Poisonous substance screening method, device and equipment based on multi-dimensional vector matching and storage medium

Abstract

The application discloses a kind of based on multidimensional vector matching's poisoning material screening method, device, equipment and storage medium, wherein method includes: the multidimensional characteristic data of substance contained in blood sample is obtained;Construct the substance characteristic matrix to be screened;Construct reference substance characteristic standard numerical matrix;For each row characteristic data in the substance characteristic matrix to be screened, calculate the distance of each row characteristic data in reference substance characteristic standard numerical matrix, select the first k row characteristic data of minimum distance;Scaling processing is carried out to the data in k row characteristic data, to obtain the upper limit, lower limit of each characteristic data;For each row characteristic data in the substance characteristic matrix to be screened, compare each row characteristic data in corresponding k row characteristic data after scaling processing, and filter out several rows of characteristic data that meet the condition to constitute fine screening data set.According to the screening of multiple characteristic data, the screening efficiency is greatly improved under the premise of ensuring accuracy.

Description

Technical Field

[0001] This invention relates to the field of medical data processing technology, and in particular to a method, apparatus, device, and storage medium for screening toxic substances based on multidimensional vector matching. Background Technology

[0002] There are many types of substances and causes of poisoning in the human body. Professional medical personnel need to compare the chemical components in human blood to screen for poisoning elements. Blood test results will contain many elemental indicators. By comparing the characteristics of the test indicators and reference indicators, the causes of poisoning can be gradually ruled out. The number of people screened may reach tens of thousands. Such steps are very tedious and complicated, resulting in very low screening efficiency for poisoning substances. Summary of the Invention

[0003] This invention provides a method, apparatus, device, and storage medium for screening toxic substances based on multidimensional vector matching, in order to solve the problem of low efficiency in existing screening methods.

[0004] Firstly, a method for screening toxic substances based on multidimensional vector matching is provided, including:

[0005] To obtain multidimensional characteristic data of substances contained in blood samples;

[0006] Construct an m×n dimensional matrix of material features to be screened, where m is the total number of material types to be screened and n is the dimension of the multidimensional feature data;

[0007] Construct an s×n dimensional reference material characteristic standard numerical matrix, where s is the total number of reference material types;

[0008] For each row of feature data in the feature matrix of the substance to be screened, calculate its distance to each row of feature data in the reference substance feature standard value matrix, and select the k rows of feature data with the smallest distance;

[0009] Scaling is applied to each column of data in the k rows of feature data to obtain the upper and lower limits of each feature data in the k rows of feature data;

[0010] For each row of feature data in the feature matrix of substances to be screened, compare it with each row of feature data in the corresponding k-row scaled feature data, and select all those that meet the criteria from the k-row feature data. The refined screening dataset consists of several rows of feature data; where x i This represents the i-th feature data in each row of the feature data in the feature matrix of the substance to be screened. Let i and n represent the upper and lower bounds of the i-th feature data in each row of the k-row feature data after scaling, respectively, where i takes the value [1, n].

[0011] Furthermore, the multidimensional feature data includes at least two of the following: standard quality, positive and negative polarity, minimum HAC drift index, minimum HAC interval, blood background AA, urine background AA, and reagent background SD.

[0012] Furthermore, the construction of the s×n-dimensional reference material characteristic standard numerical matrix includes:

[0013] Obtain s substances and their corresponding standard characteristic data from the standard database;

[0014] From the standard feature data of s substances, feature standard values ​​corresponding to the multidimensional feature data are extracted, and an s×n-dimensional reference substance feature standard value matrix is ​​constructed.

[0015] Furthermore, the distance is the Manhattan distance, which is calculated using the following formula:

[0016]

[0017] In the formula, x i y represents the i-th feature data in each row of the feature data of the substance feature matrix to be screened. i This represents the i-th feature data in each row of the reference material characteristic standard numerical matrix.

[0018] Furthermore, the value of k is a preset fixed value or s×q%, where q is a preset value.

[0019] Furthermore, the scaling process performed on each column of data in the k rows of feature data to obtain the upper and lower limits of each feature data in the k rows of feature data includes:

[0020] For each type of feature data, a corresponding scaling threshold r is set, and the value of r ranges from [0,1].

