CN117137585A - Ultrasonic osteotome control system with superposition of low-frequency pulse and high-frequency vibration - Google Patents
Ultrasonic osteotome control system with superposition of low-frequency pulse and high-frequency vibration Download PDFInfo
- Publication number
- CN117137585A CN117137585A CN202311137375.8A CN202311137375A CN117137585A CN 117137585 A CN117137585 A CN 117137585A CN 202311137375 A CN202311137375 A CN 202311137375A CN 117137585 A CN117137585 A CN 117137585A
- Authority
- CN
- China
- Prior art keywords
- time
- training
- time sequence
- module
- operation area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 239000007788 liquid Substances 0.000 claims abstract description 48
- 238000011217 control strategy Methods 0.000 claims abstract description 17
- 238000006243 chemical reaction Methods 0.000 claims abstract description 5
- 239000013598 vector Substances 0.000 claims description 111
- 238000012549 training Methods 0.000 claims description 85
- 210000000988 bone and bone Anatomy 0.000 claims description 39
- 239000011159 matrix material Substances 0.000 claims description 22
- 238000005070 sampling Methods 0.000 claims description 17
- 230000001965 increasing effect Effects 0.000 claims description 13
- 238000007781 pre-processing Methods 0.000 claims description 11
- 230000003247 decreasing effect Effects 0.000 claims description 10
- 238000000605 extraction Methods 0.000 claims description 10
- 230000011218 segmentation Effects 0.000 claims description 7
- 230000002123 temporal effect Effects 0.000 claims description 7
- 238000001356 surgical procedure Methods 0.000 claims description 6
- 238000012300 Sequence Analysis Methods 0.000 claims description 5
- 238000004458 analytical method Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 2
- 239000012530 fluid Substances 0.000 abstract description 13
- 238000004422 calculation algorithm Methods 0.000 abstract description 10
- 238000000034 method Methods 0.000 description 25
- 238000010586 diagram Methods 0.000 description 14
- 230000006870 function Effects 0.000 description 11
- 230000008569 process Effects 0.000 description 10
- 238000005520 cutting process Methods 0.000 description 6
- 238000009826 distribution Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 6
- 238000012545 processing Methods 0.000 description 6
- 230000026683 transduction Effects 0.000 description 5
- 238000010361 transduction Methods 0.000 description 5
- 239000000110 cooling liquid Substances 0.000 description 4
- 239000000243 solution Substances 0.000 description 4
- 230000001276 controlling effect Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000001816 cooling Methods 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000002347 injection Methods 0.000 description 2
- 239000007924 injection Substances 0.000 description 2
- 238000007477 logistic regression Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 210000004872 soft tissue Anatomy 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000003763 carbonization Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000005094 computer simulation Methods 0.000 description 1
- 239000002826 coolant Substances 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000001939 inductive effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000003014 reinforcing effect Effects 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
- 238000012731 temporal analysis Methods 0.000 description 1
- 230000003685 thermal hair damage Effects 0.000 description 1
- 238000000700 time series analysis Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
- 230000002792 vascular Effects 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B17/00—Surgical instruments, devices or methods, e.g. tourniquets
- A61B17/32—Surgical cutting instruments
- A61B17/320068—Surgical cutting instruments using mechanical vibrations, e.g. ultrasonic
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B17/00—Surgical instruments, devices or methods, e.g. tourniquets
- A61B17/32—Surgical cutting instruments
- A61B17/320068—Surgical cutting instruments using mechanical vibrations, e.g. ultrasonic
- A61B2017/320082—Surgical cutting instruments using mechanical vibrations, e.g. ultrasonic for incising tissue
Landscapes
- Health & Medical Sciences (AREA)
- Surgery (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Heart & Thoracic Surgery (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Mechanical Engineering (AREA)
- Biomedical Technology (AREA)
- Dentistry (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Surgical Instruments (AREA)
Abstract
An ultrasonic osteotome control system with superposition of low-frequency pulse and high-frequency vibration is disclosed. The system comprises a CPU control module, a foot switch module, a manual switch module, a power conversion module and a touch display module which are communicatively connected with the CPU control module, an ultrasonic high-frequency driving module, a low-frequency controller and a liquid flow module which are connected with the CPU control module, a transducer and a cutter bar which is drivably connected with the transducer, a working cutter head which is arranged on the cutter bar, a temperature sensor which is arranged in an operation area, and a liquid jet opening which faces to the working cutter head, wherein the liquid flow module is used for controlling a control strategy for generating liquid flow. Thus, an intelligent algorithm can be utilized to collect the temperature values of the surgical field and analyze the implicit timing characteristics therefrom to implement a control strategy for automatically determining fluid flow.
Description
Technical Field
The present disclosure relates to the field of ultrasonic osteotomes, and more particularly, to an ultrasonic osteotome control system with superposition of low frequency pulses and high frequency vibrations.
Background
Ultrasonic bone knives are a commonly used cutting tool in medical procedures that utilize the characteristics of ultrasonic waves to cut bone tissue to achieve accurate and efficient surgical procedures.
Conventional ultrasonic osteotome systems typically use only high frequency vibrations to cut bone tissue. However, the single high-frequency vibration cannot generate enough moment, so that the bone knife is difficult to penetrate hard bone, and the cutting efficiency is low.
Thus, an optimized solution is desired.
Disclosure of Invention
In view of this, the present disclosure proposes an ultrasonic osteotome control system that superimposes low frequency pulses and high frequency vibrations, and can utilize an intelligent algorithm to collect the temperature values of the surgical field and analyze implicit timing characteristics therefrom, thereby implementing a control strategy that automatically determines the flow of fluid.
