CN119607399B - Radio frequency gold microneedle control system based on subcutaneous impedance identification - Google Patents
Radio frequency gold microneedle control system based on subcutaneous impedance identificationInfo
- Publication number
- CN119607399B CN119607399B CN202510092389.5A CN202510092389A CN119607399B CN 119607399 B CN119607399 B CN 119607399B CN 202510092389 A CN202510092389 A CN 202510092389A CN 119607399 B CN119607399 B CN 119607399B
- Authority
- CN
- China
- Prior art keywords
- impedance
- radio frequency
- microneedle
- impedance distribution
- gold
- 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.)
- Active
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61M—DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY; DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY; DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR
- A61M37/00—Other apparatus for introducing media into the body; Percutany, i.e. introducing medicines into the body by diffusion through the skin
- A61M37/0015—Other apparatus for introducing media into the body; Percutany, i.e. introducing medicines into the body by diffusion through the skin by using microneedles
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/05—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves
- A61B5/053—Measuring electrical impedance or conductance of a portion of the body
- A61B5/0536—Impedance imaging, e.g. by tomography
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4836—Diagnosis combined with treatment in closed-loop systems or methods
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7271—Specific aspects of physiological measurement analysis
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61N—ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
- A61N1/00—Electrotherapy; Circuits therefor
- A61N1/40—Applying electric fields by inductive or capacitive coupling ; Applying radio-frequency signals
- A61N1/403—Applying electric fields by inductive or capacitive coupling ; Applying radio-frequency signals for thermotherapy, e.g. hyperthermia
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61M—DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY; DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY; DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR
- A61M37/00—Other apparatus for introducing media into the body; Percutany, i.e. introducing medicines into the body by diffusion through the skin
- A61M37/0015—Other apparatus for introducing media into the body; Percutany, i.e. introducing medicines into the body by diffusion through the skin by using microneedles
- A61M2037/0061—Methods for using microneedles
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Veterinary Medicine (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Heart & Thoracic Surgery (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Surgery (AREA)
- Artificial Intelligence (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physiology (AREA)
- Psychiatry (AREA)
- Signal Processing (AREA)
- Evolutionary Computation (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Dermatology (AREA)
- Anesthesiology (AREA)
- Hematology (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
本发明涉及医疗器械技术领域，提供一种基于皮下阻抗识别的射频黄金微针控制系统，射频黄金微针设有用于阻抗断层成像的电极阵列，包括：阻抗分布重构模块将电极阵列采集的皮下组织的电压信号通过基于遗传算法的阻抗分布重构模型生成重构阻抗分布；阻抗成像模块将重构阻抗分布通过基于UNet架构的阻抗率分割模型生成用于反映皮下组织阻抗分布的阻抗分布图像；深度控制模块根据阻抗分布图像确定射频黄金微针的推荐深度；射频能量控制模块根据多个阻抗分布图像调整控制所述射频黄金微针的射频能量的控制算法的控制参数。本发明实现了对于阻抗分布的准确分割判断和更优成像效果，进而实现了使阻抗成像应用于对射频黄金微针的深度和能量脉冲控制。
The present invention relates to the field of medical device technology, and provides a radiofrequency gold microneedle control system based on subcutaneous impedance identification. The radiofrequency gold microneedle is equipped with an electrode array for impedance tomography imaging. The control system comprises: an impedance distribution reconstruction module that generates a reconstructed impedance distribution by applying a genetic algorithm-based impedance distribution reconstruction model to the voltage signal of the subcutaneous tissue collected by the electrode array; an impedance imaging module that generates an impedance distribution image reflecting the impedance distribution of the subcutaneous tissue by applying an impedance ratio segmentation model based on a UNet architecture to the reconstructed impedance distribution; a depth control module that determines the recommended depth of the radiofrequency gold microneedle based on the impedance distribution image; and a radiofrequency energy control module that adjusts the control parameters of the control algorithm for controlling the radiofrequency energy of the radiofrequency gold microneedle based on multiple impedance distribution images. The present invention achieves accurate segmentation and judgment of the impedance distribution and better imaging effects, thereby enabling the application of impedance imaging to control the depth and energy pulses of the radiofrequency gold microneedle.
Description
Technical Field
The invention relates to the technical field of medical instruments, in particular to a radio frequency gold microneedle control system based on subcutaneous impedance identification.
Background
The radio frequency gold microneedle, also called gold microneedle radio frequency technology, combines microneedle and radio frequency technology. After the radio frequency Huang Jinwei needle penetrates into a specific layer of the epidermis of the skin, the radio frequency energy is emitted, so that the radio frequency energy reaches the dermis layer below the epidermis, and the energy is directly and completely injected into the dermis layer. When the tissue of the dermis is subjected to radio frequency energy, the charged particles can severely vibrate and rub under the action of electromagnetic waves to generate heat so as to stimulate the contraction and the regeneration of collagen, promote the effective denaturation, recombination and proliferation of protein, realize the younger and tightening of facial skin, improve the fish tail, head-up line, stature line, neck line and the like.