[0021] The scaling process is performed according to the following formula to obtain the upper limit of each feature data in the k rows of feature data. lower limit

[0022]

[0023]

[0024] In the formula, y i r represents the i-th feature data in each row of the reference material characteristic standard numerical matrix. i This represents the scaling threshold for the i-th feature data.

[0025] Secondly, a poisoning substance screening device based on multidimensional vector matching is provided, comprising:

[0026] The data acquisition module is used to acquire multidimensional feature data of substances contained in blood samples;

[0027] The module for constructing the feature matrix of substances to be screened is used to construct an m×n dimensional feature matrix of substances to be screened, where m is the total number of substances to be screened and n is the dimension of the multidimensional feature data.

[0028] The reference material characteristic standard numerical matrix construction module is used to construct an s×n dimensional reference material characteristic standard numerical matrix, where s is the total number of reference material types.

[0029] The coarse screening module is used to calculate the distance between each row of feature data in the feature matrix of the substance to be screened and each row of feature data in the reference material feature standard value matrix, and select the k rows of feature data with the smallest distance.

[0030] The data expansion module is used to scale each column of data in k rows of feature data to obtain the upper and lower limits of each feature data in k rows of feature data;

[0031] The fine screening module compares each row of feature data in the feature matrix of the substances to be screened with each row of feature data in the corresponding scaled k rows of feature data, and selects all substances that meet the requirements from the k rows of feature data. The refined screening dataset consists of several rows of feature data; where x i This represents the i-th feature data in each row of the feature data in the feature matrix of the substance to be screened. Let i and n represent the upper and lower bounds of the i-th feature data in each row of the k-row feature data after scaling, respectively, where i takes the value [1, n].

[0032] Furthermore, the distance is the Manhattan distance, which is calculated using the following formula:

[0033]

[0034] In the formula, x i y represents the i-th feature data in each row of the feature data of the substance feature matrix to be screened. i This represents the i-th feature data in each row of the reference material characteristic standard numerical matrix.

[0035] Furthermore, the scaling process performed on each column of data in the k rows of feature data to obtain the upper and lower limits of each feature data in the k rows of feature data includes:

[0036] For each type of feature data, a corresponding scaling threshold r is set, and the value of r ranges from [0,1].

[0037] The scaling process is performed according to the following formula to obtain the upper limit of each feature data in the k rows of feature data. lower limit

[0038]

[0039]

[0040] In the formula, y i ri represents the i-th feature data in each row of the reference material feature standard numerical matrix, where ri represents the scaling threshold of the i-th feature data.

[0041] Thirdly, an electronic device is provided, comprising:

[0042] A memory that stores computer programs;

[0043] A processor is used to implement the multidimensional vector matching-based poisoning substance screening method as described above when executing the computer program.

[0044] Fourthly, a computer-readable storage medium is provided that stores a computer program, which, when executed by a processor, implements the multidimensional vector matching-based poisoning substance screening method described above.

[0045] Beneficial effects

[0046] This invention proposes a method, apparatus, device, and storage medium for screening poisoning substances based on multidimensional vector matching. In use, the solution provided by this invention performs multidimensional vector matching on blood test data from poisoned patients. First, k possible poisoning substances are coarsely screened out by calculating distances. Then, the feature data of these k possible poisoning substances are scaled and compared to achieve fine screening. After two screening processes, tens of thousands of data points can be reduced to single or double digits, allowing professional medical personnel to make a final judgment. Medical personnel only need to complete the screening on a small dataset. This invention simultaneously performs matching and screening based on multiple feature data, greatly improving screening efficiency without affecting the accuracy of screening. Moreover, it can screen samples from multiple patients simultaneously in a short time, significantly improving screening time efficiency and saving diagnostic time for professional medical personnel. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 This is a flowchart of a poisoning substance screening method based on multidimensional vector matching provided by an embodiment of the present invention;

[0049] Figure 2 This is a schematic diagram of some feature data in the original data provided in the embodiments of the present invention;

[0050] Figure 3 This is an example of substance data to be screened in a poisoned patient provided in an embodiment of the present invention;

[0051] Figure 4 This is an example of data obtained by scaling a standard quality feature by r = 0.05% according to an embodiment of the present invention;

[0052] Figure 5 The embodiment of the present invention provides a scaling of quality features by r = 0.05%, followed by filtering to obtain the top 10 samples. Detailed Implementation

[0053] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be described in detail below. Obviously, the described embodiments are merely some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other implementation methods obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0054] Example 1

[0055] like Figure 1 As shown, this embodiment provides a method for screening toxic substances based on multidimensional vector matching, including:

[0056] S1: Obtain multidimensional feature data of substances contained in blood samples.