According to an aspect of the present disclosure, there is provided an ultrasonic osteotome control system in which low frequency pulses and high frequency vibrations are superimposed, including a CPU control module, a foot switch module, a manual switch module, a power conversion module, and a touch display module communicably connected to the CPU control module, an ultrasonic high frequency driving module, a low frequency controller, and a liquid flow module connected to the CPU control module, and a transducer and a tool bar drivably connected to the transducer, a working tool bit being mounted to the tool bar, including:
a temperature sensor mounted to the surgical field;
a liquid ejection port facing the working bit; and
the liquid flow module is used for controlling a control strategy for generating liquid flow.
According to an embodiment of the present disclosure, the system includes a CPU control module, a foot switch module, a manual switch module, a power conversion module, and a touch display module communicably connected to the CPU control module, an ultrasonic high frequency driving module, a low frequency controller, and a liquid flow module connected to the CPU control module, a transducer, and a cutter bar drivably connected to the transducer, a working cutter head mounted to the cutter bar, a temperature sensor mounted to an operation area, and a liquid injection port facing the working cutter head, the liquid flow module for controlling a control strategy for generating a liquid flow. Thus, an intelligent algorithm can be utilized to collect the temperature values of the surgical field and analyze the implicit timing characteristics therefrom to implement a control strategy for automatically determining fluid flow.
Other features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments, which proceeds with reference to the accompanying drawings.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate exemplary embodiments, features and aspects of the present disclosure and together with the description, serve to explain the principles of the disclosure.
Fig. 1 shows a block diagram of the fluid flow module in an ultrasonic bone knife control system with low frequency pulses superimposed with high frequency vibrations in accordance with an embodiment of the present disclosure.
Fig. 2 shows a block diagram of the timing analysis unit in the ultrasonic bone knife control system with superposition of low frequency pulses and high frequency vibrations according to an embodiment of the present disclosure.
Fig. 3 shows a block diagram of the data preprocessing subunit in an ultrasonic bone knife control system with low frequency pulses superimposed with high frequency vibrations in accordance with an embodiment of the present disclosure.
Fig. 4 shows a block diagram of the timing feature extraction subunit in an ultrasonic bone-knife control system with superposition of low frequency pulses and high frequency vibrations in accordance with an embodiment of the present disclosure.
Fig. 5 shows a block diagram of the fluid flow control unit in an ultrasonic bone knife control system with low frequency pulses superimposed with high frequency vibrations in accordance with an embodiment of the present disclosure.
Fig. 6 shows a flowchart of an ultrasonic osteotome control method with low frequency pulses superimposed with high frequency vibrations, in accordance with an embodiment of the present disclosure.
Fig. 7 shows a schematic architecture diagram of an ultrasonic osteotome control method with superposition of low frequency pulses and high frequency vibrations in accordance with an embodiment of the present disclosure.
Fig. 8 illustrates an application scenario diagram of an ultrasonic bone-knife control system with low frequency pulses superimposed with high frequency vibrations according to an embodiment of the present disclosure.
Fig. 9 shows a schematic view of an ultrasonic osteotome system in accordance with another embodiment of the present disclosure.
Detailed Description
The following description of the embodiments of the present disclosure will be made clearly and fully with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some, but not all embodiments of the disclosure. All other embodiments, which can be made by one of ordinary skill in the art without undue burden based on the embodiments of the present disclosure, are also within the scope of the present disclosure.
As used in this disclosure and in the claims, the terms "a," "an," "the," and/or "the" are not specific to a singular, but may include a plurality, unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that the steps and elements are explicitly identified, and they do not constitute an exclusive list, as other steps or elements may be included in a method or apparatus.
Various exemplary embodiments, features and aspects of the disclosure will be described in detail below with reference to the drawings. In the drawings, like reference numbers indicate identical or functionally similar elements. Although various aspects of the embodiments are illustrated in the accompanying drawings, the drawings are not necessarily drawn to scale unless specifically indicated.
In addition, numerous specific details are set forth in the following detailed description in order to provide a better understanding of the present disclosure. It will be understood by those skilled in the art that the present disclosure may be practiced without some of these specific details. In some instances, methods, means, elements, and circuits well known to those skilled in the art have not been described in detail in order not to obscure the present disclosure.
The present disclosure provides an ultrasonic osteotome control system with superposition of low frequency pulses and high frequency vibrations, comprising: the generation controller is used for converting the electric energy into an input signal, wherein the generation controller comprises a low-frequency pulse controller and a high-frequency vibration controller; a transducer driver communicably connected to the generation controller for receiving the input signal, resonating, and generating ultrasonic waves; a working bit communicatively connected to the transduction driver for transmitting the ultrasonic wave from an end of the working bit near the transduction driver to an end far from the transduction driver, thereby inducing resonance such that a tip of the working bit is amplitude-shifted and acts on bone tissue such that the bone tissue is cut; a liquid flow module for determining a control strategy for liquid flow; and an injection port mounted adjacent the working bit and communicatively connected to the liquid flow module for injecting a cooling liquid based on the control strategy of the liquid flow.
Accordingly, in one example, an ultrasonic osteotome control system with superimposed low frequency pulses and high frequency vibrations includes a CPU control module, a foot switch module, a manual switch module, a power conversion module, and a touch display module communicatively coupled to the CPU control module, an ultrasonic high frequency drive module, a low frequency controller, and a fluid flow module coupled to the CPU control module, a transducer, and a blade bar drivingly coupled to the transducer, a working blade mounted to the blade bar, a temperature sensor mounted to a surgical area, and a fluid ejection port facing the working blade, the fluid flow module for controlling a control strategy for generating a fluid flow.
