cortical bone and cancellous bone, the changes in the vibration rhythm increase significantly.

[0003] This method relies on the surgeon's subjective judgment; however, each surgeon's ability to perceive vibrations varies. Some surgeons can keenly detect subtle changes, while others may overlook these details. Therefore, for inexperienced surgeons, the accuracy of judging the cutting status through vibration is often low. For example, in total knee replacement surgery, if excessive penetration occurs when using a bone oscillating saw, it may accidentally cut the carotid artery in the lower leg, posing a serious surgical risk. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and solve the inaccuracy caused by relying on manual hand-sensing vibration for cutting state identification, and to provide a method for constructing and applying a cutting state classification model when a medical bone oscillating saw cuts human bone.

[0005] The objective of this invention is achieved through the following technical solution:

[0006] A method for constructing a classification model of cutting states when a medical bone oscillating saw cuts human bone, comprising:

[0007] S1. Prepare artificial cancellous bone as experimental material;

[0008] S2. Classify the cutting states into cutting preparation state, pre-cutting state, real-time cutting state, critical state, and penetration state; define the acceleration signal characteristics of each state, and determine the cutting process based on the changes in vibration and frequency.

[0009] S3. Set the sampling frequency of the acceleration signal to 4000Hz, with each sampling frame lasting 64ms. After feature extraction, fifteen-dimensional features are obtained, including one-dimensional frequency domain features, five-dimensional wavelet features, and nine-dimensional time domain features.

[0010] S4. Based on the fifteen-dimensional features obtained from feature extraction, manually divide the acceleration signal during the cutting process into the five cutting states defined in step S2, and mark the start and end points of each state according to different feature changes;

[0011] S5. Establish a classification model; use a Support Vector Machine (SVM) classification model with a multinomial kernel function.

[0012] Furthermore, the experimental material for the artificial cancellous bone described in step S1 is solid rigid polyurethane foam with a thickness of 4 mm.

[0013] Furthermore, in step S2, the definitions of each state are as follows:

[0014] In the cutting preparation state, the bone oscillating saw first accelerates to the predetermined rotational speed, and then approaches the experimental material to be cut; the acceleration signal in this stage is caused by the reciprocating motion of the bone oscillating saw itself.

[0015] The pre-cutting state marks the beginning of the cutting process, with the bone oscillating saw making its initial contact with the experimental material and gradually cutting into its interior; as the experimental material deforms and the bone oscillating saw penetrates deeper, the interaction between the two triggers vibrations;

[0016] The real-time cutting state is in which the interaction between the bone oscillating saw and the experimental material reaches its peak; the acceleration signal in this state stage has significant peak values ​​and oscillation modes, with an increase in high-frequency components.

[0017] The critical state occurs when the bone oscillating saw is about to cut out of the experimental material, and the cutting process is nearing its end. During this state, the acceleration signal shows a significant decreasing trend, accompanied by fluctuations, indicating that the cutting activity has not completely terminated.

[0018] Once the bone oscillating saw has completely detached from the experimental material, the cutting operation is complete.

[0019] Furthermore, in step S3,

[0020] One-dimensional frequency domain feature: The spectrum of a sampled frame is obtained by FFT transformation. The frequency value corresponding to the largest peak value of the sampled frame in the frequency range of 200Hz to 300Hz is taken as its fundamental frequency value. The fundamental frequency value is the selected one-dimensional frequency domain feature.

[0021] The five-dimensional wavelet features are obtained by decomposing the sampled frame using the Maximum Overlap Discrete Wavelet Transform (MODWT) and extracting five relative energies as five-dimensional wavelet features.

[0022] The nine-dimensional time-domain features include nine statistical features introduced into the sampled frames: mean, variance, peak value, RMS value, peak factor, kurtosis, impulse factor, waveform factor, and margin factor.

[0023] Furthermore, a Symlet wavelet of order 4 is used as the signal processing tool; the sampling frame is decomposed into 5 layers W1, W2, W3, W4, and W5 using MODWT, with each layer corresponding to a frequency range of 1000–2000 Hz, 500–1000 Hz, 250–500 Hz, 125–250 Hz, and 60–125 Hz, respectively; the signal energy corresponding to the decomposed layers W1, W2, W3, W4, and W5 is divided by the energy of the sampling frame signal to obtain five relative energies; these five relative energies are the five-dimensional wavelet features.

[0024] Furthermore, the calibration process for each complete cutting process in step S4 is as follows:

[0025] In the cutting preparation state calibration: During a complete cutting process, the amplitude of the spectrum of the cutting preparation state does not exceed 1. Therefore, the initial moment until the amplitude exceeds 1 is marked as the cutting preparation state.

[0026] Pre-cutting state: The end of the cutting preparation marks the start of the pre-cutting state, and the end is marked when the fundamental frequency values ​​extracted from the three sampling frames are all equal, and the last sampling frame is marked as the end of the pre-cutting state.

[0027] Real-time cutting state: This is the intermediate part between the end of the pre-cutting state and the beginning of the critical state;

[0028] Critical phase: The portion of the first 3 sampling frames before the start of the penetration state;

[0029] Penetration status: After the bone oscillating saw penetrates, the saw blade will block the optical path of the laser sensor and transmit a high-level signal to the main control chip. This high-level signal is used to mark the penetration status.

[0030] The present invention also provides a cutting status recognition device for a medical bone oscillating saw cutting human bone, including a main control chip, a storage module, an accelerometer, a communication module, a laser sensor, a display module and a power module;

[0031] The main control chip is connected to the storage module, accelerometer, communication module, laser sensor, display module and power module respectively;

[0032] The communication module uses Bluetooth for transmission, the display module uses a serial port screen, and the laser sensor is used to calibrate the penetration state with a detection frequency of 100Hz. The bone oscillating saw cuts perpendicular to the bone plate downwards, and the collected acceleration signal is the vibration direction of the bone oscillating saw.

[0033] This invention also provides an application of a classification model for the cutting state when a medical bone oscillating saw cuts human bone. Based on the method for constructing the classification model for the cutting state when a medical bone oscillating saw cuts human bone as described in any one of claims 1-6 and the cutting state recognition device of claim 7, the invention is characterized by: acquiring signals through the accelerometer of the cutting state recognition device and calculating the fundamental frequency amplitude; extracting features from the acquired signals and inputting them into a support vector machine (SVM) classification model for classification; and displaying the classification results in real time to assist doctors in judging the cutting state.

[0034] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of constructing a method for classifying the cutting state when the medical bone oscillating saw cuts human bone.

[0035] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for constructing a classification model of cutting states when the medical bone oscillating saw cuts human bone.

[0036] Compared with the prior art, the beneficial effects of the technical solution of the present invention are:

[0037] 1. Improved accuracy in identifying cutting status: This invention uses feature extraction from acceleration signals combined with an SVM classification model for identification. Compared to traditional methods that rely on the doctor's subjective perception, it quantifies the cutting status through objective characteristics of vibration and frequency, reducing human error. It is particularly suitable for inexperienced doctors and improves the accuracy of judging the cutting status.

[0038] 2. Reduced surgical risks: By accurately identifying the cutting state, especially critical and penetrating states, the risk of over-cutting is effectively avoided, reducing the possibility of accidental damage to surrounding important blood vessels or nerves. This is particularly important in surgeries such as total knee replacement, avoiding accidental damage to important tissues such as the carotid artery in the lower leg, thereby improving the safety of the surgery.

[0039] 3. Improve surgical efficiency; Automatic identification of cutting status can provide real-time feedback on the cutting process, reducing the time doctors spend observing or adjusting during the cutting process, which helps to improve overall surgical efficiency and shorten the operation time.

[0040] 4. High adaptability, capable of handling complex cutting processes; This invention divides the cutting process into five states (cutting preparation, pre-cutting, real-time cutting, critical, and penetration), and defines the characteristic changes of each state. This segmentation method can comprehensively cover different states in the cutting process, helping to address the influence of changes in bone tissue density and hardness on vibration and acceleration, thereby achieving accurate identification of complex cutting processes.

[0041] 5. Facilitates data analysis and improvement; all cutting data and feature information are stored through a storage module (SD card), enabling postoperative analysis of acceleration signal characteristics, classification model performance, and cutting status at various states. This provides data support for future model optimization and equipment improvement, further enhancing the reliability and applicability of identification.

[0042] 6. Applicable to various types of bone tissue; Through the combined extraction of frequency domain, wavelet and time domain features, this invention can meet the cutting needs of bone tissues of different densities, such as cortical bone and cancellous bone, and is suitable for various types of bone cutting, which is conducive to its widespread application in diverse orthopedic surgeries.

[0043] 7. Real-time display of cutting status enhances the doctor's operating experience; During the operation, the recognition device can display the cutting status to the doctor in real time, allowing the doctor to intuitively grasp the surgical progress, reduce reliance on hand vibration or sound judgment, and improve the surgical operation experience.

[0044] In summary, this invention, through the combination of machine learning models and sensor technology, constructs an automated, precise, and real-time feedback bone oscillating saw cutting status recognition device, realizing precise surgical assistance functions and possessing significant application value and promotion prospects. Attached Figure Description

[0045] Figure 1 This is a schematic diagram of the structure of the present invention.

[0046] Figure 2 It consists of the state diagrams and their corresponding amplitude-frequency curves after classifying the cutting states.

[0047] Figure 3a This is a sampled frame image; Figure 3b This is a signal diagram of the sampled frame after MODWT decomposition.

[0048] Figure 4 This is a schematic diagram of the test set confusion matrix.

[0049] Figure 5 This is a flowchart of the segmentation state classification model. Detailed Implementation

[0050] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only for explaining the present invention and are not intended to limit the present invention.

[0051] like Figure 1As shown, the medical bone oscillating saw cutting status recognition device provided in this embodiment includes a main control chip, a storage module, an accelerometer, a communication module, a laser sensor, a display module, and a power module; the main control chip is connected to the storage module, the accelerometer, the communication module, the laser sensor, the display module, and the power module respectively; the power module is used to supply power to each module;

[0052] The main control chip used in this embodiment is an STM32F407, the storage module is an SD card used to store acceleration signals during the implementation process, the accelerometer (ADXL375, USA, ADI) with a maximum range of 200g, the communication module uses Bluetooth transmission, the specific model is JDY-34 (China, Yourcee), and the laser sensor is LSZ-S30N1 (China, GTRIC) used to calibrate the penetration state, with a detection frequency of 100Hz.

[0053] The display module uses a serial port screen (TJC3224T124-011N, China, TAOJINGCHI);

[0054] The bone oscillating saw cuts perpendicularly downwards from the bone plate, and the collected acceleration signal represents the vibration direction of the bone oscillating saw.

[0055] Preferably, this embodiment provides a method for constructing a classification model of cutting states when a medical oscillating saw cuts human bone, including:

[0056] S1. Prepare artificial cancellous bone as experimental material; in this embodiment, the experimental material used is artificial cancellous bone (40PCF, SAWBONES, USA), which is a solid rigid polyurethane foam with a thickness of 4mm.

[0057] S2. The cutting states are classified as follows: cutting preparation state, pre-cutting state, real-time cutting state, critical state, and penetration state. The acceleration signal characteristics of each state are defined separately, and the cutting process is determined based on changes in vibration and frequency. Figure 2 As shown, specifically:

[0058] 201) In the cutting preparation state, the bone oscillating saw first accelerates to a predetermined rotational speed, and then approaches the experimental material to be cut. The acceleration signal in this stage is mainly caused by the reciprocating motion of the bone oscillating saw itself. For example... Figure 2 As shown, in this state, the amplitude of the frequency response remains at a low level, with no record of exceeding 1.

[0059] 202) Pre-cutting state, marking the start of the cutting process. The bone oscillating saw makes its initial contact with the experimental material and gradually cuts into its interior. As the experimental material deforms and the bone oscillating saw penetrates deeper, the interaction between the two triggers strong vibrations.

[0060] 203) The real-time cutting state is the core stage of the entire cutting process. During this period, the interaction between the cutting tool and the experimental material reaches its peak. Due to the strong interaction between the bone oscillating saw and the experimental material, the acceleration signal in this stage has significant peak values ​​and oscillation modes, with an increase in high-frequency components.

[0061] 204) Critical state: This stage occurs just as the bone oscillating saw is about to cut through the experimental material, and the cutting process is nearing its end. The study revealed that the acceleration signal exhibits a significant decreasing trend during this stage, although it is still accompanied by some fluctuations, indicating that the cutting activity has not completely ceased.

[0062] 205) Once the bone oscillating saw has completely detached from the experimental material, the cutting operation is complete.

[0063] S3. Feature Extraction; The sampling frequency of the acceleration signal is set to 4000Hz, with 64ms as one sampling frame, and 256 data points in each sampling frame; 15-dimensional features are obtained through feature extraction, including one-dimensional frequency domain features, five-dimensional wavelet features, and nine-dimensional time domain features; specifically:

[0064] One-dimensional frequency domain feature: The obtained sample frame is transformed by FFT to obtain its spectrum. The frequency value corresponding to the largest peak value of the sample frame in the frequency range of 200Hz to 300Hz is selected as its fundamental frequency value. This fundamental frequency value is the selected one-dimensional frequency domain feature.

[0065] Five-dimensional wavelet features: The sampled frames are decomposed using the Maximum Overlap Discrete Wavelet Transform (MODWT); in this embodiment, the Symlet wavelet (sym4) of order 4 is used as the signal processing tool. Figure 3a and Figure 3b As shown, MODWT is used to sample frames ( Figure 3a The signal is decomposed into five layers (W1, W2, W3, W4, W5), with corresponding frequency ranges of 1000–2000 Hz, 500–1000 Hz, 250–500 Hz, 125–250 Hz, and 60–125 Hz, respectively. The signal energy corresponding to each of the decomposed layers (W1, W2, W3, W4, W5) is divided by the energy of the sampled frame signal to obtain five relative energies. These five relative energies are the selected five-dimensional wavelet features.

[0066] Nine-dimensional time-domain features: The W2 layer signal obtained above is introduced with the following statistical features: mean, variance, peak value, RMS value, peak factor, kurtosis, impulse factor, waveform factor, and margin factor. These nine values ​​are the selected nine-dimensional time-domain features.

[0067] S4. Based on the fifteen-dimensional features obtained from feature extraction, the acceleration signal during the cutting process is manually divided into the five cutting states defined in step S2, and the start and end points of each state are marked according to different feature changes; specifically, the state marking process for each complete cutting process is as follows:

[0068] In the cutting preparation state calibration: During a complete cutting process, the amplitude of the spectrum of the cutting preparation state does not exceed 1. Therefore, the initial moment until the amplitude exceeds 1 is marked as the cutting preparation state.

[0069] Pre-cutting state: The end of the cutting preparation marks the start of the pre-cutting state, and the end is marked when the fundamental frequency values ​​extracted from the three sampling frames are all equal, and the last sampling frame is marked as the end of the pre-cutting state.

[0070] Real-time cutting state: This is the intermediate part between the end of the pre-cutting state and the beginning of the critical state;

[0071] Critical phase: The portion of the first 3 sampling frames before the start of the penetration state;

[0072] Penetration status: After the bone oscillating saw penetrates, the saw blade will block the optical path of the laser sensor and transmit a high-level signal to the main control chip. This high-level signal is used to mark the penetration status.

[0073] S5. Establish the classification model; the classification model used in this embodiment is SVM (Support Vector Machine). The following are the specific hyperparameter settings: BoxConstraint is set to 10. GapTolerance is set to 0. This embodiment selects polynomial as the KernelFunction. KernelScale is set to 10. PolynomialOrder is set to 2, representing the order of the polynomial kernel function, which determines the dimensionality of the data when mapped to a high-dimensional space, thus affecting the model's nonlinear fitting ability. The RemoveDuplicates parameter is set to true, meaning that duplicate observations in the dataset will be removed to reduce the impact of redundant information on model training and improve computational efficiency. ShrinkagePeriod is set to 0. Finally, ISDA (Iterative Single Data Algorithm) is selected as the Solver.

[0074] The total number of samples is 1415. The dataset consists of 5 different cutting states, each containing 283 samples. 80% of the dataset is used as the training set, and 20% is used as the test set to evaluate the performance of the classification network model. This means that the number of training and test sets are 1132 and 283 respectively, and the output of the classification network model is 5 different cutting stages.

[0075] The confusion matrix in the test set is as follows: Figure 4 As shown, the critical and penetration parameters achieved a 100% recognition rate in prediction.

[0076] Preferably, an application of a classification model for the cutting state when a medical bone oscillating saw cuts human bone is described. Based on the aforementioned method for constructing the classification model and the cutting state recognition device, the accelerometer of the cutting state recognition device collects signals and calculates the fundamental frequency amplitude. Features are extracted from the collected signals and input into a support vector machine (SVM) classification model for classification. The classification results are displayed in real time to assist doctors in judging the cutting state. Specific application methods are as follows: Figure 5 As shown, the STM32 main control chip first initializes each sensor, then starts manually. The accelerometer begins to collect acceleration signals. It first collects one frame of signal and performs FFT calculation to determine whether its fundamental frequency amplitude is greater than 1. If it is less than or equal to 1, it collects another frame of signal. If it is greater than 1, the frame of data is sent to the computer through the serial port for 15-dimensional feature extraction. The features are then fed into a pre-trained SVM classification model for classification. The classification results are then displayed in real time.

[0077] Preferably, embodiments of this application also provide a specific implementation of an electronic device capable of implementing all steps in the method for constructing a classification model of cutting states when a medical bone oscillating saw cuts human bone in the above embodiments. The electronic device specifically includes the following:

[0078] Processor, memory, communications interface, and bus;

[0079] The processor, memory, and communication interface communicate with each other via a bus; the communication interface is used to realize information transmission between server-side devices, metering devices, and user-side devices.

[0080] The processor is used to call the computer program in the memory. When the processor executes the computer program, it implements all the steps in the method for constructing the cutting state classification model when the medical bone oscillating saw cuts human bone in the above embodiment.

[0081] Embodiments of this application also provide a computer-readable storage medium capable of implementing all steps in the method for constructing a classification model of cutting state when a medical bone oscillating saw cuts human bone in the above embodiments. The computer-readable storage medium stores a computer program, which, when executed by a processor, implements all steps in the method for constructing a classification model of cutting state when a medical bone oscillating saw cuts human bone in the above embodiments.

[0082] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on its differences from other embodiments. In particular, hardware + program embodiments are relatively simple in description because they are fundamentally similar to method embodiments; relevant parts can be referred to the descriptions in the method embodiments.

[0083] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0084] While this application provides method operation steps as shown in the embodiments or flowcharts, more or fewer operation steps may be included based on conventional or non-inventive labor. The order of steps listed in the embodiments is merely one possible execution order among many and does not represent the only execution order. In actual device or client product execution, the method can be executed sequentially as shown in the embodiments or drawings, or in parallel (e.g., in a parallel processor or multi-threaded processing environment).

[0085] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0086] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0087] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0088] This invention is not limited to the embodiments described above. The above description of specific embodiments is intended to illustrate and explain the technical solutions of this invention. The specific embodiments described above are merely illustrative and not restrictive. Without departing from the spirit and scope of the claims, those skilled in the art can make many specific modifications based on the teachings of this invention, and these modifications all fall within the scope of protection of this invention.

Claims

1. A method for constructing a classification model of cutting states when a medical bone oscillating saw cuts human bone, characterized in that, include: S1. Prepare artificial cancellous bone as experimental material; S2. The cutting states are classified as follows: cutting preparation state, pre-cutting state, real-time cutting state, critical state, and penetration state. The acceleration signal characteristics of each state are defined separately, and the cutting process is determined based on the changes in vibration and frequency. The definitions of each state are as follows: In the cutting preparation state, the bone oscillating saw first accelerates to the predetermined rotational speed, and then approaches the experimental material to be cut; the acceleration signal in this stage is caused by the reciprocating motion of the bone oscillating saw itself. The pre-cutting state marks the beginning of the cutting process, where the bone oscillating saw makes its first contact with the experimental material and gradually cuts into its interior. As the experimental material deforms and the bone pendulum saw penetrates deeper, the interaction between the two triggers vibrations; The real-time cutting state is in which the interaction between the bone oscillating saw and the experimental material reaches its peak; the acceleration signal in this state stage has significant peak values ​​and oscillation modes, with an increase in high-frequency components. The critical state occurs when the bone oscillating saw is about to cut out of the experimental material, and the cutting process is nearing its end. The acceleration signal in this state phase shows a significant downward trend, accompanied by fluctuations, indicating that the cutting activity has not completely terminated; Once the bone oscillating saw has completely detached from the experimental material, the cutting operation is complete. S3. Set the sampling frequency of the acceleration signal to 4000 Hz, with each sampling frame lasting 64ms. After feature extraction, fifteen-dimensional features are obtained. Including one-dimensional frequency domain features, five-dimensional wavelet features, and nine-dimensional time domain features; One-dimensional frequency domain features: The spectrum of a sampled frame is obtained by FFT transformation. The frequency value corresponding to the largest peak value of the sampled frame in the frequency range of 200Hz~300Hz is used as its fundamental frequency value. The fundamental frequency value is a selected one-dimensional frequency domain feature; The five-dimensional wavelet features are obtained by decomposing the sampled frame using the Maximum Overlap Discrete Wavelet Transform (MODWT) and extracting five relative energies as five-dimensional wavelet features. Nine-dimensional time-domain features include nine statistical features introduced into the sampled frame: mean, variance, peak value, RMS value, peak factor, kurtosis, impulse factor, waveform factor, and margin factor; A Symlet wavelet of order 4 was used as the signal processing tool. The sampled frame was decomposed into five layers (W1, W2, W3, W4, W5) using MODWT, with each layer corresponding to a frequency range of 1000–2000 Hz, 500–1000 Hz, 250–500 Hz, 125–250 Hz, and 60–125 Hz, respectively. The signal energy corresponding to each of the decomposed layers (W1, W2, W3, W4, W5) was divided by the energy of the sampled frame signal to obtain five relative energies. These five relative energies constitute the five-dimensional wavelet features. S4. Based on the fifteen-dimensional features obtained from feature extraction, the acceleration signal during the cutting process is manually divided into the five cutting states defined in step S2, and the start and end points of each state are marked according to different feature changes; the state calibration process for each complete cutting process is as follows: In the cutting preparation state calibration: During a complete cutting process, the amplitude of the spectrum of the cutting preparation state does not exceed 1. Therefore, the initial moment until the amplitude exceeds 1 is marked as the cutting preparation state. Pre-cutting state: The end of the cutting preparation marks the start of the pre-cutting state, and the end is marked when the fundamental frequency values ​​extracted from the three sampling frames are all equal, and the last sampling frame is marked as the end of the pre-cutting state. Real-time cutting state: This is the intermediate part between the end of the pre-cutting state and the beginning of the critical state; Critical state: The portion of the first 3 sampling frames before the start of the penetration state; Penetration status: After the bone oscillating saw penetrates, the saw blade will block the optical path of the laser sensor and transmit a high-level signal to the main control chip. This high-level signal is used to mark the penetration status. S5. Establish a classification model; use a Support Vector Machine (SVM) classification model with a multinomial kernel function.

2. The method for constructing a classification model of cutting states when a medical bone oscillating saw cuts human bone according to claim 1, characterized in that, The experimental material for the artificial cancellous bone mentioned in step S1 is solid rigid polyurethane foam with a thickness of 4 mm.

3. A device for identifying the cutting state when a medical bone oscillating saw cuts human bone, used in the method for constructing a classification model of the cutting state when the medical bone oscillating saw cuts human bone according to any one of claims 1-2, characterized in that, It includes a main control chip, storage module, accelerometer, communication module, laser sensor, display module, and power supply module; The main control chip is connected to the storage module, accelerometer, communication module, laser sensor, display module and power module respectively; The communication module uses Bluetooth for transmission, the display module uses a serial port screen, and the laser sensor is used to calibrate the penetration state with a detection frequency of 100 Hz. The cutting direction of the bone oscillating saw is perpendicular to the bone plate and cuts downwards. The collected acceleration signal is the vibration direction of the bone oscillating saw.

4. An application of a cutting state classification model when a medical bone oscillating saw cuts human bone, based on the method for constructing the cutting state classification model when a medical bone oscillating saw cuts human bone as described in any one of claims 1-2 and the cutting state recognition device of claim 3, characterized in that, The accelerometer of the cutting status recognition device collects signals and calculates the fundamental frequency amplitude; the collected signals are feature extracted and input into the support vector machine (SVM) classification model for classification; the classification results are displayed in real time to assist doctors in judging the cutting status.

5. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of constructing a classification model of the cutting state when the medical bone oscillating saw cuts human bone according to any one of claims 1 to 2.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the method for constructing a classification model of the cutting state when the medical bone oscillating saw cuts human bone as described in any one of claims 1 to 2.

Images