[0057] It should be noted that the raw data comes from the physical and chemical characteristic data obtained by separating and detecting blood substances based on the mass differences of atoms, molecules, or molecular fragments using a medical mass spectrometer. The types and number of characteristic data can be selected from the raw data according to actual needs. In this embodiment, the multidimensional characteristic data can select at least two from dozens of characteristics, such as standard mass, positive and negative polarity, minimum HAC drift index (HAC is a chemical reagent), minimum HAC interval, blood background AA, urine background AA, and reagent background SD. Some characteristics of the raw data are shown below. Figure 2 As shown:

[0058] S2: Construct an m×n dimensional matrix of material features to be screened, where m is the total number of material types to be screened and n is the dimension of the multidimensional feature data.

[0059] Where m is a positive integer. When m is 1, it means screening for one substance in the blood of a poisoned patient. When m is greater than 1, it can mean screening for multiple substances in the blood of a poisoned patient, or screening for one or more poisoning substances in the blood of multiple poisoned patients. Figure 3 Example data on 10 substances to be screened for a poisoning patient, based on three characteristics: mass, drift index, and polarity.

[0060] S3: Construct an s×n dimensional reference material characteristic standard numerical matrix, where s is the total number of reference material types. Specifically, this includes:

[0061] The standard database includes several substances and their corresponding standard characteristic data. In this embodiment, s substances and their corresponding standard characteristic data are obtained from the standard database.

[0062] However, the standard feature data corresponding to each substance in the standard database is quite extensive, and not all feature data is what we need. Therefore, it is necessary to extract the feature standard values ​​corresponding to the multidimensional feature data from the standard feature data of s substances and construct an s×n-dimensional reference substance feature standard value matrix.

[0063] S4: For each row of feature data in the feature matrix of the substance to be screened, calculate its distance to each row of feature data in the reference substance feature standard value matrix, and select the k rows of feature data with the smallest distance. Specifically, this includes:

[0064] Each row of feature data in the feature matrix of the substance to be screened and each row of feature data in the reference material feature standard value matrix are treated as vectors. Then, the distance between the two vectors is calculated. In this embodiment, the Manhattan distance is used, and its calculation formula is as follows:

[0065]

[0066] In the formula, x i y represents the i-th feature data in each row of the feature data of the substance feature matrix to be screened. i This represents the i-th feature data in each row of the reference material characteristic standard numerical matrix;

[0067] After the distance calculation is completed, the data is sorted in ascending order, and the top k rows of feature data with the smallest distance are selected. The value of k can be a preset fixed value, such as 50, 80, 100, etc., or it can be calculated by the formula s×q%. The value of q can be selected according to the actual data volume. For example, for data in the tens of thousands, q can be 0.1.

[0068] S5: Scale each column of data in the k rows of feature data to obtain the upper and lower bounds of each feature data point in the k rows. Specifically, this includes:

[0069] For each type of feature data, a corresponding scaling threshold r is pre-set, and the value of r is in the range of [0,1].

[0070] The scaling process is performed according to the following formula to obtain the upper limit of each feature data in the k rows of feature data. lower limit

[0071]

[0072]

[0073] In the formula, y i r represents the i-th feature data in each row of the reference material characteristic standard numerical matrix. i This represents the scaling threshold for the i-th feature data. For example, scaling the quality feature by r = 0.05% yields data like the example shown below. Figure 4 As shown.

[0074] S6: For each row of feature data in the feature matrix of the material to be screened, compare it with each row of feature data in the corresponding scaled k rows of feature data, and select all those that meet the criteria from the k rows of feature data. The refined screening dataset consists of several rows of feature data; where x i This represents the i-th feature data in each row of the feature data in the feature matrix of the substance to be screened. Let represent the upper and lower bounds of the i-th feature data in each row of the k-row feature data after scaling, where i takes the value [1, n]. For example, scaling the quality features in the above column by r = 0.05% and then filtering, the top 10 samples are... Figure 5As shown, the values ​​in the matching rows represent the substances in rows k that meet the mass data range.

[0075] After the above processing, the selected dataset can be handed over to professional medical personnel for evaluation. The medical personnel can complete the screening from a small amount of data, which greatly improves the efficiency of screening without affecting the accuracy.

[0076] Example 2

[0077] This embodiment provides a poisoning substance screening device based on multidimensional vector matching, including:

[0078] The data acquisition module is used to acquire multidimensional feature data of substances contained in blood samples;

[0079] The module for constructing the feature matrix of substances to be screened is used to construct an m×n dimensional feature matrix of substances to be screened, where m is the total number of substances to be screened and n is the dimension of the multidimensional feature data.

[0080] The reference material characteristic standard numerical matrix construction module is used to construct an s×n dimensional reference material characteristic standard numerical matrix, where s is the total number of reference material types.

[0081] The coarse screening module is used to calculate the distance between each row of feature data in the feature matrix of the substance to be screened and each row of feature data in the reference material feature standard value matrix, and select the k rows of feature data with the smallest distance.

[0082] The data expansion module is used to scale each column of data in k rows of feature data to obtain the upper and lower limits of each feature data in k rows of feature data;

[0083] The fine screening module compares each row of feature data in the feature matrix of the substances to be screened with each row of feature data in the corresponding scaled k-row feature data, and selects all substances that meet the requirements from the k-row feature data. The refined screening dataset consists of several rows of feature data; where x i This represents the i-th feature data in each row of the feature data in the feature matrix of the substance to be screened. Let i and n represent the upper and lower bounds of the i-th feature data in each row of the k-row feature data after scaling, respectively, where i takes the value [1, n].

[0084] In this embodiment, the distance is the Manhattan distance, and its calculation formula is as follows:

[0085]

[0086] In the formula, x iy represents the i-th feature data in each row of the feature data of the substance feature matrix to be screened. i This represents the i-th feature data in each row of the reference material characteristic standard numerical matrix.

[0087] The scaling process performed on each column of the k rows of feature data to obtain the upper and lower limits of each feature data in the k rows of feature data includes:

[0088] For each type of feature data, a corresponding scaling threshold r is set, and the value of r ranges from [0,1].

[0089] The scaling process is performed according to the following formula to obtain the upper limit of each feature data in the k rows of feature data. lower limit

[0090]

[0091]

[0092] In the formula, y i r represents the i-th feature data in each row of the reference material characteristic standard numerical matrix. i This represents the scaling threshold for the i-th feature data.

[0093] Example 3

[0094] This embodiment provides an electronic device, including:

[0095] A memory that stores computer programs;

[0096] A processor is used to implement the poisoning substance screening method based on multidimensional vector matching as described in Example 1 when executing the computer program.

[0097] Example 4

[0098] This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the poisoning substance screening method based on multidimensional vector matching as described in Embodiment 1.

[0099] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0100] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0101] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0102] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0103] It is understood that the same or similar parts in the above embodiments can be referred to each other, and the contents not described in detail in some embodiments can be referred to the same or similar contents in other embodiments.

[0104] It should be understood that the functional unit modules in the various embodiments of the present invention can be concentrated in one processing unit, or each unit module can exist physically separately, or two or more unit modules can be integrated into one unit module, and can be implemented in hardware or software.

[0105] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as will be understood by those skilled in the art to which embodiments of the invention pertain.

[0106] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for screening toxic substances based on multidimensional vector matching, characterized in that, include: To obtain multidimensional characteristic data of substances contained in blood samples; The multidimensional feature data includes standard quality, and also includes at least one of positive and negative polarity, minimum HAC drift index, minimum HAC interval, blood background AA, urine background AA, and reagent background SD; Construct an m×n dimensional matrix of material features to be screened, where m is the total number of material types to be screened and n is the dimension of the multidimensional feature data; Construct an s×n dimensional reference material characteristic standard numerical matrix, where s is the total number of reference material types; For each row of feature data in the feature matrix of the substance to be screened, calculate its distance to each row of feature data in the reference substance feature standard value matrix, and select the k rows of feature data with the smallest distance; Scaling is applied to each column of data in the k rows of feature data to obtain the upper and lower limits of each feature data in the k rows of feature data; For each row of feature data in the feature matrix of substances to be screened, compare it with each row of feature data in the corresponding k-row scaled feature data, and select all those that meet the criteria from the k-row feature data. The dataset consists of several rows of feature data; among them, This represents the i-th feature data in each row of the feature data in the feature matrix of the substance to be screened. Let i and n represent the upper and lower bounds of the i-th feature data in each row of the k-row feature data after scaling, respectively, where i takes the value [1, n].

2. The method for screening toxic substances based on multidimensional vector matching according to claim 1, characterized in that, The distance mentioned is the Manhattan distance, and its calculation formula is as follows: ； In the formula, This represents the i-th feature data in each row of the feature data in the feature matrix of the substance to be screened. This represents the i-th feature data in each row of the reference material characteristic standard numerical matrix.

3. The method for screening toxic substances based on multidimensional vector matching according to claim 1, characterized in that, The value of k is a preset fixed value or s×q%, where q is a preset value.

4. The method for screening toxic substances based on multidimensional vector matching according to any one of claims 1 to 3, characterized in that, The scaling process performed on each column of data in the k rows of feature data yields the upper and lower limits of each feature data point in the k rows of feature data, including: For each type of feature data, a corresponding scaling threshold r is set, and the value of r ranges from [0,1]. The scaling process is performed according to the following formula to obtain the upper limit of each feature data in the k rows of feature data. Lower limit : ； ； In the formula, This represents the i-th feature data in each row of the reference material characteristic standard numerical matrix. This represents the scaling threshold for the i-th feature data.

5. A poisoning substance screening device based on multidimensional vector matching, characterized in that, include: The data acquisition module is used to acquire multidimensional feature data of substances contained in blood samples; The multidimensional feature data includes standard quality, and also includes at least one of positive and negative polarity, minimum HAC drift index, minimum HAC interval, blood background AA, urine background AA, and reagent background SD; The module for constructing the feature matrix of substances to be screened is used to construct an m×n dimensional feature matrix of substances to be screened, where m is the total number of substances to be screened and n is the dimension of the multidimensional feature data. The reference material characteristic standard numerical matrix construction module is used to construct an s×n dimensional reference material characteristic standard numerical matrix, where s is the total number of reference material types. The coarse screening module is used to calculate the distance between each row of feature data in the feature matrix of the substance to be screened and each row of feature data in the reference material feature standard value matrix, and select the k rows of feature data with the smallest distance. The data expansion module is used to scale each column of data in k rows of feature data to obtain the upper and lower limits of each feature data in k rows of feature data; The fine screening module compares each row of feature data in the feature matrix of the substances to be screened with each row of feature data in the corresponding scaled k rows of feature data, and selects all substances that meet the requirements from the k rows of feature data. The dataset consists of several rows of feature data; among them, This represents the i-th feature data in each row of the feature data in the feature matrix of the substance to be screened. Let i and n represent the upper and lower bounds of the i-th feature data in each row of the k-row feature data after scaling, respectively, where i takes the value [1, n].

6. The poisoning substance screening device based on multidimensional vector matching according to claim 5, characterized in that, The distance mentioned is the Manhattan distance, and its calculation formula is as follows: ； In the formula, This represents the i-th feature data in each row of the feature data in the feature matrix of the substance to be screened. This represents the i-th feature data in each row of the reference material characteristic standard numerical matrix.

7. The poisoning substance screening device based on multidimensional vector matching according to claim 5 or 6, characterized in that, The scaling process for each column of data in the k rows of feature data yields the upper and lower limits of each feature data point in the k rows of feature data, including: For each type of feature data, a corresponding scaling threshold r is set, and the value of r ranges from [0,1]. The scaling process is performed according to the following formula to obtain the upper limit of each feature data in the k rows of feature data. Lower limit : ； ； In the formula, This represents the i-th feature data in each row of the reference material characteristic standard numerical matrix. This represents the scaling threshold for the i-th feature data.

8. An electronic device, characterized in that, include: A memory that stores computer programs; A processor, configured to implement the poisoning substance screening method based on multidimensional vector matching as described in any one of claims 1 to 4 when executing the computer program.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the poisoning substance screening method based on multidimensional vector matching as described in any one of claims 1 to 4.

Images