CN111951976A - Value judgment method, system, terminal and medium based on medical data margin - Google Patents
Value judgment method, system, terminal and medium based on medical data margin Download PDFInfo
- Publication number
- CN111951976A CN111951976A CN202010849411.3A CN202010849411A CN111951976A CN 111951976 A CN111951976 A CN 111951976A CN 202010849411 A CN202010849411 A CN 202010849411A CN 111951976 A CN111951976 A CN 111951976A
- Authority
- CN
- China
- Prior art keywords
- prediction model
- medical data
- value
- data
- characteristic variable
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000011160 research Methods 0.000 claims abstract description 61
- 238000012549 training Methods 0.000 claims description 43
- 230000015654 memory Effects 0.000 claims description 15
- 201000010099 disease Diseases 0.000 claims description 8
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 7
- 206010020751 Hypersensitivity Diseases 0.000 claims description 5
- 208000026935 allergic disease Diseases 0.000 claims description 5
- 230000007815 allergy Effects 0.000 claims description 5
- 229940079593 drug Drugs 0.000 claims description 5
- 239000003814 drug Substances 0.000 claims description 5
- 239000013589 supplement Substances 0.000 claims description 2
- 238000005065 mining Methods 0.000 abstract description 13
- 238000003745 diagnosis Methods 0.000 abstract description 9
- 230000000694 effects Effects 0.000 abstract description 6
- 238000005516 engineering process Methods 0.000 abstract description 6
- 239000000284 extract Substances 0.000 abstract description 4
- 210000004279 orbit Anatomy 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 210000003484 anatomy Anatomy 0.000 description 3
- 208000018087 Orbital disease Diseases 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 238000004393 prognosis Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 230000009469 supplementation Effects 0.000 description 1
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
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
-
- 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
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Epidemiology (AREA)
- Data Mining & Analysis (AREA)
- Primary Health Care (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Pathology (AREA)
- Databases & Information Systems (AREA)
- Medical Treatment And Welfare Office Work (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The value judgment method, the system, the terminal and the medium based on the medical data allowance solve the problems that in the prior art, a solution for further analyzing and mining the value of the medical data is incomplete, a large amount of research is blindly carried out, and social resources are wasted. The invention extracts massive information in medical data by using an information mining technology, comprehensively analyzes and utilizes the data from multiple dimensions, and mines the data which is not effectively utilized before, thereby further improving the medical research level, and mining more information in the existing data to try to deeply interpret and search the internal rules of the data, the rules among the data and even the rules between the data and the specific diagnosis/treatment effect.
Description
Technical Field
The present invention relates to the medical field, and in particular, to a value determination method, system, terminal, and medium based on medical data margin.
Background
The medical data comprises patient demographic information, examination data, inspection data and the like, and the information can play a role in assisting a doctor in decision making during diagnosis, treatment effect evaluation and prognosis judgment. With the development of technologies such as artificial intelligence and the like, the existing medical data are gradually and deeply read and widely utilized, the fields of intelligent diagnosis, intelligent decision making and the like are promoted, and the medical progress is promoted. However, a large amount of manpower and material resources need to be invested in the research and development process in the field, but an effective solution for further analyzing and mining the value of medical data in the prior art is unavailable, so that a large amount of research is blindly carried out, and social resources are wasted.
With the wide development of medical artificial intelligence research, more and more teams invest in manpower and material resources to carry out AI research for auxiliary diagnosis and treatment. The AI technology can be used to find the information margin which is not mined in the clinical data to improve the accuracy of the diagnosis and treatment model, which is an important prepositive step for carrying out large-scale data acquisition, labeling and AI model training.
Disclosure of Invention
In view of the above drawbacks of the prior art, an object of the present invention is to provide a method, a system, a terminal and a medium for value determination based on medical data margin, which are used to solve the problem that the prior art has an imperfect solution for further analyzing and mining the value of medical data, thereby resulting in a large amount of blindly developed research and wasting social resources.
In order to achieve the above objects and other related objects, the present invention provides a method for value determination based on medical data margin, comprising: performing label-free clustering on the acquired medical data, and taking the category corresponding to each cluster as a characteristic variable of the medical data, wherein the characteristic variable comprises each known characteristic variable and one or more margin characteristic variables except each known characteristic variable; obtaining a new prediction model of the characteristic variables corresponding to the medical data based on a retrospective research prediction model obtained by training of each known characteristic variable in a general data database; comparing the prediction results obtained respectively based on the new prediction model and the retrospective study prediction model to obtain a comparison difference value; if the comparison difference value is larger than a preset difference threshold value, predicting the characteristic variable of the medical data by using the new prediction model to obtain a prediction result of correspondingly judging high research value of the residual characteristic variable; if the comparison difference value is not larger than the difference threshold value, obtaining a to-be-sampled supplementary result corresponding to the situation that the sample is insufficient based on the sample quantity judgment condition, and/or obtaining a prediction result with low research value of the corresponding judgment margin characteristic variable by predicting the characteristic variable of the medical data by using the retrospective research prediction model corresponding to the situation that the sample is sufficient.
