CN111407279A - Magnetoelectricity combined positioning and tracking method and device based on neural network - Google Patents
Magnetoelectricity combined positioning and tracking method and device based on neural network Download PDFInfo
- Publication number
- CN111407279A CN111407279A CN201910012954.7A CN201910012954A CN111407279A CN 111407279 A CN111407279 A CN 111407279A CN 201910012954 A CN201910012954 A CN 201910012954A CN 111407279 A CN111407279 A CN 111407279A
- Authority
- CN
- China
- Prior art keywords
- neural network
- network model
- training
- sample
- tracking method
- 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
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/06—Devices, other than using radiation, for detecting or locating foreign bodies ; Determining position of diagnostic devices within or on the body of the patient
- A61B5/061—Determining position of a probe within the body employing means separate from the probe, e.g. sensing internal probe position employing impedance electrodes on the surface of the body
- A61B5/062—Determining position of a probe within the body employing means separate from the probe, e.g. sensing internal probe position employing impedance electrodes on the surface of the body using magnetic field
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/06—Devices, other than using radiation, for detecting or locating foreign bodies ; Determining position of diagnostic devices within or on the body of the patient
- A61B5/061—Determining position of a probe within the body employing means separate from the probe, e.g. sensing internal probe position employing impedance electrodes on the surface of the body
- A61B5/063—Determining position of a probe within the body employing means separate from the probe, e.g. sensing internal probe position employing impedance electrodes on the surface of the body using impedance measurements
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Animal Behavior & Ethology (AREA)
- Medical Informatics (AREA)
- Heart & Thoracic Surgery (AREA)
- Surgery (AREA)
- Pathology (AREA)
- Human Computer Interaction (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a magnetoelectric combined positioning and tracking method and device based on a neural network, and belongs to the technical field of precise positioning of implants in living tissues. The method comprises the following steps: 1. acquiring data in real time to obtain a training sample and a sample to be detected; 2. carrying out normalization processing on the training sample, establishing a training data set, and carrying out normalization processing on the sample to be detected to obtain a normalized sample to be detected; 3. extracting a pre-stored first neural network model, training to obtain a second neural network model, and inputting a normalized sample to be tested into the first neural network model to obtain a preliminary prediction result; 4. and replacing the pre-stored first neural network model with the second neural network model, and performing inverse normalization processing on the preliminary prediction result to obtain the electrode space coordinate corresponding to the sample to be detected. The method and the device of the invention realize the real-time accurate positioning of the three-dimensional space position of the implant in the living body, and have the characteristics of high positioning accuracy, high calculation efficiency and good real-time performance.
Description
Technical Field
The invention relates to a technology for accurately positioning an implant in a living tissue, in particular to a magnetoelectric combined positioning and tracking method and device based on a neural network.
Background
At present, many medical devices are more or less related to the placement of implants such as sensors and catheters in living tissues, and the three-dimensional space positioning of the implants is realized by matching multi-angle CT collected images, but the multi-angle CT collected images increase the irradiation amount of X-rays to living bodies, and the position of the movable implants is difficult to realize the accurate measurement and display.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provide a magnetoelectric combination positioning and tracking method and device based on a neural network.
In order to achieve the above purpose, the invention provides the following technical scheme:
a magnetoelectricity combined positioning and tracking method based on a neural network comprises the following steps:
s1: collecting data in real time, wherein the data comprises a training sample and a sample to be detected, the training sample comprises an electric field impedance value and a corresponding magnetic field coordinate, and the sample to be detected comprises an electric field impedance value;
s2: carrying out normalization processing on the training sample, establishing a training data set, and carrying out normalization processing on the sample to be detected to obtain a normalized sample to be detected;
s3: extracting a pre-stored first neural network model, inputting a training data set into the first neural network model, training to obtain a second neural network model, and inputting a normalized sample to be tested into the first neural network model to obtain a preliminary prediction result;
s4: and replacing the pre-stored first neural network model with the second neural network model, storing the second neural network model, and performing inverse normalization processing on the preliminary prediction result to obtain the electrode space coordinate corresponding to the sample to be detected.
The training samples are normalized, and the establishment of the training data set means that:
s11: carrying out normalization processing on the training samples, constructing normalized training samples, and storing a normalized parameter matrix A;
s12: storing the normalized training samples into a training data set;
s13: and judging whether the number and the range of the training data set samples reach the set threshold value or not, returning to S11, and completing the establishment of the training data set when the number and the range of the training data set samples reach the set threshold value.
And the normalization parameter matrix adopted in the anti-normalization processing is a normalization parameter matrix A.
The first neural network model and the second neural network model are of three-layer structures and comprise an input layer, a hidden layer and an output layer, the input layer comprises m input units and 1 bias item, the hidden layer comprises N hidden units and 1 bias item, and the output layer comprises N output units.
The first neural network model and the second neural network model have neural network model parameters, the neural network model parameters include input layer to hidden layer parameters and hidden layer to output layer parameters, the input layer to hidden layer parameters are m × N matrices, and the hidden layer to output layer parameters are N × N matrices.
The activation function of the hidden unit is:
wherein z is the hidden unit input and e is the natural constant.
Inputting a training data set into a first neural network model, and training to obtain a second neural network model, which comprises the following specific steps:
s21: inputting the training data set into a first neural network model to obtain a training output result;
s22: calculating a mean square error value of a training output result and an expected target;
s23: comparing the mean square error value with a preset threshold value, and training to obtain a second neural network model when the mean square error value is less than or equal to the preset threshold value; and when the mean square error value is larger than the preset threshold value, returning to the step S1, and continuing the model training.
The mean square error value is calculated as:
wherein E (f; D) is the mean square error value, m is the number of samples in the training data set, f (x) is the neural network function, f (x)i) Is the electric field impedance xiBy means of neural netNumber derived prediction result, yiIs the desired target, i.e. the electric field impedance x acquired in real timeiThe corresponding magnetic field coordinates.
A device for executing a magnetoelectric combination positioning tracking method based on a neural network is characterized in that a control unit controls an excitation distribution device to sequentially distribute excitation according to the sequential circulation of an excitation source V1, an excitation source V2 and an excitation source V3, electric field impedance collectors on a catheter respectively collect electric field impedance between electrode plates of the electric field impedance collectors, the electric field impedance is amplified through an amplifier, the control unit controls a magnetic field generator to be turned on and off, and when the magnetic field generator is turned on, a magnetic field information collector on the catheter collects magnetic field coordinates And storing the second neural network model, calculating the preliminary prediction result, and finishing the inverse normalization processing of the preliminary prediction result.
Compared with the prior art, the invention has the beneficial effects that:
1. the three-dimensional space positioning of the implant is realized without adopting CT collected images, so that the human body is prevented from being damaged by X-ray irradiation.
2. The real-time accurate positioning of the three-dimensional space position of the implant in the living body is realized, and the method has the characteristics of high positioning accuracy, high calculation efficiency and good real-time property.
Description of the drawings:
FIG. 1 is a flow chart of a magneto-electric combined positioning and tracking method based on a neural network;
FIG. 2 is a diagram of an apparatus for a magnetic-electric combined localization tracking method based on neural network;
FIG. 3 is a flowchart illustrating processing and analyzing of the collected data by the processor according to embodiment 1 of the present invention;
fig. 4 is a diagram of a neural network model structure according to embodiment 1 of the present invention.
Detailed Description
The present invention will be described in further detail with reference to test examples and specific embodiments. It should be understood that the scope of the above-described subject matter is not limited to the following examples, and any techniques implemented based on the disclosure of the present invention are within the scope of the present invention.
Example 1
The invention relates to a device based on a neural network and adopting a magnetoelectric combined positioning and tracking method, as shown in figure 2, three electrode plates in a group are respectively pasted on the chest and the back of the upper half of a human body and are distributed in a triangular shape. The controller controls the data acquisition and comprises two modules which run in parallel, wherein firstly, the controller controls the excitation issuing device to circularly and sequentially issue excitation according to the sequence of excitation V1 → excitation V2 → excitation V3, and each electrical impedance sensor on the catheter acquires impedance data between the electrical impedance sensor and an electrode plate A, B, C, D, E, F, and the impedance data is amplified and then input into the operation processor; secondly, the controller controls the magnetic field generator to be opened and closed according to a certain period, and when the magnetic field generator is opened, the magnetic sensor on the catheter is used for collecting magnetic coordinate data and inputting the data into the arithmetic processor.