Impedance imaging (ELECTRICAL IMPEDANCE impedance, EIT) techniques are to place a plurality of electrodes in a set shape on the face surface, apply appropriate safe excitation currents to the electrodes, collect the voltages of the electrodes, and reconstruct an image of the impedance distribution inside the biological tissue by image reconstruction algorithms from the voltage and current data.
Currently, the impedance of the subcutaneous tissue of the dermis is a critical factor to be considered in rf gold microneedle therapy. The magnitude of the impedance is affected by a number of factors including the moisture content of the tissue, ion concentration, temperature, frequency of the radio frequency, and the like. In rf gold microneedle therapy, the depth of microneedle insertion directly affects the distribution of rf energy in the subcutaneous tissue and the heating effect. With the increase of the insertion depth of the micro-needle, the tissue layer through which the radio frequency energy needs to pass is thickened, the electrical impedance is correspondingly increased, the transmission efficiency of the radio frequency energy is reduced, and the heating effect is weakened, however, the radio frequency energy can act on subcutaneous tissue more deeply, and the stimulation of the contraction and the regeneration of collagen in deeper layers is more beneficial to improving the problems of skin relaxation, wrinkles and the like caused by deep tissues, so that the treatment effect is improved. Therefore, during treatment of the RF gold microneedle, it is necessary to determine the proper depth and depth of the RF gold microneedle according to the impedance of subcutaneous tissue
Therefore, during the treatment process, the doctor needs to precisely adjust the insertion depth of the microneedle according to the skin condition and the requirement of the patient, so as to ensure that the rf energy can precisely act on the target tissue layer, and simultaneously avoid damaging surrounding tissues.
Therefore, how to improve impedance imaging for application to depth and energy pulse control of rf gold microneedles based on subcutaneous impedance identification is a technical problem that is currently being solved.
Disclosure of Invention
Therefore, the invention provides the radio frequency gold microneedle control system based on subcutaneous impedance identification, which introduces priori knowledge of a sample set through the global searching capability of a genetic algorithm, radio frequency energy and a UNet architecture, realizes accurate segmentation judgment on impedance distribution and better imaging effect, and further realizes that impedance imaging is applied to depth and energy pulse control of the radio frequency gold microneedle based on subcutaneous impedance identification.
In order to achieve the above object, the present invention provides a rf gold microneedle control system based on subcutaneous impedance recognition, on which an electrode array for impedance tomography is disposed, comprising:
the impedance distribution reconstruction module is used for generating a reconstructed impedance distribution from the voltage signals of subcutaneous tissues acquired by the electrode array through an impedance distribution reconstruction model based on a genetic algorithm;
the impedance imaging module is used for generating an impedance distribution image for reflecting subcutaneous tissue impedance distribution through an impedance rate segmentation model based on a UNet architecture by the reconstructed impedance distribution;
The depth control module is used for determining the recommended depth of the radio frequency Huang Jinwei needle according to the impedance distribution image;
the radio frequency energy control module is used for adjusting control parameters of a control algorithm according to a plurality of impedance distribution images of a set time interval, and the control algorithm is used for controlling the radio frequency energy of the radio frequency gold microneedle.
Further, the impedance distribution reconstruction module comprises a coding unit, an initial population generation unit and a genetic operation unit;
the coding unit is used for carrying out chromosome coding on the mapping impedance of the voltage signals according to the excitation position sequence to generate a chromosome sequence;
the initial population generation unit is used for generating an impedance distribution initial population by adopting an initialization strategy for the chromosome sequence;
The genetic operation unit is used for carrying out genetic operation on the impedance distribution initial population through the impedance distribution reconstruction model to generate the reconstructed impedance distribution.
Further, the voltage signals comprise an excitation voltage signal, a ground voltage signal and an average voltage signal, the chromosome sequence comprises a first layer voltage signal sequence and a second layer position sequence, and the coding unit comprises a voltage acquisition coding subunit, a voltage position coding subunit and a chromosome sequence generating subunit;
The voltage acquisition subunit is used for controlling each electrode in the electrode array to perform adjacent excitation and constructing a voltage sequence of the excitation voltage signal, the grounding voltage signal and the average voltage signal, which are generated by corresponding to all adjacent excitation;
The sequence generation subunit is used for generating and constructing a corresponding first layer impedance sequence according to the type of the voltage signal, and constructing a corresponding second layer position sequence according to the voltage sequence;
the chromosome sequence generation subunit is configured to generate the chromosome sequence according to the first layer impedance sequence and the second layer position sequence.
Further, the impedance distribution reconstruction module further includes a fitness setting unit;
The fitness setting unit is connected with the initial population generation unit and used for constructing a fitness function of the impedance distribution reconstruction model according to the cosine similarity of the voltage sequence and the chromosome sequence and the difference value of the Manhattan norm.
In the scheme, the voltage signals acquired by the electrode array are coded, the initial population is generated, the fitness evaluation and the genetic operation are carried out in a set mode through the genetic algorithm, so that more accurate impedance distribution is generated through the global searching capability of the genetic algorithm, and further, a better imaging effect is realized.
Further, the impedance imaging module comprises a data expansion unit and a tomographic image generation unit;
The data expansion unit is used for carrying out data expansion on the reconstructed impedance distribution through a fully connected network to generate an expanded impedance distribution;
the tomographic image generation unit is configured to generate the impedance distribution image by dividing the extended impedance distribution by the impedance characteristic division model.
Further, the tomographic image generation unit includes a judgment subunit;
And the judging subunit is used for dividing the impedance characteristics of the output layer of the impedance ratio dividing model through a single-heat encoding algorithm to generate the impedance distribution image.
Further, the impedance ratio segmentation model employs an improved mixing loss function;
the improved mixing loss function is used for calculating the improved mixing loss function by carrying out weighted square sum on the cross entropy loss term and the Dice loss term.
According to the scheme, the impedance distribution is segmented and layered by introducing priori knowledge of a sample set through improving an impedance rate segmentation model based on a UNet architecture, and particularly, the overall optimization capacity of a mixed loss function and a cross entropy loss term and the local overlapping optimization capacity of a position loss term are improved, so that less obvious loss term difference in subcutaneous tissue impedance distribution is amplified, and a better imaging effect is achieved.
Further, the depth control module comprises a depth range determination module and a depth determination module;
The depth range determining module is used for determining a safety depth range matched with the transmission efficiency of the set radio frequency energy according to the impedance distribution image;
the depth determining module is used for calculating the recommended depth in the safety depth range according to the working time length of the set radio frequency energy.
Further, the control algorithm is a dynamic PID control algorithm, the control parameters are a differential coefficient and an integral coefficient, and the radio frequency energy control module comprises an adjusting unit and a control unit;
The adjusting unit is used for adjusting the differential coefficient and the integral coefficient according to a plurality of impedance intermediate values of a plurality of impedance distribution images;
the control unit is used for controlling the radio frequency energy of the radio frequency gold microneedle through the dynamic PID control algorithm.
Further, the control algorithm is a dynamic PID control algorithm, the control parameters are a differential coefficient and an integral coefficient, and the radio frequency energy control module comprises an adjusting unit and a control unit;
The adjusting unit is used for adjusting the differential coefficient and the integral coefficient according to a plurality of impedance intermediate values of a plurality of impedance distribution images;
the control unit is used for controlling the radio frequency energy of the radio frequency gold microneedle through the dynamic PID control algorithm.
Further, the adjustment unit comprises an intermediate value calculation subunit and an adjustment calculation subunit;
The intermediate value calculating subunit is used for calculating the impedance intermediate value according to the impedance median and the median area impedance reflected by the impedance distribution image;
the adjustment calculation operator unit is used for determining adjustment amounts of the differential coefficient and the integral coefficient according to differences of the impedance intermediate values.
According to the scheme, the recommended depth of the radio frequency gold microneedle is calculated in a safe depth range according to the better impedance distribution imaging, and the radio frequency energy of the radio frequency gold microneedle is adjusted more safely in a fine and sensitivity-adjustable mode through a dynamic PID control algorithm.
Compared with the prior art, the invention has the beneficial effects that,
1. The prior knowledge of the sample set is introduced through the global searching capability of the genetic algorithm, the radio frequency energy and the UNet architecture, so that accurate segmentation judgment and better imaging effect on impedance distribution are realized, and further, the application of impedance imaging to the depth and energy pulse control of the radio frequency gold microneedle based on subcutaneous impedance identification is realized.
2. The voltage signals collected by the electrode array are coded, an initial population is generated, fitness evaluation and genetic operation are carried out in a set mode through a genetic algorithm, so that more accurate impedance distribution is generated through the global searching capability of the genetic algorithm, and further, a better imaging effect is achieved.
3. The impedance distribution is segmented and layered by introducing priori knowledge of a sample set through improving an impedance rate segmentation model based on a UNet architecture, and particularly, the overall optimization capacity of a mixed loss function and a cross entropy loss term and the local overlap optimization capacity of a Dice loss term are improved, so that less obvious loss term difference in subcutaneous tissue impedance distribution is amplified, and a better imaging effect is achieved.
4. The method and the device realize the calculation of the recommended depth of the radio frequency gold microneedle in a safe depth range according to the better impedance distribution imaging, and realize the fine and sensitivity-adjustable safer adjustment of the radio frequency energy of the radio frequency gold microneedle through a dynamic PID control algorithm.
Drawings
FIG. 1 is a schematic diagram of a control system for a radio frequency gold microneedle based on subcutaneous impedance identification according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of an impedance distribution image generation flow of a radio frequency gold microneedle control system based on subcutaneous impedance identification according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of an impedance distribution image generation principle of a radio frequency gold microneedle control system based on subcutaneous impedance identification according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of an impedance classification model based on UNet architecture of a radio frequency gold microneedle control system based on subcutaneous impedance recognition according to an embodiment of the present invention.
Detailed Description
The invention will be further described with reference to examples for the purpose of making the objects and advantages of the invention more apparent, it being understood that the specific examples described herein are given by way of illustration only and are not intended to be limiting.
Preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are merely for explaining the technical principles of the present invention, and are not intended to limit the scope of the present invention.
It should be noted that, in the description of the present invention, terms such as "upper," "lower," "left," "right," "inner," "outer," and the like indicate directions or positional relationships based on the directions or positional relationships shown in the drawings, which are merely for convenience of description, and do not indicate or imply that the apparatus or elements must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention.
In addition, it should be noted that, in the description of the present invention, unless explicitly specified and limited otherwise, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, integrally connected, mechanically connected, electrically connected, directly connected, indirectly connected through an intermediate medium, or in communication between two elements. The specific meaning of the above terms in the present invention can be understood by those skilled in the art according to the specific circumstances.
As shown in fig. 1 to 4, the invention provides a radio frequency gold microneedle control system based on subcutaneous impedance identification, which introduces priori knowledge of a sample set through global searching capability of a genetic algorithm, radio frequency energy and UNet architecture, realizes accurate segmentation judgment and better imaging effect on impedance distribution, and further realizes that impedance imaging is applied to depth and energy pulse control of the radio frequency gold microneedle based on subcutaneous impedance identification.
As shown in fig. 1 to 4, the embodiment provides a subcutaneous impedance identification-based radio frequency gold microneedle control system, on which an electrode array for impedance tomography is arranged, comprising an impedance distribution reconstruction module, an impedance imaging module, a subcutaneous tissue analysis module and a subcutaneous tissue analysis module, wherein the impedance distribution reconstruction module is used for generating a reconstructed impedance distribution of a subcutaneous tissue voltage signal acquired by the electrode array through an impedance distribution reconstruction model based on a genetic algorithm;
The device comprises a radio frequency Huang Jinwei needle, a depth control module and a radio frequency energy control module, wherein the radio frequency Huang Jinwei needle is used for detecting the impedance distribution image of the gold microneedle, the depth control module is used for determining the recommended depth of the radio frequency Huang Jinwei needle according to the impedance distribution image, and the radio frequency energy control module is used for adjusting control parameters of a control algorithm according to a plurality of impedance distribution images with set time intervals, and the control algorithm is used for controlling the radio frequency energy of the radio frequency gold microneedle.
It should be noted that, referring to fig. 3, the array in the circle in the figure represents the rf gold microneedle, the array on the circle represents the electrode array, the electrode array is preferably disposed around the rf gold microneedle, the electrode array is preferably disposed in a circle shape, and an electromagnetic shielding cover is preferably disposed between the electrode array and the rf gold microneedle to avoid electromagnetic interference of interaction.
Further, as shown in fig. 2, the impedance distribution reconstruction module includes a coding unit, an initial population generation unit and a genetic operation unit, where the coding unit is configured to perform chromosome coding on the mapped impedances of the plurality of voltage signals according to their excitation position sequences to generate a chromosome sequence, the initial population generation unit is configured to generate an impedance distribution initial population for the chromosome sequence by adopting an initialization strategy, and the genetic operation unit is configured to perform genetic operation on the impedance distribution initial population through the impedance distribution reconstruction model to generate the reconstructed impedance distribution.
It will be appreciated that the electrode array field of impedance imaging (ELECTRICAL IMPEDANCE impedance) satisfies maxwell's equations and electromagnetic field theory, and that image reconstruction based on impedance imaging can be seen as a study of the mapping relationship between the relative conductivity (i.e., impedance) distribution and the voltage signal distribution within the field.
Based on this, an efficient initialization strategy is used for generating a high quality initial population of impedance distributions for the chromosome sequences, in particular by using an operational order of randomly arranging the first layer impedance sequences and the second layer impedance sequences by a random number generator. Therefore, the diversity of the initial population can be widened, and the generation of the high-quality and various initial population can ensure that the algorithm can quickly and effectively converge to the optimal solution, so that more real reconstruction impedance distribution is calculated through the genetic algorithm.
Further, as shown in fig. 2, the voltage signals include an excitation voltage signal, a ground voltage signal and an average voltage signal, the chromosome sequence includes a first layer voltage signal sequence and a second layer position sequence, and the coding unit includes a voltage acquisition coding subunit, a voltage position coding subunit and a chromosome sequence generating subunit;
The voltage acquisition subunit is used for controlling each electrode in the electrode array to perform adjacent excitation and constructing a voltage sequence of the excitation voltage signal, the grounding voltage signal and the average voltage signal, which are generated by corresponding to all adjacent excitation;
The sequence generation subunit is used for generating and constructing a corresponding first layer impedance sequence according to the type of the voltage signal, and constructing a corresponding second layer position sequence according to the voltage sequence;
the chromosome sequence generation subunit is configured to generate the chromosome sequence according to the first layer impedance sequence and the second layer position sequence.
The adjacent excitation process comprises the steps of firstly, definitely exciting and measuring, namely, 10 electrodes are marked as 1 to 10, after the ground voltage signals comprising all the electrodes are acquired, selecting two adjacent electrodes as excitation electrode pairs each time, acquiring 1 excitation voltage signal of the excitation electrode pairs and 8 average voltage signals of other electrodes, and further enabling the 10 electrodes to sequentially serve as excitation electrodes, so that the electrodes at one position can acquire 1 ground voltage signal, 1 excitation voltage signal and 8 average voltage signals for 10 voltage signals in total, and acquiring a voltage sequence of 100 voltage signals.
Specifically, the first layer impedance sequence and the second layer impedance sequence are respectively located in a two-layer vector model, the first layer impedance sequence is a plurality of impedance regions which divide the region in the circle as shown in fig. 3 into equal parts according to the circular arc of a set angle, and the second layer impedance sequence is the same order of impedance regions as the adjacent excitation order represented by the voltage sequence, so that the chromosome sequence which accords with the format and is possible to be solved is generated through encoding.
Further, as shown in fig. 2, the fitness setting unit is connected to the initial population generating unit, and is configured to construct a fitness function of the impedance distribution reconstruction model according to a difference between the cosine similarity and the manhattan norm of the voltage sequence and the chromosome sequence.
Specifically, the fitness function is expressed as:
in the formula, f represents a fitness function, P represents a chromosome sequence, V represents a voltage sequence, I represents cosine similarity of the sequence and maps to a non-negative interval, I represents Manhattan norm, and flattening the Manhattan norm can prevent the difference value from falling into the negative interval while amplifying the difference value.
Therefore, the fitness function combines the cosine similarity and the Manhattan norm, wherein the cosine similarity focuses on the direction of the vector, and the Manhattan norm measures the sum of absolute differences of the vectors, so that the genetic algorithm can find a solution similar to the direction of the target vector and simultaneously keep a certain difference with the target vector in value.
Specifically, the Pareto-based genetic algorithm framework.
In the scheme, the voltage signals acquired by the electrode array are coded, the initial population is generated, the fitness evaluation and the genetic operation are carried out in a set mode through the genetic algorithm, so that more accurate impedance distribution is generated through the global searching capability of the genetic algorithm, and further, a better imaging effect is realized.
Further, as shown in fig. 4, the impedance imaging module includes a data expansion unit and a tomographic image generation unit, wherein the data expansion unit is configured to perform data expansion on the reconstructed impedance distribution through a fully connected network to generate an expanded impedance distribution, and the tomographic image generation unit is configured to perform impedance feature segmentation on the expanded impedance distribution through the impedance rate segmentation model to generate the impedance distribution image.
Specifically, as shown in fig. 4, the nodes of the input layer, the hidden layer and the output layer of the fully connected network (Fully Connected Network, FCN) are respectively 5, 6 and 8, the input layer is used as a feature extractor, useful features are extracted from the input reconstructed impedance distribution, and the useful features are mapped to a new feature space through the nonlinear transformation capability of the hidden layer and the output layer so as to generate a new data representation, thereby realizing data expansion.
Further, as shown in fig. 4, the tomographic image generation unit includes a determination subunit, and the determination subunit is configured to generate the impedance distribution image by dividing the impedance feature of the output layer of the impedance ratio division model by a single thermal encoding algorithm.
Specifically, as shown in fig. 4, the impedance feature is accurately segmented and graded by a single thermal coding algorithm, wherein the impedance grade is determined to be one of extremely low, medium, high and extremely high, and vector representation is performed by the single thermal coding algorithm, and the vector representation is used for determining pixel values to generate an impedance distribution image.
Thus, the impedance distribution image enables visualization of impedance data.
Further, the impedance rate segmentation model employs a modified hybrid loss function to calculate the modified hybrid loss function by weighted sum of squares of cross entropy loss terms and Dice loss terms.
It will be appreciated that the improved mixing loss function may also be replaced by a conventional loss function which is more direct and insensitive to the differential representation of the predicted and true distribution and thus less optimal, expressed as:
LC＝α·LCE+β·LDice
Where L C denotes the modified mixing loss function, α, β denote the two weighting coefficients, L CE denotes the cross entropy loss term, and L Dice denotes the Dice loss term.
Specifically, the improved mixing loss function is expressed as:
where L C denotes the modified mixing loss function, α, β denote two weighting coefficients, preferably 0.6 and 0.4 to make the difference of the predicted distribution from the true distribution more pronounced, L CE denotes the cross entropy loss term, and L Dice denotes the Dice loss term.
The cross entropy loss term is expressed as:
Where L CE denotes the cross entropy loss term, N denotes the total number of pixels, C denotes the number of classes, y i,c denotes the true label of the ith pixel, the C-th class, A predictive label representing the ith pixel, the c-th class. The cross entropy loss term can measure the classification difference between the predicted distribution and the true distribution.
The expression of the Dice loss term is:
Where L Dice denotes the Dice loss term, N denotes the total number of pixels, y i denotes the true label of the ith pixel, A prediction label representing the i-th pixel,Representing the predictive label of the im+1st pixel, ε is a smooth term that prevents the denominator from taking zero. Thus, the difference loss term is used to measure the degree of overlap of the predicted distribution with the true distribution, and the closer the difference loss term is to 1, the closer the predicted result is to the true value.
According to the scheme, the impedance distribution is segmented and layered by introducing priori knowledge of a sample set through improving an impedance rate segmentation model based on a UNet architecture, and particularly, the overall optimization capacity of a mixed loss function and a cross entropy loss term and the local overlapping optimization capacity of a position loss term are improved, so that less obvious loss term difference in subcutaneous tissue impedance distribution is amplified, and a better imaging effect is achieved.
Further, as shown in fig. 3, the depth control module includes a depth range determination module and a depth determination module;
The depth range determining module is configured to determine, from the impedance distribution image, a safe depth range that matches a set transmission efficiency of radio frequency energy, and may be expressed as:
Hmax＝H+Zmaxη,Zmaxη≤T1
Hmin＝H-Zminη,Zminη≥T2
Wherein, H min、Hmax is the lower limit and the upper limit of the safety depth range, H is the set depth, which is determined by the doctor according to the superficial treatment, the middle treatment or the deep treatment of the patient and the treatment mode, and is between 0.5mm and 5.0mm, Z max、Zmin is the maximum impedance value of the impedance distribution image and the minimum impedance value with the area larger than the set limit, and T 1、T2 is the upper limit safety value and the lower limit safety value, preferably 8 x 10 -3 ohms per square meter and 5 x 10 -5 ohms per square meter.
It will be appreciated that during rf treatment of rf gold microneedles, the impedance of the subcutaneous tissue may change as the treatment depth increases, and identifying the extremum of the impedance may help the physician to understand the distribution of rf energy in the subcutaneous tissue. Avoiding overheating and potential damage to the tissue when the radiofrequency energy encounters high impedance tissue, the energy is concentrated at the tissue layer, resulting in an increase in the temperature of the tissue.
The depth determining module is configured to calculate the recommended depth within the safe depth range according to the working duration of the set radio frequency energy, and may be expressed as:
Hrec＝Hmin+(Hmax-Hmin)(Tw-Trec)
Wherein H rec、Trec represents the recommended depth and the recommended working time respectively, H min、Hmax represents the lower limit and the upper limit of the safety depth range respectively, and T w represents the working time.
Therefore, the scheme can carry out fine adjustment control on the depth of the radio frequency Huang Jinwei needle set by a doctor according to the impedance distribution diagram.
Further, as shown in fig. 3, the control algorithm is a dynamic PID control algorithm, the control parameters are a differential coefficient and an integral coefficient, the rf energy control module includes an adjusting unit and a control unit, the adjusting unit is configured to adjust the differential coefficient and the integral coefficient according to a plurality of impedance intermediate values of a plurality of impedance distribution images, and the control unit is configured to control the rf energy of the rf gold microneedle by using the dynamic PID control algorithm.
It can be appreciated that the radio frequency gold microneedle control with real-time Intelligent Feedback System (IFS) is formed by a dynamic PID control algorithm, and the aim of accurately controlling tissue reaction is achieved by measuring tissue impedance in real time, and by feeding back radio frequency energy including radio frequency and radio frequency power to deliver constant energy to the tissue of the subcutaneous dermis of a patient, maintaining the temperature of the dermis tissue, and making the treatment process independent of individual impedance changes. In addition, as the treatment is carried out, the impedance characteristic of subcutaneous tissue can be changed, and the dynamic PID parameter adjustment can automatically optimize the control parameters according to the changes, so that the reaction process is more stable and efficient.
Further, the adjustment unit comprises an intermediate value calculation subunit and an adjustment calculation subunit;
The intermediate value calculating subunit, configured to calculate the impedance intermediate value according to the impedance median and the median area impedance reflected by the impedance distribution image, may be expressed as:
In the formula, p' med、pmed、pmeds represents an impedance intermediate value, an impedance rate median and a median area impedance respectively, and I represents an absolute value.
The adjustment amount calculating unit is configured to determine an adjustment amount of the differential coefficient and the integral coefficient according to a difference value of the plurality of impedance intermediate values, and may be expressed as:
Wherein K i′、K′d represents the differential coefficient and the integral coefficient after adjustment, K i、Kd represents the differential coefficient and the integral coefficient after adjustment, Δk i、ΔKd represents the adjustment amount of the differential coefficient and the adjustment amount of the integral coefficient, p' medi-p′medi-1 represents the difference between two impedance intermediate values, R is the number of the impedance intermediate values, and 0.1 and 0.125 are adjustment coefficients, respectively, so that the dynamic PID is more sensitive, and the differential coefficient and the integral coefficient are adjusted in proportion.
According to the scheme, the recommended depth of the radio frequency gold microneedle is calculated in a safe depth range according to the better impedance distribution imaging, and the radio frequency energy of the radio frequency gold microneedle is adjusted more safely in a fine and sensitivity-adjustable mode through a dynamic PID control algorithm.
In the embodiment, the prior knowledge of the sample set is introduced through the global searching capability of the genetic algorithm, the radio frequency energy and the UNet architecture, so that accurate segmentation judgment and better imaging effect on impedance distribution are realized, and further, the impedance imaging is applied to depth and energy pulse control of the radio frequency gold microneedle based on subcutaneous impedance identification. The voltage signals collected by the electrode array are coded, an initial population is generated, fitness evaluation and genetic operation are carried out in a set mode through a genetic algorithm, so that more accurate impedance distribution is generated through the global searching capability of the genetic algorithm, and further, a better imaging effect is achieved. The impedance distribution is segmented and layered by introducing priori knowledge of a sample set through improving an impedance rate segmentation model based on a UNet architecture, and particularly, the overall optimization capacity of a mixed loss function and a cross entropy loss term and the local overlap optimization capacity of a Dice loss term are improved, so that less obvious loss term difference in subcutaneous tissue impedance distribution is amplified, and a better imaging effect is achieved. The method and the device realize the calculation of the recommended depth of the radio frequency gold microneedle in a safe depth range according to the better impedance distribution imaging, and realize the fine and sensitivity-adjustable safer adjustment of the radio frequency energy of the radio frequency gold microneedle through a dynamic PID control algorithm.
Thus far, the technical solution of the present invention has been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of protection of the present invention is not limited to these specific embodiments. Equivalent modifications and substitutions for related technical features may be made by those skilled in the art without departing from the principles of the present invention, and such modifications and substitutions will be within the scope of the present invention.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, and various modifications and variations of the present invention will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (7)
1. A radio frequency gold microneedle control system based on subcutaneous impedance identification, characterized in that an electrode array for impedance tomography is arranged on the radio frequency gold microneedle, comprising:
the impedance distribution reconstruction module is used for generating a reconstructed impedance distribution from the voltage signals of subcutaneous tissues acquired by the electrode array through an impedance distribution reconstruction model based on a genetic algorithm;
the impedance imaging module is used for generating an impedance distribution image for reflecting subcutaneous tissue impedance distribution through an impedance rate segmentation model based on a UNet architecture by the reconstructed impedance distribution;
The depth control module is used for determining the recommended depth of the radio frequency Huang Jinwei needle according to the impedance distribution image;
The radio frequency energy control module is used for adjusting control parameters of a control algorithm according to a plurality of impedance distribution images of a set time interval, and the control algorithm is used for controlling the radio frequency energy of the radio frequency gold microneedle;
the impedance distribution reconstruction module comprises a coding unit, an initial population generation unit and a genetic operation unit;
the coding unit is used for carrying out chromosome coding on the mapping impedance of the voltage signals according to the excitation position sequence to generate a chromosome sequence;
the initial population generation unit is used for generating an impedance distribution initial population by adopting an initialization strategy for the chromosome sequence;
The genetic operation unit is used for carrying out genetic operation on the initial population of the impedance distribution to generate the reconstructed impedance distribution;
the voltage signals comprise an excitation voltage signal, a ground voltage signal and an average voltage signal, the chromosome sequence comprises a first layer voltage signal sequence and a second layer position sequence, and the coding unit comprises a voltage acquisition coding subunit, a voltage position coding subunit and a chromosome sequence generating subunit;
The voltage position coding subunit is used for controlling each electrode in the electrode array to perform adjacent excitation and constructing a voltage sequence of the excitation voltage signal, the grounding voltage signal and the average voltage signal, which are generated by corresponding to all adjacent excitation;
The chromosome sequence generation subunit is used for generating and constructing a corresponding first layer impedance sequence according to the type of the voltage signal, constructing a corresponding second layer position sequence according to the voltage sequence, and generating the chromosome sequence according to the first layer impedance sequence and the second layer position sequence;
the impedance distribution reconstruction module further comprises a fitness setting unit;
The fitness setting unit is used for constructing a fitness function of the impedance distribution reconstruction model according to the difference value of the cosine similarity and the Manhattan norm of the voltage sequence and the chromosome sequence.
2. The radio frequency gold microneedle control system based on subcutaneous impedance recognition according to claim 1, wherein the impedance imaging module comprises a data expansion unit and a tomographic image generation unit;
The data expansion unit is used for carrying out data expansion on the reconstructed impedance distribution through a fully connected network to generate an expanded impedance distribution;
the tomographic image generation unit is configured to generate the impedance distribution image by dividing the extended impedance distribution by the impedance characteristic division model.
3. The radio frequency gold microneedle control system based on subcutaneous impedance recognition according to claim 2, wherein the tomographic image generation unit includes a judgment subunit;
And the judging subunit is used for dividing the impedance characteristics of the output layer of the impedance ratio dividing model through a single-heat encoding algorithm to generate the impedance distribution image.
4. The subcutaneous impedance-based radio frequency gold microneedle control system of claim 1, wherein the impedance-rate segmentation model employs an improved mixing loss function;
The improved mixing loss function is determined by weighted sum-of-squares calculation of cross entropy loss terms and Dice loss terms.
5. The rf gold microneedle control system of any one of claims 1-4, wherein the depth control module comprises a depth range determination module and a depth determination module;
The depth range determining module is used for determining a safety depth range matched with the transmission efficiency of the set radio frequency energy according to the impedance distribution image;
The depth determining module is used for calculating and determining the recommended depth in the safety depth range according to the working time length of the set radio frequency energy.
6. The radio frequency gold microneedle control system based on subcutaneous impedance recognition according to claim 5, wherein the control algorithm is a dynamic PID control algorithm, the control parameters are a differential coefficient and an integral coefficient, and the radio frequency energy control module comprises an adjusting unit and a control unit;
The adjusting unit is used for adjusting the differential coefficient and the integral coefficient according to a plurality of impedance intermediate values of a plurality of impedance distribution images;
the control unit is used for controlling the radio frequency energy of the radio frequency gold microneedle through the dynamic PID control algorithm.
