CN114566276A - Lung color Doppler ultrasound-based training method and device for child pneumonia auxiliary diagnosis model - Google Patents
Lung color Doppler ultrasound-based training method and device for child pneumonia auxiliary diagnosis model Download PDFInfo
- Publication number
- CN114566276A CN114566276A CN202210167256.6A CN202210167256A CN114566276A CN 114566276 A CN114566276 A CN 114566276A CN 202210167256 A CN202210167256 A CN 202210167256A CN 114566276 A CN114566276 A CN 114566276A
- Authority
- CN
- China
- Prior art keywords
- training
- model
- diagnosis
- database
- accuracy
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Public Health (AREA)
- Medical Informatics (AREA)
- Pathology (AREA)
- Databases & Information Systems (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Ultra Sonic Daignosis Equipment (AREA)
Abstract
The application provides a training method and a training device of a lung color Doppler ultrasound-based children pneumonia auxiliary diagnosis model, wherein the training method comprises the following steps: acquiring a diagnosis and treatment information database comprising color Doppler ultrasound images of a plurality of children pneumonia patients; acquiring a plurality of sample sets according to a database; training the training set in each sample set to obtain a training model; testing the training model according to the test set of each sample set to obtain the accuracy of the training model; determining a diagnostic model based on the accuracy of the plurality of training models; each sample set comprises a training set and a corresponding test set. The diagnosis model obtained by the training method is high in accuracy, can provide help for the diagnosis of clinicians, reduces missed diagnosis and misdiagnosis rate, can be used as a tool for the study and diagnosis of doctors, and can provide a huge driving force for the rapid growth of the doctors.
Description
Technical Field
The application relates to the technical field of medical computers, in particular to a training method and a training device of a child pneumonia auxiliary diagnosis model based on lung color Doppler ultrasound.
Background
According to the World Health Organization (WHO), infantile pneumonia is the leading cause of death in children. In 2015, WHO estimated that about 92 million children died from pneumonia worldwide, with children under 5 years of age accounting for about 15%. Pneumonia is sometimes difficult to diagnose because symptoms vary according to the age of children and the cause of infection. In addition, some symptoms are not only directed to pneumonia in children, but also to the associated clinical manifestations of other diseases. Pneumonia is used as a disease with extremely high threat to infants, once discovered, accurate diagnosis should be given, and then symptomatic treatment is carried out, wherein the earlier the diagnosis is, the higher the accuracy is, the better the diagnosis is, the improvement of the cure rate of the infant patients is facilitated, and important attention should be given to the selection of the infant diagnosis method. Particularly, the community needs accurate pneumonia diagnosis means for the diagnosis of infantile pneumonia.
However, the conventional methods for examining pediatric pneumonia mainly include chest X-ray and CT examination, and the medical images of the same chest X-ray and CT examination have great observer differences due to high inconsistency of the judgment results of different time points or different doctors.
Disclosure of Invention
In order to solve or at least partially solve the technical problem, the present application provides a training method for a lung color ultrasound-based children pneumonia auxiliary diagnosis model, which comprises: acquiring a diagnosis and treatment information database comprising color Doppler ultrasound images of a plurality of children pneumonia patients; acquiring a plurality of sample sets according to a database; training the training set in each sample set to obtain a training model; testing the training model according to the test set of each sample set to obtain the accuracy of the training model; determining a diagnostic model based on the accuracy of the plurality of training models; each sample set comprises a training set and a corresponding test set.
The training method of the lung color Doppler ultrasound-based children pneumonia auxiliary diagnosis model comprises the following steps: the diagnosis and treatment information database of the color Doppler ultrasound images of a plurality of children pneumonia patients is obtained, the source of the database can be the department of pediatrics of a certain hospital, the diagnosis information of the children pneumonia patients obtained during the diagnosis period can be provided, the database can provide data support for the training of a diagnosis model, and the accuracy of the diagnosis model is improved as much as possible.
Next, a plurality of sample sets are obtained according to the database, and for each sample set, each sample set has a training set and a corresponding test set, and one sample set may include all data of the database or only partial data. The diagnosis and treatment information in the training set is used for model training, the number of the test sets is used for accurate test of the model, the training model obtained after training is tested by the test sets, and therefore the accuracy of the training model can be obtained.
Training the training set in each sample set to obtain a training model, and then testing the training model according to the test set of each sample set to obtain the accuracy of the training model. Because the number of the sample sets is multiple, when each sample set is trained, multiple training models are obtained, and then the accuracy of the multiple training models is obtained after the multiple training models are respectively tested.
And finally, determining a diagnosis model based on the accuracy of the plurality of training models, wherein the diagnosis model has higher accuracy, can provide help for the diagnosis of doctors, reduce the working pressure of pediatricians, improve the diagnosis rate of pneumonia in hospitals at all levels, assist clinicians in reducing missed diagnosis and misdiagnosis rate, and can be used as a tool for the study and diagnosis of the doctors and provide great driving force for the rapid growth of the doctors.
It is worth mentioning that the training method of the lung color ultrasound-based children pneumonia auxiliary diagnosis model provided by the application relates to data acquisition and is not used as a diagnosis means.
Optionally, each of the clinical information includes an ultrasound image, a corresponding test result, and corresponding diagnostic information.
Optionally, the step of obtaining a database composed of diagnosis and treatment information of a plurality of children pneumonia patients specifically includes: acquiring diagnosis and treatment information of a plurality of children pneumonia patients; carrying out image preprocessing on ultrasonic images in the plurality of diagnosis and treatment information to obtain a database; the image preprocessing comprises cutting, turning, rotating and scaling processing.
Optionally, the step of obtaining a plurality of sample sets according to the database specifically includes: randomly acquiring a plurality of sample sets in a database; wherein, the proportion of the training set and the test set in any two sample sets is different.
Optionally, the step of determining the diagnostic model based on the accuracy of the plurality of training models includes: and the accuracy of the first training model is greater than that of the second training model, and the first training model is determined to be a diagnosis model.
Optionally, the network structure of the diagnostic model comprises an AlexNet network structure, a Resnet18 network structure, or a Resnet50 network structure.
The application also provides a children pneumonia aided diagnosis model's trainer based on lung color Doppler ultrasound, its characterized in that includes:
the system comprises an acquisition module, a data acquisition module and a data processing module, wherein the acquisition module is used for acquiring a diagnosis and treatment information database comprising color Doppler ultrasound images of a plurality of children pneumonia patients and acquiring a plurality of sample sets according to the database;
the training module is used for training the training set in each sample set to obtain a training model;
the test module is used for testing the training model according to the test set of each sample set to obtain the accuracy of the training model;
a determination module that determines a diagnostic model based on accuracies of the plurality of training models;
each sample set comprises a training set and a corresponding test set.
The child pneumonia auxiliary diagnosis model training device comprises an acquisition module, a training module, a testing module and a determination module, wherein the acquisition module is used for acquiring a diagnosis and treatment information database comprising color Doppler ultrasound images of a plurality of child pneumonia patients, the source of the database can be the department of pediatrics of a certain hospital, the diagnosis information of the child pneumonia patients acquired during the diagnosis period is provided, the database can provide data support for the training of the diagnosis model, and the accuracy of the diagnosis model is improved as much as possible.
Next, the obtaining module can further obtain a plurality of sample sets according to the database, each sample set has a training set and a corresponding test set, and one sample set may include all data of the database or only a part of the data. The diagnosis and treatment information in the training set is used for model training, the number of the test sets is used for accurate test of the model, the training model obtained after training is tested by the test sets, and therefore the accuracy of the training model can be obtained.
The training module can train the training set in each sample set to obtain a training model, and then the training module is used for testing the training model according to the test set of each sample set to obtain the accuracy of the training model. Because the number of the sample sets is multiple, when each sample set is trained, multiple training models are obtained, and then the accuracy of the multiple training models is obtained after the multiple training models are respectively tested.
Finally, the determining module can determine the diagnosis model based on the accuracy of the plurality of training models, namely, the training model with higher accuracy is selected from the plurality of training models to serve as the diagnosis model, the accuracy of the diagnosis model is higher, the diagnosis model can provide help for the diagnosis of doctors, the working pressure of pediatricians is reduced, the diagnosis rate of all levels of hospitals on pneumonia is improved, the missed diagnosis rate and the misdiagnosis rate of clinicians are reduced, the diagnosis model can be used as a tool for the study and diagnosis of doctors, and a great driving force can be provided for the rapid growth of the doctors.
Optionally, the obtaining module is further configured to: and randomly acquiring a plurality of sample sets in the database, wherein the proportion of the training set to the testing set in any two sample sets is different.
The present application also provides a computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein: the processor is used for executing the steps of the training method of the lung color ultrasound-based children pneumonia auxiliary diagnosis model.
In the computer device of the present application, the processor included in the computer device is configured to execute the steps of any one of the above-mentioned designs of the training method for the lung color ultrasound-based aided diagnosis model of children pneumonia, so that the computer device can achieve all the beneficial effects of the training method for the lung color ultrasound-based aided diagnosis model of children pneumonia, and details are not repeated herein.
The present application also provides a computer-readable storage medium having stored thereon a computer program characterized in that: the computer program when executed by the processor realizes the steps of the training method of the lung color ultrasound-based children pneumonia auxiliary diagnosis model.
The computer-readable storage medium in the present application, when being executed by a processor, implements the steps of the training method for a lung color ultrasound-based aided diagnosis model of children pneumonia in any design as described above, so that the computer-readable storage medium can implement all the beneficial effects of the training method, and is not described in detail herein.
Drawings
In order to more clearly illustrate the embodiments of the present application, reference will now be made briefly to the accompanying drawings. It is to be understood that the drawings in the following description are only intended to illustrate some embodiments of the present application, and that a person skilled in the art may also derive from these drawings many other technical features and connections etc. not mentioned herein.
Fig. 1 is a flowchart illustrating a training method of a lung color ultrasound-based children pneumonia auxiliary diagnosis model according to an embodiment of the present application.
Fig. 2 is a flowchart illustrating a training method of a lung color ultrasound-based children pneumonia auxiliary diagnosis model according to another embodiment of the present application.
Fig. 3 is a schematic block diagram of a training device of a lung color ultrasound-based children pneumonia auxiliary diagnosis model according to an embodiment of the present application.
FIG. 4 is a schematic block diagram of a computer device in one embodiment in accordance with the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be described in detail below with reference to the drawings in the embodiments of the present application.
The inventors of the present application have found that, in the related art, the diagnosis of pneumonia in children mainly depends on medical images, but the results of determination by different observers are highly inconsistent for the same medical image, and thus missed diagnosis and misdiagnosis may occur.
In view of this, the present application provides a training method for a children pneumonia auxiliary diagnosis model based on lung color ultrasound, so as to assist a clinician to reduce missed diagnosis and misdiagnosis rate, and simultaneously, reduce the working strength of the clinician, and improve the diagnosis accuracy and the diagnosis efficiency.
Implementation mode one
Fig. 1 is a schematic flow chart of a training method of a lung color ultrasound-based children pneumonia auxiliary diagnosis model according to an embodiment of the present application. The training method comprises the following steps:
s102, acquiring a diagnosis and treatment information database comprising color Doppler ultrasound images of a plurality of children pneumonia patients;
s104, acquiring a plurality of sample sets according to the database;
s106, training the training set in each sample set to obtain a training model;
s108, testing the training model according to the test set of each sample set to obtain the accuracy of the training model;
s110, determining a diagnosis model based on the accuracy of the training models;
each sample set comprises a training set and a corresponding test set.
The training method of the lung color Doppler ultrasound-based children pneumonia auxiliary diagnosis model comprises the following steps: the diagnosis and treatment information database of the color Doppler ultrasound images of a plurality of children pneumonia patients is obtained, the source of the database can be the department of pediatrics of a certain hospital, the diagnosis information of the children pneumonia patients obtained during the diagnosis period can be provided, the database can provide data support for the training of a diagnosis model, and the accuracy of the diagnosis model is improved as much as possible.
Next, a plurality of sample sets are obtained according to the database, and for each sample set, each sample set has a training set and a corresponding test set, and one sample set may include all data of the database or only partial data. The diagnosis and treatment information in the training set is used for model training, the number of the test sets is used for accurate test of the model, the training model obtained after training is tested by the test sets, and therefore the accuracy of the training model can be obtained.
Training the training set in each sample set to obtain a training model, and then testing the training model according to the test set of each sample set to obtain the accuracy of the training model. Because the number of the sample sets is multiple, when each sample set is trained, multiple training models are obtained, and then the accuracy of the multiple training models is obtained after the multiple training models are respectively tested.
And finally, determining a diagnosis model based on the accuracy of the plurality of training models, wherein the diagnosis model has higher accuracy, can provide help for the diagnosis of doctors, reduce the working pressure of pediatricians, improve the diagnosis rate of pneumonia in hospitals at all levels, assist clinicians in reducing missed diagnosis and misdiagnosis rate, and can be used as a tool for the study and diagnosis of the doctors and provide great driving force for the rapid growth of the doctors.
Optionally, each medical information includes an ultrasound image, a corresponding test result, and corresponding diagnostic information.
In this embodiment, the database contains a plurality of medical information, each medical information corresponds to one patient, and each medical information includes an ultrasound image, a test result and diagnosis information corresponding to each patient.
Wherein the ultrasound image is obtained using an ultrasound imaging technique. Specifically, the ultrasound imaging is to scan a human body by using ultrasound, and to receive and process a reflection signal to obtain an image of an internal organ. There are a number of commonly used ultrasound instruments: the type a (amplitude modulation type) indicates the strength of the reflected signal with the amplitude, and a "echo diagram" is shown. The ultrasonic image has the characteristics of no wound and no radioactivity, can effectively improve the condition that a patient is damaged by radiation, and has feasibility, and the ultrasonic examination has the advantages of simplicity, convenience, rapidness, multi-directionality and less limitation.
Wherein the test results comprise blood test results during hospitalization of the patient.
The diagnosis information comprises a diagnosis conclusion made by the doctor according to the ultrasonic image and the inspection result. Specifically, the diagnostic information includes pneumothorax, interstitial lung syndrome, lung consolidation, acute lung injury, respiratory distress syndrome, transient tachypnea of newborn, pneumonia and the like.
Specifically, the lung tissue of a healthy human body contains a large amount of gas and a small amount of water, the interface of the pleura and the lung tissue forms an interface between soft tissue and gas, strong reflection is generated when sound waves penetrate through the interface, the main pathological change of the infantile pneumonia disease is that the lung tissue is exudative, and the normal air inflation of the alveoli is filled with exudates, inflammatory cells and the like, which is the basis of ultrasonic pneumonia diagnosis.
When the lung is inflamed, the air content in the lung is reduced, the exudation of the inflammation is increased, the acoustic impedance between gas and liquid is increased, and the ultrasound generates strong reverberation at the gas-liquid junction to form a comet tail which extends to a far field to be a B line. The number of B-lines depends on the degree of pulmonary ventilation loss, with the intensity of the echo increasing with inspiratory motion. Among them, a plurality of B lines with a pitch of 3mm or less, called B3 lines, which are associated with the frosty glass shadows displayed by CT of the chest, indicate the possibility of alveolar pulmonary edema, and a plurality of B lines with a pitch of > 7mm, called B7 lines, indicate the possibility of thickening of the lobular spaces. If the disease condition of the infant patient is developed until the gas in the alveolus disappears and a large amount of exudates such as fibrin, red blood cells, white blood cells and the like are filled, the disease becomes the pathological basis of the lung excess change through ultrasonic diagnosis. The ultrasonic examination provides conditions for the ultrasonic examination when the consolidation lung is in direct contact with the pleura or passes through water to form an acoustic window.
Second embodiment
Fig. 2 is a schematic flow chart of a training method of a lung color ultrasound-based children pneumonia auxiliary diagnosis model according to another embodiment of the present application. The training method comprises the following steps:
s202, acquiring diagnosis and treatment information of a plurality of children pneumonia patients;
s204, carrying out image preprocessing on the ultrasonic images in the plurality of diagnosis and treatment information to obtain a database;
s206, acquiring a plurality of sample sets according to the database;
s208, training the training set in each sample set to obtain a training model;
s210, testing the training model according to the test set of each sample set to obtain the accuracy of the training model;
s212, determining a diagnosis model based on the accuracy of the plurality of training models.
In this embodiment, the step of obtaining a database including diagnosis and treatment information of a plurality of children pneumonia patients specifically includes: the diagnosis and treatment information of a plurality of children pneumonia patients is obtained, and the diagnosis and treatment information of each children pneumonia patient comprises an ultrasonic image, a corresponding test result and corresponding diagnosis information. In the subsequent training process, a sample set composed of a plurality of diagnosis and treatment information is used as a training basis, and therefore, the accuracy of the diagnosis and treatment information is particularly important.
The examination result and the diagnosis information in the diagnosis and treatment information are both composed of data texts, the precision is high, the training processing can be directly carried out, however, due to the fact that the ultrasonic images are limited by different imaging devices or are influenced by factors such as the operation proficiency of imaging doctors, the precision of the ultrasonic images has a space for improving.
In this embodiment, the ultrasound images in each diagnosis and treatment information are subjected to image preprocessing, so that noise in the ultrasound images can be reduced, data defects of the ultrasound images are made up, the accuracy of the ultrasound images is improved, and the efficiency of training processing is improved.
It should be noted that, in an ultrasound image, the main noise is speckle (speckle), which is caused by interference due to scattering of an acoustic beam on uneven fine tissue, and the noise appears as particles in the image, does not reflect the actual tissue structure, but affects the detail resolution capability of the image. This is not conducive to quantitative analysis of the image, and therefore speckle noise in the image needs to be suppressed.
Optionally, the image pre-processing comprises cropping, flipping, rotating, scaling processing.
Before the training process is carried out, in order to reduce the workload of data calculation and improve the training speed, the original picture can be cut for the interested characteristic part, so that each ultrasonic image not only retains the characteristic set to be extracted, but also reduces the overall size, the time consumed by the training model can be effectively shortened, and the training efficiency is improved.
Optionally, the step of obtaining a plurality of sample sets according to the database specifically includes: randomly acquiring a plurality of sample sets in a database; wherein, the proportion of the training set and the test set in any two sample sets is different.
In this embodiment, the step of obtaining a plurality of sample sets according to a database specifically includes: a plurality of sample sets are randomly obtained from the database, so that the characteristic properties of the training set in each sample set during training are ensured as much as possible, different training models are obtained as much as possible, and more choices are provided for obtaining a final diagnosis model.
Optionally, the proportion of the training set and the test set in any two sample sets is different, that is, the composition of any two sample sets is different, and in this case, the obtained training models for the two sample sets are also different.
For example, a first sample set, a second sample set and a third sample set are obtained from a database, wherein the ratio of a training set to a test set in the first sample set is 5:5, the ratio of the training set to the test set in the second sample set is 8:2, and the ratio of the training set to the test set in the third sample set is 9:1, then a first training model can be obtained by training the test set of the first sample set, a second training model can be obtained by training the test set of the second sample set, a third training model can be obtained by training the test set of the third sample set, then, the accuracy of the obtained training models is tested through the respective test sets to obtain the respective corresponding accuracy, and the diagnostic model is determined according to the accuracy of the multiple training models.
When the diagnosis model is applied to the actual ultrasonic examination process, the lung ultrasonic image of the patient is input into the diagnosis model, so that the diagnosis model can assist a doctor to reduce missed diagnosis and misdiagnosis rate and achieve the purposes of early diagnosis and early treatment.
Optionally, the step of determining the diagnostic model based on the accuracy of the plurality of training models includes: and the accuracy of the first training model is greater than that of the second training model, and the first training model is determined to be a diagnosis model.
In this embodiment, the step of determining the diagnostic model based on the accuracy of the plurality of training models includes: and the accuracy of the first training model is higher than that of the second training model, the first training model is determined to be a diagnosis model, and the diagnosis model is a training model with higher accuracy.
The higher accuracy of the training model indicates that the efficiency is excellent, and the diagnostic information can be well obtained according to the ultrasonic image, so that a clinician can be effectively assisted, and the working intensity is reduced.
It should be noted that the number of the plurality of training models may be a positive integer, without setting an upper limit, the number of the sample sets may be selected according to an actual situation, so as to obtain a corresponding number of training models, and then the accuracy is preferentially selected from the plurality of training models to be a final diagnosis model.
Optionally, the network structure of the diagnostic model comprises an AlexNet network structure, a Resnet18 network structure, or a Resnet50 network structure.
Third embodiment
Fig. 3 shows a schematic block diagram of a training device of a child pneumonia auxiliary diagnosis model based on a lung color Doppler ultrasound according to an embodiment of the present application. As shown in fig. 3, the training apparatus 300 includes:
an obtaining module 302, configured to obtain a diagnosis and treatment information database including color Doppler ultrasound images of multiple children pneumonia patients, and obtain multiple sample sets according to the database;
a training module 304, configured to perform training processing on a training set in each sample set to obtain a training model;
the test module 306 is used for testing the training model according to the test set of each sample set to obtain the accuracy of the training model;
a determination module 308 that determines a diagnostic model based on the accuracy of the plurality of training models;
each sample set comprises a training set and a corresponding test set.
The training device 300 of the lung color ultrasound-based auxiliary diagnosis model for children pneumonia in the application comprises an acquisition module 302, a training module 304, a testing module 306 and a determination module 308, wherein the acquisition module 302 is used for acquiring a diagnosis and treatment information database comprising color ultrasound images of a plurality of children pneumonia patients, the source of the database can be the department of pediatrics of a certain hospital, the children pneumonia patients can see the doctor during the doctor, the database can provide data support for the training of the diagnosis model, and the accuracy of the diagnosis model is improved as much as possible.
Next, the obtaining module 302 can further obtain a plurality of sample sets according to the database, where each sample set has a training set and a corresponding testing set, and a sample set may include all data of the database or only a part of the data. The diagnosis and treatment information in the training set is used for model training, the number of the test sets is used for accurate test of the model, the training model obtained after training is tested by the test sets, and therefore the accuracy of the training model can be obtained.
The training module 304 can perform training processing on the training set in each sample set to obtain a training model, and then the testing module 306 tests the training model according to the testing set of each sample set to obtain the accuracy of the training model. Because the number of the sample sets is multiple, when each sample set is trained, multiple training models are obtained, and then the accuracy of the multiple training models is obtained after the multiple training models are respectively tested.
Finally, the determining module 308 can determine a diagnosis model based on the accuracy of the training models, the diagnosis model has high accuracy, can provide help for the diagnosis of doctors, reduce the working pressure of pediatricians, improve the diagnosis rate of pneumonia in hospitals at all levels, assist clinicians in reducing missed diagnosis and misdiagnosis rates, can also use the diagnosis model as a tool for the study and diagnosis of doctors, and can also provide great driving force for the rapid growth of doctors.
Optionally, the obtaining module 302 is further configured to: and randomly acquiring a plurality of sample sets in the database, wherein the proportion of the training set to the testing set in any two sample sets is different.
Optionally, each of the clinical information includes an ultrasound image, a corresponding test result, and corresponding diagnostic information.
Optionally, the obtaining module 302 is further configured to: acquiring diagnosis and treatment information of a plurality of children pneumonia patients; carrying out image preprocessing on ultrasonic images in the plurality of diagnosis and treatment information to obtain a database; the image preprocessing comprises cutting, turning, rotating and zooming.
Optionally, the plurality of training models includes a first training model and a second training model, and the determining module 308 is further configured to: and the accuracy of the first training model is greater than that of the second training model, and the first training model is determined to be a diagnosis model.
Optionally, the network structure of the diagnostic model comprises an AlexNet network structure, a Resnet18 network structure, or a Resnet50 network structure.
Embodiment IV
Fig. 4 shows a schematic block diagram of a computer device 400 in an embodiment of the present application. The computer device 400 includes a memory 402, a processor 404, and a computer program stored in the memory 402 and executable on the processor 404, wherein: the processor 404 is configured to perform the steps of the training method for the lung color ultrasound-based aided diagnosis model of pneumonia in children as described above.
In the computer device 400 of the present application, the processor 404 included in the computer device is configured to execute the steps of any one of the above-mentioned methods for training the lung color ultrasound-based aided diagnosis model of children pneumonia, so that the computer device 400 can achieve all the beneficial effects of the method for training the lung color ultrasound-based aided diagnosis model of children pneumonia, and details thereof are not repeated herein.
Fifth embodiment
The present application also provides a computer-readable storage medium having stored thereon a computer program characterized in that: the computer program when executed by the processor realizes the steps of the training method of the lung color ultrasound-based children pneumonia auxiliary diagnosis model.
The computer-readable storage medium in the present application, when being executed by a processor, implements the steps of the training method for a lung color ultrasound-based aided diagnosis model of children pneumonia in any design as described above, so that the computer-readable storage medium can implement all the beneficial effects of the training method, and is not described in detail herein.
It will be evident to those skilled in the art that the application is not limited to the details of the foregoing illustrative embodiments, and that the present application may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.
Claims (10)
1. A training method of a lung color Doppler ultrasound-based children pneumonia auxiliary diagnosis model is characterized by comprising the following steps:
acquiring a diagnosis and treatment information database comprising color Doppler ultrasound images of a plurality of children pneumonia patients;
acquiring a plurality of sample sets according to the database;
training the training set in each sample set to obtain a training model;
testing the training model according to the test set of each sample set to obtain the accuracy of the training model;
determining the diagnostic model based on the accuracy of a plurality of the training models;
each sample set comprises a training set and a corresponding test set.
2. Training method according to claim 1,
each diagnosis and treatment information comprises an ultrasonic image, a corresponding test result and corresponding diagnosis information.
3. A training method according to claim 1, wherein the step of obtaining a database of diagnosis and treatment information of a plurality of children pneumonia patients specifically comprises:
acquiring diagnosis and treatment information of a plurality of children pneumonia patients;
carrying out image preprocessing on the ultrasonic images in the diagnosis and treatment information to obtain a database;
the image preprocessing comprises cutting, turning, rotating and zooming.
4. The training method according to claim 1, wherein the step of obtaining a plurality of sample sets from the database specifically comprises:
randomly acquiring a plurality of sample sets in the database;
and the proportion of the training set to the test set in any two sample sets is different.
5. The training method of claim 1, wherein the plurality of training models comprises a first training model and a second training model, and wherein the step of determining the diagnostic model based on the accuracy of the plurality of training models comprises:
and the accuracy of the first training model is greater than that of the second training model, and the first training model is determined to be the diagnosis model.
6. Training method according to claim 1,
the network structure of the diagnostic model comprises an AlexNet network structure, a Resnet18 network structure or a Resnet50 network structure.
7. A children pneumonia aided diagnosis model training device based on lung color Doppler ultrasound is characterized by comprising:
the system comprises an acquisition module, a data acquisition module and a data processing module, wherein the acquisition module is used for acquiring a diagnosis and treatment information database comprising color Doppler ultrasound images of a plurality of children pneumonia patients and acquiring a plurality of sample sets according to the database;
the training module is used for training the training set in each sample set to obtain a training model;
the testing module is used for testing the training model according to the testing set of each sample set to obtain the accuracy of the training model;
a determination module that determines the diagnostic model based on an accuracy of a plurality of the training models;
each sample set comprises a training set and a corresponding test set.
8. The training device of claim 7,
the acquisition module is further configured to: randomly acquiring a plurality of sample sets in the database, wherein the proportion of a training set to a testing set in any two sample sets is different.
9. A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein: the processor is used for executing the steps of the training method of the lung color ultrasound-based children pneumonia auxiliary diagnosis model according to any one of claims 1 to 6.
10. A computer-readable storage medium having stored thereon a computer program, characterized in that: the computer program when executed by a processor realizes the steps of the training method of the lung color ultrasound-based children pneumonia auxiliary diagnosis model according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210167256.6A CN114566276A (en) | 2022-02-23 | 2022-02-23 | Lung color Doppler ultrasound-based training method and device for child pneumonia auxiliary diagnosis model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210167256.6A CN114566276A (en) | 2022-02-23 | 2022-02-23 | Lung color Doppler ultrasound-based training method and device for child pneumonia auxiliary diagnosis model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114566276A true CN114566276A (en) | 2022-05-31 |
Family
ID=81714527
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210167256.6A Pending CN114566276A (en) | 2022-02-23 | 2022-02-23 | Lung color Doppler ultrasound-based training method and device for child pneumonia auxiliary diagnosis model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114566276A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11977154B2 (en) | 2016-10-28 | 2024-05-07 | Ppg Industries Ohio, Inc. | Coatings for increasing near-infrared detection distances |
| US12001034B2 (en) | 2019-01-07 | 2024-06-04 | Ppg Industries Ohio, Inc. | Near infrared control coating, articles formed therefrom, and methods of making the same |
| US12050950B2 (en) | 2018-11-13 | 2024-07-30 | Ppg Industries Ohio, Inc. | Method of detecting a concealed pattern |
-
2022
- 2022-02-23 CN CN202210167256.6A patent/CN114566276A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11977154B2 (en) | 2016-10-28 | 2024-05-07 | Ppg Industries Ohio, Inc. | Coatings for increasing near-infrared detection distances |
| US12050950B2 (en) | 2018-11-13 | 2024-07-30 | Ppg Industries Ohio, Inc. | Method of detecting a concealed pattern |
| US12001034B2 (en) | 2019-01-07 | 2024-06-04 | Ppg Industries Ohio, Inc. | Near infrared control coating, articles formed therefrom, and methods of making the same |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114566276A (en) | Lung color Doppler ultrasound-based training method and device for child pneumonia auxiliary diagnosis model | |
| Demi | Lung ultrasound: The future ahead and the lessons learned from COVID-19 | |
| Mento et al. | On the influence of imaging parameters on lung ultrasound B-line artifacts, in vitro study | |
| JP4402436B2 (en) | Method and system for airway measurement | |
| Musolino et al. | Ten years of pediatric lung ultrasound: a narrative review | |
| Tonni et al. | Role of 3-D ultrasound in clinical obstetric practice: evolution over 20 years | |
| Abuhamad et al. | First trimester ultrasound diagnosis of fetal abnormalities | |
| CN111134619B (en) | Multi-mode photoacoustic/ultrasonic imaging rheumatoid arthritis scoring system and application | |
| Ammirabile et al. | Lung ultrasound in pediatrics and neonatology: an update | |
| Chen et al. | Application of lung ultrasonography in the diagnosis of childhood lung diseases | |
| JP2000107185A (en) | Ultrasonograph | |
| Hata et al. | Four‐dimensional volume‐rendered imaging of the fetal ventricular outflow tracts and great arteries using inversion mode for detection of congenital heart disease | |
| Ma et al. | A computer-aided diagnosis scheme for detection of fatty liver in vivo based on ultrasound kurtosis imaging | |
| Ticinesi et al. | The geriatric patient: the ideal one for chest ultrasonography? A review from the chest ultrasound in the elderly Study Group (GRETA) of the Italian Society of gerontology and geriatrics (SIGG) | |
| Ge et al. | Clinical value of real‑time elastography quantitative parameters in evaluating the stage of liver fibrosis and cirrhosis | |
| Sandberg et al. | Ultrasound shear wave elastography: does it add value to gray-scale ultrasound imaging in differentiating biliary atresia from other causes of neonatal jaundice? | |
| Cheong et al. | Comparison of inter‐and intraobserver agreement between three types of fetal volume measurement technique (XI VOCAL™, VOCAL™ and multiplanar) | |
| Ostras et al. | Diagnostic ultrasound imaging of the lung: A simulation approach based on propagation and reverberation in the human body | |
| US20120078101A1 (en) | Ultrasound system for displaying slice of object and method thereof | |
| JP5390149B2 (en) | Ultrasonic diagnostic apparatus, ultrasonic diagnostic support program, and image processing apparatus | |
| Schenk et al. | Measurement of real-time tissue elastography in a phantom model and comparison with transient elastography in pediatric patients with liver diseases | |
| You et al. | Reference ranges of fetal spleen biometric parameters and volume assessed by three‐dimensional ultrasound and their applicability in spleen malformations | |
| Sklansky | New dimensions and directions in fetal cardiology | |
| Leote et al. | Influence of ultrasound settings on laboratory vertical artifacts | |
| Maged et al. | The role of three-dimensional ultrasonography fetal lung volume measurement in the prediction of neonatal respiratory function outcome |
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 |