Wherein, the cooling liquid plays an important role in the ultrasonic osteotome system. It is mainly used for cooling the working tool bit. When ultrasonic waves are transmitted to the working bit and cause resonance, a certain amount of heat is generated. Without cooling by the coolant, the working bits may overheat, resulting in reduced cutting effectiveness or thermal damage. Generally, the flow rate of the cooling liquid needs to be controlled to ensure that enough cooling liquid flows through the working tool bit in the cutting process, so that the temperature is effectively reduced, and the cutting operation is not influenced by excessive flow rate. Conventional control of the flow of liquid is typically performed manually, i.e., by an operator who is required to adjust the flow of the injected liquid based on experience and observation. This approach may be subjective and inconsistent in control results and may not allow precise control of fluid flow. In this regard, an optimized liquid flow module is to be designed to solve this technical problem in the technical solution of the present disclosure.
Specifically, the technical concept of the present disclosure is to utilize an intelligent algorithm to collect a surgical area temperature value and analyze implicit timing characteristics therefrom, thereby implementing a control strategy to automatically determine a fluid flow.
Based on this, fig. 1 shows a block diagram schematic of the liquid flow module in an ultrasonic bone knife control system with low frequency pulses superimposed with high frequency vibrations according to an embodiment of the present disclosure. As shown in fig. 1, in the ultrasonic osteotome control system according to the embodiment of the disclosure, in which the low frequency pulse is superimposed with the high frequency vibration, the liquid flow module 100 includes: a temperature value acquisition unit 110 for acquiring operation region temperature values at a plurality of predetermined time points within a predetermined period of time acquired by the temperature sensor; a time sequence analysis unit 120, configured to perform time sequence analysis on the surgical field temperature values at the plurality of predetermined time points to obtain a time dimension reinforced surgical field temperature time sequence correlation feature vector; and a liquid flow control unit 130 for determining a control strategy of the liquid flow based on the time-dimension intensive surgery area temperature time sequence correlation feature vector.
Accordingly, in the technical solution of the present disclosure, first, the operation region temperature values at a plurality of predetermined time points within a predetermined period of time acquired by the temperature sensor are acquired. Then, data preprocessing is carried out on the operation area temperature values at a plurality of preset time points to obtain a sequence of operation area temperature local time sequence input vectors.
In one specific example of the present disclosure, the encoding process of performing data preprocessing on the surgical area temperature values at the plurality of predetermined time points to obtain a sequence of surgical area temperature local time sequence input vectors includes: firstly, arranging the operation area temperature values of the plurality of preset time points into operation area temperature time sequence input vectors according to a time dimension; then, up-sampling the surgical area temperature time sequence input vector based on linear interpolation to obtain an up-sampled surgical area temperature time sequence input vector; and vector segmentation is carried out on the up-sampling operation area temperature time sequence input vector so as to obtain a sequence of operation area temperature local time sequence input vectors.
Here, since the actually acquired surgical field temperature data may have a low sampling frequency, the time interval between data points is large, which may cause the model to lose some important time series information while learning. By upsampling the data, the data points can be increased in the time dimension, enabling the model to better capture the time dependence and dynamic changes in liquid flow. Upsampling may be achieved by interpolation methods, such as linear interpolation or spline interpolation, to fill in the values between the original data points, resulting in denser time series data.
And then, carrying out time sequence feature extraction on the sequence of the local time sequence input vectors of the temperature of the operation area to obtain the time dimension reinforced operation area temperature time sequence related feature vector. In a specific example of the present disclosure, the encoding process of performing time sequence feature extraction on the sequence of the surgical region temperature local time sequence input vectors to obtain the time dimension enhanced surgical region temperature time sequence correlation feature vector includes: firstly, the sequence of the local time sequence input vector of the temperature of the operation area is respectively passed through a time sequence feature extractor based on a one-dimensional convolution layer to obtain the sequence of the local time sequence feature vector of the temperature of the operation area; and then passing the sequence of the local time sequence characteristic vectors of the temperature of the operation area through a time attention module to obtain a time dimension reinforced operation area temperature time sequence related characteristic vector.
Accordingly, as shown in fig. 2, the timing analysis unit 120 includes: a data preprocessing subunit 121, configured to perform data preprocessing on the surgical area temperature values at the plurality of predetermined time points to obtain a sequence of surgical area temperature local time sequence input vectors; and a time sequence feature extraction subunit 122, configured to perform time sequence feature extraction on the sequence of the local time sequence input vectors of the surgical region temperature to obtain the time dimension enhanced surgical region temperature time sequence associated feature vector.
More specifically, as shown in fig. 3, the data preprocessing subunit 121 includes: an input vector arrangement secondary sub-unit 1211 for arranging the surgical field temperature values at the plurality of predetermined time points in a time dimension as a surgical field temperature time sequence input vector; an upsampling secondary subunit 1212 configured to upsample the surgical field temperature timing input vector based on linear interpolation to obtain an upsampled surgical field temperature timing input vector; and a vector slicing second-level subunit 1213, configured to perform vector slicing on the upsampled surgical area temperature time-series input vector to obtain a sequence of the surgical area temperature local time-series input vector. It is worth mentioning that upsampling based on linear interpolation is a data processing technique for increasing the sampling rate of data or improving the time resolution of data. Given input data at a lower sampling rate or lower temporal resolution, linear interpolation can estimate the value of the intermediate point in time by linearly interpolating between the data points. In the data preprocessing subunit, the upsampling secondary subunit 1212 upsamples the surgical field temperature timing input vector using a linear interpolation-based method, which means that it inserts new time points between existing time points and uses linear interpolation to estimate the values of these new time points, and by increasing the number of time points, upsampling can increase the time resolution of the data so that finer time variations can be captured. The purpose of the upsampling is to better capture subtle changes in the temperature of the surgical field to provide more detailed and accurate input data. This may be beneficial for subsequent data analysis, model training, or other tasks, as higher temporal resolution may provide more information that helps reveal patterns and trends in the data.