7. The radio frequency gold microneedle control system based on subcutaneous impedance recognition according to claim 6, wherein the adjustment unit comprises an intermediate value calculation subunit and an adjustment calculation subunit;
The intermediate value calculating subunit is used for calculating and determining the intermediate value of the impedance according to the median impedance and the median area impedance reflected by the impedance distribution image;
the adjustment calculation operator unit is used for determining adjustment amounts of the differential coefficient and the integral coefficient according to differences of the impedance intermediate values.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510092389.5A CN119607399B (en) | 2025-01-21 | 2025-01-21 | Radio frequency gold microneedle control system based on subcutaneous impedance identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510092389.5A CN119607399B (en) | 2025-01-21 | 2025-01-21 | Radio frequency gold microneedle control system based on subcutaneous impedance identification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN119607399A CN119607399A (en) | 2025-03-14 |
| CN119607399B true CN119607399B (en) | 2025-08-19 |
Family
ID=94908340
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202510092389.5A Active CN119607399B (en) | 2025-01-21 | 2025-01-21 | Radio frequency gold microneedle control system based on subcutaneous impedance identification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119607399B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1965755A (en) * | 2006-11-17 | 2007-05-23 | 清华大学 | Impedance tomography apparatus based on microneedle electrodes and micro-traumatic measuring method thereof |
| CN114288555A (en) * | 2022-01-26 | 2022-04-08 | 云南贝泰妮生物科技集团股份有限公司 | Radio frequency beauty instrument self-adaptive frequency modulation system based on skin impedance |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20210077350A1 (en) * | 2008-08-22 | 2021-03-18 | Viol Co., Ltd. | Method, system, and apparatus for dermatological treatment |
| US8588884B2 (en) * | 2010-05-28 | 2013-11-19 | Emkinetics, Inc. | Microneedle electrode |
| CN111938641B (en) * | 2020-08-14 | 2023-08-08 | 河北工业大学 | Optimized self-adaptive extended Kalman filtering bioelectrical impedance imaging method |
-
2025
- 2025-01-21 CN CN202510092389.5A patent/CN119607399B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1965755A (en) * | 2006-11-17 | 2007-05-23 | 清华大学 | Impedance tomography apparatus based on microneedle electrodes and micro-traumatic measuring method thereof |
| CN114288555A (en) * | 2022-01-26 | 2022-04-08 | 云南贝泰妮生物科技集团股份有限公司 | Radio frequency beauty instrument self-adaptive frequency modulation system based on skin impedance |
Also Published As
| Publication number | Publication date |
|---|---|
| CN119607399A (en) | 2025-03-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11701161B2 (en) | Optimizing treatment using TTFields by changing the frequency during the course of long term tumor treatment | |
| EP3824805B1 (en) | Apparatus for treating a tumor with an alternating electric field and for selecting a treatment frequency based on estimated cell size | |
| CN114343820B (en) | Visual high-voltage high-frequency steep pulse of image melts tumour cell system | |
| US12201831B2 (en) | Impedance tomography using electrodes of a tumor treating fields (TTFields) system | |
| TWI461180B (en) | Method for improving imaging resolution of electrical impedance tomography | |
| CN103027673B (en) | Stimulation target range calibration device | |
| CN111466904B (en) | Placement of electrodes near the heart | |
| CN119607399B (en) | Radio frequency gold microneedle control system based on subcutaneous impedance identification | |
| CN115737111A (en) | Energy control method and system based on wall thickness and impedance and ablation system | |
| JP7480083B2 (en) | Apparatus and method for treating tumors with alternating electric fields and selecting treatment frequencies based on estimated cell size - Patents.com | |
| CN118634422B (en) | Individualized non-invasive deep brain stimulation devices and related products | |
| CN115153829A (en) | Accuracy optimization method of electrical pulse ablation model based on non-uniform reconstruction of active tissue electrical impedance information | |
| CN115054829B (en) | Intelligent plasma brush system, implementation method and device thereof, and storage medium | |
| CN119606425B (en) | High-resolution peripheral nerve imaging system based on acoustic-electric coupling | |
| HK40042528B (en) | Apparatus for treating a tumor with an alternating electric field and for selecting a treatment frequency based on estimated cell size | |
| HK40042528A (en) | Apparatus for treating a tumor with an alternating electric field and for selecting a treatment frequency based on estimated cell size | |
| HK40017598B (en) | Apparatus for treating a tumor with an alternating electric field and for selecting a treatment frequency based on estimated cell size | |
| CN120412911B (en) | Device for improving needle electrode treatment effect | |
| HK40017598A (en) | Apparatus for treating a tumor with an alternating electric field and for selecting a treatment frequency based on estimated cell size | |
| HK1216596B (en) | Apparatus for treating a tumor with an alternating electric field and for selecting a treatment frequency based on estimated cell size | |
| CN120605085A (en) | Electric field application control method and system for multi-lesion dynamic non-invasive non-contact treatment device | |
| HK40095950A (en) | Impedance tomography using electrodes of a tumor treating fields (ttfields) system | |
| CN119367039A (en) | Neurological diagnosis and treatment system, control method of neurological diagnosis and treatment system, and electronic equipment | |
| CN119367040A (en) | Neurological diagnosis and treatment system, control method of neurological diagnosis and treatment system, and electronic equipment |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |