CN108735295B - Blood analysis method and terminal equipment based on regression tree model - Google Patents
Blood analysis method and terminal equipment based on regression tree model Download PDFInfo
- Publication number
- CN108735295B CN108735295B CN201810457009.3A CN201810457009A CN108735295B CN 108735295 B CN108735295 B CN 108735295B CN 201810457009 A CN201810457009 A CN 201810457009A CN 108735295 B CN108735295 B CN 108735295B
- Authority
- CN
- China
- Prior art keywords
- blood
- value
- regression tree
- tree model
- detection
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000004159 blood analysis Methods 0.000 title claims abstract description 32
- 210000004369 blood Anatomy 0.000 claims abstract description 383
- 239000008280 blood Substances 0.000 claims abstract description 383
- 238000001514 detection method Methods 0.000 claims abstract description 141
- 208000024891 symptom Diseases 0.000 claims abstract description 91
- 238000012549 training Methods 0.000 claims abstract description 20
- 230000006870 function Effects 0.000 claims description 48
- 238000004590 computer program Methods 0.000 claims description 22
- 230000008569 process Effects 0.000 claims description 15
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000004458 analytical method Methods 0.000 abstract description 9
- 238000012545 processing Methods 0.000 abstract description 6
- 239000000306 component Substances 0.000 description 18
- 210000000265 leukocyte Anatomy 0.000 description 9
- 102000001554 Hemoglobins Human genes 0.000 description 6
- 108010054147 Hemoglobins Proteins 0.000 description 6
- 238000012360 testing method Methods 0.000 description 6
- 230000036541 health Effects 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 210000004072 lung Anatomy 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 3
- 238000007670 refining Methods 0.000 description 3
- 208000006545 Chronic Obstructive Pulmonary Disease Diseases 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 210000003743 erythrocyte Anatomy 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000012503 blood component Substances 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000004141 dimensional analysis Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
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
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Medical Informatics (AREA)
- Health & Medical Sciences (AREA)
- Public Health (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Computational Mathematics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Probability & Statistics with Applications (AREA)
- Evolutionary Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Primary Health Care (AREA)
- General Health & Medical Sciences (AREA)
- Operations Research (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Epidemiology (AREA)
- Algebra (AREA)
- Pathology (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The invention is applicable to the technical field of data processing, and provides a regression tree model-based blood analysis method, terminal equipment and a computer-readable storage medium, which comprise the following steps: obtaining a plurality of blood information samples, each blood information sample comprising a blood value and a symptom characteristic value; fitting the plurality of blood information samples with a regression tree model, and taking the regression tree model after fitting as a detection model; obtaining a blood value of blood to be detected, and inputting the blood value of the blood to be detected into the detection model to obtain a detection value; and if the detection value is larger than the detection threshold value, outputting a first alarm prompt. According to the invention, through training the regression tree model, comprehensive analysis of the blood value in the blood to be tested is realized, and the reliability and accuracy of blood analysis are improved.
Description
Technical Field
The invention belongs to the technical field of data processing, and particularly relates to a regression tree model-based blood analysis method, terminal equipment and a computer-readable storage medium.
Background
Blood is an important component of the human body, and along with the continuous progress of science and technology, the analysis of blood is also developing toward refinement and diversification. In general, in a blood analysis process, numerical information of each component in blood needs to be obtained first, specifically, the blood to be measured is placed in a testing instrument, and then an analyst obtains the numerical information of the blood according to a testing result displayed by the testing instrument. Then, the computer judges whether the numerical information exceeds the corresponding threshold value based on the statistical theory, the mathematical theory and the preset threshold value, and obtains the analysis result of the blood according to the judgment result.
In order to prevent erroneous judgment, the value of the threshold is generally set to be out of a large range of normal values. If a problem of a certain blood is represented by a certain number of numerical information being higher or lower, but the numerical information of each component does not exceed the corresponding threshold value, the analysis result obtained by analyzing the blood according to the existing analysis method is obviously wrong. In summary, the existing blood analysis method only can analyze the numerical information of each component in blood alone, and cannot comprehensively judge a plurality of numerical information, namely, the reliability and accuracy of analyzing the blood are low.
Disclosure of Invention
In view of the above, embodiments of the present invention provide a regression tree model-based blood analysis method, a terminal device, and a computer-readable storage medium, so as to solve the problem that in the prior art, multiple components of blood cannot be integrated for analysis, and reliability and accuracy of analysis of blood are low.
A first aspect of an embodiment of the present invention provides a blood analysis method based on a regression tree model, including:
obtaining a plurality of blood information samples, each blood information sample comprising a blood value and a symptom characteristic value;
fitting the plurality of blood information samples with a regression tree model, and taking the regression tree model after fitting as a detection model;
Obtaining a blood value of blood to be detected, and inputting the blood value of the blood to be detected into the detection model to obtain a detection value;
and if the detection value is larger than the detection threshold value, outputting a first alarm prompt.
A second aspect of an embodiment of the present invention provides a terminal device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the following steps when executing the computer program:
obtaining a plurality of blood information samples, each blood information sample comprising a blood value and a symptom characteristic value;
fitting the plurality of blood information samples with a regression tree model, and taking the regression tree model after fitting as a detection model;
obtaining a blood value of blood to be detected, and inputting the blood value of the blood to be detected into the detection model to obtain a detection value;
and if the detection value is larger than the detection threshold value, outputting a first alarm prompt.
A third aspect of the embodiments of the present invention provides a computer readable storage medium storing a computer program which when executed by a processor performs the steps of:
Obtaining a plurality of blood information samples, each blood information sample comprising a blood value and a symptom characteristic value;
fitting the plurality of blood information samples with a regression tree model, and taking the regression tree model after fitting as a detection model;
obtaining a blood value of blood to be detected, and inputting the blood value of the blood to be detected into the detection model to obtain a detection value;
and if the detection value is larger than the detection threshold value, outputting a first alarm prompt.
Compared with the prior art, the embodiment of the invention has the beneficial effects that:
according to the embodiment of the invention, a plurality of blood information samples are obtained, each blood information sample comprises a blood value and a symptom characteristic value, the plurality of blood information samples are fitted with the regression tree model, the fitted regression tree model is used as a detection model, then the blood to be detected is analyzed, the blood value of the blood to be detected is firstly obtained, the blood value is input into the detection model, the output result of the detection model is obtained as a detection value, and finally, if the detection value is judged to be greater than the detection threshold value, a first alarm prompt is output.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments or the description of the prior art will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart of a regression tree model-based blood analysis method according to an embodiment of the present invention;
FIG. 2 is a flowchart of a blood analysis method based on a regression tree model according to a second embodiment of the present invention;
FIG. 3 is a flowchart of a blood analysis method based on a regression tree model according to a third embodiment of the present invention;
FIG. 4 is a flowchart of a blood analysis method based on a regression tree model according to a fourth embodiment of the present invention;
FIG. 5 is a flowchart of a regression tree model-based blood analysis method according to a fifth embodiment of the present invention;
fig. 6 is a block diagram of a terminal device according to a sixth embodiment of the present invention;
fig. 7 is a schematic diagram of a terminal device according to a seventh embodiment of the present invention.
Detailed Description
In the following description, for purposes of explanation and not limitation, specific details are set forth such as the particular system architecture, techniques, etc., in order to provide a thorough understanding of the embodiments of the present invention. It will be apparent, however, to one skilled in the art that the present invention may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present invention with unnecessary detail.
In order to illustrate the technical scheme of the invention, the following description is made by specific examples.
Fig. 1 shows an implementation flow of a regression tree model-based blood analysis method according to an embodiment of the present invention, which is described in detail below:
in S101, a plurality of blood information samples are acquired, each of which includes a blood value and a symptom characteristic value.
In the embodiment of the invention, a plurality of blood information samples are firstly obtained before analysis of blood to be tested, and each blood information sample comprises a blood value and a symptom characteristic value. The blood value is the relevant value of one or more components in the existing blood to which the blood information sample belongs, and can be determined by the test result obtained by analyzing the existing blood by a test instrument, wherein the existing blood is the blood which is analyzed by the test instrument and is provided with a file. For example, the blood value may be the red blood cell content of the existing blood, or may be a collection of the red blood cell content, the white blood cell content, and the hemoglobin content of the existing blood. The symptom characteristic value is set for the affected condition of the existing blood, and is determined by judging whether the existing blood is affected by the symptom, generally, the symptom characteristic value under the blood information sample of the existing blood affected by the symptom is set to be 1, and the symptom characteristic value under the blood information sample of the existing blood not affected by the symptom is set to be 0, so that calculation is convenient. In order to improve accuracy of blood analysis to be tested, blood information samples are obtained according to preset orders of magnitude, namely the number of the obtained blood information samples is in the preset orders of magnitude, and the orders of magnitude can be freely set according to accuracy requirements. For example, on the order of one thousand.
Preferably, a plurality of blood information samples are obtained from a health record or hospital database. Because a plurality of blood information samples of different crowds are stored in the health file or the hospital database, the number of the blood information samples is large, and the accuracy of symptom characteristic values under the blood information samples is high, the required blood information samples can be directly obtained from the health file or the hospital database. Optionally, a plurality of blood information samples are obtained according to specific conditions. In an actual application scene, acquiring a plurality of blood information samples according to scene requirements, for example, on the premise that the blood to be detected of a human body of a certain town is required to be analyzed, setting specific conditions, and acquiring a plurality of blood information samples only from a health file or a hospital database related to the town; for example, if the blood to be detected of a human body with age above sixty years is required to be analyzed, corresponding specific conditions are set, a plurality of blood information samples corresponding to the crowd with age above sixty years are obtained from a health file or a hospital database, specific conditions can be set to obtain a plurality of blood information samples with specific orders of magnitude, and the like, so that the applicability of different application scenes is improved.
In S102, the plurality of blood information samples are fitted to a regression tree model, and the regression tree model after the fitting is used as a detection model.
Since the blood value in a blood information sample may contain a plurality of related values of a plurality of components in the existing blood, the plurality of related values of the plurality of components are related to whether the existing blood is affected by symptoms. For example, the blood value of the blood information sample includes a white blood cell content, a hemoglobin content and a platelet content, and if the existing blood to which the blood information sample belongs is affected by the chronic obstructive pulmonary disease, an increase in the white blood cell content, an increase in the hemoglobin content and a decrease in the platelet content may occur in the blood value of the blood information sample compared to the blood value of the normal blood not affected by the chronic obstructive pulmonary disease. However, in general, the change rule of a plurality of relevant values of a plurality of components in the blood values affected by symptoms cannot be known, so in the embodiment of the invention, a plurality of blood information samples with determined symptom characteristic values are fitted with a regression tree model, the regression tree model is continuously trained in the fitting process, and finally the fitted regression tree model is output as a detection model. The regression tree model applies a classification and regression tree (Classification and regression tree, CART) algorithm, the CART algorithm uses a plurality of relevant values in blood values as a plurality of features, two branches of yes and no are established for each relevant value, normally the left branch is the branch with the value of yes, the right branch is the branch with the value of no, and the specific dividing method is based on symptom feature values corresponding to the relevant values. And recursively dividing each related value by the CART algorithm, dividing an input space formed by a plurality of related values into a plurality of limited units, determining the unit value of each unit by the symptom characteristic value corresponding to the plurality of related values, and finally generating a binary tree. The multiple blood information samples are input into a regression tree model, and a binary tree is essentially established and modified based on the multiple blood information samples, and the finally generated binary tree is the detection model, and the specific calculation process is explained later.
In S103, a blood value of the blood to be measured is obtained, and the blood value of the blood to be measured is input to the detection model to obtain a detection value.
After a detection model is generated by a plurality of blood information samples, analyzing the blood to be detected, obtaining the blood value of the blood to be detected, inputting the blood value of the blood to be detected into the detection model, and obtaining the output result of the detection model, namely the detection value. It should be noted that, the components corresponding to the blood values of the blood to be measured are the same as the components corresponding to the blood values of the blood information sample, or the components corresponding to the blood values of the blood to be measured are part of the components corresponding to the blood values of the blood information sample, and in both cases, after the blood values of the blood to be measured are input into the detection model, the output result of the detection model can be used as an effective detection value.
In S104, if the detection value is greater than the detection threshold, a first alarm prompt is output.
The detection value is calculated by the blood value of the blood to be detected and is used for judging whether the blood to be detected is affected by symptoms or not. If the symptom characteristic value of 1 indicates that the corresponding blood is affected by symptoms, and the symptom characteristic value of 0 indicates that the corresponding blood is not affected by symptoms, the closer the detection value is to 1, the greater the possibility that the blood to be tested is affected by symptoms. In order to improve the accuracy of judgment, a detection threshold value is set, if the detection value is larger than the detection threshold value, the blood to be tested is determined to be affected by symptoms, and a first alarm prompt is output, so that the user can check conveniently; if the detection value is smaller than or equal to the detection threshold value, determining that the blood to be detected is not affected by symptoms, and outputting a normal prompt. The detection threshold can be determined according to a plurality of blood information samples and a plurality of symptom characteristic values of the plurality of blood information samples, or a plurality of blood information samples can be input into the detection model and determined according to a plurality of output results output by the plurality of blood information samples according to the detection model.
As can be seen from the embodiment shown in fig. 1, in the embodiment of the present invention, by obtaining a plurality of existing blood information samples, where the blood information samples include blood values and symptom characteristic values, fitting the plurality of blood information samples with a regression tree model, outputting the fitted regression tree model as a detection model, finally performing analysis on blood to be detected, obtaining a blood value of the blood to be detected, inputting the blood value of the blood to be detected into the detection model, using an output result of the detection model as a detection value, comparing the detection value with a detection threshold, if the detection value is greater than the detection threshold, then proving that the blood to be detected is affected by symptoms, outputting a first alarm prompt, and according to the embodiment of the present invention, the blood analysis on different situations can be improved by training the regression tree model, and accuracy of the blood analysis is improved.
Fig. 2 shows a step of refining a process of fitting a plurality of blood information samples to a regression tree model on the basis of the first embodiment of the present invention. The embodiment of the invention provides a flow chart for realizing a blood analysis method based on a regression tree model, as shown in fig. 2, the blood analysis method can comprise the following steps:
In S201, the plurality of blood information samples are input to the regression tree model to train the regression tree model, wherein the blood values in the blood information samples are used as input parameters of the regression tree model, and the symptom characteristic values in the blood information samples are used as label parameters of the regression tree model.
Since there are a plurality of Blood information samples, a Blood sample set is constructed based on a plurality of Blood values and a plurality of symptom characteristic values in the plurality of Blood information samples, which is (Blood value1 ，Symptom value1 )，(Blood value2 ，Symptom value2 )，(Blood value1 ，Symptom value1 )……(Blood valuen ，Symptom valuen ) Wherein Blood is valuei Blood value indicating the ith blood information sample, symptom valuei Indicating a symptom characteristic value of the ith blood information sample. After the construction is completed, the blood sample set is input into a regression tree model, the blood value is used as an input parameter of the regression tree model, the symptom characteristic value is used as a label parameter of the regression tree model, and the regression tree model is fitted, and in the embodiment of the invention, the calculation formula of the regression tree model is as follows:
in the above-mentioned formula(s),representing the input parameter as Blood valuei Is to be Blood valuei And (3) inputting the input parameters into a regression tree model, and outputting a result after calculation of the regression tree model. f () indicates a function that exists in a function space, which refers to a set of functions of a given kind from one set to another, i.e., the f () function is initially in an unknown state, and K indicates that K of the above-mentioned f () functions exist in the regression tree model.
After determining the calculation formula of the regression tree model, training the regression tree model based on the blood sample set so that K f () functions in the final regression tree model maximally conform to the data in the blood sample set. In the embodiment of the invention, the f () function is learned by adopting a sequential learning method to reduce errors in the learning process, for example, when the input parameter is Blood valuei On the basis of the above, the prediction value prediction of the t round is performed, and the prediction value prediction result of the t-1 th round is reserved when the prediction value prediction of the t round is performed, namely each regression tree (each f () function) in the regression tree model is trained in turn, specifically as follows:
in the above formulaIs to give the input parameter as Blood valuei On the basis of (3), the predicted value after the t-th round of prediction is performed. In order to determine the f () function required in the sequential learning process, i.e. to determine the f () function that corresponds to the blood sample set, an objective function of the regression tree model is constructed, specifically as follows:
in the above formula, symptom valuei Is Blood sample set and input parameter Blood valuei The corresponding label parameter is the symptom characteristic value corresponding to the blood value in the blood information sample. In the above formula Is a regular term, D is a constant term, wherein the regular term controls an objective function Obj (t) The training degree of (2) preventing the blood sample set and the regression tree model from being over-fitted, and specifically describing the regular term later; the constant term is a constant, and in the embodiment of the invention, the constant term is mainly set for controlling Obj (t) Is a numerical range of (c). In addition, a->As an error function, for an objective function Obj (t) Performing optimizationThe conversion, i.e. finding the appropriate f () function causesThe value of (2) is reduced as much as possible.
In the embodiment of the invention, the optimization of the objective function is conveniently carried outUnfolding and defining:
the expanded objective function is:
extracting all constant items in the expanded objective function, and generating a training function, wherein the training function is specifically as follows:
in the training function, the output value obtained by the training function depends on g i And h i The simplicity of training is improved, so in the embodiment of the invention, the regression tree model can be trained through the finally generated training function.
Alternatively, in defining the f () function asω∈R k ，q:R d On the basis of {1,2, …, K }, a canonical term +.>Will f t (Blood valuei ) Seen as a regression tree, f t (Blood valuei ) The value of (1) is the Blood value Blood of the regression tree valuei Is a predicted value of (a). The regression tree is set to have K leaf nodes, and the values of the K leaf nodes form a K-dimensional vector omega. Q (Blood) in f () function valuei ) For mapping, in particular Blood of a certain Blood value valuei Mapping to a value of 1 to K, i.e. to Blood valuei To a node of the regression tree. On the basis of the above, a regularization term is defined +.>Wherein H is the same as D in the objective function and is a constant term; gamma and lambda are training coefficients, and a user can customize the training coefficients for an actual application scene to adjust the structure (scale of a detection model) of the generated regression tree, and any increase of gamma and lambda can lead to the simplification of the structure of the regression tree. Through the defined regular terms, the training effect of the regression tree model can be remarkably improved.
In S202, the trained regression tree model is output as the detection model.
And when all the blood values and symptom characteristic values in the blood sample set are input into the regression tree model, and after the regression tree model is trained, outputting the trained regression tree model as a detection model. When the Blood to be tested needs to be analyzed, blood value Blood of the Blood to be tested valuex Inputting the detection model, namely, the optimized calculation formula in the detection model can be usedObtaining a detection value +.>
As can be seen from the embodiment shown in fig. 2, in the embodiment of the present invention, a plurality of existing blood information samples are input into a regression tree model, wherein a blood value in the blood information sample is used as an input parameter of the regression tree model, a symptom characteristic value in the blood information sample is used as a label parameter of the regression tree model, the regression tree model is trained, and a trained regression tree model is obtained as a detection model.
Fig. 3 shows a specific process of refining the generation of the detection threshold on the basis of the first embodiment of the present invention. The embodiment of the invention provides a flow chart for realizing a blood analysis method based on a regression tree model, and as shown in fig. 3, the blood analysis method can comprise the following steps:
in S301, the blood values of the plurality of blood information samples are input to the detection model, and a plurality of output values output by the detection model are acquired.
After the detection model is generated, the blood value of the blood to be detected can be input into the detection model to obtain a detection value, and whether the blood to be detected is affected by symptoms or not is judged according to the detection value. In order to facilitate the above-mentioned determination, in the embodiment of the present invention, after the detection model is obtained, the blood values of the plurality of blood information samples are input as input parameters to the detection model, and a plurality of output values corresponding to the plurality of blood information samples output by the detection model are obtained.
In S302, the plurality of output values are ordered to generate a sequence of values.
The detection model is generated based on the blood values and the symptom characteristic values in the plurality of blood information samples, but when the blood values in the plurality of blood information samples are input into the detection model again, the operation of training the model is not executed, so that the output values generated by the detection model are different from the original symptom characteristic values. After a plurality of output values are obtained, the plurality of output values are ordered according to the size, a value sequence is generated, and the column head of the value sequence is the output value with the largest value. It should be noted that, if the first output value and the second output value in the plurality of output values are the same, a value sequence is generated according to the input order of the blood information samples corresponding to the first output value and the second output value. Specifically, a sequence writing mechanism is set, when a certain output value needs to be written into a value sequence, whether the existing output value which is the same as the output value exists in the value sequence is judged, if the existing output value which is the same as the output value does not exist, the smallest existing output value which is larger than the output value in the value sequence is searched out according to the value size of the output value, and the output value is written into a position behind the existing output value; if there are the existing output values identical to the output value, the output value is written in the position behind the existing output value, if there are a plurality of existing output values identical to the output value, the existing output value at the end of the plurality of existing output values is searched, and the output value is written in the position behind the existing output value.
In S303, an output value located at a preset position in the sequence of values is used as the detection threshold.
And after the numerical value sequence is generated, taking the output numerical value at a preset position in the numerical value sequence as a detection threshold. For example, 300 output values have been written in the value sequence, and the preset position is the 50 th bit, and the output value of the value sequence located in the 50 th bit is extracted as the detection threshold. Optionally, the preset position is determined according to a preset ratio. The preset ratio may be determined according to a big data analysis method, for example, a plurality of nationwide blood samples in a nationwide range may be counted, each of the nationwide blood samples includes a blood value and a symptom characteristic value, and a ratio of the nationwide blood samples having a symptom characteristic value of 1 among the plurality of nationwide blood samples is determined and taken as the preset ratio. For example, the ratio is 10%, 300 output values exist in the numerical value sequence, the preset position is the 30 th position, the output value of the numerical value sequence at the 30 th position is extracted as the detection threshold, and the set detection threshold has higher universal applicability through a big data analysis method. Of course, the preset ratio or preset position may be manually preset.
As can be seen from the embodiment shown in fig. 3, in the embodiment of the present invention, by inputting the blood values of a plurality of blood information samples as input parameters to the detection model, obtaining a plurality of output values output by the detection model, sorting the plurality of output values according to the size, generating a value sequence, extracting the output values at preset positions in the value sequence, using the output values as detection thresholds, and setting the detection thresholds according to the detection model, the applicability of the detection thresholds is improved.
Fig. 4 shows another specific process of refining the generation of the detection threshold on the basis of the first embodiment of the present invention. The embodiment of the invention provides a flow chart for realizing a blood analysis method based on a regression tree model, as shown in fig. 4, the blood analysis method can comprise the following steps:
in S401, a first sample number of people whose symptom characteristic value is the first characteristic value and a second sample number of people whose symptom characteristic value is the second characteristic value are determined according to the plurality of blood information samples.
Generally, there are two values for the symptom characteristic value in the blood information sample, i.e. 1 or 0, but it should be understood that other values for the symptom characteristic value can be applied in the embodiment of the present invention. For convenience of explanation, the symptom characteristic value of 1 is taken as a first characteristic value, the symptom characteristic value of 0 is taken as a second characteristic value, the first characteristic value represents that the existing blood to which the blood information sample belongs is affected by symptoms, and the second characteristic value represents that the existing blood to which the blood information sample belongs is not affected by symptoms. In the embodiment of the invention, the number of the blood information samples with the symptom characteristic value being the first characteristic value in the plurality of blood information samples is taken as the first sample number, and the number of the blood information samples with the symptom characteristic value being the second characteristic value in the plurality of blood information samples is taken as the second sample number.
In S402, a first duty ratio and a second duty ratio are determined according to the first sample population and the second sample population, and the detection threshold is determined according to the first duty ratio, the second duty ratio, the first characteristic value, and the second characteristic value.
After the first sample number and the second sample number are determined, a first proportion of the first sample number occupied by the total number of the plurality of blood information samples and a second proportion of the second sample number occupied by the total number of the plurality of blood information samples are determined, and a detection threshold value is determined according to the first proportion, the second proportion, the first characteristic value and the second characteristic value. For example, if the first characteristic value is greater than the second characteristic value, one method for determining the detection threshold value is to obtain a difference value between the first characteristic value and the second characteristic value, multiply the difference value by a first duty ratio to obtain a boundary value, and subtract the boundary value from the first characteristic value to obtain a result as the detection threshold value. For example, if the first characteristic value is 2, the second characteristic value is 1, and the first ratio is 30%, the detection threshold is 2- (2-1) ×30% =1.7. Of course, the above is only an example of the method for determining the detection threshold, and is not limited to the embodiment of the present invention, and more methods for determining the detection threshold may be applied according to different actual application scenarios.
As can be seen from the embodiment shown in fig. 4, in the embodiment of the present invention, by determining the first number of samples with the symptom characteristic value being the first characteristic value and the second number of samples with the symptom characteristic value being the second characteristic value in the plurality of blood information samples when the symptom characteristic value is the first characteristic value or the second characteristic value, determining the first duty ratio and the second duty ratio according to the number of the first samples and the number of the second samples, and finally obtaining the detection threshold according to the first duty ratio, the second duty ratio, the first characteristic value and the second characteristic value, since the plurality of blood information samples may be collected in a certain region, the embodiment of the present invention improves the accuracy of the detection threshold for a certain region.
Fig. 5 shows a process of comparing a blood value of blood to be measured with a plurality of blood threshold templates, which is added on the basis of the first embodiment of the present invention. The embodiment of the invention provides a flow chart for realizing a blood analysis method based on a regression tree model, and as shown in the figure, the blood analysis method can comprise the following steps:
in S501, a preset plurality of blood threshold templates are acquired, which are determined by the blood value of the alert blood.
Because the influence on the blood value caused by different symptoms is different, a plurality of blood threshold templates corresponding to a plurality of symptoms are obtained in advance, the blood threshold templates are the same as the blood value of the alarm blood affected by a certain symptom, wherein the blood threshold templates possibly comprise the related values of a certain or a plurality of blood components in the alarm blood, and the types of the related values can be selected according to practical application scenes. Optionally, a plurality of alarm blood affected by a symptom is obtained, the average value of a plurality of blood values of the alarm blood is evaluated, the generated average value is used as a blood threshold template, the influence of unexpected factors generated by the blood threshold template can be reduced, and the accuracy of the blood threshold template is improved. In addition, a plurality of alarm blood which is affected by symptoms and reaches a preset time can be obtained, for example, a plurality of alarm blood which is affected by symptoms and reaches six months is obtained, and a blood threshold template is generated from a plurality of blood values of the plurality of alarm blood by the same average evaluation method, so that the accuracy of judgment based on the blood threshold template in the time dimension is further improved.
In S502, a comparison is made between the blood value of the blood to be tested and the plurality of blood threshold templates.
Since multiple blood threshold templates are for different symptoms and the blood threshold templates may contain multiple related values, the alignment direction is set for the multiple related values of the symptom versus blood threshold templates. Taking the symptoms as slow-blocking lung as an example, since the blood affected by the slow-blocking lung has increased white blood cell content, increased hemoglobin content and reduced platelet content, when the blood threshold template corresponding to the slow-blocking lung contains three related values of white blood cell content, hemoglobin content and platelet content, the comparison direction is set up for the white blood cell content to be compared upwards, the comparison direction is set up for the hemoglobin content to be compared upwards, and the comparison direction is set up for the platelet content to be compared downwards. The upward comparison means that if the content of the white blood cells in the blood value of the blood to be detected exceeds or is equal to the content of the white blood cells in the blood threshold template, the comparison is confirmed to be successful, otherwise, the comparison is confirmed to be failed; the downward comparison means that if the platelet content in the blood value of the blood to be tested is smaller than the leukocyte content in the blood threshold template, the comparison is confirmed to be successful, otherwise, the comparison is confirmed to be failed. In the embodiment of the invention, a setting mechanism can be established for different symptoms, and the setting of the comparison direction in a plurality of blood threshold templates can be automatically completed through the setting mechanism.
In S503, if the comparison between the blood value of the blood to be tested and one of the blood threshold templates is successful, a second alarm prompt is output.
After the comparison of the plurality of blood threshold templates is completed, comparing the blood value of the blood to be tested with the plurality of blood threshold templates, and outputting a second alarm prompt if the blood value of the blood to be tested is successfully compared with a certain blood threshold template, wherein the second alarm prompt can comprise symptoms corresponding to the blood threshold templates, so that a user can conveniently and quickly check the blood value; if the comparison of the blood value of the blood to be detected and the plurality of blood threshold templates fails, outputting a normal prompt.
As can be seen from the embodiment shown in fig. 5, in the embodiment of the present invention, by acquiring a plurality of preset blood threshold templates for a plurality of symptoms, the blood threshold templates are taken from blood values of alarm blood affected by the symptoms, the blood values of the blood to be tested are compared with the plurality of blood threshold templates, if the blood values of the blood to be tested are successfully compared with one of the plurality of blood threshold templates, a second alarm prompt is output to perform an alarm, and multi-dimensional analysis of the blood to be tested is realized by setting the plurality of blood threshold templates, thereby improving reliability of the blood analysis to be tested.
It should be understood that the sequence number of each step in the foregoing embodiment does not mean that the execution sequence of each process should be determined by the function and the internal logic, and should not limit the implementation process of the embodiment of the present invention.
Fig. 6 shows a block diagram of a terminal device according to an embodiment of the present invention, where the terminal device includes units for performing the steps in the corresponding embodiment of fig. 1. Please refer to fig. 1 and the related description of the embodiment corresponding to fig. 1. For convenience of explanation, only the portions related to the present embodiment are shown.
Referring to fig. 6, the terminal device includes:
a first acquisition unit 61 for acquiring a plurality of blood information samples, each of which includes a blood value and a symptom characteristic value;
a fitting unit 62, configured to fit the plurality of blood information samples to a regression tree model, and use the regression tree model after the fitting as a detection model;
a second obtaining unit 63, configured to obtain a blood value of blood to be tested, and input the blood value of the blood to be tested to the detection model to obtain a detection value;
and the output unit 64 is configured to output a first alarm prompt if the detection value is greater than the detection threshold.
Optionally, the fitting unit 62 includes:
an input unit configured to input the plurality of blood information samples to the regression tree model to train the regression tree model, wherein a blood value in the blood information sample is used as an input vector of the regression tree model, and a symptom characteristic value in the blood information sample is used as a label vector of the regression tree model;
and the model output unit is used for outputting the regression tree model after training as the detection model.
Optionally, the output unit 64 further includes:
a sample input unit for inputting the blood values of the plurality of blood information samples to the detection model and obtaining a plurality of output values output by the detection model;
the sorting unit is used for sorting the plurality of output values to generate a value sequence;
and the threshold value output unit is used for taking the output value positioned at a preset position in the value sequence as the detection threshold value.
Optionally, the output unit 64 further includes:
a first determining unit configured to determine, from the plurality of blood information samples, a first number of samples in which the symptom characteristic value is the first characteristic value and a second number of samples in which the symptom characteristic value is the second characteristic value;
And the second determining unit is used for determining a first duty ratio and a second duty ratio according to the first sample number of people and the second sample number of people, and determining the detection threshold according to the first duty ratio, the second duty ratio, the first characteristic value and the second characteristic value.
Optionally, the terminal device further includes:
the template acquisition unit is used for acquiring a plurality of preset blood threshold templates, and the blood threshold templates are determined by the blood value of the alarm blood;
the comparison unit is used for comparing the blood value of the blood to be detected with the plurality of blood threshold templates;
and the alarm output unit is used for outputting a second alarm prompt if the comparison between the blood value of the blood to be detected and one of the blood threshold templates is successful.
Therefore, the terminal equipment provided by the embodiment of the invention can comprehensively analyze the blood value in the blood to be tested by training the regression tree model, so that the reliability and the accuracy of blood analysis are improved.
Fig. 7 is a schematic diagram of a terminal device according to an embodiment of the present invention. As shown in fig. 7, the terminal device 7 of this embodiment includes: a processor 70, a memory 71 and a computer program 72 stored in said memory 71 and executable on said processor 70, for example a control program of a terminal device. The processor 70, when executing the computer program 72, implements the steps of the various regression tree model-based blood analysis method embodiments described above, such as steps S101 through S104 shown in fig. 1. Alternatively, the processor 70, when executing the computer program 72, performs the functions of the units in the above-described device embodiments, such as the functions of the units 61 to 64 shown in fig. 6.
By way of example, the computer program 72 may be divided into one or more units, which are stored in the memory 71 and executed by the processor 70 to accomplish the present invention. The one or more units may be a series of computer program instruction segments capable of performing a specific function for describing the execution of the computer program 72 in the terminal device 7. For example, the computer program 72 may be divided into a first acquisition unit, a fitting unit, a second acquisition unit, and an output unit, each unit functioning specifically as follows:
a first acquisition unit configured to acquire a plurality of blood information samples, each of the blood information samples including a blood value and a symptom characteristic value;
the fitting unit is used for fitting the plurality of blood information samples with a regression tree model, and taking the regression tree model after fitting as a detection model;
the second acquisition unit is used for acquiring the blood value of the blood to be detected and inputting the blood value of the blood to be detected into the detection model to obtain a detection value;
and the output unit is used for outputting a first alarm prompt if the detection value is larger than the detection threshold value.
The terminal device 7 may be a computing device such as a desktop computer, a notebook computer, a palm computer, a cloud server, etc. The terminal device 7 may include, but is not limited to, a processor 70, a memory 71. It will be appreciated by those skilled in the art that fig. 7 is merely an example of the terminal device 7 and does not constitute a limitation of the terminal device 7, and may include more or less components than illustrated, or may combine certain components, or different components, e.g. the terminal device 7 may further include input-output devices, network access devices, buses, etc.
The processor 70 may be a central processing unit (Central Processing Unit, CPU), or may be another general purpose processor, a digital signal processor (Digital Signal Processor, DSP), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), an off-the-shelf programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory 71 may be an internal storage unit of the terminal device 7, such as a hard disk or a memory of the terminal device 7. The memory 71 may be an external storage device of the terminal device 7, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card) or the like, which are provided on the terminal device 7. Further, the memory 71 may also include both an internal storage unit and an external storage device of the terminal device 7. The memory 71 is used for storing the computer program as well as other programs and data required by the terminal device 7. The memory 71 may also be used for temporarily storing data that has been output or is to be output.
It will be apparent to those skilled in the art that the above-described functional units are merely illustrated in terms of division for convenience and brevity, and that in practical applications, the above-described functional units may be allocated to different functional units, i.e., the internal structure of the apparatus may be divided into different functional units, so as to perform all or part of the above-described functions. The functional units in the embodiment may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit, where the integrated units may be implemented in a form of hardware or a form of a software functional unit. In addition, the specific names of the functional units are also only for distinguishing from each other, and are not used to limit the protection scope of the present application. The specific working process of the units in the above system may refer to the corresponding process in the foregoing method embodiment, which is not described herein again.
In the foregoing embodiments, the descriptions of the embodiments are emphasized, and in part, not described or illustrated in any particular embodiment, reference is made to the related descriptions of other embodiments.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus/terminal device and method may be implemented in other manners. For example, the apparatus/terminal device embodiments described above are merely illustrative, e.g., the division of the units is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection via interfaces, devices or units, which may be in electrical, mechanical or other forms.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the present invention may implement all or part of the flow of the method of the above embodiment, or may be implemented by a computer program to instruct related hardware, where the computer program may be stored in a computer readable storage medium, and when the computer program is executed by a processor, the computer program may implement the steps of each of the method embodiments described above. Wherein the computer program comprises computer program code which may be in source code form, object code form, executable file or some intermediate form etc. The computer readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), an electrical carrier signal, a telecommunications signal, a software distribution medium, and so forth. It should be noted that the computer readable medium contains content that can be appropriately scaled according to the requirements of jurisdictions in which such content is subject to legislation and patent practice, such as in certain jurisdictions in which such content is subject to legislation and patent practice, the computer readable medium does not include electrical carrier signals and telecommunication signals.
The above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention, and are intended to be included in the scope of the present invention.
Claims (8)
1. A regression tree model-based blood analysis method, comprising:
obtaining a plurality of blood information samples, each blood information sample comprising a blood value and a symptom characteristic value;
fitting the plurality of blood information samples with a regression tree model, and taking the regression tree model after fitting as a detection model;
obtaining a blood value of blood to be detected, and inputting the blood value of the blood to be detected into the detection model to obtain a detection value;
if the detection value is larger than the detection threshold, outputting a first alarm prompt;
the calculation formula of the regression tree model is as follows:
In the above-mentioned formula(s),representing the input parameter as Blood valuei F () indicates a function that exists in a function space, said function space referring to a set of functions of a given kind from one set to another, K indicating the presence of K f () functions in said regression tree model;
fitting the plurality of blood information samples with a regression tree model, and taking the regression tree model with the fitted regression tree model as a detection model, wherein the method comprises the following steps:
inputting the plurality of blood information samples into the regression tree model to train the regression tree model, wherein the blood values in the blood information samples are used as input parameters of the regression tree model, and the symptom characteristic values in the blood information samples are used as label parameters of the regression tree model; the process of training the regression tree model includes: and (3) learning the f () function by adopting a sequential learning method, wherein the formula is as follows:
……
in the above-mentioned formula(s),is to give the input parameter as Blood valuei On the basis of the above, the predicted value after the t-th round of prediction is carried out; the regression tree model has the objective function of:
in the above formula, symptom valuei Is Blood sample set and input parameter Blood valuei The corresponding tag parameter(s) are used,is a regular term, D is a constant term, wherein the regular term controls an objective function Obj (t) Is a training level of (a);
and outputting the trained regression tree model as the detection model.
2. The method of claim 1, wherein before outputting the first alert if the detected value is greater than the detection threshold, further comprising:
inputting the blood values of the blood information samples into the detection model, and obtaining a plurality of output values output by the detection model;
sorting the plurality of output values to generate a sequence of values;
and taking the output value at a preset position in the value sequence as the detection threshold.
3. The method of claim 1, wherein the symptom characteristic value is a first characteristic value or a second characteristic value, and the method further comprises, before outputting the first alert if the detected value is greater than a detection threshold value:
determining a first number of samples of which the symptom characteristic value is the first characteristic value and a second number of samples of which the symptom characteristic value is the second characteristic value according to the plurality of blood information samples;
And determining a first duty ratio and a second duty ratio according to the first sample number and the second sample number, and determining the detection threshold according to the first duty ratio, the second duty ratio, the first characteristic value and the second characteristic value.
4. The blood analysis method of claim 1, further comprising:
acquiring a plurality of preset blood threshold templates, wherein the blood threshold templates are determined by blood values of alarm blood;
comparing the blood value of the blood to be detected with the plurality of blood threshold templates;
and if the comparison between the blood value of the blood to be detected and one of the blood threshold templates is successful, outputting a second alarm prompt.
5. A terminal device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the following steps when executing the computer program:
obtaining a plurality of blood information samples, each blood information sample comprising a blood value and a symptom characteristic value;
fitting the plurality of blood information samples with a regression tree model, and taking the regression tree model after fitting as a detection model;
Obtaining a blood value of blood to be detected, and inputting the blood value of the blood to be detected into the detection model to obtain a detection value;
if the detection value is larger than the detection threshold, outputting a first alarm prompt;
the calculation formula of the regression tree model is as follows:
in the above-mentioned formula(s),representing the input parameter as Blood valuei F () indicates a function that exists in a function space, said function space referring to a set of functions of a given kind from one set to another, K indicating the presence of K f () functions in said regression tree model; fitting the plurality of blood information samples with a regression tree model, and taking the regression tree model with the fitted regression tree model as a detection model, wherein the method comprises the following steps:
inputting the plurality of blood information samples into the regression tree model to train the regression tree model, wherein the blood values in the blood information samples are used as input parameters of the regression tree model, and the symptom characteristic values in the blood information samples are used as label parameters of the regression tree model; the process of training the regression tree model includes: and (3) learning the f () function by adopting a sequential learning method, wherein the formula is as follows:
……
In the above-mentioned formula(s),is to give the input parameter as Blood valuei On the basis of the above, the predicted value after the t-th round of prediction is carried out; the regression tree model has the objective function of:
in the above formula, symptom valuei Is Blood sample set and input parameter Blood valuei The corresponding tag parameter(s) are used,is a regular term, D is a constant term, wherein the regular term controls an objective function Obj (t) Is a training level of (a);
and outputting the trained regression tree model as the detection model.
6. The terminal device of claim 5, wherein before outputting the first alert prompt if the detection value is greater than the detection threshold, further comprising:
inputting the blood values of the blood information samples into the detection model, and obtaining a plurality of output values output by the detection model;
sorting the plurality of output values to generate a sequence of values;
and taking the output value at a preset position in the value sequence as the detection threshold.
7. The terminal device according to claim 5, wherein the symptom characteristic value is a first characteristic value or a second characteristic value, and before outputting the first alert prompt if the detected value is greater than a detection threshold value, further comprising:
Determining a first number of samples of which the symptom characteristic value is the first characteristic value and a second number of samples of which the symptom characteristic value is the second characteristic value according to the plurality of blood information samples;
and determining a first duty ratio and a second duty ratio according to the first sample number and the second sample number, and determining the detection threshold according to the first duty ratio, the second duty ratio, the first characteristic value and the second characteristic value.
8. A computer readable storage medium storing a computer program, characterized in that the computer program when executed by a processor implements the steps of the blood analysis method according to any one of claims 1 to 4.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810457009.3A CN108735295B (en) | 2018-05-14 | 2018-05-14 | Blood analysis method and terminal equipment based on regression tree model |
| PCT/CN2018/097562 WO2019218483A1 (en) | 2018-05-14 | 2018-07-27 | Regression tree model-based blood analysis method and apparatus, terminal device and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810457009.3A CN108735295B (en) | 2018-05-14 | 2018-05-14 | Blood analysis method and terminal equipment based on regression tree model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108735295A CN108735295A (en) | 2018-11-02 |
| CN108735295B true CN108735295B (en) | 2024-01-26 |
Family
ID=63937452
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810457009.3A Active CN108735295B (en) | 2018-05-14 | 2018-05-14 | Blood analysis method and terminal equipment based on regression tree model |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN108735295B (en) |
| WO (1) | WO2019218483A1 (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105550507A (en) * | 2015-12-09 | 2016-05-04 | 上海斐讯数据通信技术有限公司 | Intelligent terminal for measuring blood type and measuring method |
| CN106456078A (en) * | 2013-10-17 | 2017-02-22 | 西门子保健有限责任公司 | Method and system for machine learning based assessment of fractional flow reserve |
| CN106934223A (en) * | 2017-02-28 | 2017-07-07 | 思派（北京）网络科技有限公司 | A kind of blood disease intelligent classification system and method based on big data |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101923648B (en) * | 2009-06-15 | 2015-04-29 | 深圳迈瑞生物医疗电子股份有限公司 | Clustering method and device for support vector machine |
| EP2884888A4 (en) * | 2012-08-16 | 2016-04-20 | Ginger Io Inc | Method for modeling behavior and health changes |
| CN104715129A (en) * | 2013-12-13 | 2015-06-17 | 中国科学院心理研究所 | Mental health state assessment system and method based on mobile equipment using behavior |
-
2018
- 2018-05-14 CN CN201810457009.3A patent/CN108735295B/en active Active
- 2018-07-27 WO PCT/CN2018/097562 patent/WO2019218483A1/en active Application Filing
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106456078A (en) * | 2013-10-17 | 2017-02-22 | 西门子保健有限责任公司 | Method and system for machine learning based assessment of fractional flow reserve |
| CN105550507A (en) * | 2015-12-09 | 2016-05-04 | 上海斐讯数据通信技术有限公司 | Intelligent terminal for measuring blood type and measuring method |
| CN106934223A (en) * | 2017-02-28 | 2017-07-07 | 思派（北京）网络科技有限公司 | A kind of blood disease intelligent classification system and method based on big data |
Non-Patent Citations (1)
| Title |
|---|
| 统计预测模型在医院感染中的应用研究;李娜 等;四川医学(第01期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108735295A (en) | 2018-11-02 |
| WO2019218483A1 (en) | 2019-11-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Molnar et al. | Pitfalls to avoid when interpreting machine learning models | |
| CN109817339B (en) | Patient grouping method and device based on big data | |
| CN111460250B (en) | Image data cleaning method, image data cleaning device, image data cleaning medium, and electronic apparatus | |
| US20110131551A1 (en) | Graphical user interface input element identification | |
| Liu et al. | Clinical time series prediction: Toward a hierarchical dynamical system framework | |
| JP7299002B2 (en) | Discriminator and machine learning method | |
| WO2023280229A1 (en) | Image processing method, electronic device, and storage medium | |
| CN112613617A (en) | Uncertainty estimation method and device based on regression model | |
| CN113705092B (en) | Disease prediction method and device based on machine learning | |
| CN111161884A (en) | Disease prediction method, device, equipment and medium for unbalanced data | |
| US20200279148A1 (en) | Material structure analysis method and material structure analyzer | |
| CN116091963B (en) | Quality evaluation method and device for clinical test institution, electronic equipment and storage medium | |
| CN111368837B (en) | Image quality evaluation method and device, electronic equipment and storage medium | |
| CN108735295B (en) | Blood analysis method and terminal equipment based on regression tree model | |
| US20230385690A1 (en) | Computer-readable recording medium storing determination program, determination apparatus, and method of determining | |
| CN110929877A (en) | Model establishing method, device, equipment and storage medium based on transfer learning | |
| JP6930195B2 (en) | Model identification device, prediction device, monitoring system, model identification method and prediction method | |
| CN113392086B (en) | Medical database construction method, device and equipment based on Internet of things | |
| CN111582404B (en) | Content classification method, device and readable storage medium | |
| CN114943695A (en) | Medical sequence image anomaly detection method, device, equipment and storage medium | |
| CN107995181B (en) | Gait-based identity authentication method, device, equipment and storage medium | |
| CN111276248A (en) | State determination system and electronic device | |
| KR20190078846A (en) | Abnormal sequence identification method based on intron and exon | |
| CN111027607B (en) | Unsupervised high-dimensional data feature importance assessment and selection method and device | |
| JP2018151913A (en) | Information processing system, information processing method, and program |
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 |