In an embodiment of the present invention, the obtaining a new prediction model of the feature variables corresponding to the medical data based on the retrospective study prediction model trained from each known feature variable in the general data database includes: training the retrospective research prediction model obtained by training the known characteristic variables in a general data database by using the characteristic variables extracted from the medical data and the association degree information between the characteristic variables as training data to obtain a new prediction model.
In an embodiment of the invention, the training mode of the retrospective study prediction model includes: and training a prediction model by using the known characteristic variables extracted from the general data in the general data database and the correlation degree information among the known characteristic variables as training data to obtain the retrospective research prediction model.
In an embodiment of the invention, the comparing the prediction results obtained based on the new prediction model and the retrospective study prediction model respectively to obtain a comparison difference value includes: and comparing the prediction results respectively obtained by the new prediction model and the retrospective prediction model which input the same known characteristic variable to obtain a comparison difference value.
In an embodiment of the present invention, the sample number determination condition includes: comparing the number of samples with the characteristic variables of the medical data with a preset sample number threshold; if the number of samples with the characteristic variables of the medical data is greater than or equal to the sample number threshold value, determining that the samples are sufficient; and if the number of the samples with the characteristic variables of the medical data is less than the sample number threshold value, determining that the samples are insufficient.
To achieve the above and other related objects, the present invention provides a value determination system based on a remaining amount of medical data, the system comprising: the characteristic variable acquisition module is used for performing label-free clustering on the acquired medical data and taking the category corresponding to each cluster as a characteristic variable of the medical data, wherein the characteristic variable comprises each known characteristic variable and one or more margin characteristic variables except for each known characteristic variable; the new prediction model module is connected with the characteristic variable acquisition module and used for acquiring a new prediction model of the characteristic variable corresponding to the medical data based on a retrospective research prediction model acquired by training of each known characteristic variable in a general data database; the comparison module is connected with the new prediction model module and is used for comparing prediction results obtained respectively based on the new prediction model and the retrospective research prediction model to obtain a comparison difference value; the residual value judgment high module is connected with the comparison module and used for predicting the characteristic variable of the medical data by using the new prediction model to obtain a prediction result corresponding to the residual value judgment high research value of the characteristic variable if the comparison difference value is larger than a preset difference threshold; and the judgment residual amount value low module is connected with the comparison module and is used for obtaining a to-be-sample supplement result corresponding to the condition that the sample is insufficient based on the sample quantity judgment condition and/or a prediction result corresponding to the condition that the sample is sufficient and obtained by predicting the characteristic variable of the medical data by using the retrospective research prediction model, wherein the prediction result corresponds to the condition that the sample quantity judgment condition is not larger than the difference threshold value and is low in research value of the characteristic variable of the judgment residual amount.
In an embodiment of the present invention, the obtaining a new prediction model of the feature variables corresponding to the medical data based on the retrospective study prediction model trained from each known feature variable in the general data database includes: training the retrospective research prediction model obtained by training the known characteristic variables in a general data database by using the characteristic variables extracted from the medical data and the association degree information between the characteristic variables as training data to obtain a new prediction model.
To achieve the above and other related objects, the present invention provides a value determination terminal based on medical data margin, comprising: a memory for storing a computer program; and the processor is used for executing the value judgment method based on the medical data allowance.
To achieve the above and other related objects, the present invention provides a computer storage medium storing a computer program, which when executed, implements the method for value determination based on remaining amount of medical data.
As described above, the value determination method, system, terminal and medium based on medical data margin according to the present invention have the following advantages: the invention extracts massive information in medical data by using an information mining technology, comprehensively analyzes and utilizes the data from multiple dimensions, and mines the data which is not effectively utilized before, thereby further improving the medical research level, and mining more information in the existing data to try to deeply interpret and search the internal rules of the data, the rules among the data and even the rules between the data and the specific diagnosis/treatment effect.
Drawings
Fig. 1 is a flow chart illustrating a method for determining a value based on a remaining amount of medical data according to an embodiment of the invention.
Fig. 2 is a flowchart illustrating a method for determining a value based on a residual amount of medical data of an orbital disease according to an embodiment of the invention.
FIG. 3 is a block diagram of a system for value determination based on residual medical data according to an embodiment of the present invention.
Fig. 4 is a schematic structural diagram of a value determination terminal based on the remaining amount of medical data according to an embodiment of the invention.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict.