More specifically, as shown in fig. 4, the timing feature extraction subunit 122 includes: a one-dimensional convolution encoding secondary subunit 1221, configured to obtain a sequence of local time sequence feature vectors of the surgical region temperature by using a time sequence feature extractor based on a one-dimensional convolution layer, where the sequence of local time sequence input vectors of the surgical region temperature is respectively; and a temporal attention encoding secondary subunit 1222 for passing the sequence of surgical field temperature local time series feature vectors through a temporal attention module to obtain the temporal dimension enhanced surgical field temperature time series correlation feature vector.
It should be noted that the one-dimensional convolution layer is a neural network layer commonly used in deep learning, and is used for processing one-dimensional sequence data, such as time sequence or signal data. It applies a convolution operation on one-dimensional input data, extracting local features of the input data by learning the weights of the convolution kernel (also called a filter). The one-dimensional convolution layer has the following effects and advantages in time series data: 1. feature extraction: the one-dimensional convolution layer performs convolution operation on the input sequence in a sliding window mode, local features are extracted, and a weight sharing mechanism of the convolution kernel enables the one-dimensional convolution layer to capture local modes and structures of the input data, such as trends, periodicity or other important time-related features in the time sequence. 2. Parameter sharing: the convolution kernels in the one-dimensional convolution layer share the same weight over the entire input sequence, and the feature of such parameter sharing reduces the number of parameters of the model, making the model lighter and capable of better processing long sequence data. 3. Translation invariance: the one-dimensional convolution layer has the characteristic of translational invariance, namely, for the same mode in the input sequence, no matter where the same mode appears in the sequence, the convolution layer can extract similar characteristic representation, and the characteristic enables the one-dimensional convolution layer to have better robustness for the local mode in the time sequence. 4. Dimension reduction: by adjusting the size and stride of the convolution kernel, the one-dimensional convolution layer can reduce the dimension of the input sequence while preserving important features, which helps reduce the computational complexity of the model, and can extract higher level abstract features. In the data processing process, the one-dimensional convolution coding subunit 1221 performs convolution operation on the sequence of the operation region temperature local time sequence input vectors by applying the time sequence feature extractor of the one-dimensional convolution layer, so as to obtain the sequence of the operation region temperature local time sequence feature vectors, thus being capable of converting original temperature data into more abstract and meaningful feature representation and providing more information input for subsequent data analysis and model training.
It should be noted that the time attention module is a neural network module, which is used for weighting different time steps in the time sequence data so as to strengthen the time dimension association characteristic of the time sequence data. The time attention module has the following effects and advantages when processing time series data: 1. time weight learning: the time attention module weights the features of the different time steps by learning weights that can be used to highlight or suppress features of certain time steps to capture important points in time or time periods in the time series data. 2. Modeling time sequence association: by weighting the different time steps of the time series data, the time attention module can better model the association between the time series data, which can help the model better understand the time dependence and dynamic change in the time series data, thereby extracting more accurate and meaningful time series association characteristics. 3. Improving the model performance: the time attention module can introduce richer information interaction between different time steps of the time sequence data, the model can pay more attention to important time steps by weighting the time sequence data, and reduce dependence on irrelevant time steps, so that the performance and generalization capability of the model can be improved, and the model can be better adapted to the characteristics of the time sequence data. In other words, the time attention module can extract and strengthen the associated features in the time sequence data by weighting the time dimension of the time sequence data, so that the understanding and modeling capability of the model on the time sequence data is improved.
Further, the time-dimension intensive surgery area temperature time sequence associated feature vector is passed through a classifier to obtain a classification result, wherein the classification result is used for representing that the liquid flow value of the current time point should be increased or decreased.
Accordingly, as shown in fig. 5, the liquid flow control unit 130 includes: a flow value judging subunit 131, configured to pass the time dimension intensive surgery area temperature time sequence related feature vector through a classifier to obtain a classification result, where the classification result is used to indicate that the liquid flow value at the current time point should be increased or decreased; and a control subunit 132 for taking the classification result as a control strategy of the liquid flow rate.
More specifically, the flow value determining subunit 131 is further configured to: performing full-connection coding on the time dimension reinforced operation area temperature time sequence associated feature vector by using a full-connection layer of the classifier to obtain a coded classification feature vector; and inputting the coding classification feature vector into a Softmax classification function of the classifier to obtain the classification result.
That is, in the technical solution of the present disclosure, the labels of the classifier include a first label to which the liquid flow value at the current time point should be increased, and a second label to which the liquid flow value at the current time point should be decreased, wherein the classifier determines, through a soft maximum function, to which classification label the time-dimension intensive surgery region temperature time sequence association feature vector belongs. It should be noted that the first tag p1 and the second tag p2 do not include the concept of artificial setting, and in fact, during the training process, the computer model does not have the concept of "the liquid flow value at the current time point should be increased or should be decreased", which is only two kinds of classification tags, and the probability that the output feature is under the two kinds of classification tags, that is, the sum of p1 and p2 is one. Therefore, the classification result that the liquid flow value at the current time point should be increased or decreased is actually converted into the classified probability distribution conforming to the natural rule through classifying the tag, and the physical meaning of the natural probability distribution of the tag is essentially used instead of the language text meaning that the liquid flow value at the current time point should be increased or decreased.
It should be appreciated that the role of the classifier is to learn the classification rules and classifier using a given class, known training data, and then classify (or predict) the unknown data. Logistic regression (logistics), SVM, etc. are commonly used to solve the classification problem, and for multi-classification problems (multi-class classification), logistic regression or SVM can be used as well, but multiple bi-classifications are required to compose multiple classifications, but this is error-prone and inefficient, and the commonly used multi-classification method is the Softmax classification function.
Further, the ultrasonic osteotome control system with the superposition of the low-frequency pulse and the high-frequency vibration further comprises a training module for training the time sequence feature extractor based on the one-dimensional convolution layer, the time attention module and the classifier. The training module plays an important role in the ultrasonic osteotome control system, and is used for training a time sequence feature extractor, a time attention module and a classifier of a one-dimensional convolution layer in the system so that the training module can effectively learn and extract relevant features in input data and accurately classify or predict the data. Specifically, the main functions of the training module include: 1. parameter optimization: the training module adjusts and optimizes parameters in the model through an optimization algorithm (such as gradient descent) to minimize a loss function of the model on training data, and enables the model to gradually learn better feature representation and classification capability through continuous iteration and parameter updating. 2. And (3) feature learning: the training module compares the predicted result of the model with the real label through a back propagation algorithm, and then transmits an error signal back to each component of the model, including a time sequence feature extractor and a time attention module of a one-dimensional convolution layer, so that the components can learn more distinguishable and discriminant feature representations to better distinguish different types of data. 3. Model evaluation: the training module can also evaluate and monitor the model in the training process, and evaluate the performance of the model on unseen data by using a verification set or a cross-verification method, which is helpful for judging whether the model is over-fitted or under-fitted, and correspondingly adjusting and improving. Through the training process of the training module, the ultrasonic bone knife control system can learn a characteristic extraction and classification model suitable for the task according to ultrasonic bone knife signals overlapped by low-frequency pulses and high-frequency vibrations in training data. Therefore, the system can accurately classify and control the ultrasonic bone knife signals according to the input ultrasonic bone knife signals, and high-precision and reliable ultrasonic bone knife operation is realized.
Wherein, more specifically, the training module comprises: a training data acquisition unit for acquiring training data including training operation area temperature values at a plurality of predetermined time points within a predetermined period of time acquired by the temperature sensor, and a true value at which a liquid flow value at a current time point should be increased or decreased; the training input vector arrangement unit is used for arranging the training operation area temperature values of the plurality of preset time points into training operation area temperature time sequence input vectors according to the time dimension; the training up-sampling unit is used for up-sampling the temperature time sequence input vector of the training operation area based on linear interpolation to obtain the temperature time sequence input vector of the training up-sampling operation area; the training vector segmentation unit is used for carrying out vector segmentation on the training up-sampling operation area temperature time sequence input vector so as to obtain a sequence of training operation area temperature local time sequence input vector; the training one-dimensional convolution coding unit is used for enabling the sequence of the training operation area temperature local time sequence input vector to pass through the time sequence feature extractor based on the one-dimensional convolution layer respectively so as to obtain the sequence of the training operation area temperature local time sequence feature vector; the training time attention unit is used for passing the sequence of the training operation area temperature local time sequence feature vectors through the time attention module to obtain training time dimension reinforced operation area temperature time sequence associated feature vectors; the classification loss function value calculation unit is used for enabling the training time dimension reinforced operation area temperature time sequence associated feature vector to pass through a classifier to obtain a classification loss function value; and an iteration unit, configured to train the one-dimensional convolutional layer-based time sequence feature extractor, the time attention module and the classifier with the classification loss function value, where, in each iteration of the training, an external boundary constraint iteration based on a reference annotation is performed on a weight matrix of the classifier.
In the technical scheme of the disclosure, each training operation area temperature local time sequence feature vector in the sequence of training operation area temperature local time sequence feature vectors can express time sequence local associated features of training operation area temperature values under a segmented time domain, therefore, after passing through a time attention module, time sequence associated feature distribution under certain local time domains can be further enhanced, so that feature expression significance under the local time domains is enhanced, the expression effect of the training time dimension enhanced operation area temperature time sequence associated feature vectors is improved, but at the same time, the strengthening of the time sequence associated feature distribution under the local time domains also enables the overall feature distribution of the training time dimension enhanced operation area temperature sequence associated feature vectors to deviate from the global time domain source domain feature distribution of the training operation area temperature values, so that feature source domain deviation of class probability mapping of the training time dimension enhanced operation area temperature time sequence associated feature vectors is caused in a weight matrix iteration process of a classifier under a classification scene, and further weight matrix is based on source domain feature fitting divergence of the training time dimension enhanced operation area temperature associated feature vectors, so that the training effect of the training time dimension enhanced operation area temperature associated feature vectors is influenced, and the training time dimension enhanced operation area temperature associated feature vector is well obtained. Based on the above, the applicant of the present disclosure performs external boundary constraint of the weight matrix based on the reference annotation in the training process of the training time dimension reinforced operation region temperature time sequence associated feature vector through the classifier.
Accordingly, in one specific example, in each iteration of the training, performing an external boundary constraint iteration based on a reference annotation on the weight matrix of the classifier includes: performing external boundary constraint iteration based on reference annotation on the weight matrix of the classifier according to the following iteration formula; wherein, the iterative formula is:
wherein,and->The weight matrix of the last iteration and the current iteration are respectively adopted, wherein, during the first iteration, different initialization strategies are adopted to set +.>And->(e.g.)>Set as a unitary matrix->Set as the diagonal matrix of the mean value of the feature vector to be classified),>is the training time dimension reinforced operation area temperature time sequence related characteristic vector, and +.>In the form of column vectors>Representing matrix multiplication +.>Representing matrix addition, ++>Is a transpose operation->Is the weight matrix of the classifier after iteration.
Here, by reinforcing the surgical field temperature timing related feature vector in the training time dimensionThe iterative association representation in the weight space is used as the external association boundary constraint of the weight matrix iteration, so that the time sequence association characteristic vector of the temperature of the operation area is strengthened by the training time dimension in the weight space iteration process under the condition that the previous weight matrix is used as the reference annotation (benchmark annotation) in the iteration process>Is used as an anchor point, thereby carrying out the directional mismatching (oriented mismatch) of the weight matrix in the iterative process relative to the training time dimension to strengthen the time sequence associated characteristic vector of the operation region temperature>Is used for compensating source domain characteristic offset of the class probability mapping, and further enhancing the weight matrix to strengthen the time sequence associated characteristic vector of the operation area temperature based on the training time dimensionThe source domain features are fitted and aggregated to improve the training effect of the model and the accuracy of the classification result of the time dimension reinforced operation area temperature time sequence associated feature vector obtained by the trained model.
In summary, an ultrasonic osteotome control system based on superposition of low frequency pulses and high frequency vibrations in accordance with embodiments of the present disclosure is illustrated that may utilize intelligent algorithms to collect surgical field temperature values and analyze implicit timing characteristics therefrom to automatically determine a control strategy for fluid flow.
As described above, the ultrasonic bone knife control system of the low-frequency pulse and high-frequency vibration superposition according to the embodiment of the present disclosure may be implemented in various terminal devices, for example, a server or the like having an ultrasonic bone knife control algorithm of the low-frequency pulse and high-frequency vibration superposition. In one example, the ultrasonic bone-knife control system with low frequency pulses superimposed with high frequency vibrations may be integrated into the terminal device as a software module and/or hardware module. For example, the ultrasonic bone knife control system with the superposition of low frequency pulses and high frequency vibrations may be a software module in the operating system of the terminal device or may be an application developed for the terminal device; of course, the ultrasonic bone knife control system with the superposition of the low-frequency pulse and the high-frequency vibration can be one of a plurality of hardware modules of the terminal equipment.
Alternatively, in another example, the low frequency pulse and high frequency vibration superimposed ultrasonic bone knife control system and the terminal device may be separate devices, and the low frequency pulse and high frequency vibration superimposed ultrasonic bone knife control system may be connected to the terminal device through a wired and/or wireless network and transmit interactive information in accordance with a agreed data format.
Fig. 6 shows a flowchart of an ultrasonic osteotome control method with low frequency pulses superimposed with high frequency vibrations, in accordance with an embodiment of the present disclosure. Fig. 7 shows a schematic diagram of a system architecture of an ultrasonic bone-knife control method of superposition of low frequency pulses and high frequency vibrations in accordance with an embodiment of the present disclosure. As shown in fig. 6 and 7, the ultrasonic osteotome control method according to an embodiment of the present disclosure, in which a low frequency pulse is superimposed with a high frequency vibration, includes: s110, acquiring operation area temperature values of a plurality of preset time points in a preset time period acquired by the temperature sensor; s120, performing time sequence analysis on the operation area temperature values of the plurality of preset time points to obtain time dimension reinforced operation area temperature time sequence association feature vectors; and S130, determining a control strategy of the liquid flow based on the time dimension reinforced operation area temperature time sequence correlation characteristic vector.
In one possible implementation, performing a time-series analysis on the surgical area temperature values at the plurality of predetermined time points to obtain a time-dimension enhanced surgical area temperature time-series correlation feature vector, including: performing data preprocessing on the surgical area temperature values at a plurality of preset time points to obtain a sequence of surgical area temperature local time sequence input vectors; and extracting time sequence characteristics of the sequence of the local time sequence input vectors of the temperature of the operation area to obtain the time dimension reinforced operation area temperature time sequence related characteristic vector.
Here, it will be understood by those skilled in the art that the specific operations of the respective steps in the above-described ultrasonic bone knife control method of superimposing the low-frequency pulse and the high-frequency vibration have been described in detail in the above description of the ultrasonic bone knife control system of superimposing the low-frequency pulse and the high-frequency vibration with reference to fig. 1 to 5, and thus, repetitive descriptions thereof will be omitted.
Fig. 8 illustrates an application scenario diagram of an ultrasonic bone-knife control system with low frequency pulses superimposed with high frequency vibrations according to an embodiment of the present disclosure. As shown in fig. 8, in this application scenario, first, surgical area temperature values at a plurality of predetermined time points within a predetermined period of time acquired by the temperature sensor (for example, D illustrated in fig. 8) are acquired, and then, the surgical area temperature values at the plurality of predetermined time points are input to a server (for example, S illustrated in fig. 8) where an ultrasonic bone knife control algorithm in which a low-frequency pulse and a high-frequency vibration are superimposed is deployed, wherein the server is capable of processing the surgical area temperature values at the plurality of predetermined time points using the ultrasonic bone knife control algorithm in which the low-frequency pulse and the high-frequency vibration are superimposed to obtain a classification result indicating that the liquid flow value at the current time point should be increased or decreased.
Further, as shown in connection with fig. 9, in another embodiment of the present disclosure, an ultrasonic osteotome system is comprised of a generation controller, a control switch, a transducer driver, and a working head, wherein the controller includes a control low frequency pulse, a dither controller, a fluid flow controller, and the like. The working principle is that the current bottom pulse gear, the energy level (the example is set to 5 gear levels) during high-frequency resonance and the adjustment of the liquid flow can be regulated and displayed through the display controller. When the ultrasonic surgical instrument works, the generator converts electric energy into signals, the signals are input to the transduction driver to resonate, ultrasonic waves generated by the transduction driver are transmitted from the near end to the far end by the working tool bit to cause resonance, the tip of the working tool bit generates certain amplitude displacement under the resonance frequency, and when the working tool bit is contacted with bone tissue, the bone tissue is cut off due to cavitation effect generated by high-frequency vibration, and meanwhile, the soft tissue cannot be damaged due to the resonance frequency.
The high-frequency vibration refers to vibration of the ultrasonic bone knife with the resonance frequency in the range of 20KHz-50 KHz. The low frequency pulse refers to the resonance of the osteotome occurring in the form of intermittent pulses. The time period is one pulse period before resonance occurs next time plus stopping. The pulse duty cycle refers to the ratio of the time at which resonance occurs to the time of one cycle in one pulse period. The pulse period and the pulse duty cycle are both adjustable. As shown in table 1 below:
the high-frequency vibration can cut bone tissues, and meanwhile, soft tissues such as vascular nerves and the like cannot be damaged at the frequency according to the high-frequency vibration characteristic; the low-frequency pulse can improve the speed of cutting bone tissues, and hard bone tissues are encountered, so that carbonization is not easy to occur.
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The foregoing description of the embodiments of the present disclosure has been presented for purposes of illustration and description, and is not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the various embodiments described. The terminology used herein was chosen in order to best explain the principles of the embodiments, the practical application, or the improvement of technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
Claims (8)
1. An ultrasonic osteotome control system with superimposed low-frequency pulses and high-frequency vibrations, which comprises a CPU control module, a foot switch module, a manual switch module, a power conversion module and a touch display module which are communicably connected with the CPU control module, an ultrasonic high-frequency driving module, a low-frequency controller and a liquid flow module which are connected with the CPU control module, a transducer and a cutter bar which is drivably connected with the transducer, and a working cutter head which is arranged on the cutter bar, wherein the ultrasonic osteotome control system comprises:
a temperature sensor mounted to the surgical field;
a liquid ejection port facing the working bit; and
the liquid flow module is used for controlling a control strategy for generating liquid flow.
2. The ultrasonic bone knife control system of claim 1, wherein the liquid flow module comprises:
a temperature value acquisition unit for acquiring operation region temperature values at a plurality of predetermined time points within a predetermined period of time acquired by the temperature sensor;
the time sequence analysis unit is used for performing time sequence analysis on the operation area temperature values of the plurality of preset time points to obtain time dimension reinforced operation area temperature time sequence association characteristic vectors; and
and the liquid flow control unit is used for determining a control strategy of the liquid flow based on the time dimension reinforced operation area temperature time sequence correlation characteristic vector.
3. The ultrasonic bone knife control system of claim 2, wherein the timing analysis unit comprises:
the data preprocessing subunit is used for preprocessing the data of the operation area temperature values at a plurality of preset time points to obtain a sequence of operation area temperature local time sequence input vectors; and
and the time sequence feature extraction subunit is used for extracting time sequence features of the sequence of the local time sequence input vectors of the temperature of the operation area so as to obtain the time dimension reinforced operation area temperature time sequence associated feature vectors.
4. The ultrasonic bone knife control system of claim 3, wherein the data preprocessing subunit comprises:
an input vector arrangement secondary subunit, configured to arrange the surgical area temperature values at the plurality of predetermined time points into a surgical area temperature time sequence input vector according to a time dimension;
an up-sampling secondary subunit, configured to perform up-sampling based on linear interpolation on the surgical area temperature time sequence input vector to obtain an up-sampling surgical area temperature time sequence input vector; and
and the vector segmentation secondary subunit is used for carrying out vector segmentation on the up-sampling operation area temperature time sequence input vector so as to obtain a sequence of the operation area temperature local time sequence input vector.
5. The ultrasonic bone knife control system of claim 4, wherein the timing feature extraction subunit comprises:
the one-dimensional convolution coding secondary subunit is used for enabling the sequence of the operation region temperature local time sequence input vector to respectively pass through a time sequence feature extractor based on a one-dimensional convolution layer to obtain the sequence of the operation region temperature local time sequence feature vector; and
and the time attention coding secondary subunit is used for enabling the sequence of the local time sequence characteristic vectors of the temperature of the operation area to pass through a time attention module to obtain the time dimension reinforced operation area temperature time sequence related characteristic vectors.
6. The ultrasonic bone knife control system of claim 5, wherein the liquid flow control unit comprises:
the flow value judging subunit is used for enabling the time dimension intensive surgery area temperature time sequence related characteristic vector to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating that the liquid flow value of the current time point should be increased or decreased; and
and the control subunit is used for taking the classification result as a control strategy of the liquid flow.
7. The ultrasonic bone knife control system of claim 6, further comprising a training module for training the one-dimensional convolutional layer based temporal feature extractor, the temporal attention module, and the classifier;
wherein, training module includes:
a training data acquisition unit for acquiring training data including training operation area temperature values at a plurality of predetermined time points within a predetermined period of time acquired by the temperature sensor, and a true value at which a liquid flow value at a current time point should be increased or decreased;
the training input vector arrangement unit is used for arranging the training operation area temperature values of the plurality of preset time points into training operation area temperature time sequence input vectors according to the time dimension;
the training up-sampling unit is used for up-sampling the temperature time sequence input vector of the training operation area based on linear interpolation to obtain the temperature time sequence input vector of the training up-sampling operation area;
the training vector segmentation unit is used for carrying out vector segmentation on the training up-sampling operation area temperature time sequence input vector so as to obtain a sequence of training operation area temperature local time sequence input vector;
the training one-dimensional convolution coding unit is used for enabling the sequence of the training operation area temperature local time sequence input vector to pass through the time sequence feature extractor based on the one-dimensional convolution layer respectively so as to obtain the sequence of the training operation area temperature local time sequence feature vector;
the training time attention unit is used for passing the sequence of the training operation area temperature local time sequence feature vectors through the time attention module to obtain training time dimension reinforced operation area temperature time sequence associated feature vectors;
the classification loss function value calculation unit is used for enabling the training time dimension reinforced operation area temperature time sequence associated feature vector to pass through a classifier to obtain a classification loss function value; and
and the iteration unit is used for training the time sequence feature extractor based on the one-dimensional convolution layer, the time attention module and the classifier by the classification loss function value, wherein in each round of iteration of training, the weight matrix of the classifier is subjected to external boundary constraint iteration based on reference annotation.
8. The ultrasonic bone-knife control system of claim 7, wherein, in each iteration of the training, performing an external boundary constraint iteration based on reference annotations on the weight matrix of the classifier, comprises:
performing external boundary constraint iteration based on reference annotation on the weight matrix of the classifier according to the following iteration formula;
wherein, the iterative formula is:
wherein,and->The weight matrix of last and current iteration, respectively,/->Is the training time dimension reinforced operation area temperature time sequence related characteristic vector, and +.>In the form of column vectors>Representing matrix multiplication +.>Representing matrix addition, ++>Is a transpose operation->Is the weight matrix of the classifier after iteration.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311137375.8A CN117137585A (en) | 2023-09-05 | 2023-09-05 | Ultrasonic osteotome control system with superposition of low-frequency pulse and high-frequency vibration |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311137375.8A CN117137585A (en) | 2023-09-05 | 2023-09-05 | Ultrasonic osteotome control system with superposition of low-frequency pulse and high-frequency vibration |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117137585A true CN117137585A (en) | 2023-12-01 |
Family
ID=88886495
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311137375.8A Pending CN117137585A (en) | 2023-09-05 | 2023-09-05 | Ultrasonic osteotome control system with superposition of low-frequency pulse and high-frequency vibration |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117137585A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117574741A (en) * | 2024-01-17 | 2024-02-20 | 西安稀有金属材料研究院有限公司 | Method for adjusting ultrasonic bone knife frequency by secondary structure under finite element analysis |
-
2023
- 2023-09-05 CN CN202311137375.8A patent/CN117137585A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117574741A (en) * | 2024-01-17 | 2024-02-20 | 西安稀有金属材料研究院有限公司 | Method for adjusting ultrasonic bone knife frequency by secondary structure under finite element analysis |
| CN117574741B (en) * | 2024-01-17 | 2024-04-30 | 西安稀有金属材料研究院有限公司 | Method for adjusting ultrasonic bone knife frequency by secondary structure under finite element analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117137585A (en) | Ultrasonic osteotome control system with superposition of low-frequency pulse and high-frequency vibration | |
| US9898825B2 (en) | Segmentation of magnetic resonance imaging data | |
| CN111539467A (en) | GAN network architecture and method for data augmentation of medical image data set based on generation of countermeasure network | |
| EP3665703A1 (en) | Computer-aided diagnostics using deep neural networks | |
| US10664972B2 (en) | Systems and methods of dynamic image segmentation | |
| Cammilleri et al. | POD-spectral decomposition for fluid flow analysis and model reduction | |
| CN111402216B (en) | Three-dimensional broken bone segmentation method and device based on deep learning | |
| CN107610165A (en) | The 3 D shearing multi-modal medical image sequence fusion methods of wave zone based on multiple features | |
| Vavekanand | A Deep Learning Approach for Medical Image Segmentation Integrating Magnetic Resonance Imaging to Enhance Brain Tumor Recognition | |
| CN108986909A (en) | Soft tissue elasticity and viscoelastic characterizations' method and device based on ultrasonic elastograph imaging | |
| Dinh et al. | Facial soft tissue thicknesses prediction using anthropometric distances | |
| CN111914479A (en) | Grouting power mixed interval prediction method for dam bedrock | |
| CN117074529A (en) | Highly integrated cavitation and temperature monitoring method and system | |
| CN113951988B (en) | Grinding method, device and system for ultrasonic bone knife | |
| Aalto et al. | Algorithmic surface extraction from MRI data-modelling the human vocal tract | |
| Martínez-Montes et al. | Penalized PARAFAC analysis of spontaneous EEG recordings | |
| CN103099642A (en) | Ultrasonic blood flow signal quality real-time analysis method | |
| CN115081476A (en) | Method and device for bone recognition for bone tissue grinding | |
| KR20220108677A (en) | Arrhythmia classification method using densenet based on convolution neural network | |
| CN117297769A (en) | Bone layer identification method in hard bone tissue operation | |
| CN116585007A (en) | Laminectomy safety control method, system and equipment based on multimode information fusion | |
| CN109223124A (en) | One kind puncturing dual boot control system based on computer orthopaedics | |
| US20240041391A1 (en) | Method and Apparatus for Orthopedic Procedure State Indication | |
| CN109583323A (en) | Subway Vibration signal recognition method based on door control cycling element | |
| Tafsast et al. | Enhanced Ultrasound Classification of Microemboli Using Convolutional Neural Network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |