WO2019218483A1 - Regression tree model-based blood analysis method and apparatus, terminal device and readable storage medium - Google Patents

Abstract

The present solution is applicable to the field of data processing technology, and provides a regression tree model-based blood analysis method, a terminal device and a computer readable storage medium, the method comprising: acquiring a plurality of blood information samples, each of the blood information samples comprising a blood numerical value and a symptom feature value; fitting a regression tree model to the plurality of blood information samples, and using the fitted regression tree model as a test model; acquiring the blood numerical value of blood to be tested, and inputting the blood numerical value of said blood into the test model to obtain a test value; and if the test value is greater than a test threshold, outputting a first alarm prompt. By training the regression tree model, the present solution implements a comprehensive analysis of the blood numerical value of blood to be tested, improving the reliability and accuracy of blood analysis.

Description

Blood analysis method, device, terminal device and readable storage medium based on regression tree model

This application claims priority to Chinese Patent Application No. 201810457009.3, entitled "Regression Tree Model Based Blood Analysis Method and Terminal Equipment", submitted to the Chinese Patent Office on May 14, 2018, the entire contents of which are incorporated by reference. In this application.

Technical field

The present application belongs to the field of data processing technologies, and in particular, to a blood analysis method, apparatus, terminal device and computer readable storage medium based on a regression tree model.

Background technique

Blood is an important part of the human body. With the continuous advancement of science and technology, the analysis of blood is also progressing towards refinement and diversification. Generally speaking, in the blood analysis process, first, it is necessary to obtain numerical information of each component in the blood. The specific process is to place the blood to be tested on the test instrument, and then the analyst obtains the numerical information of the blood according to the test result displayed by the test instrument. Then, based on the statistical theory, the mathematical theory, and the preset threshold, the computer judges whether the numerical information exceeds the corresponding threshold, and obtains the blood analysis result according to the judgment result.

In order to prevent erroneous judgment, the value of the threshold is generally set to a range larger than the normal value. If the problem of blood in a certain problem is that some numerical information appears to be higher or lower, but the numerical information of each component does not exceed the corresponding threshold, then the analysis result obtained by analyzing according to the existing analytical method is obviously wrong. In summary, the existing blood analysis method can only analyze the numerical information of each component in the blood alone, and cannot comprehensively judge a plurality of numerical information, that is, the reliability and accuracy of analyzing the blood are low.

technical problem

In view of this, the embodiments of the present application provide a blood analysis method, device, terminal device, and computer readable storage medium based on a regression tree model, so as to solve the analysis of multiple components in the prior art that cannot integrate blood, and analyze blood. The problem of low reliability and accuracy.

Technical solution

A first aspect of the embodiments of the present application provides a blood analysis method based on a regression tree model, including:

Obtaining a plurality of blood information samples, each of the blood information samples including a blood value and a symptom characteristic value;

Comparing the plurality of blood information samples with a regression tree model, and fitting the completed regression tree model as a detection model;

Obtaining a blood value of the blood to be tested, and inputting the blood value of the blood to be tested into the detection model to obtain a detection value;

If the detected value is greater than the detection threshold, the first alarm prompt is output.

A second aspect of an embodiment of the present application provides a blood analysis method apparatus based on a regression tree model, which may include means for implementing the steps of the above-described regression tree model based blood analysis method.

A third aspect of the embodiments of the present application provides a terminal device, including a memory and a processor, where the computer stores computer readable instructions executable on the processor, the processor executing the computer The steps of the above-described regression analysis based blood analysis method are implemented when the instruction is read.

A fourth aspect of the embodiments of the present application provides a computer readable storage medium storing computer readable instructions, the computer readable instructions being executed by a processor to implement the above-described regression tree model based The steps of the blood analysis method.

Beneficial effect

In the embodiment of the present application, a plurality of blood information samples are obtained, each blood information sample includes a blood value and a symptom feature value, and a plurality of blood information samples are fitted to a regression tree model, and the fitted regression tree model is used as a detection. The model, then the blood to be analyzed, first obtain the blood value of the blood to be tested, and input the blood value to the detection model, obtain the output of the detection model as the detection value, and finally, if it is determined that the detection value is greater than the detection threshold, the output is An alarm prompts that the embodiment of the present application trains the regression tree model based on a plurality of blood information samples, realizes comprehensive analysis of blood values in the blood to be measured, and improves the reliability and accuracy of the blood analysis.

DRAWINGS

1 is a flowchart showing an implementation of a blood analysis method based on a regression tree model in the first embodiment of the present application;

2 is a flowchart showing an implementation of a blood analysis method based on a regression tree model in Embodiment 2 of the present application;

3 is a flowchart showing an implementation of a blood analysis method based on a regression tree model in Embodiment 3 of the present application;

4 is a flowchart showing an implementation of a blood analysis method based on a regression tree model in Embodiment 4 of the present application;

5 is a flowchart showing an implementation of a blood analysis method based on a regression tree model in Embodiment 5 of the present application;

6 is a structural block diagram of a blood analysis apparatus based on a regression tree model in Embodiment 6 of the present application;

FIG. 7 is a schematic diagram of a terminal device in Embodiment 7 of the present application.

Embodiments of the invention

In order to more clearly understand the technical features, objects and effects of the present application, the specific embodiments of the present application will be described in detail with reference to the accompanying drawings.

Please refer to FIG. 1. FIG. 1 is a flowchart of implementing a blood analysis method based on a regression tree model according to an embodiment of the present application. As shown in Figure 1, the blood analysis method comprises the following steps:

S101: Acquire a plurality of blood information samples, each of the blood information samples including a blood value and a symptom characteristic value.

In the embodiment of the present application, before analyzing the blood to be tested, a plurality of blood information samples are first acquired, and each blood information sample includes a blood value and a symptom characteristic value. The blood value is the relevant value of one or several components in the existing blood of the blood information sample, and can be determined by the test result obtained by the existing blood test instrument, and the existing blood refers to the analyzed instrument and the file is established. Blood. For example, the blood value can be the red blood cell content of the existing blood, or a collection of red blood cell content, white blood cell content, and hemoglobin content of the existing blood. The symptom characteristic value is the affected condition of the existing blood, and is specifically set for a certain symptom, and the symptom characteristic value is determined by judging whether the existing blood is affected by the symptom, and generally, the existing blood affected by the symptom is determined. The symptom characteristic value under the blood information sample is set to 1, and the symptom characteristic value under the blood information sample of the existing blood that is not affected by the symptom is set to 0 to facilitate calculation, and the purpose of the embodiment of the present application is to judge Whether the blood to be tested is affected by the symptoms. In order to improve the accuracy of the blood analysis to be measured, the blood information sample is obtained according to a preset order of magnitude, that is, the obtained blood information sample quantity is at a preset order of magnitude, and the order of magnitude can be freely set according to the accuracy requirement. For example, the order of magnitude can be one thousand.

Preferably, a plurality of blood information samples are obtained from a health file or a hospital database. Since a plurality of blood information samples of different populations are stored in the health file or the hospital database, the number of blood information samples is large, and the symptom characteristic values under the blood information samples are highly accurate, so the need can be directly obtained from the health file or the hospital database. Multiple blood information samples. Optionally, multiple blood information samples are acquired according to specific conditions. In the actual application scenario, the acquisition condition is set according to the scene requirement to acquire multiple blood information samples, for example, on the premise that the blood to be tested of the human body in a certain town needs to be analyzed, specific conditions are set, only relevant to the town. The health file or hospital database obtains multiple blood information samples; for example, if the blood to be tested is aged for people over 60 years old, the corresponding specific conditions are set, and the age is 60 years old from the health file or hospital database. The plurality of blood information samples corresponding to the above population may also set specific conditions to obtain a plurality of blood information samples of a certain order of magnitude, etc., thereby improving applicability to different application scenarios.

S102: Fit the plurality of blood information samples with a regression tree model, and use the regression tree model that is fitted to be a detection model.

Since the blood value under the blood information sample may contain a plurality of related values of the 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 the symptoms. For example, the blood value of the blood information sample includes the white blood cell content, the hemoglobin content, and the platelet content. If the blood of the blood information sample belongs to the chronic obstructive pulmonary disease, it is affected by the chronic obstructive pulmonary disease. The blood value of normal blood, the blood value of the blood information sample may appear to increase white blood cell content, increase hemoglobin content and decrease platelet content. However, in a general case, it is not known that a plurality of related values of a plurality of components in the blood value are affected by the symptoms. Therefore, in the embodiment of the present application, a plurality of blood information samples having the symptom characteristic values have been determined and the regression tree. The model is fitted, and the regression tree model is continuously trained during the fitting process. Finally, the fitted regression tree model is output as the detection model. Among them, the regression tree model applies the classification and regression tree (CART) algorithm, and the CART algorithm takes multiple correlation values in blood values as multiple features, and establishes "yes" and "no" for each relevant value. Two forks, usually the left branch is a branch with a value of "yes", and the right branch is a branch with a value of "no". The specific division method is based on the symptom characteristic values corresponding to multiple related values. Recursively divide each relevant value by the CART algorithm, divide the input space composed of multiple related values into a limited number of units, and determine the unit value of each unit by the symptom feature values corresponding to the multiple related values, and finally generate Binary tree. The input of multiple blood information samples into the regression tree model essentially establishes and corrects the binary tree based on multiple blood information samples, and the resulting binary tree is the detection model. The specific calculation process is described later.

S103: Acquire a blood value of the blood to be tested, and input the blood value of the blood to be tested into the detection model to obtain a detection value.

After the detection model is generated from the plurality of blood information samples, the blood to be measured is analyzed, the blood value of the blood to be tested is obtained, and the blood value of the blood to be tested is input to the detection model, and the output result of the detection model is obtained, which is the detection value. It is worth mentioning that the plurality of components corresponding to the blood value of the blood to be tested are the same as the plurality of components corresponding to the blood value of the blood information sample, or the blood component corresponding to the blood value of the blood to be tested is the blood value of the blood information sample. A part of the corresponding plurality of components, in the above two cases, after the blood value of the blood to be tested is input to the detection model, the output of the detection model can be regarded as an effective detection value.

S104: If the detected value is greater than the detection threshold, output a first alarm prompt.

The detection value is calculated from the blood value of the blood to be tested, and is used to judge whether the blood to be tested is affected by the symptoms. If the symptom characteristic value is 1 indicating that the corresponding blood has been affected by the symptoms, and the symptom characteristic value is 0 indicating that the corresponding blood is not affected by the symptoms, the closer the detection value is to 1, the more likely the blood to be tested is affected by the symptoms. In order to improve the accuracy of the judgment, the detection threshold is set. If the detection value is greater than the detection threshold, it is determined that the blood to be tested is affected by the symptom, and the first alarm prompt is output, which is convenient for the user to view; if the detection value is less than or equal to the detection threshold, it is determined to be The blood is not affected by the symptoms, and the output is normal. The detection threshold may 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 may be input to the detection model, and the plurality of blood information samples are output according to the detection model. The output is determined.

It can be seen from the embodiment shown in FIG. 1 that, in the embodiment of the present application, a plurality of existing blood information samples are obtained, wherein the blood information samples include blood values and symptom feature values, and the plurality of blood information samples and the regression tree model are obtained. The fitting is performed, and the fitted regression tree model is output as a detection model, and finally, the blood to be measured is analyzed, the blood value of the blood to be tested is obtained, and the blood value of the blood to be tested is input to the detection model, and the model is detected. The output result is used as a detection value, and the detection value is compared with the detection threshold. If the detection value is greater than the detection threshold, it is proved that the blood to be tested is affected by the symptom, and the first alarm prompt is output. The embodiment of the present application can perform blood values for different components. The analysis of the blood to be tested improves the applicability of the blood analysis in different scenarios, and improves the accuracy of the blood analysis by training the regression tree model.

Please refer to FIG. 2. FIG. 2 is a flowchart of an implementation of a blood analysis method based on a regression tree model according to Embodiment 2 of the present application. With respect to the embodiment corresponding to FIG. 1, this embodiment refines S102 to obtain S201-S202, which are as follows:

S201: 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 parameter of the regression tree model, The symptom feature value in the blood information sample is used as the tag parameter 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, (Blood _value1 , Symptom _value1 ), (Blood _value2 , Symptom _value2 ), ( Blood _value1 , Symptom _value1 )...(Blood _valuen , Symptom _valuen ), where Blood _valuei indicates the blood value of the i-th blood information sample, and Symptom _valuei indicates the symptom characteristic value of the i-th blood information sample. After the construction is completed, the blood sample set is input to the regression tree model, the blood value is used as an input parameter of the regression tree model, and the symptom feature value is used as a label parameter of the regression tree model, and is fitted to the regression tree model, in the embodiment of the present application. The calculation formula for the regression tree model is:

In the above formula,

Indicates the predicted value of the input parameter for the Blood _valuei , that is, the output result after the regression tree model is input by inputting the Blood _valuei as an input parameter to the regression tree model. f() indicates a function that exists in the function space. The function space refers to a set of functions of a given kind from one set to another, that is, the f() function is initially in an unknown state, and K is in the regression tree model. There are K of the above f() functions.

After determining the calculation formula of the regression tree model, the regression tree model is trained based on the blood sample set, so that the K f() functions in the final regression tree model maximize the data in the blood sample set. In the embodiment of the present application, the f() function is learned by a sequential learning method to reduce errors in the learning process. For example, on the basis of the input parameter is a Blood _valuei , the predicted value of the t-round is predicted. And when predicting the prediction value of the t-th round, the prediction result of the prediction value of the t-1th round is retained, that is, each regression tree (each f() function) in the regression tree model is sequentially trained, as follows:

In the above formula

The prediction value after the prediction of the t-th round is given on the basis that the input parameter is the Blood _valuei . In order to judge the f() function required in the sequential learning process, that is, to determine the f() function that conforms to the blood sample set, the objective function of the regression tree model is constructed, as follows:

In the above formula, Symptom _valuei concentration is a blood sample _Blood valuei input parameter corresponding to the tag parameter, i.e. the value of the blood sample and the information value corresponding to the symptoms characteristic of the blood. In the above formula

For the regular term, D is a constant term, wherein the regular term controls the training degree of the objective function Obj ^(t) to prevent over-fitting of the blood sample set and the regression tree model, and the regular term is specifically described later; the constant term is one Constant, the constant term is set in the embodiment of the present application mainly for controlling the numerical range of Obj ^(t) . In addition,

For the error function, the objective function Obj ^(t) is optimized, that is, the appropriate f() function is found.

The value is minimized.

In the embodiment of the present application, in order to facilitate optimization of the objective function,

Expand and define:

The expanded target function is:

Extract all the constant items in the expanded target function to generate a training function, as follows:

In the training function, the output value obtained by the training function depends on the values of g _i and h _i , which improves the simplicity of the training. Therefore, in the embodiment of the present application, the regression tree model can be trained by the finally generated training function.

Optionally, after defining the f() function as

ω∈R ^K , q:R ^d →{1,2,...,K}, based on the definition of the regular term

The f _{t (Blood} valuei) as a regression tree, the value of f _{t (Blood} valuei) of the regression prediction value is the tree of blood _Blood valuei value. There are K leaf nodes in the regression tree, and the values of the above K leaf nodes constitute a K-dimensional vector ω. The q(Blood _valuei ) in the f() function is used to map, specifically mapping a blood value Blood _valuei to a value from 1 to K, that is, dividing the Blood _valuei into a node of the regression tree. On the basis of the foregoing, define the regular term

Where H is the same as D in the objective function and is a constant term; γ and λ are training coefficients, which can be customized by the user for the actual application scenario to adjust the structure of the generated regression tree (the size of the detection model), γ and Any increase in λ will result in a streamlined structure of the regression tree. Through the defined regular items, the training effect of the regression tree model can be significantly improved.

S202: Output the trained regression tree model as the detection model.

When all the blood values and symptom characteristic values in the blood sample set are input into the regression tree model, and the regression tree model training is completed, the trained regression tree model is output as the detection model. When the test needs to be taken for analysis of blood, the blood test values _Blood valuex blood detection model input, the calculation formula can be optimized by detecting a Model

After calculation, the detected value is obtained.

As shown in the embodiment shown in FIG. 2, in the embodiment of the present application, the existing plurality of blood information samples are input to a regression tree model, wherein the blood value in the blood information sample is used as an input parameter of the regression tree model, The symptom feature value in the blood information sample is used as the tag parameter of the regression tree model, the regression tree model is trained, and the trained regression tree model is obtained as the detection model. The embodiment of the present application embodies the specific training process of the regression tree model, and Training accuracy and training effects are improved by setting the objective function.

Referring to FIG. 3, FIG. 3 is a flowchart of an implementation of a blood analysis method based on a regression tree model according to Embodiment 3 of the present application. With respect to the embodiment corresponding to FIG. 1, the embodiment refines the process before S104 to obtain S301-S303, which are as follows:

S301: Input blood values of the plurality of blood information samples to the detection model, and acquire a plurality of output values output by the detection model.

After the detection model is generated, the blood value of the blood to be tested can be input to the detection model to obtain a detection value, and according to the detection value, it is judged whether the blood to be tested has been affected by the symptom. In order to facilitate the above determination, in the embodiment of the present application, after the detection model is acquired, the blood value of the plurality of blood information samples is input as an input parameter to the detection model, and the blood sample of the detection model and the plurality of blood information samples are acquired. Corresponding multiple output values.

S302: Sort the plurality of output values to generate a sequence of values.

The detection model is generated based on the blood value and the symptom feature value in the plurality of blood information samples, but when the blood value in the plurality of blood information samples is re-entered into the detection model, the operation of the training model is not performed, so the output value generated by the detection model is There is a difference from the original symptom characteristic value. After obtaining multiple output values, multiple output values are sorted according to size to generate a sequence of values, and the first column of the sequence of values is the output value with the largest value. It is worth mentioning that if the first output value of the plurality of output values is the same as the second output value, the sequence of values is generated according to the input order of the blood information samples corresponding to the first output value and the second output value. Specifically, the order writing mechanism is set. When an output value needs to be written into the numerical sequence, it is determined whether there is an existing output value in the numerical sequence that is the same as the output value, if there is no existing output that is the same as the output value. The value is used to find the smallest output value in the numerical sequence that is greater than the output value according to the value of the output value, and write the output value at the position after the existing output value; if there is the same value as the output value The existing output value is written to the output value after the existing output value. If there are multiple existing output values that are the same as the output value, the existing output at the end of the existing output value is searched. The value is written to the output value after the existing output value.

S303: Use an output value of the numerical sequence located at a preset position as the detection threshold.

After the numerical sequence is generated, the output value at the preset position in the numerical sequence is used as the detection threshold. For example, if 300 output values have been written in the numerical sequence and the preset position is the 50th position, the output value of the 50th digit of the numerical sequence is extracted as the detection threshold. Optionally, the preset position is determined according to a preset ratio. The preset ratio can be determined according to the big data analysis method. For example, a plurality of national blood samples can be counted nationwide, and each national blood sample includes blood value and symptom characteristic value, and the symptom characteristic value of 1 is determined in a plurality of national blood samples. The proportion of blood samples nationwide, and the ratio is used as a preset ratio. For example, if the ratio is 10% and there are 300 output values in the numerical sequence, the preset position is the 30th position, and the output value of the 30th digit of the extracted numerical sequence is used as the detection threshold, and the detection threshold is set by the big data analysis method. Generally universal applicability. Of course, the preset ratio or preset position can also be pre-set by humans.

As shown in the embodiment shown in FIG. 3, in the embodiment of the present application, the blood value of the plurality of blood information samples is input as an input parameter to the detection model, and the plurality of output values output by the detection model are acquired, and the plurality of output values are obtained. Sorting according to the size, generating a numerical sequence, extracting the output value at the preset position in the numerical sequence, using the output value as the detection threshold, and setting the detection threshold according to the detection model, thereby improving the applicability of the detection threshold.

Please refer to FIG. 4. FIG. 4 is a flowchart of an implementation of a blood analysis method based on a regression tree model according to Embodiment 4 of the present application. With respect to the embodiment corresponding to FIG. 1, in the embodiment, after the symptom feature value is the first feature value or the second feature value, the process before S104 is refined to obtain S401 to S402, which are as follows:

S401: Determine, according to the plurality of blood information samples, the first sample number of the symptom characteristic value as the first feature value and the second sample number of the symptom feature value as the second feature value.

In general, the symptom characteristic values in the blood information sample have two values, namely 1 or 0, but it should be understood that other characteristic symptom characteristic values can also be applied in the embodiments of the present application. For convenience of explanation, the symptom characteristic value of 1 is taken as the first characteristic value, and the symptom characteristic value of 0 is taken as the second characteristic value, and the first characteristic value represents that the existing blood to which the blood information sample belongs has been affected by the symptom, The second eigenvalue represents that the existing blood to which the blood information sample belongs is not affected by the symptoms. In the embodiment of the present application, the number of blood information samples in which the symptom characteristic value is the first characteristic value in the plurality of blood information samples is taken as the first sample number, and the symptom characteristic value in the plurality of blood information samples is the second value. The number of blood information samples of the feature value is taken as the number of second samples.

S402: Determine a first proportion ratio and a second proportion ratio according to the first sample number of people and the second sample number, and according to the first proportion ratio, the second proportion ratio, the The first feature value and the second feature value determine the detection threshold.

After the first sample number and the second sample number are determined, determining a first proportion of the first sample number in a total number of the plurality of blood information samples, and the second sample number is plural a second proportion of the total number of blood information samples, and the detection threshold is determined according to the first ratio, the second ratio, the first characteristic value, and the second characteristic value. For example, if the first eigenvalue is greater than the second eigenvalue, one method of determining the detection threshold is to obtain a difference between the first eigenvalue and the second eigenvalue, and the first proportion ratio and the difference A multiplication operation is performed to obtain a boundary value, and a result obtained by subtracting the boundary value from the first feature value is used as a detection threshold. For example, if the first feature value is 2, the second feature value is 1, and the first ratio is 30%, the detection threshold is 2-(2-1)×30%=1.7. Certainly, the above is only a specific example of the method for determining the detection threshold, and does not constitute a limitation on the embodiment of the present application. According to different actual application scenarios, more methods for determining the detection threshold may be applied.

It can be seen from the embodiment shown in FIG. 4 that, in the embodiment of the present application, the symptom characteristic value in the plurality of blood information samples is determined as the first characteristic value by using the symptom characteristic value as the first characteristic value or the second characteristic value. The first sample number and the symptom characteristic value are the second sample number of the second feature value, and the first proportion ratio and the second proportion ratio are determined according to the first sample number of people and the second sample number, and finally according to The first proportion ratio, the second proportion ratio, the first eigenvalue, and the second eigenvalue are obtained by the detection threshold. Since the plurality of blood information samples may be collected in a certain area, the embodiment of the present application improves the detection threshold for a certain Geographical accuracy.

Please refer to FIG. 5. FIG. 5 is a flowchart of implementing a blood analysis method based on a regression tree model according to Embodiment 5 of the present application. The embodiment further includes the following steps with respect to the embodiment corresponding to FIG. 1 :

S501: Acquire a preset plurality of blood threshold templates, wherein the blood threshold template is determined by a blood value of the alarm blood.

Because different symptoms have different effects on blood values, multiple blood threshold templates corresponding to multiple symptoms are acquired in advance, and the blood threshold template is the same as the blood value of the alarm blood that has been affected by a certain symptom, wherein the blood threshold template may Contains the relevant values of one or several blood components in the alarm blood. The types of related values can be selected according to the actual application scenario. Optionally, obtaining a plurality of alarm bloods that have been affected by a symptom, and performing average evaluation of the plurality of blood values of the plurality of alarm blood, and using the generated average value as a blood threshold template, which can reduce the generation of the blood threshold template. The effect of unexpected factors increases the accuracy of the blood threshold template. In addition, it is also possible to obtain a plurality of alarm bloods that are affected by the symptoms for a preset time, such as obtaining a plurality of alarm bloods that have been affected by the symptoms for six months, and by the same average number evaluation method, multiple alarm bloods are provided. The blood values generate a blood threshold template to further improve the accuracy in the time dimension based on the blood threshold template.

S502: Perform a one-to-one comparison of 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 template may contain multiple related values, the alignment direction is set for multiple related values of the blood threshold template for the symptoms. Taking the symptoms of chronic obstructive pulmonary disease as an example, blood levels affected by chronic obstructive pulmonary disease increase, hemoglobin content increases and platelet content decreases, so the blood threshold template corresponding to chronic obstructive pulmonary disease contains leukocyte content and hemoglobin content. When the three values are related to the platelet content, the white blood cell content is set to the upward alignment, and the hemoglobin content is also set to the opposite direction, and the platelet content is set to the opposite direction. The upward alignment means that if the white blood cell content in the blood value of the blood to be tested exceeds or equals the white blood cell content in the blood threshold template, the comparison is determined to be successful, otherwise the comparison fails; the downward comparison refers to the blood to be tested. If the platelet content in the blood value is less than the white blood cell content in the blood threshold template, the comparison is deemed successful, otherwise the comparison is deemed to have failed. In the embodiment of the present application, a setting mechanism may be created for different symptoms, and the setting of the matching direction in the plurality of blood threshold templates is automatically completed by the setting mechanism.

S503: If the blood value of the blood to be tested is successfully aligned with one of the plurality of blood threshold templates, outputting a second alarm prompt.

After the setting of the alignment direction of the plurality of blood threshold templates is completed, the blood value of the blood to be tested is compared with a plurality of blood threshold templates, and if the blood value of the blood to be tested is successfully compared with a blood threshold template, the output is The second alarm prompt, wherein the second alarm prompt may include a symptom corresponding to the blood threshold template, which is convenient for the user to quickly view; if the blood blood value of the blood to be tested and the plurality of blood threshold templates fail to match, the normal prompt is output.

According to the embodiment shown in FIG. 5, in the embodiment of the present application, by acquiring a plurality of preset blood threshold templates for a plurality of symptoms, the blood threshold template is taken from the blood value of the alarm blood affected by the symptom, and is to be tested. The blood blood value is compared with the plurality of blood threshold templates one by one. If the blood blood value of the blood to be tested is successfully compared with one of the plurality of blood threshold templates, the second alarm prompt is output, and the alarm is generated, and the alarm is set. A blood threshold template enables multi-dimensional analysis of the blood to be measured and improves the reliability of the blood analysis to be tested.

Corresponding to a regression tree model-based blood analysis method described in the above embodiments, FIG. 6 is a structural block diagram of a regression tree model-based blood analysis apparatus provided by an embodiment of the present application. Referring to FIG. 6, The device includes:

a first obtaining unit 61, configured to acquire a plurality of blood information samples, each of the blood information samples including a blood value and a symptom characteristic value;

a fitting unit 62, configured to fit the plurality of blood information samples with a regression tree model, and use the regression tree model that is completed as a detection model;

a second obtaining unit 63, configured to acquire a blood value of the blood to be tested, and input the blood value of the blood to be tested into the detection model to obtain a detection value;

The output unit 64 is configured to output a first alarm prompt if the detected 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 using the symptom feature value in the blood information sample as a label vector of the regression tree model;

And a model output unit, configured to output the trained regression tree model as the detection model.

Optionally, the output unit 64 further includes:

a sample input unit, configured to input blood values of the plurality of blood information samples to the detection model, and acquire a plurality of output values output by the detection model;

a sorting unit, configured to sort the plurality of output values to generate a sequence of values;

And a threshold output unit configured to use an output value of the numerical sequence at a preset position as the detection threshold.

Optionally, the symptom characteristic value is a first feature value or a second feature value, and the output unit 64 further includes:

a first determining unit, configured to determine, according to the plurality of blood information samples, that the symptom feature value is the first sample number of the first feature value and the symptom feature value is the second feature value Two sample numbers;

a second determining unit, configured to determine, according to the first sample number of people and the second sample number of people, a first proportion ratio and a second proportion ratio, and according to the first proportion ratio, the second proportion The ratio, the first eigenvalue, and the second eigenvalue determine the detection threshold.

Optionally, the blood analysis device further includes:

a template acquiring unit, configured to acquire a preset plurality of blood threshold templates, wherein the blood threshold template is determined by a blood value of the alarm blood;

An aligning unit, configured to compare the blood value of the blood to be tested and the plurality of blood threshold templates one by one;

The alarm output unit is configured to output a second alarm prompt if the blood value of the blood to be tested is successfully matched with one of the plurality of blood threshold templates.

FIG. 7 is a schematic diagram of a terminal device according to an embodiment of the present application. As shown in FIG. 7, the terminal device 7 of this embodiment includes a processor 70 and a memory 71 in which computer readable instructions 72 executable on the processor 70 are stored, for example, based on a regression tree model. Blood analysis program. The processor 70 executes the computer readable instructions 72 to implement the steps in the various embodiments of the regression tree model based blood analysis method described above, such as steps S101 through S104 shown in FIG. Alternatively, the processor 70, when executing the computer readable instructions 72, implements the functions of the various units in the apparatus embodiments described above, such as the functions of the units 61 through 64 shown in FIG.

Illustratively, the computer readable instructions 72 may be partitioned into one or more modules/units that are stored in the memory 71 and executed by the processor 70, To complete this application. The one or more modules/units may be a series of computer readable instruction segments capable of performing a particular function for describing the execution of the computer readable instructions 72 in the terminal device 7. For example, the computer readable instructions 72 may be segmented into a first acquisition unit, a fitting unit, a second acquisition unit, and an output unit, each unit having a specific function as described above.

The terminal device may include, but is not limited to, a processor 70 and a memory 71. It will be understood by those skilled in the art that FIG. 7 is only 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 those illustrated, or combine some components or different components. For example, the terminal device may further include an input/output device, a network access device, a bus, and the like.

The processor 70 may be a central processing unit (CPU), or may be other general-purpose processors, a digital signal processor (DSP), an application specific integrated circuit (ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, etc. The general purpose processor may be a microprocessor or the processor or 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 also be an external storage device of the terminal device 7, for example, a plug-in hard disk provided on the terminal device 7, a smart memory card (SMC), and a secure digital (SD). Card, flash card, etc. Further, the memory 71 may also include both an internal storage unit of the terminal device 7 and an external storage device. The memory 71 is for storing the computer readable instructions and other programs and data required by the terminal device. The memory 71 can also be used to temporarily store data that has been output or is about to be output.

In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

The integrated unit, if implemented in the form of a software functional unit and sold or used as a standalone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application, in essence or the contribution to the prior art, or all or part of the technical solution may be embodied in the form of a software product stored in a storage medium. A number of instructions are included to cause a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and the like, which can store program codes. .

The above embodiments are only used to explain the technical solutions of the present application, and are not limited thereto; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still The technical solutions described in the embodiments are modified, or the equivalents of the technical features are replaced by the equivalents. The modifications and substitutions of the embodiments do not depart from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims (20)

1. A blood analysis method based on a regression tree model, comprising:

   Obtaining a plurality of blood information samples, each of the blood information samples including a blood value and a symptom characteristic value;

   Comparing the plurality of blood information samples with a regression tree model, and fitting the completed regression tree model as a detection model;

   Obtaining a blood value of the blood to be tested, and inputting the blood value of the blood to be tested into the detection model to obtain a detection value;

   If the detected value is greater than the detection threshold, the first alarm prompt is output.
2. The blood analysis method according to claim 1, wherein the fitting the plurality of blood information samples to a regression tree model, and fitting the completed regression tree model as a detection model comprises:

   Inputting the plurality of blood information samples to the regression tree model to train the regression tree model, wherein the blood value in the blood information sample is used as an input parameter of the regression tree model, the blood is The symptom feature value in the information sample is used as a tag parameter of the regression tree model;

   The trained regression tree model is output as the detection model.
3. The blood analysis method according to claim 1, wherein if the detection value is greater than the detection threshold, before the outputting the first alarm prompt, the method further includes:

   And inputting blood values of the plurality of blood information samples to the detection model, and acquiring a plurality of output values output by the detection model;

   Sorting the plurality of output values to generate a sequence of values;

   An output value at the preset position in the numerical sequence is used as the detection threshold.
4. The blood analysis method according to claim 1, wherein the symptom characteristic value is a first characteristic value or a second characteristic value, and if the detection value is greater than a detection threshold, before the first alarm prompt is output, Also includes:

   Determining, according to the plurality of blood information samples, the first sample number of the symptom characteristic value of the first feature value and the second sample number of the symptom feature value of the second feature value;

   Determining, according to the first sample number of people and the second sample number of people, a first proportion ratio and a second proportion ratio, and according to the first ratio ratio, the second proportion ratio, the first The feature value and the second feature value determine the detection threshold.
5. The blood analysis method according to claim 1, further comprising:

   Obtaining a preset plurality of blood threshold templates, wherein the blood threshold template is determined by a blood value of the alarm blood;

   Comparing the blood value of the blood to be tested and the plurality of blood threshold templates one by one;

   And if the blood value of the blood to be tested is successfully aligned with one of the plurality of blood threshold templates, the second alarm prompt is output.
6. A blood analysis device based on a regression tree model, comprising:

   a first acquiring 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;

   a fitting unit, configured to fit the plurality of blood information samples with a regression tree model, and use the regression tree model that is completed as a detection model;

   a second acquiring unit, configured to acquire a blood value of the blood to be tested, and input the blood value of the blood to be tested into the detection model to obtain a detection value;

   And an output unit, configured to output a first alarm prompt if the detected value is greater than a detection threshold.
7. The blood analysis device according to claim 6, wherein the fitting unit comprises:

   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 using the symptom feature value in the blood information sample as a label vector of the regression tree model;

   And a model output unit, configured to output the trained regression tree model as the detection model.
8. The blood analysis device according to claim 6, wherein the output unit further comprises:

   a sample input unit, configured to input blood values of the plurality of blood information samples to the detection model, and acquire a plurality of output values output by the detection model;

   a sorting unit, configured to sort the plurality of output values to generate a sequence of values;

   And a threshold output unit configured to use an output value of the numerical sequence at a preset position as the detection threshold.
9. The blood analysis device according to claim 6, wherein the symptom characteristic value is a first feature value or a second feature value, and the output unit further comprises:

   a first determining unit, configured to determine, according to the plurality of blood information samples, that the symptom feature value is the first sample number of the first feature value and the symptom feature value is the second feature value Two sample numbers;

   a second determining unit, configured to determine, according to the first sample number of people and the second sample number of people, a first proportion ratio and a second proportion ratio, and according to the first proportion ratio, the second proportion The ratio, the first eigenvalue, and the second eigenvalue determine the detection threshold.
10. The blood analysis device according to claim 6, further comprising:

    a template acquiring unit, configured to acquire a preset plurality of blood threshold templates, wherein the blood threshold template is determined by a blood value of the alarm blood;

    An aligning unit, configured to compare the blood value of the blood to be tested and the plurality of blood threshold templates one by one;

    The alarm output unit is configured to output a second alarm prompt if the blood value of the blood to be tested is successfully matched with one of the plurality of blood threshold templates.
11. A terminal device, comprising: a memory and a processor, wherein the memory stores computer readable instructions executable on the processor, and the processor implements the following steps when the computer readable instructions are executed :

    Obtaining a plurality of blood information samples, each of the blood information samples including a blood value and a symptom characteristic value;

    Comparing the plurality of blood information samples with a regression tree model, and fitting the completed regression tree model as a detection model;

    Obtaining a blood value of the blood to be tested, and inputting the blood value of the blood to be tested into the detection model to obtain a detection value;

    If the detected value is greater than the detection threshold, the first alarm prompt is output.
12. The terminal device according to claim 11, wherein the fitting the plurality of blood information samples with a regression tree model, and fitting the completed regression tree model as a detection model comprises:

    Inputting the plurality of blood information samples to the regression tree model to train the regression tree model, wherein the blood value in the blood information sample is used as an input parameter of the regression tree model, the blood is The symptom feature value in the information sample is used as a tag parameter of the regression tree model;

    The trained regression tree model is output as the detection model.
13. The terminal device according to claim 11, wherein if the detection value is greater than the detection threshold, before the outputting the first alarm prompt, the method further includes:

    And inputting blood values of the plurality of blood information samples to the detection model, and acquiring a plurality of output values output by the detection model;

    Sorting the plurality of output values to generate a sequence of values;

    An output value at the preset position in the numerical sequence is used as the detection threshold.
14. The terminal device according to claim 11, wherein the symptom characteristic value is a first feature value or a second feature value, and if the detection value is greater than a detection threshold, before the first alarm prompt is output, include:

    Determining, according to the plurality of blood information samples, the first sample number of the symptom characteristic value of the first feature value and the second sample number of the symptom feature value of the second feature value;

    Determining, according to the first sample number of people and the second sample number of people, a first proportion ratio and a second proportion ratio, and according to the first ratio ratio, the second proportion ratio, the first The feature value and the second feature value determine the detection threshold.
15. The terminal device according to claim 11, further comprising:

    Obtaining a preset plurality of blood threshold templates, wherein the blood threshold template is determined by a blood value of the alarm blood;

    Comparing the blood value of the blood to be tested and the plurality of blood threshold templates one by one;

    And if the blood value of the blood to be tested is successfully aligned with one of the plurality of blood threshold templates, the second alarm prompt is output.
16. A computer readable storage medium storing computer readable instructions, wherein the computer readable instructions, when executed by at least one processor, implement the following steps:

    Obtaining a plurality of blood information samples, each of the blood information samples including a blood value and a symptom characteristic value;

    Comparing the plurality of blood information samples with a regression tree model, and fitting the completed regression tree model as a detection model;

    Obtaining a blood value of the blood to be tested, and inputting the blood value of the blood to be tested into the detection model to obtain a detection value;

    If the detected value is greater than the detection threshold, the first alarm prompt is output.
17. The computer readable storage medium of claim 16, wherein the computer readable instructions are executed by at least one processor to implement the following steps:

    Inputting the plurality of blood information samples to the regression tree model to train the regression tree model, wherein the blood value in the blood information sample is used as an input parameter of the regression tree model, the blood is The symptom feature value in the information sample is used as a tag parameter of the regression tree model;

    The trained regression tree model is output as the detection model.
18. The computer readable storage medium of claim 16, wherein the computer readable instructions are executed by at least one processor to implement the following steps:

    And inputting blood values of the plurality of blood information samples to the detection model, and acquiring a plurality of output values output by the detection model;

    Sorting the plurality of output values to generate a sequence of values;

    An output value at the preset position in the numerical sequence is used as the detection threshold.
19. The computer readable storage medium according to claim 16, wherein the symptom characteristic value is a first feature value or a second feature value, and the computer readable instructions are further executed as follows when executed by at least one processor :

    Determining, according to the plurality of blood information samples, the first sample number of the symptom characteristic value of the first feature value and the second sample number of the symptom feature value of the second feature value;

    Determining, according to the first sample number of people and the second sample number of people, a first proportion ratio and a second proportion ratio, and according to the first ratio ratio, the second proportion ratio, the first The feature value and the second feature value determine the detection threshold.
20. The computer readable storage medium of claim 16, wherein the computer readable instructions are further executed by the at least one processor to:

    Obtaining a preset plurality of blood threshold templates, wherein the blood threshold template is determined by a blood value of the alarm blood;

    Comparing the blood value of the blood to be tested and the plurality of blood threshold templates one by one;

    And if the blood value of the blood to be tested is successfully aligned with one of the plurality of blood threshold templates, the second alarm prompt is output.

Images