It is noted that in the following description, reference is made to the accompanying drawings which illustrate several embodiments of the present invention. It is to be understood that other embodiments may be utilized and that mechanical, structural, electrical, and operational changes may be made without departing from the spirit and scope of the present invention. The following detailed description is not to be taken in a limiting sense, and the scope of embodiments of the present invention is defined only by the claims of the issued patent. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. Spatially relative terms, such as "upper," "lower," "left," "right," "lower," "below," "lower," "over," "upper," and the like, may be used herein to facilitate describing one element or feature's relationship to another element or feature as illustrated in the figures.
Throughout the specification, when a part is referred to as being "connected" to another part, this includes not only a case of being "directly connected" but also a case of being "indirectly connected" with another element interposed therebetween. In addition, when a certain part is referred to as "including" a certain component, unless otherwise stated, other components are not excluded, but it means that other components may be included.
The terms first, second, third, etc. are used herein to describe various elements, components, regions, layers and/or sections, but are not limited thereto. These terms are only used to distinguish one element, component, region, layer or section from another element, component, region, layer or section. Thus, a first element, component, region, layer or section discussed below could be termed a second element, component, region, layer or section without departing from the scope of the present invention.
Also, as used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context indicates otherwise. It will be further understood that the terms "comprises," "comprising," and/or "comprising," when used in this specification, specify the presence of stated features, operations, elements, components, items, species, and/or groups, but do not preclude the presence, or addition of one or more other features, operations, elements, components, items, species, and/or groups thereof. The terms "or" and/or "as used herein are to be construed as inclusive or meaning any one or any combination. Thus, "A, B or C" or "A, B and/or C" means "any of the following: a; b; c; a and B; a and C; b and C; A. b and C ". An exception to this definition will occur only when a combination of elements, functions or operations are inherently mutually exclusive in some way.
The embodiment of the invention provides a value judgment method based on medical data allowance, and solves the problems that in the prior art, the solution for further analyzing and mining the value of medical data is imperfect, a large amount of research is blindly developed, and social resources are wasted. The invention extracts massive information in medical data by using an information mining technology, comprehensively analyzes and utilizes the data from multiple dimensions, and mines the data which is not effectively utilized before, thereby further improving the medical research level, and mining more information in the existing data to try to deeply interpret and search the internal rules of the data, the rules among the data and even the rules between the data and the specific diagnosis/treatment effect.
Embodiments of the present invention will be described in detail below with reference to the accompanying drawings so that those skilled in the art can easily implement the embodiments of the present invention. The present invention may be embodied in many different forms and is not limited to the embodiments described herein.
Fig. 1 shows a flow chart of a value determination method based on medical data margin according to an embodiment of the present invention.
The method comprises the following steps:
step S11: and performing label-free clustering on the acquired medical data, and taking the category corresponding to each cluster as a characteristic variable of the medical data, wherein the characteristic variable comprises each known characteristic variable and one or more margin characteristic variables except each known characteristic variable.
Optionally, the obtained medical data is subjected to label-free clustering, and the category corresponding to each cluster is respectively used as a characteristic variable of the medical data. Wherein the characteristic variables include known characteristic variables and one or more margin characteristic variables other than the known characteristic variables.
Optionally, the known characteristic variables are characteristic variables commonly used in the prior retrospective research method.
Alternatively, the medical data may be all examination information and/or self-information related to a patient, wherein the examination information is medical examination information in any field, and is not limited in the present invention.
Optionally, the medical data includes: gender, age, race, duration of disease, treatment history, other ailments, birth history, allergy history, surgical history, medication history, and treatment history data.
Step S12: and obtaining a new prediction model of the characteristic variables corresponding to the medical data based on the retrospective research prediction model obtained by training the known characteristic variables in the general data database.
Optionally, the obtaining a new prediction model of the feature variables corresponding to the medical data based on the retrospective research prediction model obtained by training of each known feature variable in the general data database includes: training the retrospective research prediction model obtained by training the known characteristic variables in a general data database by using the characteristic variables extracted from the medical data and the association degree information between the characteristic variables as training data to obtain a new prediction model.
Specifically, each feature variable extracted from the medical data and the information of the degree of association between the feature variables are input as training data into the retrospective study prediction model obtained by training with each piece of known feature variable information in a general data database, and a new prediction model related to each known feature variable and one or more residual feature variables other than each known feature variable is obtained.
Optionally, the relationship between the characteristic variables is abstracted and summarized, when some characteristic variable factor changes, other characteristic variable factors affecting the characteristic variable factor can be predicted, and then a new prediction model is obtained through deep learning of a computer and the high-dimensional characteristic based on the retrospective research prediction model.
Optionally, the general data may be a set of public medical knowledge and/or medical record information in one or more medical fields, which is not limited in the present invention.
Optionally, the training mode of the retrospective study prediction model includes: and training a prediction model by using the known characteristic variables extracted from the general data in the general data database and the correlation degree information among the known characteristic variables as training data to obtain the retrospective research prediction model.
Specifically, the relationship between different known characteristic variables is abstracted and summarized through massive general data, and when one known characteristic variable factor is changed, other known characteristic variable factors influencing the known characteristic variable factor can be predicted to obtain a retrospective prediction model.
Optionally, the general data database includes: gender, age, race, duration of disease, treatment history, other ailments, birth history, allergy history, surgical history, medication history, and treatment history data.
Step S13: and comparing the prediction results obtained respectively based on the new prediction model and the retrospective research prediction model to obtain a comparison difference value.
Optionally, the comparing the prediction results obtained based on the new prediction model and based on the retrospective study prediction model respectively to obtain a comparison difference value includes: and comparing the prediction results respectively obtained by the new prediction model and the retrospective prediction model which input the same known characteristic variable to obtain a comparison difference value.
Optionally, the new prediction model and the retrospective prediction model which input the same known characteristic variable respectively obtain prediction results related to the characteristic variable associated with the known characteristic variable, and compare the prediction results to obtain a comparison difference value. Wherein, based on the new prediction model, other known characteristic variables associated with the known characteristic variable and the prediction result related to the residual characteristic can be obtained; based on the retrospective predictive model, a prediction result related to other known characteristic variables associated with the known characteristic variable may be obtained.
Step S14: if the comparison difference value is larger than a preset difference threshold value, predicting the characteristic variable of the medical data by using the new prediction model to obtain a prediction result with high research value of the corresponding judgment allowance characteristic variable.
Optionally, the preset difference threshold is determined according to specific requirements, and is not limited in the present invention.
Optionally, if the comparison difference value is greater than a preset difference threshold value, it is indicated that the residual characteristic variable of the medical data is large and the research value of the residual characteristic variable is high, and the feature variable of the medical data needs to be predicted by using the new prediction model including the residual characteristic variable and each known feature to obtain a prediction result, so as to mine residual information.
Step S15: if the comparison difference value is not larger than the difference threshold value, obtaining a to-be-sampled supplementary result corresponding to the situation that the sample is insufficient based on the sample quantity judgment condition, and/or obtaining a prediction result with low research value of the corresponding judgment margin characteristic variable by predicting the characteristic variable of the medical data by using the retrospective research prediction model corresponding to the situation that the sample is sufficient.
Optionally, if the comparison difference value is less than or equal to the difference threshold, first determining whether the result difference is caused by insufficient samples based on the sample number determination condition, and if the samples are insufficient, obtaining a result to be supplemented to the sample corresponding to the case of insufficient samples, so as to perform re-verification after the samples are sufficient; if the sample is sufficient, the residual characteristics of the medical data are few and the research value of the residual characteristic variables is low, a prediction result obtained by predicting the characteristic variables of the medical data by using the retrospective research prediction model under the condition that the sample is sufficient is obtained, namely, the residual information does not need to be further mined for the current medical data.
Optionally, the sample number judging condition includes: comparing the number of samples with the characteristic variables of the medical data with a preset sample number threshold; if the number of samples with the characteristic variables of the medical data is greater than or equal to the sample number threshold value, determining that the samples are sufficient; and if the number of the samples with the characteristic variables of the medical data is less than the sample number threshold value, determining that the samples are insufficient.
In order to better describe the value judgment method based on the medical data margin, a specific embodiment is provided.
Example 1: fig. 2 is a schematic flow chart of the method for determining the value based on the residual amount of the orbital disease medical data.
The method comprises the following steps:
performing label-free clustering on the acquired orbit disease medical data, and taking the category corresponding to each cluster as a characteristic variable of the medical data, wherein the characteristic variable comprises each known characteristic variable and one or more margin characteristic variables except each known characteristic variable; wherein the margin characteristic variable comprises: acquiring orbit key point set space position characteristic data, a two-dimensional image high-dimensional characteristic parameter automatically identified and labeled by a meridian anatomical structure, a demographic index and a clinical examination index; TAO orbital CT high-dimensional characteristic parameters and TAO orbital MRI high-dimensional characteristic parameters.
A retrospective research prediction model in a general data database, namely a standard image training model, is transferred to diagnosis and treatment judgment related feature recognition of CT, MRI, two-dimensional pictures and space position variables of orbit diseases so as to establish a new prediction model of a corresponding depth network.
Comparing prediction results obtained respectively based on the new prediction model and the standard image training model to obtain a comparison difference value;
if the contrast difference value is larger than a preset difference threshold value, predicting the feature variable of the orbit disease medical data by using the new prediction model to obtain orbit key point set space position feature data in the corresponding judgment residual feature variable, a two-dimensional image high-dimensional feature parameter automatically identified and labeled by the anatomical structure, a demographic index and a clinical examination index; and (4) prediction results with high research value of TAO orbital CT high-dimensional characteristic parameters and TAO orbital MRI high-dimensional characteristic parameters.
If the contrast difference value is not greater than the difference threshold value, obtaining a to-be-sampled supplementary result corresponding to the situation of insufficient samples based on the sample quantity judgment condition, and/or obtaining eye socket key point set spatial position feature data, a meridian anatomy structure automatic identification and labeling two-dimensional image high-dimensional feature parameter, a demographic index and a clinical examination index in corresponding judgment allowance feature variables obtained by predicting the feature variables of the medical data by using the retrospective study prediction model corresponding to the situation of sufficient samples; and (4) prediction results with low research value of TAO orbital CT high-dimensional characteristic parameters and TAO orbital MRI high-dimensional characteristic parameters.
Similar to the principle of the above embodiment, the present invention provides a value determination system based on the remaining amount of medical data.
Specific embodiments are provided below in conjunction with the attached figures:
fig. 3 shows a schematic structural diagram of a system of a value determination method based on medical data margin according to an embodiment of the present invention.
The system comprises:
the characteristic variable acquiring module 31 is configured to perform label-free clustering on the acquired medical data, and use a category corresponding to each cluster as a characteristic variable of the medical data, where the characteristic variable includes each known characteristic variable and one or more margin characteristic variables except for each known characteristic variable;
a new prediction model module 32 connected to the characteristic variable obtaining module 31, configured to obtain a new prediction model of the characteristic variables corresponding to the medical data based on a retrospective research prediction model obtained through training of each known characteristic variable in a general data database;
a comparison module 33, connected to the new prediction model module 32, for comparing the prediction results obtained based on the new prediction model and the retrospective study prediction model respectively to obtain a comparison difference value;
the residual value high judgment module 34 is connected to the comparison module 33, and is configured to predict the characteristic variable of the medical data by using the new prediction model if the comparison difference value is greater than a preset difference threshold value, so as to obtain a prediction result corresponding to high research value of the residual characteristic variable;
and a residual quantity value judging module 35, connected to the comparison module 33, for obtaining a to-be-sample supplementation result corresponding to a sample shortage condition based on the sample quantity judging condition if the comparison difference value is not greater than the difference threshold value, and/or obtaining a prediction result corresponding to a low research value of the residual quantity characteristic variable by predicting the characteristic variable of the medical data by using the retrospective research prediction model corresponding to a sample sufficiency condition.
Optionally, the characteristic variable acquiring module 31 performs label-free clustering on the acquired medical data, and uses a category corresponding to each cluster as a characteristic variable of the medical data. Wherein the characteristic variables include known characteristic variables and one or more margin characteristic variables other than the known characteristic variables.
Optionally, the known characteristic variables are characteristic variables commonly used in the prior retrospective research method.
Alternatively, the medical data may be all examination information and/or self-information related to a patient, wherein the examination information is medical examination information in any field, and is not limited in the present invention.
Optionally, the medical data includes: gender, age, race, duration of disease, treatment history, other ailments, birth history, allergy history, surgical history, medication history, and treatment history data.
Optionally, the new prediction model module 32 trains the retrospective research prediction model obtained by training the characteristic variables extracted from the medical data and the correlation degree information between the characteristic variables as training data based on the known characteristic variables in the general data database, so as to obtain a new prediction model.
Specifically, the new prediction model module 32 inputs feature variables extracted from the medical data and information on the degree of association between the feature variables as training data into the retrospective study prediction model trained from information on known feature variables in a general data database, and obtains a new prediction model associated with the known feature variables and one or more residual feature variables other than the known feature variables.
Optionally, the new prediction model module 32 refines and summarizes the relationship between the characteristic variables, and when some characteristic variable factor changes, it can predict other characteristic variable factors affecting the characteristic variable factor, and then obtains a new prediction model based on the retrospective study prediction model through deep learning and high-dimensional features of the computer.
Optionally, the training mode of the retrospective study prediction model includes: and training a prediction model by using the known characteristic variables extracted from the general data in the general data database and the correlation degree information among the known characteristic variables as training data to obtain the retrospective research prediction model.
Specifically, the relationship between different known characteristic variables is abstracted and summarized through massive general data, and when one known characteristic variable factor is changed, other known characteristic variable factors influencing the known characteristic variable factor can be predicted to obtain a retrospective prediction model.
Optionally, the general data database includes: gender, age, race, duration of disease, treatment history, other ailments, birth history, allergy history, surgical history, medication history, and treatment history data.
Optionally, the comparison module 33 compares the prediction results obtained by the new prediction model and the retrospective prediction model, which are input with the same known characteristic variable, to obtain a comparison difference value.
Optionally, the comparison module 33 compares the prediction results related to the characteristic variables associated with the known characteristic variables obtained by the new prediction model and the retrospective prediction model, which are input with the same known characteristic variable, to obtain a comparison difference value. Wherein, based on the new prediction model, other known characteristic variables associated with the known characteristic variable and the prediction result related to the residual characteristic can be obtained; based on the retrospective predictive model, a prediction result related to other known characteristic variables associated with the known characteristic variable may be obtained.
Optionally, the preset difference threshold is determined according to specific requirements, and is not limited in the present invention.
Optionally, if the comparison difference value is greater than a preset difference threshold value, it indicates that the residual characteristic variable of the medical data is more and the research value of the judgment residual characteristic variable is high, and the judgment residual value high module 34 needs to predict the characteristic variable of the medical data by using the new prediction model including the residual characteristic variable and each known characteristic to obtain a prediction result, so as to mine residual information.
Optionally, if the comparison difference value is less than or equal to the difference threshold, the module 35 for determining that the remaining value is low first determines whether the result difference is caused by insufficient samples based on the sample number determination condition, and if the samples are insufficient, obtains a result to be supplemented for the samples corresponding to the case of insufficient samples, and performs re-verification after the samples are sufficient; if the sample is sufficient, the residual characteristics of the medical data are few and the research value of the residual characteristic variables is low, a prediction result obtained by predicting the characteristic variables of the medical data by using the retrospective research prediction model under the condition that the sample is sufficient is obtained, namely, the residual information does not need to be further mined for the current medical data.
Optionally, the sample number judging condition includes: comparing the number of samples with the characteristic variables of the medical data with a preset sample number threshold; if the number of samples with the characteristic variables of the medical data is greater than or equal to the sample number threshold value, determining that the samples are sufficient; and if the number of the samples with the characteristic variables of the medical data is less than the sample number threshold value, determining that the samples are insufficient.
Fig. 4 shows a schematic structural diagram of a value determination terminal 40 based on the remaining amount of medical data in the embodiment of the invention.
The value determination terminal 40 based on the medical data margin includes: a memory 41 and a processor 42, the memory 41 being for storing computer programs; the processor 42 runs a computer program to implement the value determination method based on the remaining amount of medical data as shown in fig. 1.
Alternatively, the number of the memories 41 may be one or more, the number of the processors 42 may be one or more, and fig. 4 illustrates one example.
Optionally, the processor 42 in the value determining terminal 40 based on the remaining amount of medical data loads one or more instructions corresponding to the process of the application program into the memory 41 according to the steps shown in fig. 1, and the processor 42 runs the application program stored in the first memory 41, so as to implement various functions in the value determining method based on the remaining amount of medical data shown in fig. 1.
Optionally, the memory 41 may include, but is not limited to, a high speed random access memory, a non-volatile memory. Such as one or more magnetic disk storage devices, flash memory devices, or other non-volatile solid-state storage devices; the Processor 42 may include, but is not limited to, a Central Processing Unit (CPU), a Network Processor (NP), and the like; the Integrated Circuit may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, or a discrete hardware component.
Optionally, the Processor 42 may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; the Integrated Circuit may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, or a discrete hardware component.
The present invention also provides a computer-readable storage medium storing a computer program which, when executed, implements the value determination method based on the remaining amount of medical data as shown in fig. 1. The computer-readable storage medium may include, but is not limited to, floppy diskettes, optical disks, CD-ROMs (compact disc-read only memories), magneto-optical disks, ROMs (read-only memories), RAMs (random access memories), EPROMs (erasable programmable read only memories), EEPROMs (electrically erasable programmable read only memories), magnetic or optical cards, flash memory, or other type of media/machine-readable medium suitable for storing machine-executable instructions. The computer readable storage medium may be a product that is not accessed by the computer device or may be a component that is used by an accessed computer device.
In summary, the value judgment method, system, terminal and medium based on the medical data margin are used for solving the problems that in the prior art, the solution for further analyzing and mining the value of the medical data is incomplete, so that a large amount of research is blindly carried out and social resources are wasted. The invention extracts massive information in medical data by using an information mining technology, comprehensively analyzes and utilizes the data from multiple dimensions, and mines the data which is not effectively utilized before, thereby further improving the medical research level, and mining more information in the existing data to try to deeply interpret and search the internal rules of the data, the rules among the data and even the rules between the data and the specific diagnosis/treatment effect. Therefore, the invention effectively overcomes various defects in the prior art and has high industrial utilization value.
The foregoing embodiments are merely illustrative of the principles of the present invention and its efficacy, and are not to be construed as limiting the invention. Any person skilled in the art can modify or change the above-mentioned embodiments without departing from the spirit and scope of the present invention. Accordingly, it is intended that all equivalent modifications or changes which can be made by those skilled in the art without departing from the spirit and technical spirit of the present invention be covered by the claims of the present invention.
Claims (10)
1. A value judgment method based on medical data margin is characterized by comprising the following steps:
performing label-free clustering on the acquired medical data, and taking the category corresponding to each cluster as a characteristic variable of the medical data, wherein the characteristic variable comprises each known characteristic variable and one or more margin characteristic variables except each known characteristic variable;
obtaining a new prediction model of the characteristic variables corresponding to the medical data based on a retrospective research prediction model obtained by training of each known characteristic variable in a general data database;
comparing the prediction results obtained respectively based on the new prediction model and the retrospective study prediction model to obtain a comparison difference value;
if the comparison difference value is larger than a preset difference threshold value, predicting the characteristic variable of the medical data by using the new prediction model to obtain a prediction result of correspondingly judging high research value of the residual characteristic variable;
if the comparison difference value is not larger than the difference threshold value, obtaining a to-be-sampled supplementary result corresponding to the situation that the sample is insufficient based on the sample quantity judgment condition, and/or obtaining a prediction result with low research value of the corresponding judgment margin characteristic variable by predicting the characteristic variable of the medical data by using the retrospective research prediction model corresponding to the situation that the sample is sufficient.
2. The method for value determination based on medical data margin according to claim 1, wherein the obtaining a new prediction model of the feature variables corresponding to the medical data based on the retrospective study prediction model obtained by training each known feature variable in the general data database comprises:
training the retrospective research prediction model obtained by training the known characteristic variables in a general data database by using the characteristic variables extracted from the medical data and the association degree information between the characteristic variables as training data to obtain a new prediction model.
3. The method for value determination based on medical data residual according to claim 1 or 2, wherein the training mode of the retrospective study prediction model comprises:
and training a prediction model by using the known characteristic variables extracted from the general data in the general data database and the correlation degree information among the known characteristic variables as training data to obtain the retrospective research prediction model.
4. The method of claim 1, wherein the comparing the prediction results obtained based on the new prediction model and the retrospective study prediction model to obtain a difference value comprises:
and comparing the prediction results respectively obtained by the new prediction model and the retrospective prediction model which input the same known characteristic variable to obtain a comparison difference value.
5. The method of claim 1, wherein the sample number determination condition comprises:
comparing the number of samples with the characteristic variables of the medical data with a preset sample number threshold;
if the number of samples with the characteristic variables of the medical data is greater than or equal to the sample number threshold value, determining that the samples are sufficient;
and if the number of the samples with the characteristic variables of the medical data is less than the sample number threshold value, determining that the samples are insufficient.
6. A value judgment method based on medical data margin as claimed in claim 1, wherein the general data database comprises: gender, age, race, duration of disease, treatment history, other ailments, birth history, allergy history, surgical history, medication history, and treatment history data.
7. A value determination system based on a remaining amount of medical data, the system comprising:
the characteristic variable acquisition module is used for performing label-free clustering on the acquired medical data and taking the category corresponding to each cluster as a characteristic variable of the medical data, wherein the characteristic variable comprises each known characteristic variable and one or more margin characteristic variables except for each known characteristic variable;
the new prediction model module is connected with the characteristic variable acquisition module and used for acquiring a new prediction model of the characteristic variable corresponding to the medical data based on a retrospective research prediction model acquired by training of each known characteristic variable in a general data database;
the comparison module is connected with the new prediction model module and is used for comparing prediction results obtained respectively based on the new prediction model and the retrospective research prediction model to obtain a comparison difference value;
the residual value judgment high module is connected with the comparison module and used for predicting the characteristic variable of the medical data by using the new prediction model to obtain a prediction result corresponding to the residual value judgment high research value of the characteristic variable if the comparison difference value is larger than a preset difference threshold;
and the judgment residual amount value low module is connected with the comparison module and is used for obtaining a to-be-sample supplement result corresponding to the condition that the sample is insufficient based on the sample quantity judgment condition and/or a prediction result corresponding to the condition that the sample is sufficient and obtained by predicting the characteristic variable of the medical data by using the retrospective research prediction model, wherein the prediction result corresponds to the condition that the sample quantity judgment condition is not larger than the difference threshold value and is low in research value of the characteristic variable of the judgment residual amount.
8. The system of claim 7, wherein the obtaining a new prediction model corresponding to the feature variables of the medical data based on a retrospective study prediction model trained from each known feature variable in a general data base comprises:
training the retrospective research prediction model obtained by training the known characteristic variables in a general data database by using the characteristic variables extracted from the medical data and the association degree information between the characteristic variables as training data to obtain a new prediction model.
9. A value judging terminal based on medical data allowance is characterized by comprising:
a memory for storing a computer program;
a processor for executing the method for value determination based on the remaining amount of medical data according to any one of claims 1 to 6.
10. A computer storage medium, characterized in that a computer program is stored, which when running implements the method for value determination based on the residual amount of medical data according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010849411.3A CN111951976B (en) | 2020-08-21 | 2020-08-21 | Value judging method, system, terminal and medium based on medical data allowance |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010849411.3A CN111951976B (en) | 2020-08-21 | 2020-08-21 | Value judging method, system, terminal and medium based on medical data allowance |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111951976A true CN111951976A (en) | 2020-11-17 |
| CN111951976B CN111951976B (en) | 2024-03-22 |
Family
ID=73359509
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010849411.3A Active CN111951976B (en) | 2020-08-21 | 2020-08-21 | Value judging method, system, terminal and medium based on medical data allowance |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111951976B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109767263A (en) * | 2018-12-19 | 2019-05-17 | 平安科技（深圳）有限公司 | Business revenue data predication method, device, computer equipment and storage medium |
| CN109830303A (en) * | 2019-02-01 | 2019-05-31 | 上海众恒信息产业股份有限公司 | Clinical data mining analysis and aid decision-making method based on internet integration medical platform |
| CN110021439A (en) * | 2019-03-07 | 2019-07-16 | 平安科技（深圳）有限公司 | Medical data classification method, device and computer equipment based on machine learning |
-
2020
- 2020-08-21 CN CN202010849411.3A patent/CN111951976B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109767263A (en) * | 2018-12-19 | 2019-05-17 | 平安科技（深圳）有限公司 | Business revenue data predication method, device, computer equipment and storage medium |
| CN109830303A (en) * | 2019-02-01 | 2019-05-31 | 上海众恒信息产业股份有限公司 | Clinical data mining analysis and aid decision-making method based on internet integration medical platform |
| CN110021439A (en) * | 2019-03-07 | 2019-07-16 | 平安科技（深圳）有限公司 | Medical data classification method, device and computer equipment based on machine learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111951976B (en) | 2024-03-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Allugunti | A machine learning model for skin disease classification using convolution neural network | |
| KR102583103B1 (en) | Systems and methods for processing electronic images for computational detection methods | |
| Özbay | An active deep learning method for diabetic retinopathy detection in segmented fundus images using artificial bee colony algorithm | |
| US20220058839A1 (en) | Translation of images of stained biological material | |
| CN111899252B (en) | Pathological image processing method and device based on artificial intelligence | |
| CN109770903A (en) | The classification prediction technique of functional magnetic resonance imaging, system, device | |
| Coupé et al. | LesionBrain: an online tool for white matter lesion segmentation | |
| Masoud Abdulhamid et al. | New auxiliary function with properties in nonsmooth global optimization for melanoma skin cancer segmentation | |
| Hameurlaine et al. | Survey of brain tumor segmentation techniques on magnetic resonance imaging | |
| CN111882538A (en) | Processing system and information processing method for distinguishing pulmonary tuberculosis information and tumor information | |
| Sunil Babu et al. | An effective approach for sub-acute Ischemic stroke lesion segmentation by adopting meta-heuristics feature selection technique along with hybrid Naive Bayes and sample-weighted random forest classification | |
| Patil et al. | Generating region of interests for invasive breast cancer in histopathological whole-slide-image | |
| Wolcott et al. | Automated classification of estrous stage in rodents using deep learning | |
| Ong et al. | Detection of subtle white matter lesions in MRI through texture feature extraction and boundary delineation using an embedded clustering strategy | |
| Kasim | Ensemble classification based optimized transfer learning feature method for early stage diagnosis of diabetic retinopathy | |
| Li et al. | [Retracted] Artificial Intelligence‐Based Semisupervised Self‐Training Algorithm in Pathological Tissue Image Segmentation | |
| Saranyaraj et al. | Early prediction of breast cancer based on the classification of HER‐2 and ER biomarkers using deep neural network | |
| Han et al. | Multimodal 3D convolutional neural networks for classification of brain disease using structural MR and FDG-PET images | |
| CN111951976A (en) | Value judgment method, system, terminal and medium based on medical data margin | |
| Tyagi et al. | Machine learning approaches for the detection of schizophrenia using structural mri | |
| Chan et al. | Tissue section image-based liver scar detection | |
| CN114170415A (en) | TMB classification method and system based on histopathology image depth domain adaptation | |
| Marwa et al. | Comparative study of two classification methods for the detection of Alzheimer's disease | |
| Menagadevi et al. | An Epidemic of Neurodegenerative Disease Analysis Using Machine Learning Techniques | |
| Hamim et al. | Enhancing Brain Tumor MRI Segmentation Accuracy and Efficiency with Optimized U-Net Architecture |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |