CN116415686A - Temperature rise model calibration method and device, ultrasonic imaging equipment and storage medium - Google Patents
Temperature rise model calibration method and device, ultrasonic imaging equipment and storage medium Download PDFInfo
- Publication number
- CN116415686A CN116415686A CN202211580256.5A CN202211580256A CN116415686A CN 116415686 A CN116415686 A CN 116415686A CN 202211580256 A CN202211580256 A CN 202211580256A CN 116415686 A CN116415686 A CN 116415686A
- Authority
- CN
- China
- Prior art keywords
- temperature rise
- deep learning
- learning model
- value
- data set
- 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
- 238000003384 imaging method Methods 0.000 title claims abstract description 55
- 238000000034 method Methods 0.000 title claims abstract description 48
- 238000013136 deep learning model Methods 0.000 claims abstract description 103
- 238000012360 testing method Methods 0.000 claims abstract description 31
- 238000012549 training Methods 0.000 claims description 31
- 239000000523 sample Substances 0.000 claims description 27
- 238000012795 verification Methods 0.000 claims description 24
- 238000012285 ultrasound imaging Methods 0.000 claims description 22
- 230000015654 memory Effects 0.000 claims description 18
- 230000006870 function Effects 0.000 claims description 15
- 238000011156 evaluation Methods 0.000 claims description 13
- 238000010276 construction Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 5
- 230000003044 adaptive effect Effects 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 4
- 239000012528 membrane Substances 0.000 description 4
- 238000013527 convolutional neural network Methods 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 230000007787 long-term memory Effects 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- BASFCYQUMIYNBI-UHFFFAOYSA-N platinum Chemical compound [Pt] BASFCYQUMIYNBI-UHFFFAOYSA-N 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 230000006403 short-term memory Effects 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 229910052697 platinum Inorganic materials 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/003—Environmental or reliability tests
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E30/00—Energy generation of nuclear origin
- Y02E30/30—Nuclear fission reactors
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Medical Informatics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Environmental & Geological Engineering (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Ultra Sonic Daignosis Equipment (AREA)
Abstract
The invention provides a temperature rise model calibration method, a device, ultrasonic imaging equipment and a storage medium, wherein the method comprises the following steps: acquiring a temperature rise data set of an ultrasonic imaging system to be detected based on a temperature rise testing device; constructing an initial temperature rise deep learning model of the ultrasonic imaging system to be detected; and calibrating the initial temperature rise deep learning model based on the temperature rise data set to obtain a target temperature rise deep learning model. According to the invention, the initial temperature rise deep learning model is calibrated based on the temperature rise data set, so that the accuracy and efficiency of the target temperature rise deep learning model are improved.
Description
Technical Field
The invention relates to the technical field of temperature testing, in particular to a temperature rise model calibration method and device, ultrasonic imaging equipment and a storage medium.
Background
In the research and development stage and before marketing, the current ultrasonic imaging system needs to perform a series of temperature rise tests based on safety consideration, and the temperature rise model is calibrated by taking the evaluation result as feedback, so that the temperature rise index of the ultrasonic imaging system is ensured to meet domestic/international standards.
The specific process of the calibration temperature rise model in the prior art is as follows: firstly, a calibrator sets an initial temperature rise model according to an empirical value, then based on a test result obtained by temperature rise test, when the test result does not meet the temperature rise index requirement, the calibrator modifies model coefficients (parameter adjustment) in the initial temperature rise model according to the test result and experience, and calibrates the initial temperature rise model to obtain a temperature rise calibration model.
The method for obtaining the temperature rise calibration model in the prior art has the following problems: if a more accurate temperature rise calibration model is to be obtained, a large amount of temperature rise test data is required, and each time the temperature rise test data is obtained, a calibrator needs to modify model coefficients according to test results and experience, so that the acquisition period of the temperature rise calibration model is longer, the efficiency of calibrating the initial temperature rise model is lower, the accuracy of the temperature rise calibration model depends on the calibration level of the calibrator, and the accuracy of calibrating the initial temperature rise model is lower.
Disclosure of Invention
In view of the foregoing, it is necessary to provide a temperature rise model calibration method, apparatus, ultrasonic imaging device and storage medium, so as to solve the technical problems of low efficiency and accuracy in calibrating an initial temperature rise model in the prior art.
In one aspect, the invention provides a temperature rise model calibration method, which comprises the following steps:
acquiring a temperature rise data set of an ultrasonic imaging system to be detected based on a temperature rise testing device;
constructing an initial temperature rise deep learning model of the ultrasonic imaging system to be detected;
and calibrating the initial temperature rise deep learning model based on the temperature rise data set to obtain a target temperature rise deep learning model.
In some possible implementations, the temperature rise data set includes a plurality of sets of system parameters of the ultrasound imaging system under test and a plurality of sets Wen Shengzhi corresponding to the plurality of sets of system parameters; the ultrasonic imaging system to be detected comprises a host and a probe connected with the host; the temperature rise testing device comprises a temperature sensing device for sensing the temperature of the probe and a temperature acquisition device which is connected with the temperature sensing device and is used for acquiring the temperature of the temperature sensing device;
the temperature rise data set of the ultrasonic imaging system to be detected is obtained based on the temperature rise testing device, and the temperature rise data set comprises:
acquiring a plurality of groups of initial temperature values when the probe is in an unoperated state based on the temperature sensing device and the temperature acquisition device;
acquiring the multiple groups of system parameters, and acquiring multiple groups of maximum temperature values of the probe after the probe works for a preset period of time under the control of the multiple groups of system parameters based on the temperature sensing device and the temperature acquisition device;
the plurality of sets of temperature rise values are determined based on the plurality of sets of initial temperature values and the plurality of sets of maximum temperature values.
In some possible implementations, the temperature rise data set includes a first set of temperature rise data including a first system parameter and a first initial temperature value corresponding to the first system parameter;
before the initial temperature rise deep learning model is calibrated based on the temperature rise data set, the method further comprises:
judging whether the first initial temperature value is smaller than a preset temperature value or not;
and when the first initial temperature value is greater than or equal to the preset temperature value, eliminating the first group of temperature rise data from the temperature rise data set.
In some possible implementations, the calibrating the initial temperature rise deep learning model based on the temperature rise data set to obtain a target temperature rise deep learning model includes:
selecting a temperature rise training set from the temperature rise data set, wherein the temperature rise training set comprises a plurality of groups of temperature rise training data, and the temperature rise training data comprises training temperature rise values;
determining initial model parameters of the initial temperature rise deep learning model;
inputting the temperature rise training data into the initial temperature rise deep learning model to obtain a preset temperature rise value;
determining a loss value based on a preset loss function, the training temperature rise value and the predicted temperature rise value;
judging whether the loss value is smaller than a preset loss value, and when the loss value is smaller than the preset loss value, the initial temperature rise deep learning model is the target temperature rise deep learning model; and when the loss value is greater than or equal to the preset loss value, adjusting the initial model parameters, and continuing to calibrate.
In some possible implementations, after the initial temperature rise deep learning model is calibrated based on the temperature rise data set, a target temperature rise deep learning model is obtained, the method further includes:
selecting a temperature rise verification set from the temperature rise data set, wherein the temperature rise verification set comprises at least one group of temperature rise verification data, and the temperature rise verification data comprises verification temperature rise values;
inputting the temperature rise verification data into the target temperature rise deep learning model to obtain a temperature rise value to be verified;
judging whether the temperature rise difference value between the temperature rise value to be verified and the verification temperature rise value is smaller than a preset difference value, and when the temperature rise difference value is smaller than the preset difference value, verifying the target temperature rise deep learning model; and when the temperature rise difference value is larger than or equal to the preset difference value, the initial temperature rise deep learning model is calibrated again based on the temperature rise training set.
In some possible implementations, the temperature rise model calibration method further includes:
acquiring a historical temperature rise data set of the ultrasonic imaging system to be detected;
and calibrating the initial temperature rise deep learning model based on the temperature rise data set and the historical temperature rise data set to obtain a target temperature rise deep learning model.
In some possible implementations, the temperature rise model calibration method further includes:
acquiring system parameters to be evaluated of the ultrasonic imaging system to be tested;
acquiring a temperature rise value to be evaluated of the ultrasonic imaging system to be evaluated based on the system parameter to be evaluated and the target temperature rise deep learning model;
and generating a temperature rise evaluation report based on the temperature rise value to be evaluated and the evaluation index.
On the other hand, the invention also provides a temperature rise model calibration device, which comprises:
the temperature rise data set acquisition unit is used for acquiring a temperature rise data set of the ultrasonic imaging system to be detected based on the temperature rise test device;
the initial temperature rise deep learning model construction unit is used for constructing an initial temperature rise deep learning model of the ultrasonic imaging system to be detected;
and the target temperature rise deep learning model determining unit is used for calibrating the initial temperature rise deep learning model based on the temperature rise data set to obtain a target temperature rise deep learning model.
In another aspect, the present invention also provides an ultrasound imaging apparatus comprising a memory and a processor, wherein,
the memory is used for storing programs;
the processor is coupled to the memory and is configured to execute the program stored in the memory to implement the steps in the temperature rise model calibration method described in any one of the possible implementations.
In another aspect, the present invention also provides a computer readable storage medium, configured to store a computer readable program or instructions, where the program or instructions, when executed by a processor, implement the steps in the temperature rise model calibration method described in any one of the possible implementation manners.
The beneficial effects of adopting the embodiment are as follows: according to the temperature rise model calibration method, the initial temperature rise deep learning model can be calibrated based on the temperature rise data set, the target temperature rise deep learning model is obtained, the autonomous learning capacity of the initial temperature rise deep learning model on the temperature rise data set is utilized, calibration personnel are not required to calibrate the initial temperature rise model according to experience values, dependence on the calibration personnel is eliminated, and therefore accuracy of the obtained target temperature rise deep learning model is improved. And moreover, a calibrator is not required to participate in the calibration process of the initial temperature rise model, so that the labor cost is reduced, and the calibration efficiency is improved.
Furthermore, the initial temperature rise deep learning model is calibrated based on the temperature rise data set obtained by testing the ultrasonic imaging system to be tested by the temperature rise testing device, rather than the initial temperature rise deep learning model based on the historical data of the ultrasonic imaging system to be tested, compared with the historical data, the real-time performance of the temperature rise data set is better, the influence of time and space changes on the target temperature rise deep learning model is eliminated, and the accuracy of the obtained target temperature rise deep learning model is further improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of an embodiment of a temperature rise model calibration method provided by the present invention;
FIG. 2 is a schematic structural diagram of an embodiment of a temperature rise testing device according to the present invention;
FIG. 3 is a flow chart of one embodiment of S101 of FIG. 1 according to the present invention;
FIG. 4 is a schematic flow chart of an embodiment of the present invention for processing a temperature rise data set;
FIG. 5 is a flow chart of the embodiment of S103 in FIG. 1 according to the present invention;
FIG. 6 is a schematic flow chart of an embodiment of verifying a target temperature rise deep learning model according to the present invention;
FIG. 7 is a schematic flow chart of another embodiment of calibrating an initial temperature rise deep learning model according to the present invention;
FIG. 8 is a schematic flow chart diagram of an embodiment of generating a temperature rise assessment report according to the present invention;
FIG. 9 is a schematic structural diagram of an embodiment of a temperature rise model calibration device provided by the present invention;
fig. 10 is a schematic structural diagram of an ultrasonic imaging apparatus according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. It will be apparent that the described embodiments are only some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be understood that the schematic drawings are not drawn to scale. A flowchart, as used in this disclosure, illustrates operations implemented according to some embodiments of the present invention. It should be appreciated that the operations of the flow diagrams may be implemented out of order and that steps without logical context may be performed in reverse order or concurrently. Moreover, one or more other operations may be added to or removed from the flow diagrams by those skilled in the art under the direction of the present disclosure. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software or in one or more hardware modules or integrated circuits or in different networks and/or processor systems and/or microcontroller systems.
References to "first," "second," etc. in the embodiments of the present invention are for descriptive purposes only and are not to be construed as indicating or implying a relative importance or the number of technical features indicated. Thus, a technical feature defining "first", "second" may include at least one such feature, either explicitly or implicitly.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the invention. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.
The invention provides a temperature rise model calibration method, a temperature rise model calibration device, ultrasonic imaging equipment and a storage medium, and the temperature rise model calibration method, the device, the ultrasonic imaging equipment and the storage medium are respectively described below.
Fig. 1 is a schematic flow chart of an embodiment of a temperature rise model calibration method provided by the present invention, as shown in fig. 2, the temperature rise model calibration method includes:
s101, acquiring a temperature rise data set of an ultrasonic imaging system to be tested based on a temperature rise testing device;
s102, constructing an initial temperature rise deep learning model of an ultrasonic imaging system to be detected;
and S103, calibrating the initial temperature rise deep learning model based on the temperature rise data set to obtain a target temperature rise deep learning model.
Compared with the prior art, the temperature rise model calibration method provided by the embodiment of the invention can calibrate the initial temperature rise deep learning model based on the temperature rise data set, obtain the target temperature rise deep learning model, utilize the autonomous learning capability of the initial temperature rise deep learning model on the temperature rise data set, calibrate the initial temperature rise model according to the experience value without a calibrator, eliminate the dependence on the calibrator, and further improve the accuracy of the obtained target temperature rise deep learning model. And moreover, a calibrator is not required to participate in the calibration process of the initial temperature rise model, so that the labor cost is reduced, and the calibration efficiency is improved.
Furthermore, the embodiment of the invention calibrates the initial temperature rise deep learning model based on the temperature rise data set obtained by testing the ultrasonic imaging system to be tested by the temperature rise testing device, but not calibrates the initial temperature rise deep learning model based on the historical data of the ultrasonic imaging system to be tested.
It should be understood that: the model structure of the initial temperature rise Deep learning model includes, but is not limited to, a Deep neural network model (Deep Neural Networks, DNN), a recurrent neural network model (Recurrent Neural Networks, RNN), a convolutional network model (Convolutional Neural Networks, CNN), a Deep generation model (Deep Generative Models, DGM), a generated countermeasure network (Generative Adversarial Networks, GAN), a Long/short term memory network model (Long/short term memory, LSTM), a support vector machine (Support vector machines, SVM), a Deep cross model (Deep cross), and the like.
In some embodiments of the present invention, as shown in fig. 2, the ultrasound imaging system 10 to be tested includes a main unit 11 and a probe 12 connected to the main unit 11, and since the probe 12 is in contact with the body surface of the human body in practical application, the temperature rise test device 20 includes a body membrane 21, a temperature sensing device 22 and a temperature acquisition device 23, wherein the body membrane 21 is used for simulating the body surface of the human body, the probe 12 is connected to the body membrane 21, the temperature sensing device 22 is connected to the probe 12 for sensing the temperature of the probe 12, and the temperature acquisition device 23 is connected to the temperature sensing device 22 for acquiring the temperature of the temperature sensing device 22.
It should be noted that: the connection mode in the embodiment of the invention can be set to be wired connection or wireless connection according to the actual application scene, and the connection can be set to be fixed connection or movable connection according to the actual application scene.
In particular embodiments of the present invention, the temperature sensing device 22 may be any one of a thermocouple, a thermistor, and a platinum resistor.
In order to avoid that the temperature and humidity changes in the external environment affect the temperature detected by the probe 12, resulting in inaccurate temperature rise data set, in some embodiments of the present invention, as shown in fig. 2, the temperature rise testing device 20 includes a constant temperature and humidity box 24, and the probe 12 and the body membrane 21 are placed in the constant temperature and humidity box 24, so that the influence of the temperature and humidity changes in the environment on the temperature detected by the probe 12 is eliminated, thereby improving the accuracy of the obtained temperature rise data set.
Because the target temperature rise deep learning model is used to determine the relationship between the system parameters and the temperature rise values of the ultrasonic imaging system to be measured, the temperature rise data set includes multiple groups of system parameters of the ultrasonic imaging system to be measured and multiple groups of temperature rise values corresponding to the multiple groups of system parameters, and in some embodiments of the present invention, as shown in fig. 3, step S101 includes:
s301, acquiring a plurality of groups of initial temperature values when the probe 12 is in an unoperated state based on the temperature sensing device 22 and the temperature acquisition device 23;
s302, acquiring a plurality of groups of system parameters, and acquiring a plurality of groups of maximum temperature values of the probe 12 after working for a preset period of time under the control of the plurality of groups of system parameters based on the temperature sensing device 22 and the temperature acquisition device 23;
s303, determining a plurality of groups of temperature rise values based on the plurality of groups of initial temperature values and the plurality of groups of maximum temperature values.
It should be understood that: each set of temperature rise values is the difference between each set of maximum temperature values and each set of initial temperature values.
It should be noted that: the preset time period should be adjusted according to the actual application scenario, for example: when the average duration of the probe 12 for detecting the body surface of the human body in the actual application scene is 30 minutes, the preset time period is 30 minutes.
Also to be described is: system parameters include, but are not limited to, probe type, transmit center frequency, transducer element size and number, electronic depth of focus, sampling frequency, imaging mode of the ultrasound imaging system under test, and the like.
In order to avoid damage to the human body caused by the fact that the maximum temperature rise value reached by the probe 12 under the control of the system parameter is higher than the sustainable temperature of the human body surface when the initial temperature value of the probe 12 is higher in the practical application scene, the initial temperature value of the probe 12 should be set to be substantially the same as the ambient temperature value.
However, in the process of acquiring the temperature rise data set, when the initial temperature value is not reduced to the ambient temperature value, the probe 12 is controlled to work for a preset period of time under multiple groups of system parameters, so that temperature rise data with excessively high maximum temperature rise value exists in the acquired temperature rise data set, namely: in order to solve this technical problem, in some embodiments of the present invention, the temperature rise data set is described by taking an example that the temperature rise data set includes a first set of temperature rise data, specifically, the first set of temperature rise data includes a first system parameter and a first initial temperature value corresponding to the first system parameter;
then, as shown in fig. 4, before step S102, the method further includes:
s401, judging whether the first initial temperature value is smaller than a preset temperature value or not;
s402, when the first initial temperature value is greater than or equal to a preset temperature value, eliminating a first group of temperature rise data from the temperature rise data set.
According to the embodiment of the invention, the availability of the first group of temperature rise data in the temperature rise data set can be ensured by eliminating the temperature rise data with the initial temperature value being greater than or equal to the preset temperature value, and the rationality and the availability of the temperature rise data set are further ensured.
It should be understood that: other temperature rise data except the first set of temperature rise data in the temperature rise data set should also be judged on its initial temperature value to ensure the rationality and availability of all temperature rise data in the temperature rise data set.
It should be noted that: the preset temperature value may be adjusted or defined according to the ambient temperature value, in particular, the difference between the preset temperature value and the ambient temperature value should be less than 5 ℃.
In some embodiments of the present invention, as shown in fig. 5, step S103 includes:
s501, selecting a temperature rise training set from the temperature rise data set, wherein the temperature rise training set comprises a plurality of groups of temperature rise training data, and the temperature rise training data comprises training temperature rise values;
s502, determining initial model parameters of an initial temperature rise deep learning model;
s503, inputting temperature rise training data into an initial temperature rise deep learning model to obtain a preset temperature rise value;
s504, determining a loss value based on a preset loss function, a training temperature rise value and a predicted temperature rise value;
s505, judging whether the loss value is smaller than a preset loss value, and when the loss value is smaller than the preset loss value, the initial temperature rise deep learning model is a target temperature rise deep learning model; and when the loss value is greater than or equal to the preset loss value, adjusting the initial model parameters, and continuing to calibrate.
In some embodiments of the present invention, the initial model parameters in step S502 include, but are not limited to, the number of layers of the neural network, the size and number of convolution kernels in each layer of the neural network, and the like. The loss function in step S504 includes, but is not limited to, a regression loss function and a classification loss function, wherein the regression loss function may be any one of a mean square error and an average absolute error, and the classification loss function may be any one of a cross entropy loss function, a focus loss function and a relative entropy loss function.
In the embodiment of the present invention, when the loss value is greater than or equal to the preset loss value in step S505, the initial model parameters are adjusted, and the specific flow of continuing calibration is as follows: adjusting initial model parameters (forward calculation), and inputting another group of temperature rise training data into the initial temperature rise deep learning model after the initial model parameters are adjusted to obtain a preset temperature rise value; determining a loss value based on a preset loss function, a training temperature rise value and a predicted temperature rise value; judging whether the loss value is smaller than a preset loss value, when the loss value is smaller than the preset loss value, the transition temperature rise deep learning model is a target temperature rise deep learning model (backward propagation), and when the loss value is larger than or equal to the preset loss value, continuing to adjust initial model parameters until the calculated loss value is smaller than the preset loss value.
It should be noted that: the initial model parameters are adjusted by using optimizers including, but not limited to, random gradient descent optimizers (Stochastic Gradient Descent, SGD), batch gradient descent optimizers (Batch Gradient Descent, BGD), adaptive moment estimation optimizers (Adaptive Moment Estimation, adam), and adaptive gradient optimizers (Adaptive Gradient, adagard).
It should be understood that: the preset loss value can be set or adjusted according to the actual application scene, and is not particularly limited herein.
In order to improve the generalization capability of the calibrated target temperature rise deep learning model, namely: the adaptation of the target temperature rise deep learning model to the fresh sample, in some embodiments of the present invention, as shown in fig. 6, further includes, after step S103:
s601, selecting a temperature rise verification set from the temperature rise data set, wherein the temperature rise verification set comprises at least one group of temperature rise verification data, and the temperature rise verification data comprises verification temperature rise values;
s602, inputting temperature rise verification data into a target temperature rise deep learning model to obtain a temperature rise value to be verified;
s603, judging whether the temperature rise difference between the temperature rise value to be verified and the verification temperature rise value is smaller than a preset difference value, and when the temperature rise difference is smaller than the preset difference value, verifying the target temperature rise deep learning model; and when the temperature rise difference value is greater than or equal to the preset difference value, the initial temperature rise deep learning model is calibrated again based on the temperature rise training set.
According to the embodiment of the invention, the target temperature rise deep learning model is verified through the temperature rise verification set, and when the generalization capability of the target temperature rise deep learning model does not pass, the initial temperature rise deep learning model is calibrated again based on the temperature rise training set, so that the generalization capability of the target temperature rise deep learning model can be ensured.
It should be understood that: because generalization capability refers to the adaptability of the target temperature rise deep learning model to fresh samples, the temperature rise training set and the temperature rise verification set are mutually exclusive sets, namely: the temperature rise training set and the temperature rise verification set do not have the same temperature rise data.
In a specific embodiment of the invention, the ratio of the data volume of the temperature rise training set to the temperature rise data set is 8:2.
It should be noted that: the preset difference value should be set or adjusted according to the actual application scenario or the empirical value, which is not limited herein.
Because the time required for acquiring the temperature rise data set by the temperature rise testing device is longer, in order to consider the accuracy and the determination efficiency of the target temperature rise deep learning model, in some embodiments of the present invention, the temperature rise model calibration method as shown in fig. 7 further includes:
s701, acquiring a historical temperature rise data set of an ultrasonic imaging system to be detected;
s702, calibrating the initial temperature rise deep learning model based on the temperature rise data set and the historical temperature rise data set to obtain the target temperature rise deep learning model.
According to the embodiment of the invention, the initial temperature rise deep learning model is calibrated based on the temperature rise data set and the historical temperature rise data set, so that the training data amount can be increased, and the accuracy of the calibrated target temperature rise deep learning model is further improved. And the historical temperature rise data set is only required to be called from the storage medium storing the historical temperature rise data set, so that the acquisition time of the historical temperature rise data set is shorter, and the determination efficiency of the model for the target temperature rise deep learning is improved.
It should be noted that: to ensure real-time performance of the historical temperature rise data set, the historical temperature rise data set should be data within a preset period of time, for example: the temperature rise data of the ultrasonic imaging system to be measured in the last year.
In some embodiments of the present invention, as shown in fig. 8, the temperature rise model calibration method further includes:
s801, acquiring system parameters to be evaluated of an ultrasonic imaging system to be tested;
s802, obtaining a temperature rise value to be evaluated of an ultrasonic imaging system to be evaluated based on parameters of the system to be evaluated and a target temperature rise deep learning model;
s803, generating a temperature rise evaluation report based on the temperature rise value to be evaluated and the evaluation index.
The embodiment of the invention can be used for a worker to check and store in real time by generating the temperature rise evaluation report, so that the worker can know the temperature rise evaluation result of the ultrasonic imaging system to be tested conveniently.
Wherein, the evaluation index can be the maximum temperature rise extremum, namely: and when the temperature rise value to be evaluated is larger than the maximum temperature rise extremum, the temperature rise evaluation result of the ultrasonic imaging system to be evaluated is not passed, and when the temperature rise extremum to be evaluated is smaller than or equal to the maximum temperature rise extremum, the temperature rise evaluation result of the ultrasonic imaging system to be evaluated is passed.
It should be noted that: the evaluation index can also be set according to the actual application scene, and will not be described in detail here.
In order to better implement the temperature rise model calibration method in the embodiment of the present invention, correspondingly, as shown in fig. 9, the embodiment of the present invention further provides a temperature rise model calibration device 900, which includes:
a temperature rise data set acquisition unit 901, configured to acquire a temperature rise data set of an ultrasonic imaging system to be tested based on a temperature rise testing device;
an initial temperature rise deep learning model construction unit 902, configured to construct an initial temperature rise deep learning model of the ultrasound imaging system to be tested;
and the target temperature rise deep learning model determining unit 903 is configured to calibrate the initial temperature rise deep learning model based on the temperature rise data set, and obtain a target temperature rise deep learning model.
The temperature rise model calibration device 900 provided in the foregoing embodiment may implement the technical solution described in the foregoing embodiment of the temperature rise model calibration method, and the specific implementation principle of each module or unit may refer to the corresponding content in the foregoing embodiment of the temperature rise model calibration method, which is not described herein again.
As shown in fig. 10, the present invention also correspondingly provides an ultrasonic imaging apparatus 1000. The ultrasound imaging device 1000 includes a processor 1001, a memory 1002, and a display 1003. Fig. 10 shows only some of the components of the ultrasound imaging device 1000, but it should be understood that not all of the illustrated components are required to be implemented and that more or fewer components may be implemented instead.
The memory 1002 may be an internal storage unit of the ultrasound imaging device 1000 in some embodiments, such as a hard disk or memory of the ultrasound imaging device 1000. The memory 1002 may also be an external storage device of the ultrasound imaging apparatus 1000 in other embodiments, such as a plug-in hard disk, smart Media Card (SMC), secure Digital (SD) Card, flash memory Card (Flash Card) or the like, which are provided on the ultrasound imaging apparatus 1000.
Further, the memory 1002 may also include both internal and external storage units of the ultrasound imaging device 1000. The memory 1002 is used for storing application software for installing the ultrasonic imaging apparatus 1000 and various types of data.
The processor 1001 may in some embodiments be a central processing unit (Central Processing Unit, CPU), microprocessor or other data processing chip for executing program code or processing data stored in the memory 1002, such as the temperature rise model calibration method of the present invention.
The display 1003 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, or the like in some embodiments. The display 1003 is used to display information at the ultrasound imaging device 1000 and to display a visual user interface. The components 1001-1003 of the ultrasound imaging device 1000 communicate with each other via a system bus.
In some embodiments of the present invention, when the processor 1001 executes the temperature rise model calibration program in the memory 1002, the following steps may be implemented:
acquiring a temperature rise data set of an ultrasonic imaging system to be detected based on a temperature rise testing device;
constructing an initial temperature rise deep learning model of the ultrasonic imaging system to be detected;
and calibrating the initial temperature rise deep learning model based on the temperature rise data set to obtain the target temperature rise deep learning model.
It should be understood that: the processor 1001 may perform other functions in addition to the above functions when executing the temperature rise model calibration routine in the memory 1002, see in particular the description of the corresponding method embodiments above.
Further, the type of the ultrasonic imaging apparatus 1000 is not particularly limited in the embodiment of the present invention, and the ultrasonic imaging apparatus 1000 may be a portable ultrasonic imaging apparatus such as a mobile phone, a tablet computer, a personal digital assistant (personal digital assistant, PDA), a wearable apparatus, a laptop computer (laptop), and the like. Exemplary embodiments of portable ultrasound imaging devices include, but are not limited to, portable ultrasound imaging devices that are onboard IOS, android, microsoft or other operating systems. The portable ultrasound imaging device described above may also be other portable ultrasound imaging devices, such as a laptop computer (laptop) or the like having a touch-sensitive surface (e.g., a touch panel). It should also be appreciated that in other embodiments of the invention, the ultrasound imaging device 1000 may be a desktop computer having a touch-sensitive surface (e.g., a touch panel) instead of a portable ultrasound imaging device.
Correspondingly, the embodiment of the application also provides a computer readable storage medium, which is used for storing a computer readable program or instruction, and when the program or instruction is executed by a processor, the steps in the temperature rise model calibration method or the functions in the temperature rise model calibration device provided by the embodiments of the method can be realized.
Those skilled in the art will appreciate that all or part of the flow of the methods of the embodiments described above may be accomplished by way of a computer program stored in a computer readable storage medium to instruct related hardware (e.g., a processor, a controller, etc.). The computer readable storage medium is a magnetic disk, an optical disk, a read-only memory or a random access memory.
The temperature rise model calibration method, the device, the ultrasonic imaging equipment and the storage medium provided by the invention are described in detail, and specific examples are applied to the principle and the implementation mode of the invention, and the description of the examples is only used for helping to understand the method and the core idea of the invention; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in light of the ideas of the present invention, the present description should not be construed as limiting the present invention.
Claims (10)
1. The temperature rise model calibration method is characterized by comprising the following steps of:
acquiring a temperature rise data set of an ultrasonic imaging system to be detected based on a temperature rise testing device;
constructing an initial temperature rise deep learning model of the ultrasonic imaging system to be detected;
and calibrating the initial temperature rise deep learning model based on the temperature rise data set to obtain a target temperature rise deep learning model.
2. The method of claim 1, wherein the temperature rise data set includes a plurality of sets of system parameters of the ultrasound imaging system to be measured and a plurality of sets Wen Shengzhi of the plurality of sets of system parameters; the ultrasonic imaging system to be detected comprises a host and a probe connected with the host; the temperature rise testing device comprises a temperature sensing device for sensing the temperature of the probe and a temperature acquisition device which is connected with the temperature sensing device and is used for acquiring the temperature of the temperature sensing device;
the temperature rise data set of the ultrasonic imaging system to be detected is obtained based on the temperature rise testing device, and the temperature rise data set comprises:
acquiring a plurality of groups of initial temperature values when the probe is in an unoperated state based on the temperature sensing device and the temperature acquisition device;
acquiring the multiple groups of system parameters, and acquiring multiple groups of maximum temperature values of the probe after the probe works for a preset period of time under the control of the multiple groups of system parameters based on the temperature sensing device and the temperature acquisition device;
the plurality of sets of temperature rise values are determined based on the plurality of sets of initial temperature values and the plurality of sets of maximum temperature values.
3. The temperature rise model calibration method of claim 2, wherein the temperature rise data set comprises a first set of temperature rise data comprising a first system parameter and a first initial temperature value corresponding to the first system parameter;
before the initial temperature rise deep learning model is calibrated based on the temperature rise data set, the method further comprises:
judging whether the first initial temperature value is smaller than a preset temperature value or not;
and when the first initial temperature value is greater than or equal to the preset temperature value, eliminating the first group of temperature rise data from the temperature rise data set.
4. The method for calibrating a temperature rise model according to claim 1, wherein the step of calibrating the initial temperature rise deep learning model based on the temperature rise data set to obtain a target temperature rise deep learning model comprises the steps of:
selecting a temperature rise training set from the temperature rise data set, wherein the temperature rise training set comprises a plurality of groups of temperature rise training data, and the temperature rise training data comprises training temperature rise values;
determining initial model parameters of the initial temperature rise deep learning model;
inputting the temperature rise training data into the initial temperature rise deep learning model to obtain a preset temperature rise value;
determining a loss value based on a preset loss function, the training temperature rise value and the predicted temperature rise value;
judging whether the loss value is smaller than a preset loss value, and when the loss value is smaller than the preset loss value, the initial temperature rise deep learning model is the target temperature rise deep learning model; and when the loss value is greater than or equal to the preset loss value, adjusting the initial model parameters, and continuing to calibrate.
5. The method according to claim 4, further comprising, after said initial temperature rise deep learning model is calibrated based on said temperature rise data set to obtain a target temperature rise deep learning model:
selecting a temperature rise verification set from the temperature rise data set, wherein the temperature rise verification set comprises at least one group of temperature rise verification data, and the temperature rise verification data comprises verification temperature rise values;
inputting the temperature rise verification data into the target temperature rise deep learning model to obtain a temperature rise value to be verified;
judging whether the temperature rise difference value between the temperature rise value to be verified and the verification temperature rise value is smaller than a preset difference value, and when the temperature rise difference value is smaller than the preset difference value, verifying the target temperature rise deep learning model; and when the temperature rise difference value is larger than or equal to the preset difference value, the initial temperature rise deep learning model is calibrated again based on the temperature rise training set.
6. The temperature rise model calibration method of claim 1, further comprising:
acquiring a historical temperature rise data set of the ultrasonic imaging system to be detected;
and calibrating the initial temperature rise deep learning model based on the temperature rise data set and the historical temperature rise data set to obtain the target temperature rise deep learning model.
7. The temperature rise model calibration method of claim 1, further comprising:
acquiring system parameters to be evaluated of the ultrasonic imaging system to be tested;
acquiring a temperature rise value to be evaluated of the ultrasonic imaging system to be evaluated based on the system parameter to be evaluated and the target temperature rise deep learning model;
and generating a temperature rise evaluation report based on the temperature rise value to be evaluated and the evaluation index.
8. A temperature rise model calibration device, comprising:
the temperature rise data set acquisition unit is used for acquiring a temperature rise data set of the ultrasonic imaging system to be detected based on the temperature rise test device;
the initial temperature rise deep learning model construction unit is used for constructing an initial temperature rise deep learning model of the ultrasonic imaging system to be detected;
and the target temperature rise deep learning model determining unit is used for calibrating the initial temperature rise deep learning model based on the temperature rise data set to obtain a target temperature rise deep learning model.
9. An ultrasound imaging apparatus comprising a memory and a processor, wherein,
the memory is used for storing programs;
the processor, coupled to the memory, for executing the program stored in the memory to implement the steps of the temperature rise model calibration method of any of the preceding claims 1 to 7.
10. A computer readable storage medium storing a computer readable program or instructions which when executed by a processor is capable of carrying out the steps of the temperature rise model calibration method according to any one of the preceding claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211580256.5A CN116415686A (en) | 2022-12-09 | 2022-12-09 | Temperature rise model calibration method and device, ultrasonic imaging equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211580256.5A CN116415686A (en) | 2022-12-09 | 2022-12-09 | Temperature rise model calibration method and device, ultrasonic imaging equipment and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116415686A true CN116415686A (en) | 2023-07-11 |
Family
ID=87052149
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211580256.5A Pending CN116415686A (en) | 2022-12-09 | 2022-12-09 | Temperature rise model calibration method and device, ultrasonic imaging equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116415686A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116662763A (en) * | 2023-07-25 | 2023-08-29 | 华夏天信智能物联股份有限公司 | Data processing method and system for temperature rise test of frequency converter |
-
2022
- 2022-12-09 CN CN202211580256.5A patent/CN116415686A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116662763A (en) * | 2023-07-25 | 2023-08-29 | 华夏天信智能物联股份有限公司 | Data processing method and system for temperature rise test of frequency converter |
CN116662763B (en) * | 2023-07-25 | 2023-10-20 | 华夏天信智能物联股份有限公司 | Data processing method and system for temperature rise test of frequency converter |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108541363A (en) | Technology for management of sensor exception | |
CN105866781B (en) | Data processing method and electronic equipment | |
CN110188862B (en) | Searching method, device and system for model hyper-parameters for data processing | |
CN116415686A (en) | Temperature rise model calibration method and device, ultrasonic imaging equipment and storage medium | |
CN115841046B (en) | Accelerated degradation test data processing method and device based on wiener process | |
CN112362971A (en) | Power module equivalent resistance testing method, device, equipment and storage medium | |
US9930462B2 (en) | System and method for on-site microphone calibration | |
CN113487019A (en) | Circuit fault diagnosis method and device, computer equipment and storage medium | |
CN112985867B (en) | Steering engine testing method, device, equipment and storage medium | |
CN117391036B (en) | Printed circuit board simulation method, device, equipment and storage medium | |
CN117405075B (en) | Intelligent settlement monitoring method and system | |
CN111950605A (en) | Meter identification model learning method, device and equipment and meter identification method | |
US20110130989A1 (en) | System and method for identifying a peripheral component interconnect express signal | |
CN107832935B (en) | Method and device for determining hydrologic variable design value | |
JP2021012078A (en) | Temperature distribution estimation device and temperature distribution estimation method | |
CN115452101A (en) | Instrument verification method, device, equipment and medium | |
JP2005164836A5 (en) | ||
CN110967446B (en) | Parameter targeting method and device, storage medium and electronic equipment | |
US20160070632A1 (en) | Power profiling method, power profiling system, and processor-readable storage medium | |
KR101753686B1 (en) | Simulation system for circuit analysis and verification | |
CN110134265A (en) | Slide-mouse and its setup parameter adquisitiones | |
CN116087622B (en) | Detection accuracy prediction method, device, processing equipment and storage medium | |
TW202020866A (en) | Device and method for diagnosing fault of circuit board | |
KR102456898B1 (en) | Method and apparatus for generating norm pattern of data signal | |
CN111743569B (en) | Instrument testing method, device and 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 |