The arithmetic processor performs a series of processing and analysis on the acquired data, and the flow chart is shown in fig. 3. The method comprises the following specific steps:
s1: the method comprises the steps of collecting magnetic field coordinates and electric field impedance of the same position in living body tissues in real time, establishing a first training sample and a sample to be tested, adding the first training sample into a pre-stored data training set to form a second training set, and carrying out normalization processing on the sample to be tested to obtain a normalized sample to be tested.
Adding the first training sample into a pre-stored data training set, and forming a second training set by the following steps: comparing the first training sample with a preset list of a data training set, wherein the list of the data training set comprises a pre-stored first electric field impedance set and a corresponding first magnetic field coordinate set, if the magnetic field coordinate and the corresponding electric field impedance in the first training sample are not found in the list of the data training set, normalizing the magnetic field coordinate and the corresponding electric field impedance in the first training sample and adding the normalized magnetic field coordinate and the corresponding electric field impedance into the list of the data training set, and if the magnetic field coordinate and the corresponding electric field impedance in the first training sample are already in the list of the data training set, adding operation is not performed.
And when the number and the range of the samples in the list of the data training set reach the set threshold value, forming a second training set, and storing the normalization parameter matrix A formed in the normalization processing process.
S2: inputting the second training set into a prestored neural network model for training, and outputting a training result; and inputting the normalized sample to be detected into a pre-stored neural network model to obtain a preliminary result.
The neural network model used is a three-layer neural network model, as shown in FIG. 4, the input layer is m input units X and 1 bias term B1The hidden layer is N hidden units and 1 bias item B2The output layer is n output units Y. Wherein the input data X is [ X ]1，x2，...xm]The output data Y is [ Y1，y2，...ym]Bias term B1Is [ b ]11，b12，...b1N]Bias term B2Is [ b ]21，b22，...b2N]The input layer to hidden layer parameter W1 is stored by using m × N matrix as
The hidden layer-to-output layer parameter W2 is stored by using an N × N matrix
The hidden unit activation function is:
where z is the hidden unit input, which is the product of the input data X and the input layer to hidden layer parameter W1, and e is a natural constant.
In the initialization model, weightsW1 and W2 are random numbers of-1 to 1, and offset term b1、b2Is 1. And training the pre-stored neural network model by using the second training set when the set threshold is reached by using the data quantity of the training set and the input data quantity as indexes.
The output of the training result is performed in parallel with the neural network model training, the performance of the trained network model is evaluated by adopting a mean square error, and a mean square error calculation formula is as follows:
wherein E (f; D) is mean square error, m is the number of samples in the training sample set, f (x) is a neural network function, f (x)i) Is the electric field impedance xiMagnetic field coordinates, y, obtained by a neural network functioniIs the electric field impedance x acquired in real timeiThe corresponding magnetic field coordinates.
Comparing the mean square error value with a preset threshold value, and obtaining an updated neural network model after training when the mean square error value is less than or equal to the preset threshold value; and when the mean square error value is larger than the preset threshold value, continuing the model training until the mean square error value of the training result is smaller than or equal to the preset threshold value.
After the neural network is updated every time, performance evaluation needs to be carried out on the neural network, and uncertainty of training time is caused, so that when the spatial coordinate of the electric field impedance of the sample to be measured is calculated, the pre-stored neural network model is adopted, the normalized sample to be measured is input into the pre-stored neural network model, and a preliminary prediction result is obtained. The pre-stored neural network model is the neural network model stored for the last time, the model is the model closest to the neural network model updated at this time, and the condition that the calculated mean square error value is less than or equal to the preset threshold value is met.
S4: and storing the updated neural network model, and performing inverse normalization processing on the preliminary result to obtain the electrode space coordinate corresponding to the sample to be detected.
And storing the updated neural network model and the parameters thereof for updating the neural network model next time and calculating the space coordinate corresponding to the electric field impedance of the sample to be measured.
And performing inverse normalization processing on the preliminary result by adopting the normalization parameter matrix A to obtain a space coordinate corresponding to the electric field impedance of the sample to be measured.
The above reference numbers do not represent a sequential order, the steps may be performed in parallel, and no necessary order is relied upon, which order can be changed by a person skilled in the art without departing from the scope of the invention.
Claims (9)
1. A magnetoelectricity combined positioning and tracking method based on a neural network is characterized by comprising the following steps:
s1: acquiring data in real time, wherein the data comprises a training sample and a sample to be detected, the training sample comprises an electric field impedance value and a corresponding magnetic field coordinate, and the sample to be detected comprises an electric field impedance value;
s2: carrying out normalization processing on the training sample, establishing a training data set, and carrying out normalization processing on the sample to be tested to obtain a normalized sample to be tested;
s3: extracting a pre-stored first neural network model, inputting the training data set into the first neural network model, training to obtain a second neural network model, and inputting the normalized sample to be tested into the first neural network model to obtain a preliminary prediction result;
s4: and replacing the pre-stored first neural network model with the second neural network model, storing the second neural network model, and performing inverse normalization processing on the preliminary prediction result to obtain the electrode space coordinate corresponding to the sample to be detected.
2. The magnetoelectric combination positioning and tracking method based on the neural network as claimed in claim 1, wherein the step of normalizing the training samples to establish the training data set comprises the steps of:
s11: carrying out normalization processing on the training samples, constructing normalized training samples, and storing a normalized parameter matrix A;
s12: storing the normalized training samples into a training data set;
s13: and judging whether the number and the range of the training data set samples reach the set threshold value or not, returning to S11, and completing the establishment of the training data set when the number and the range of the training data set samples reach the set threshold value.
3. The magnetoelectric combination localization tracking method based on the neural network according to claim 2, characterized in that the normalization parameter matrix adopted in the inverse normalization processing is the normalization parameter matrix a.
4. The magnetoelectric combination localization tracking method based on the neural network according to claim 1, characterized in that the first neural network model and the second neural network model are both of a three-layer structure including an input layer, a hidden layer and an output layer, the input layer includes m input units and 1 bias term, the hidden layer includes N hidden units and 1 bias term, and the output layer includes N output units.
5. The magnetoelectric combination localization tracking method based on the neural network according to claim 4, characterized in that the first neural network model and the second neural network model have neural network model parameters, the neural network model parameters include input layer to hidden layer parameters and hidden layer to output layer parameters, the input layer to hidden layer parameters are m × N matrix, and the hidden layer to output layer parameters are N × N matrix.
6. The magnetoelectric combination localization tracking method based on the neural network according to any one of claims 4 or 5, characterized in that the activation function of the hidden unit is:
wherein z is the hidden unit input and e is the natural constant.
7. The magnetoelectric combination positioning and tracking method based on the neural network as claimed in claim 1, wherein the training data set is input into the first neural network model, a second neural network model is obtained by training, and the specific steps are as follows:
s21: inputting the training data set into the first neural network model to obtain a training output result;
s22: calculating a mean square error value of the training output result and an expected target;
s23: comparing the mean square error value with a preset threshold value, and training to obtain a second neural network model when the mean square error value is less than or equal to the preset threshold value; and when the mean square error value is larger than the preset threshold value, returning to the step S1, and continuing the model training.
8. The magnetoelectric combination localization tracking method based on the neural network according to claim 7, characterized in that the calculation formula of the mean square error value is:
wherein E (f; D) is the mean square error value, m is the number of samples of the training data set, f (x) is the neural network function, f (x)i) Is the electric field impedance xiPrediction result, y, obtained by a neural network functioniIs the desired target, i.e. the electric field impedance x acquired in real timeiThe corresponding magnetic field coordinates.
9. A device of a magnetoelectric combination positioning and tracking method based on a neural network is used for executing the magnetoelectric combination positioning and tracking method based on the neural network in any one of claims 1 to 8, a control unit controls an excitation issuing device to sequentially issue excitation according to the sequence circulation of an excitation source V1, an excitation source V2 and an excitation source V3, electric field impedance collectors on a conduit respectively collect electric field impedance between the electric field impedance collectors and electrode plates, the electric field impedance is amplified by an amplifier, the control unit controls a magnetic field generator to be turned on and off, and when the magnetic field generator is turned on, a magnetic field information collector on the conduit collects magnetic field coordinates The method comprises the steps of completing normalization processing of a sample to be tested, extracting a pre-stored first neural network model, training a second neural network model, storing the second neural network model, calculating a preliminary prediction result and completing inverse normalization processing of the preliminary prediction result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910012954.7A CN111407279A (en) | 2019-01-07 | 2019-01-07 | Magnetoelectricity combined positioning and tracking method and device based on neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910012954.7A CN111407279A (en) | 2019-01-07 | 2019-01-07 | Magnetoelectricity combined positioning and tracking method and device based on neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111407279A true CN111407279A (en) | 2020-07-14 |
Family
ID=71484874
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910012954.7A Pending CN111407279A (en) | 2019-01-07 | 2019-01-07 | Magnetoelectricity combined positioning and tracking method and device based on neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111407279A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112578463A (en) * | 2020-12-22 | 2021-03-30 | 罗普特科技集团股份有限公司 | Underwater metal detection positioning method and device based on electric field |
| CN113253204A (en) * | 2021-03-08 | 2021-08-13 | 同济大学 | Positioning method, system and device based on pyroelectric infrared sensor |
| CN114543256A (en) * | 2022-02-09 | 2022-05-27 | 青岛海尔空调电子有限公司 | Household charging method and device for multi-split air conditioner and multi-split air conditioner |
| FR3120302A1 (en) * | 2021-03-08 | 2022-09-09 | Bonetag | method of imaging an implanted implant |
| CN115688610A (en) * | 2022-12-27 | 2023-02-03 | 泉州装备制造研究所 | Wireless electromagnetic six-dimensional positioning method and system, storage medium and electronic equipment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101657153A (en) * | 2007-03-09 | 2010-02-24 | 圣朱德医疗有限公司房颤分公司 | The system and method that is used for correction of inhomogeneous fields |
| CN105726010A (en) * | 2014-12-31 | 2016-07-06 | 韦伯斯特生物官能(以色列)有限公司 | Systems and methods for visualizing electrophysiological data |
| CN106709565A (en) * | 2016-11-16 | 2017-05-24 | 广州视源电子科技股份有限公司 | Neural network optimization method and device |
| CN107633301A (en) * | 2017-08-28 | 2018-01-26 | 广东工业大学 | The training method of testing and its application system of a kind of BP neural network regression model |
| CN108898218A (en) * | 2018-05-24 | 2018-11-27 | 阿里巴巴集团控股有限公司 | A kind of training method of neural network model, device and computer equipment |
| CN108918137A (en) * | 2018-06-08 | 2018-11-30 | 华北水利水电大学 | Fault Diagnosis of Gear Case devices and methods therefor based on improved WPA-BP neural network |
-
2019
- 2019-01-07 CN CN201910012954.7A patent/CN111407279A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101657153A (en) * | 2007-03-09 | 2010-02-24 | 圣朱德医疗有限公司房颤分公司 | The system and method that is used for correction of inhomogeneous fields |
| CN105726010A (en) * | 2014-12-31 | 2016-07-06 | 韦伯斯特生物官能(以色列)有限公司 | Systems and methods for visualizing electrophysiological data |
| CN106709565A (en) * | 2016-11-16 | 2017-05-24 | 广州视源电子科技股份有限公司 | Neural network optimization method and device |
| CN107633301A (en) * | 2017-08-28 | 2018-01-26 | 广东工业大学 | The training method of testing and its application system of a kind of BP neural network regression model |
| CN108898218A (en) * | 2018-05-24 | 2018-11-27 | 阿里巴巴集团控股有限公司 | A kind of training method of neural network model, device and computer equipment |
| CN108918137A (en) * | 2018-06-08 | 2018-11-30 | 华北水利水电大学 | Fault Diagnosis of Gear Case devices and methods therefor based on improved WPA-BP neural network |
Non-Patent Citations (1)
| Title |
|---|
| 黄志强 等, 上海三联书店, pages: 206 - 210 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112578463A (en) * | 2020-12-22 | 2021-03-30 | 罗普特科技集团股份有限公司 | Underwater metal detection positioning method and device based on electric field |
| CN113253204A (en) * | 2021-03-08 | 2021-08-13 | 同济大学 | Positioning method, system and device based on pyroelectric infrared sensor |
| FR3120302A1 (en) * | 2021-03-08 | 2022-09-09 | Bonetag | method of imaging an implanted implant |
| WO2022189106A1 (en) * | 2021-03-08 | 2022-09-15 | Bonetag | Method for imaging an implanted implant |
| CN114543256A (en) * | 2022-02-09 | 2022-05-27 | 青岛海尔空调电子有限公司 | Household charging method and device for multi-split air conditioner and multi-split air conditioner |
| CN115688610A (en) * | 2022-12-27 | 2023-02-03 | 泉州装备制造研究所 | Wireless electromagnetic six-dimensional positioning method and system, storage medium and electronic equipment |
| CN115688610B (en) * | 2022-12-27 | 2023-08-15 | 泉州装备制造研究所 | A wireless electromagnetic six-dimensional positioning method, system, storage medium and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111407279A (en) | Magnetoelectricity combined positioning and tracking method and device based on neural network | |
| CN106197999B (en) | A kind of planetary gear method for diagnosing faults | |
| AU2016313274A1 (en) | Hand held devices for magnetic induction tomography | |
| JP6987063B2 (en) | A method and system for determining at least one type and / or state of cells. | |
| JP2012179352A (en) | System and method for constructing current dipole | |
| EP3110325A1 (en) | Single coil magnetic induction tomographic imaging | |
| CN107137082A (en) | A kind of human cell tissue the cannot-harm-detection device and its detection method | |
| CN110850244A (en) | Time-domain atlas diagnosis method, system and medium for partial discharge defect based on deep learning | |
| CN101980304A (en) | A method for measuring deformation of three-dimensional digital volume images | |
| JP2019505772A (en) | Process and measurement system for data acquisition and processing in soft tomography surveys | |
| WO2015128705A1 (en) | Coil for magnetic induction tomography imaging | |
| CN109965875A (en) | A kind of internal 3 D positioning system and method based on multiresolution mapping | |
| Rogers et al. | Free-moment current dipoles in inverse electrocardiography | |
| CN111345909B (en) | Method and device for determining magnetoelectric mapping relation | |
| WO2023184598A1 (en) | Artificial intelligence-based heart simulator data correction system and method | |
| CN110716998A (en) | A spatialization method for fine-scale population data | |
| CN112508890B (en) | A method for detecting body fat percentage of dairy cows based on a two-level evaluation model | |
| CN115830156A (en) | Accurate electrical impedance tomography method, apparatus, system, medium and device | |
| CN210742393U (en) | A System for Determining the Optimal Frequency of Magnetic Fields Based on Biological Indicators of Animal Cells | |
| JP4390459B2 (en) | Position detection device for lesions in biological tissue | |
| Morcelles et al. | Impedance Tomographic Sensor for Monitoring Bioprinted Cell Cultures | |
| Ru et al. | A Wood Anomaly Detection System Based on Electrical Resistivity Tomography and Tiny Machine Learning | |
| Eichardt et al. | Sensitivity comparisons of cylindrical and hemi-spherical coil setups for magnetic induction tomography | |
| CN113509164B (en) | Multi-frequency magnetic induction tomography reconstruction method based on blind source separation | |
| Trabelsi et al. | Microcontroller-Based Implementation of the DRT for Bioimpedance Spectroscopy |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: No. 5, Wuke East 3rd Road, Wuhou District, Chengdu, Sichuan 610000 Applicant after: Sichuan Jinjiang Electronic Medical Device Technology Co.,Ltd. Address before: No.5, Wuke East 3rd road, Wuhou Science Park, Chengdu hi tech Industrial Development Zone, Sichuan 610045 Applicant before: SICHUAN JINJIANG ELECTRONIC SCIENCE AND TECHNOLOGY Co.,Ltd. |