CN114611568A - Cross-equipment generalization detection method of medical image diagnosis and treatment product and related equipment - Google Patents
Cross-equipment generalization detection method of medical image diagnosis and treatment product and related equipment Download PDFInfo
- Publication number
- CN114611568A CN114611568A CN202210056389.6A CN202210056389A CN114611568A CN 114611568 A CN114611568 A CN 114611568A CN 202210056389 A CN202210056389 A CN 202210056389A CN 114611568 A CN114611568 A CN 114611568A
- Authority
- CN
- China
- Prior art keywords
- data set
- style
- diagnosis
- initial data
- medical image
- 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
- 238000003745 diagnosis Methods 0.000 title claims abstract description 119
- 238000001514 detection method Methods 0.000 title claims abstract description 28
- 238000002059 diagnostic imaging Methods 0.000 claims abstract description 51
- 238000000034 method Methods 0.000 claims abstract description 40
- 230000005012 migration Effects 0.000 claims abstract description 36
- 238000013508 migration Methods 0.000 claims abstract description 36
- 238000010835 comparative analysis Methods 0.000 claims abstract description 7
- 238000012549 training Methods 0.000 claims description 30
- 230000006870 function Effects 0.000 claims description 16
- 230000008859 change Effects 0.000 claims description 10
- 238000005070 sampling Methods 0.000 claims description 5
- 206010070834 Sensitisation Diseases 0.000 claims description 4
- 230000008313 sensitization Effects 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000002591 computed tomography Methods 0.000 claims description 3
- 238000004590 computer program Methods 0.000 claims description 3
- 229940127554 medical product Drugs 0.000 claims 1
- 238000002604 ultrasonography Methods 0.000 claims 1
- 238000004891 communication Methods 0.000 description 10
- 238000006243 chemical reaction Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000005481 NMR spectroscopy Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The application provides a cross-device generalization detection method of a medical image diagnosis and treatment product and related equipment. The method comprises the following steps: acquiring an initial data set output by medical imaging equipment; respectively inputting the initial data set into at least two trained data style migration models, and outputting a generation data set with a corresponding style through each data style migration model; inputting each generated data set in all the generated data sets into a medical image diagnosis and treatment product respectively to obtain a diagnosis result corresponding to each generated data set; and performing comparative analysis on all diagnosis results to obtain a detection result of the cross-equipment generalization of the medical image diagnosis and treatment product. The method can expand the difference between the style of the output generated data set and the style of the initial data set as much as possible on the premise of ensuring that the style of the output generated data set is as close to a new style as possible, solves the problem of data shortage in the field of medical images, and can well detect the diagnosis generalization of medical image diagnosis and treatment products.
Description
Technical Field
The application relates to the technical field of medical image processing, in particular to a cross-equipment generalization detection method of medical image diagnosis and treatment products.
Background
Modern medicine increasingly depends on medical image data for diagnosis, and as intelligent image diagnosis algorithms tend to mature, a plurality of methods for processing medical images based on artificial intelligence methods appear, however, due to the problems of different types of equipment, different set parameters of the equipment, different shooting light of the equipment and the like, image data generated by different equipment have certain difference, once the intelligent image diagnosis algorithms are tested by using data sets generated by other equipment, the diagnosis performance of the algorithms can be obviously reduced, the generalization performance is very poor, and the intelligent image diagnosis algorithms have very large problems and hidden dangers in actual use and deployment. In addition, because of the difference of some parameters, data generated by different devices are difficult to share and use, so that the medical image data is in short supply.
Based on the above situation, in the prior art, the diagnosis effect of the intelligent image diagnosis algorithms is detected through the style migration model, but many models with style migration need to be trained by paired data sets, and it is difficult to find paired pictures generated by different device styles in the field of medical imaging at present, for example, different devices image the same part of the same person, so in practical situations, it is obviously very difficult to acquire such paired data sets.
Disclosure of Invention
In view of the above, an objective of the present application is to provide a method and related apparatus for detecting cross-apparatus generalization of medical image diagnosis and treatment products, so as to solve the above technical problems.
Based on the above object, a first aspect of the present application provides a method for detecting cross-device generalization of medical imaging diagnosis and treatment products, including:
acquiring an initial data set output by medical imaging equipment;
respectively inputting the initial data set into at least two trained data style migration models, and outputting a generation data set with a corresponding style through each data style migration model;
inputting each generated data set in all the generated data sets into a medical image diagnosis and treatment product respectively to obtain a diagnosis result corresponding to each generated data set;
and performing comparative analysis on all the diagnosis results to obtain a detection result of the cross-equipment diagnosis generalization of the medical image diagnosis and treatment product.
A second aspect of the present application provides a detection apparatus for product is diagnose to medical image that strides equipment generalization, includes:
an initial data acquisition module configured to acquire an initial data set output by a medical imaging device;
a data style migration module configured to input the initial data set into at least two trained data style migration models, respectively, and output a generated data set of a corresponding style via each of the data style migration models;
the diagnosis result acquisition module is configured to input each of the generated data sets into a medical image diagnosis and treatment product respectively to obtain a diagnosis result corresponding to each of the generated data sets;
and the detection result acquisition module is configured to perform comparative analysis on all the diagnosis results to obtain a detection result of the cross-equipment diagnosis generalization of the medical image diagnosis and treatment product.
A third aspect of the application provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method of the first aspect when executing the program.
A fourth aspect of the present application provides a non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the method of the first aspect.
From the above, the method for detecting the cross-device generalization of the medical image diagnosis and treatment product and the related device provided by the application can expand the difference between the style of the output generated data set and the style of the initial data set as much as possible on the premise of ensuring that the style of the output generated data set is close to a new style as much as possible through the data style migration model, so that the style conversion can be realized between medical image devices with small style difference, and the problem of data shortage in the medical image field is solved at the same time. And then inputting the medical image diagnosis and treatment product through the generated data sets in the plurality of styles, and performing contrastive analysis on all diagnosis results to obtain a detection result of the cross-equipment diagnosis and treatment product of the medical image diagnosis and treatment product, so that the diagnosis and treatment product of the medical image diagnosis and treatment product can be well detected.
Drawings
In order to more clearly illustrate the technical solutions in the present application or the related art, the drawings needed to be used in the description of the embodiments or the related art will be briefly introduced below, and it is obvious that the drawings in the following description are only embodiments of the present application, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without creative efforts.
Fig. 1 is a flowchart of a method for detecting cross-device generalization of a medical image diagnosis and treatment product according to an embodiment of the present application;
FIG. 2 is a flowchart of a training method of a data style migration model according to an embodiment of the present application;
fig. 3 is a schematic structural diagram of a detection apparatus for cross-device generalization of medical image diagnosis and treatment products according to an embodiment of the present application;
fig. 4 is a schematic diagram of an electronic device according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is further described in detail below with reference to the accompanying drawings in combination with specific embodiments.
It should be noted that technical terms or scientific terms used in the embodiments of the present application should have a general meaning as understood by those having ordinary skill in the art to which the present application belongs, unless otherwise defined. The use of "first," "second," and similar terms in the embodiments of the present application is not intended to indicate any order, quantity, or importance, but rather is used to distinguish one element from another. The word "comprising" or "comprises", and the like, means that the element or item listed before the word covers the element or item listed after the word and its equivalents, but does not exclude other elements or items. The terms "connected" or "coupled" and the like are not restricted to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "upper", "lower", "left", "right", and the like are used merely to indicate relative positional relationships, and when the absolute position of the object being described is changed, the relative positional relationships may also be changed accordingly.
In the related art, more and more medical image diagnosis and treatment products appear, some of the medical image diagnosis and treatment products have very high diagnosis accuracy on an initial data set generated by the same medical imaging device, but have the problem of insufficient generalization performance, once the medical image diagnosis and treatment products are deployed on the initial data sets generated by other medical imaging devices, the diagnosis accuracy of the medical image diagnosis and treatment products can be greatly reduced, and the medical image diagnosis and treatment products have very large problems and hidden dangers in application.
In addition, medical image data in related art is very limited, however, different medical imaging devices have different initial data sets due to sampling rate, exposure period, color space, and the like, and cannot be shared in medical image.
The embodiment of the application provides a method for detecting cross-device generalization of medical image diagnosis and treatment products, and by means of a data style migration model, on the premise that the style of an output generated data set is close to a new style as much as possible, the difference between the style of the output generated data set and the style of an initial data set can be enlarged as much as possible, so that style conversion can be achieved between medical image devices with small style difference, meanwhile, the problem of data shortage in the field of medical images is solved, and the diagnosis generalization of the medical image diagnosis and treatment products can be well detected.
As shown in fig. 1, the method of the present embodiment includes:
In this step, the initial data set is an image data set composed of image data output by a medical imaging apparatus.
And 102, respectively inputting the initial data set into at least two trained data style migration models, and outputting a generated data set with a corresponding style through each data style migration model.
In the step, the initial data set is respectively input into two or more trained data style migration models, and the generated data sets of corresponding styles are output through different data style migration models.
The style conversion module is used for converting the styles of the initial data sets output by different medical imaging equipment to obtain a required generated data set corresponding to the styles of the medical imaging equipment, expanding the difference between the style of the output generated data set and the style of the initial data set as much as possible, and simultaneously obtaining the generated data set with the same style as the initial data set, thereby solving the problem of data shortage in the field of medical imaging.
103, inputting each of the generated data sets into a medical image diagnosis and treatment product respectively to obtain a diagnosis result corresponding to each generated data set.
In the step, the generated data sets of the corresponding styles output by the different data style migration models are respectively input into the same medical image diagnosis and treatment product to be detected, so that each diagnosis result is ensured to be obtained under the same condition.
And 104, performing comparative analysis on all the diagnosis results to obtain a detection result of the cross-equipment diagnosis generalization of the medical image diagnosis and treatment product.
In the step, all the diagnosis results obtained under the same condition are compared and analyzed, and the accuracy of the detection result of the diagnosis generalization of the medical image diagnosis and treatment product to be detected is further ensured according to the same condition and the number of the diagnosis results.
According to the scheme, the initial data sets output by the medical imaging equipment are respectively input into different data style migration models, the generated data sets with corresponding styles are output, the style of the output generated data sets is expanded as much as possible on the premise that the style of the output generated data sets is close to a new style as much as possible, so that style conversion can be achieved between the medical imaging equipment with small style difference, the problem of data shortage in the medical imaging field is solved, in addition, all the generated data sets are input into the same medical imaging diagnosis and treatment product needing to be detected, the diagnosis result corresponding to each generated data set is obtained, then all the diagnosis results are compared and analyzed, the detection result of the diagnosis generalization of the medical imaging diagnosis and treatment product needing to be detected is obtained, and the diagnosis generalization detection result of the medical imaging diagnosis and treatment product needing to be detected is further ensured on the basis of the same conditions and the number of the diagnosis generalization of the medical imaging diagnosis and treatment product needing to be detected The accuracy of the measured result can well detect the diagnosis generalization of the medical image diagnosis and treatment product.
In some embodiments, as shown in FIG. 2, the data style migration model in step 102 is trained by:
In the step, the training model is constructed based on the deep learning architecture, so that the training model can be quickly constructed.
In this step, corresponding initial data sets output by two different medical imaging devices are acquired, and the initial data sets may be acquired from corresponding medical image databases, where the manner of acquiring the first initial data set and the second initial data set is not particularly limited.
In the step, a first initial data set is input into a training model, a data set is generated by a first generator, the data set is output by the training model, the first initial data set generated by a first medical imaging device is converted into a first generated data set by the first generator, then a first discriminator is used for judging whether the first generated data set accords with the style type of a second medical imaging device, the first generator, the second generator and the first discriminator are continuously traversed and circularly optimized to enable the generator and the discriminator to reach Nash equilibrium, finally the data style migration can obtain good performance effect, the first generated data set accords with the style type of the second medical imaging device, meanwhile, the first generated data set generates a second generated data set with the same style as the first initial data set by the second generator, and the second generated data set can be used as a training set for training, the problem of data shortage in the medical imaging field is solved, and meanwhile the problem of model overfitting caused by data shortage in the training process of a training model is solved.
Wherein the first generator, the second generator and the first arbiter are typically comprised of a multi-layer network comprising convolutional and/or fully-connected layers.
In the step, a second initial data set is input into a training model, the second initial data set generated by a second medical imaging device is generated by a second generator, the training model is output, the second initial data set generated by the second medical imaging device is converted into a third generated data set by the second generator, then a second discriminator is used for judging whether the third generated data set accords with the style type of the first medical imaging device, the first generator, the second generator and the second discriminator are continuously traversed and circularly optimized to enable the generator and the discriminator to reach Nash equilibrium, finally the data style migration can obtain good performance effect, the second generated data set accords with the style type of the first medical imaging device, meanwhile, the third generated data set generates a fourth generated data set with the same style as the second initial data set by the first generator, and the fourth generated data set can be used as the training set for training, the problem of data shortage in the medical imaging field is solved, and meanwhile the problem of model overfitting caused by data shortage in the training process of a training model is solved.
Wherein the first generator, the second generator and the second discriminator are typically formed by a multi-layer network comprising convolutional and/or fully-connected layers.
In the step, a style loss function and a content loss function are determined according to the first initial data set, the second initial data set, the first generator, the second generator, the first arbiter and the second arbiter, a final loss function is determined according to the style loss function and the content loss function, and adjustment and optimization are simultaneously performed from the aspects of style and content.
And step 206, performing minimization calculation on the loss function to obtain the data style migration model.
In the step, the model parameters are corrected through the loss function, so that the style of the newly generated data set is closer to the style of the initial data set output by another medical imaging device, and the style of the newly generated data set is farther from the style of the initial data set output by the original medical imaging device, thereby improving the conversion success rate. And then, the data set is retransformed to generate the style of the original medical imaging equipment, and the style of the retransformed data set is enabled to be similar to the style of the initial data set output by the original medical imaging equipment as much as possible by optimizing a loss function.
By the scheme, the built training model is trained, the first generator, the second generator, the first discriminator and the second discriminator are continuously traversed and circularly optimized, so that the generators and the discriminators can reach Nash balance, and finally, the data style migration can obtain a good performance effect. And then, the data set is retransformed to generate the style of the original medical imaging equipment, and the style of the retransformed data set is enabled to be similar to the style of the initial data set output by the original medical imaging equipment as much as possible by optimizing a loss function.
In some embodiments, the training model generates a confrontation network model for the loop.
In the step, the training model generates the confrontation network model in a circulating mode, the confrontation network model generated in a circulating mode does not need to be trained by paired data sets, and the difficulty of data collection for training the confrontation network model generated in a circulating mode is reduced.
In some embodiments, the style of the initial data set output by the medical imaging device includes at least a sampling rate, a sensitization period, and a color space.
In the step, the style of the initial data set output by the medical imaging device at least comprises a sampling rate, a sensitization period and a color space, and the effect of data style migration is ensured from multiple aspects.
In some embodiments, the medical image diagnosis and treatment product comprises at least a B-mode ultrasonic detector, a computer tomography camera and a nuclear magnetic resonance instrument.
In this step, the diagnosis generalization of various types of medical image diagnosis and treatment products can be detected.
In some embodiments, the indicators of the diagnostic result include at least a recall indicator, a precision indicator, and an F1 score indicator.
In this step, each diagnosis result is more accurate through data such as recall index, precision index, F1 score index and the like.
In some examples, the detection result of the cross-device diagnosis generalization of the medical image diagnosis and treatment product is obtained based on the change of the recall rate index, the change of the precision rate index and the change of the F1 score index corresponding to all the diagnosis results.
In this step, the detection result of the diagnosis generalization of the medical image diagnosis and treatment product is obtained according to the variation difference between the recall rate index, the precision index and the F1 score index of each diagnosis result, so that the detection result of the diagnosis generalization of the medical image diagnosis and treatment product is further more accurate.
It should be noted that the method of the embodiment of the present application may be executed by a single device, such as a computer or a server. The method of the embodiment can also be applied to a distributed scene and completed by the mutual cooperation of a plurality of devices. In such a distributed scenario, one of the multiple devices may only perform one or more steps of the method of the embodiment, and the multiple devices interact with each other to complete the method.
It should be noted that the foregoing describes some embodiments of the present application. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments described above and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
Based on the same inventive concept, the application also provides a detection device for the diagnosis generalization of the medical image diagnosis equipment, which corresponds to the method of any embodiment.
Referring to fig. 3, the apparatus for detecting cross-device generalization of medical imaging diagnosis and treatment products includes:
an initial data acquisition module 301 configured to acquire an initial data set output by the medical imaging device;
a data style migration module 302 configured to input the initial data set into at least two trained data style migration models, respectively, and output a generated data set of a corresponding style via each of the data style migration models;
a diagnosis result obtaining module 303, configured to input each of the generated data sets into a medical image diagnosis and treatment product, respectively, to obtain a diagnosis result corresponding to each of the generated data sets;
a detection result obtaining module 304, configured to perform comparative analysis on all the diagnosis results to obtain a detection result of the cross-device diagnosis generalization of the medical image diagnosis and treatment product.
In some embodiments, data style migration module 302 is specifically configured to:
constructing a training model;
acquiring a first initial data set output by a first medical imaging device, and acquiring a second initial data set output by a second medical imaging device, wherein the styles of the first initial data set and the second initial data set are different;
inputting the first initial data set into a training model, generating a first generated data set through a first generator, generating a second generated data set with the same style as the first initial data set through a second generator based on the first generated data set, and judging that the style of the first generated data set is the same as that of the second initial data set through a first discriminator;
inputting the second initial data set into the training model, generating a third generated data set via the second generator, generating a fourth generated data set by the first generator based on the second generated data set, the fourth generated data set having a same style as the second initial data set, the third generated data set determined to be the same style as the first initial data set via a second discriminator;
determining a loss function based on the first initial data set, the second initial data set, the first generator, the second generator, the first discriminator, and the second discriminator;
and carrying out minimum calculation on the loss function to obtain the data style migration model.
In some embodiments, the training model generates a confrontation network model for the loop.
In some embodiments, the style of the initial data set output by the medical imaging device includes at least a sampling rate, a sensitization period, and a color space.
In some embodiments, the medical image diagnosis and treatment product comprises at least a B-mode ultrasonic detector, a computer tomography camera and a nuclear magnetic resonance instrument.
In some embodiments, the indicators of the diagnostic result include at least a recall indicator, a precision indicator, and an F1 score indicator.
In some embodiments, the detection result of the cross-device diagnosis generalization of the medical image diagnosis and treatment product is obtained based on the change of the recall rate index, the change of the precision rate index and the change of the F1 score index corresponding to all the diagnosis results.
For convenience of description, the above devices are described as being divided into various modules by functions, and are described separately. Of course, the functionality of the various modules may be implemented in the same one or more software and/or hardware implementations as the present application.
The apparatus of the foregoing embodiment is used to implement the method for detecting the diagnosis generalization of the medical image diagnosis device in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiment, which are not described herein again.
Based on the same inventive concept, corresponding to the method of any embodiment, the present application further provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the program, the method for detecting the diagnostic generalization of the medical image diagnostic device according to any embodiment is implemented.
Fig. 4 is a schematic diagram illustrating a more specific hardware structure of an electronic device according to this embodiment, where the electronic device may include: a processor 1010, a memory 1020, an input/output interface 1030, a communication interface 1040, and a bus 1050. Wherein the processor 1010, memory 1020, input/output interface 1030, and communication interface 1040 are communicatively coupled to each other within the device via bus 1050.
The processor 1010 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an Application Specific Integrated Circuit (ASIC), or one or more Integrated circuits, and is configured to execute related programs to implement the technical solutions provided in the embodiments of the present disclosure.
The Memory 1020 may be implemented in the form of a ROM (Read Only Memory), a RAM (Random Access Memory), a static Memory device, a dynamic Memory device, or the like. The memory 1020 may store an operating system and other application programs, and when the technical solution provided by the embodiments of the present specification is implemented by software or firmware, the relevant program codes are stored in the memory 1020 and called to be executed by the processor 1010.
The input/output interface 1030 is used for connecting an input/output module to input and output information. The i/o module may be configured as a component in a device (not shown) or may be external to the device to provide a corresponding function. The input devices may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output devices may include a display, a speaker, a vibrator, an indicator light, etc.
The communication interface 1040 is used for connecting a communication module (not shown in the drawings) to implement communication interaction between the present apparatus and other apparatuses. The communication module can realize communication in a wired mode (such as USB, network cable and the like) and also can realize communication in a wireless mode (such as mobile network, WIFI, Bluetooth and the like).
It should be noted that although the above-mentioned device only shows the processor 1010, the memory 1020, the input/output interface 1030, the communication interface 1040 and the bus 1050, in a specific implementation, the device may also include other components necessary for normal operation. In addition, those skilled in the art will appreciate that the above-described apparatus may also include only those components necessary to implement the embodiments of the present description, and not necessarily all of the components shown in the figures.
The electronic device of the above embodiment is used to implement the method for detecting the diagnosis generalization of the medical image diagnosis device in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiment, which are not described herein again.
Based on the same inventive concept, corresponding to any of the above-described embodiment methods, the present application also provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute the method for detecting the diagnosis generalization of the medical image diagnosis apparatus according to any of the above-described embodiments.
Computer-readable media of the present embodiments, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device.
The computer instructions stored in the storage medium of the above embodiment are used to enable the computer to execute the method for detecting the diagnosis generalization of the medical image diagnosis device according to any one of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which are not described herein again.
Those of ordinary skill in the art will understand that: the discussion of any embodiment above is meant to be exemplary only, and is not intended to intimate that the scope of the disclosure, including the claims, is limited to these examples; within the context of the present application, features from the above embodiments or from different embodiments may also be combined, steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present application as described above, which are not provided in detail for the sake of brevity.
In addition, well-known power/ground connections to Integrated Circuit (IC) chips and other components may or may not be shown in the provided figures for simplicity of illustration and discussion, and so as not to obscure the embodiments of the application. Furthermore, devices may be shown in block diagram form in order to avoid obscuring embodiments of the application, and this also takes into account the fact that specifics with respect to implementation of such block diagram devices are highly dependent upon the platform within which the embodiments of the application are to be implemented (i.e., specifics should be well within purview of one skilled in the art). Where specific details (e.g., circuits) are set forth in order to describe example embodiments of the application, it should be apparent to one skilled in the art that the embodiments of the application can be practiced without, or with variation of, these specific details. Accordingly, the description is to be regarded as illustrative instead of restrictive.
While the present application has been described in conjunction with specific embodiments thereof, many alternatives, modifications, and variations of these embodiments will be apparent to those of ordinary skill in the art in light of the foregoing description. For example, other memory architectures (e.g., dynamic ram (dram)) may use the discussed embodiments.
The present embodiments are intended to embrace all such alternatives, modifications and variances which fall within the broad scope of the appended claims. Therefore, any omissions, modifications, substitutions, improvements, and the like that may be made without departing from the spirit and principles of the embodiments of the present application are intended to be included within the scope of the present application.
Claims (10)
1. A cross-equipment generalization detection method of medical image diagnosis and treatment products is characterized by comprising the following steps:
acquiring an initial data set output by medical imaging equipment;
respectively inputting the initial data set into at least two trained data style migration models, and outputting a generation data set with a corresponding style through each data style migration model;
inputting each generated data set in all the generated data sets into a medical image diagnosis and treatment product respectively to obtain a diagnosis result corresponding to each generated data set;
and performing comparative analysis on all the diagnosis results to obtain a detection result of the cross-equipment diagnosis generalization of the medical image diagnosis and treatment product.
2. The method of claim 1, wherein the data style migration model is trained by:
constructing a training model;
acquiring a first initial data set output by a first medical imaging device, and acquiring a second initial data set output by a second medical imaging device, wherein the styles of the first initial data set and the second initial data set are different;
inputting the first initial data set into a training model, generating a first generated data set through a first generator, generating a second generated data set with the same style as the first initial data set through a second generator based on the first generated data set, and judging that the style of the first generated data set is the same as that of the second initial data set through a first discriminator;
inputting the second initial data set into the training model, generating a third generated data set via the second generator, generating a fourth generated data set by the first generator based on the second generated data set, the fourth generated data set having a same style as the second initial data set, the third generated data set determined to be the same style as the first initial data set via a second discriminator;
determining a loss function based on the first initial data set, the second initial data set, the first generator, the second generator, the first discriminator, and the second discriminator;
and carrying out minimum calculation on the loss function to obtain the data style migration model.
3. The method of claim 2, wherein the training model is a loop-generated countermeasure network model.
4. The method of claim 1, wherein the style of the initial data set output by the medical imaging device comprises at least a sampling rate, a sensitization period, and a color space.
5. The method of claim 1, wherein the medical imaging medical product comprises at least one of a B-mode ultrasound machine, a computed tomography machine, and a magnetic resonance machine.
6. The method of claim 1, wherein the indicators of the diagnostic result comprise at least a recall indicator, a precision indicator, and an F1 score indicator.
7. The method according to claim 6, wherein the detection result of the cross-device diagnosis generalization of the medical image diagnosis and treatment product is obtained based on the change of the recall rate index, the change of the precision rate index and the change of the F1 score index corresponding to all the diagnosis results.
8. The utility model provides a product is diagnose to medical science image cross equipment generalization's detection device which characterized in that includes:
an initial data acquisition module configured to acquire an initial data set output by a medical imaging device;
a data style migration module configured to input the initial data set into at least two trained data style migration models, respectively, and output a generated data set of a corresponding style via each of the data style migration models;
the diagnosis result acquisition module is configured to input each of the generated data sets into a medical image diagnosis and treatment product respectively to obtain a diagnosis result corresponding to each of the generated data sets;
and the detection result acquisition module is configured to perform comparative analysis on all the diagnosis results to obtain a detection result of the cross-equipment diagnosis generalization of the medical image diagnosis and treatment product.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method according to any of claims 1 to 7 when executing the program.
10. A non-transitory computer readable storage medium storing computer instructions for causing a computer to perform the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210056389.6A CN114611568A (en) | 2022-01-18 | 2022-01-18 | Cross-equipment generalization detection method of medical image diagnosis and treatment product and related equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210056389.6A CN114611568A (en) | 2022-01-18 | 2022-01-18 | Cross-equipment generalization detection method of medical image diagnosis and treatment product and related equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114611568A true CN114611568A (en) | 2022-06-10 |
Family
ID=81857658
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210056389.6A Pending CN114611568A (en) | 2022-01-18 | 2022-01-18 | Cross-equipment generalization detection method of medical image diagnosis and treatment product and related equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114611568A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108389614A (en) * | 2018-03-02 | 2018-08-10 | 西安交通大学 | The method for building medical image collection of illustrative plates based on image segmentation and convolutional neural networks |
| CN109166087A (en) * | 2018-09-29 | 2019-01-08 | 上海联影医疗科技有限公司 | Style conversion method, device, medical supply, image system and the storage medium of medical image |
| CN112635013A (en) * | 2020-11-30 | 2021-04-09 | 泰康保险集团股份有限公司 | Medical image information processing method and device, electronic equipment and storage medium |
-
2022
- 2022-01-18 CN CN202210056389.6A patent/CN114611568A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108389614A (en) * | 2018-03-02 | 2018-08-10 | 西安交通大学 | The method for building medical image collection of illustrative plates based on image segmentation and convolutional neural networks |
| CN109166087A (en) * | 2018-09-29 | 2019-01-08 | 上海联影医疗科技有限公司 | Style conversion method, device, medical supply, image system and the storage medium of medical image |
| CN112635013A (en) * | 2020-11-30 | 2021-04-09 | 泰康保险集团股份有限公司 | Medical image information processing method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3963516B1 (en) | Teaching gan (generative adversarial networks) to generate per-pixel annotation | |
| CN112435341B (en) | Training method and device for three-dimensional reconstruction network, and three-dimensional reconstruction method and device | |
| CN111462020A (en) | Method, system, storage medium and device for correcting motion artifact of heart image | |
| CN114387317B (en) | CT image and MRI three-dimensional image registration method and device | |
| CN109948280B (en) | Method, device, equipment and readable storage medium for generating cuff-free blood pressure model | |
| WO2022000027A1 (en) | Identifying anomalous brain data | |
| CN111445546A (en) | Image reconstruction method and device, electronic equipment and storage medium | |
| CN112150376A (en) | Blood vessel medical image analysis method and device, computer equipment and storage medium | |
| CN113240606B (en) | Traditional Chinese medicine inspection image color correction method and system | |
| Mouches et al. | Unifying brain age prediction and age-conditioned template generation with a deterministic autoencoder | |
| US20220122257A1 (en) | Medical image processing apparatus and medical image processing system | |
| CN109147927B (en) | Man-machine interaction method, device, equipment and medium | |
| CN112419277B (en) | Three-dimensional vascular centerline synthesis method, system and storage medium | |
| US11460528B2 (en) | MRI reconstruction with image domain optimization | |
| CN114611568A (en) | Cross-equipment generalization detection method of medical image diagnosis and treatment product and related equipment | |
| CN116628600A (en) | Unbalanced data sampling method and device based on random forest | |
| CN111862255A (en) | Regularization image reconstruction method, system, readable storage medium and device | |
| US20220211283A1 (en) | Methods for blood pressure calibration selection and modeling methods thereof | |
| CN114359133A (en) | Hand bone image analysis method based on feature extraction and related equipment | |
| CN114550282A (en) | Multi-person three-dimensional attitude estimation method and device and electronic equipment | |
| CN116090458A (en) | Medical information extraction method, device, electronic equipment and storage medium | |
| CN114494484A (en) | Training method of data recognition model, data recognition method, device and equipment | |
| CN114299046A (en) | Medical image registration method, device, equipment and storage medium | |
| CN114565623A (en) | Pulmonary vessel segmentation method, device, storage medium and electronic equipment | |
| CN113706606A (en) | Method and device for determining position coordinates of spaced gestures |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220610 |
|
| RJ01 | Rejection of invention patent application after publication |