Large-scale sparse sample mixed testing method and system that significantly reduces the number of tests

Abstract

The present invention discloses a large-scale sparse sample mixed inspection method and system that significantly reduces the number of detection times, and belongs to the field of early epidemic infection screening. The method includes: setting mixing parameters and counting vectors; performing multiple sample mixing operations in each mixed detection stage; selecting a set number of samples for mixing according to the counting vector during each mixing operation; detecting the virus concentration of the samples obtained in each mixing operation; performing mathematical modeling according to the selection of samples in each mixing operation and the corresponding virus concentration detection results; converting the detection problem into a linear sparse reconstruction problem, and finding a sparse solution. If the sparse solution meets the requirements, the virus infected person is located, otherwise the next mixed detection stage is continued. The present invention is suitable for large-scale virus detection scenarios, and the number of detections is significantly reduced by gradually increasing the number of mixed inspections. It has the characteristics of strong operability and scalability.

Description

Technical Field

[0001] The present invention belongs to the technical field of early epidemic infection screening, and specifically relates to a large-scale sparse sample mixed testing method and system that significantly reduces the number of tests. Background Art

[0002] In the early stages of a pandemic, there are very few infected people, which can be regarded as a sparse signal. Virus samples from subjects, of which the number of positive samples does not exceed ( ), in order to reduce reagent consumption and detection costs, and to detect with high probability The most direct mixing strategy is to select an appropriate number of samples to mix (10 mixed with 1), and then test the mixed samples, which can reduce the number of tests to This essentially corresponds to an observation matrix ( ), Each row of represents a mixed sample, and the mixed test sample is , a total of If the measurement matrix is ​​a Gaussian random matrix, since the Gaussian random matrix has the restricted isometry property (RIP), according to the compressed sensing theory, High probability of detection with one test positive samples. When it is very big, , which means that the number of tests is greatly reduced, the consumption of test reagents is greatly reduced, and thus the test cost is greatly reduced. However, using a Gaussian random matrix as the observation matrix means that all samples must be weighted summed with the Gaussian distribution coefficient as the weight, which is completely impossible to implement in actual operation.

[0003] In contrast, the sparse Bernoulli matrix is ​​more practical, with elements of 0 and 1, which can directly correspond to the mixed detection of samples. When the probability of a sample being collected in a single mixed test is low, in order to ensure that the Positive samples, mixed detection times must be substantially increased, which would result in the number of sample recollections (i.e., the number of times each sample is mixed, and The large-scale mixed detection scheme often results in a large number of unnecessary mixing operations, which greatly increases the detection burden and resource consumption.

[0004] Therefore, existing technologies still face significant challenges in controlling the number of sample re-collection times and avoiding too many invalid mixed testing operations, and it is difficult to meet the requirements of efficient detection in practical applications. Summary of the invention

[0005] In order to solve the problems in the prior art, the present invention proposes a mixed testing method and system for large-scale sparse samples that significantly reduces the number of tests.

[0006] The technical solution adopted by the present invention is as follows:

[0007] In a first aspect, the present invention discloses a method for mixed testing of large-scale sparse samples that significantly reduces the number of tests, comprising the following steps:

[0008] S1: For the collected The samples to be tested are numbered and an equal amount of share;

[0009] S2: Set the number of mixing operations in each mixing detection stage , the number of samples mixed each time , the upper limit of the mixing times of each sample is , define the calibration reserve , used to verify the consistency of the sparse solution; define the count vector , used to record the number of times each sample has been mixed; define the mixing matrix Used to record the mixed detection process; define vector Used to record the virus concentration of mixed samples;

[0010] S3: In the current mixed detection phase, execute Sub-sample mixing operation;

[0011] Each time a sample is mixed, the samples whose current mixing times are less than A sample set is then randomly selected from the obtained sample set. non-repeated samples and mix them, and update the count vector;

[0012] Detect the virus concentration of the mixed sample obtained from each mixing operation and update the vector ;

[0013] S4: Update the mixing matrix based on the mixing operation used in S3 and the corresponding virus concentration detection results , and conduct mathematical modeling;

[0014] S5: Based on the results of mathematical modeling, the detection problem is converted into a sparse reconstruction problem, and a sparse solution is found. If the sparse solution meets the requirements, the mixed detection operation is terminated and the mixed detection result is output; if it does not meet the requirements, return to S3 and continue to the next mixed detection stage.

[0015] In the second aspect, the present invention provides a large-scale sparse sample mixed testing system for implementing the above method, which includes: a sample preprocessing module, a parameter setting module, a sample mixing module, a virus concentration detection module, a mathematical modeling module and an infected person determination module; the sample preprocessing module is used to number the collected samples to be tested, and prepare an equal amount of each sample The parameter setting module sets the number of mixed tests in each mixed test stage. , the number of samples mixed each time , the upper limit of the number of mixing times for each sample , define the calibration reserve , count vector, mixing matrix , mixed sample virus concentration vector ; The sample mixing module performs Sample mixing operations are performed each time; in each mixing operation, samples with current mixing times less than 1 are selected from all samples. A sample set is then randomly selected from the obtained sample set. Non-repetitive samples are mixed and the counting vector is updated; virus concentration detection is used to detect the virus concentration of the mixed sample obtained by each mixing operation and update the vector ; Mathematical modeling module, which updates the mixing matrix based on the mixing operation adopted by the sample mixing module and the test results of the virus concentration test , and perform mathematical modeling; the infected person determination module obtains a model based on the mathematical modeling module, converts the detection problem into a sparse reconstruction problem, and solves the sparse solution. If the sparse solution meets the requirements, the mixed inspection operation is terminated and the positive sample is successfully located; otherwise, the sample mixing module is called to continue to the next mixed detection stage.

[0016] In a third aspect, the present invention discloses an electronic device, including a processor and a memory, wherein the memory stores machine executable instructions that can be executed by the processor, and the processor executes the machine executable instructions to implement the mixed testing method of large-scale sparse samples that significantly reduces the number of detection times.

[0017] In a fourth aspect, the present invention discloses a machine-readable storage medium storing machine-executable instructions, which, when called and executed by a processor, are used to implement the mixed testing method for large-scale sparse samples that significantly reduces the number of detection times.

[0018] Compared with the prior art, the beneficial effects of the present invention include:

[0019] The present invention adopts a sparse binary random matrix as the observation matrix, and makes clear restrictions on the number of sample mixing times and the number of samples in each mixing. Compared with the traditional sparse binary Bernoulli matrix, the number of sample re-collection times is reduced from (estimated value) has dropped to This is because in the actual detection process, the virus concentration in the collected positive sample is limited. If the number of samples prepared is m, the virus concentration of each portion may be reduced to the detection limit (the larger m is, the lower the virus concentration is), resulting in the problem that the positive sample cannot be detected. In a preferred embodiment of the present invention, m is limited to an integer between 2 and 10 (the specific value of m is selected according to the detection limit of the virus detection equipment, and it should be ensured that the positive sample can still be detected normally after being divided into m portions). This method meets the needs of actual detection scenarios and can avoid missed detection of positive samples. While ensuring that no missed detection is found, the method of the present invention significantly reduces the number of re-collection of samples, and can correspond to the hybrid scheme that can be implemented in the virus sample detection link, while reducing the complexity of data calculations and improving the operability of sample mixing.

[0020] The present invention adopts a method of gradually increasing the number of mixed tests. Compared with a one-time large-scale detection scheme, this phased detection method avoids excessive and unnecessary mixed test operations, effectively reduces the reagents, instruments and time required for detection, and significantly improves the economy and efficiency of the detection process.

[0021] The parameter setting of the present invention is flexible and can be adjusted according to different sample quantities and detection capabilities. It has strong adaptability and wide application value.

[0022] Therefore, the present invention can significantly reduce the number of detection times and has the characteristics of strong operability and scalability. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 A flow chart of mixed testing of large-scale sparse samples that significantly reduces the number of tests described in the present invention.

[0024] Figure 2 This is a flow chart of the hybrid detection phase for large-scale sparse samples that significantly reduces the number of detection times described in the present invention.

[0025] Figure 3 This is the experimental result of the embodiment of the present invention when the upper limit of the number of mixing times of the sample is 5.

[0026] Figure 4 This is the experimental result of the embodiment of the present invention when the upper limit of the number of mixing times of the sample is 10.

[0027] Figure 5 Schematic diagram of the hybrid detection system for large-scale sparse samples of the present invention. DETAILED DESCRIPTION

[0028] The present invention is further described and illustrated below in conjunction with specific embodiments. The technical features of each embodiment of the present invention can be combined accordingly without conflicting with each other.

[0029] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the present invention. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the attached claims.

[0030] like Figure 1 The flow chart of the hybrid detection method in a specific embodiment of the present invention is shown. Virus samples were collected from subjects, and the virus concentration was , where the number of positive samples does not exceed ( ), usually, in order to ensure the number of mixing operations Small number of positive samples It should be less than This requirement meets the characteristics of the early stage of an epidemic, so the method of the present invention is suitable for screening infected persons in the early stage of an epidemic. The specific process of the embodiment of the present invention is as follows:

[0031] from Virus samples were collected from subjects to ensure that each sample was not contaminated by the outside world during the collection process. The virus samples of the subjects are numbered The virus concentration of all samples was ,in , prepare an equal amount of each sample Each portion has the same virus concentration and is used for subsequent mixing operations. The recommended value range is .

[0032] Set the number of samples for each mix to , should be less than , the number of mixing times for each sample does not exceed , the number of mixed tests in each mixed test stage is ,suggestion The value is Nearby; define calibration reserve , used to verify the consistency of the sparse solution; define the count vector , used to record the number of times each sample has been mixed, with an initial value of all zeros; define the mixing matrix , used to record the detection process, initially a full zero matrix; define vector , used to record the virus concentration of the mixed sample, with the initial value being all zero.

[0033] In the A hybrid detection phase is performed Sample mixing test operation.

[0034] like Figure 2 As shown, for each times of sample mixing operations, where , filter out from all samples the samples with mixing times less than A sample collection of:

[0035] (1)

[0036] In formula (1): Indicates The candidate sample set available for mixing in the sample mixing operation.

[0037] Indicates The number of times samples have been mixed.

[0038] From the candidate sample set Random selection ( ) non-repeating samples are mixed to obtain mixed samples, and the counting vector is updated; repeat the above steps times, a total of Mixed samples; The mixed samples were tested one by one and the virus concentration was recorded as ,in Representation vector No. to elements.

[0039] Update the mixing matrix according to the above sample mixing scheme No. to Rows (updated at each hybrid detection stage rows), the mixing matrix Each row of corresponds to a mixing operation, and each row includes elements, each element position corresponds to a sample number to be tested, and the element value , where the element value is It means that the sample to be tested corresponding to the element position participates in the mixing. If the element value is 0, it means that it does not participate in the mixing. The mixed test sample can be expressed as:

[0040] (2)

[0041] In formula (2): Mixing matrix Corresponding to the sample mixing scheme, the dimension is , each row has non-zero elements, and the number of non-zero elements in each column shall not exceed indivual, Represents the mixing matrix No. to rows, with dimensions , Represents the virus concentration of all samples, with dimensions , Represents the virus concentration of all mixed samples, with dimensions , Represents vector No. to elements, with dimensions .

[0042] The detection problem is converted into a linear sparse reconstruction problem, the goal is to find a sparse solution At the same time:

[0043] (3)

[0044] (4)

[0045] In formula (3): Represents the mixing matrix Before rows, with dimensions , Represents vector Before elements, with dimensions , where: Represents the mixing matrix Before rows, with dimensions , Represents vector Before elements, with dimensions .

[0046] The orthogonal matching pursuit algorithm is used to solve the optimization problems (3) and (4) and obtain the sparse solution and . Relatively sparse solution and If the consistency ,in is a small constant (in this embodiment, Take ), then it is determined that the accurate positioning is Virus infected people, The position of the non-zero element in corresponds to the number of the virus-infected person, and the mixed testing operation is terminated; if , then return to the sample mixing detection step and continue to the next mixing detection stage.

[0047] In order to verify the feasibility and effectiveness of the technical solution proposed in this invention, a simulation experiment was conducted. samples to be tested (recorded as sample numbers 1 to 1000), randomly set samples were positive, and the rest were negative. The virus concentration of positive samples followed a log-normal distribution. The virus concentration of negative samples is usually set to 0. Considering the presence of detection noise, the virus concentration of negative samples is set to a value close to 0. , the number of mixed tests in each mixed test stage Set to 10 to check the reserve Set to 5. If the total number of mixed tests is greater than In the case of mixed sample concentration difference Reconstruct the sparse , that is, to identify those infected with the virus, among which represents the minimum number of hybrid detection stages required, which is a positive integer. Represents the reconstructed sparse signal.

[0048] Figure 3 The experimental results of the embodiment of the present invention when the upper limit of the number of mixing times of samples is 5 are given. The reconstruction accuracy corresponding to different total times of mixed inspection is expressed as , ,The reconstruction accuracy indicates the number of times all virus-infected people can be accurately located by repeating the above process 200 times. Figure 4 This is the experimental result of the embodiment of the present invention under the same experimental conditions when the upper limit of the number of mixing times of the sample is 10. Figure 3 and Figure 4 The experimental results show that the sample mixing detection method of the embodiment of the present invention not only has good detection performance, but also can significantly reduce the number of detection times, and has great practical application value. Figure 5FIG. 1 is a schematic diagram of a hybrid detection system for large-scale sparse samples of the present invention. The hybrid detection system of the present invention is used to implement the hybrid detection method in the aforementioned embodiment, and includes:

[0049] The sample preprocessing module is used to number the collected samples and prepare equal amounts of each sample. share;

[0050] Parameter setting module, set the number of mixed tests in each mixed test stage , the number of samples mixed each time , the upper limit of the number of mixing times for each sample , define the calibration reserve , count vector, mixing matrix , mixed sample virus concentration vector ;

[0051] The sample mixing module performs Sample mixing operations are performed each time; in each mixing operation, samples with current mixing times less than 1 are selected from all samples. A sample set is then randomly selected from the obtained sample set. non-repeated samples and mix them, and update the count vector;

[0052] Virus concentration detection, used to detect the virus concentration of the mixed sample obtained from each mixing operation and update the vector ;

[0053] Mathematical modeling module, which updates the mixing matrix based on the mixing operation adopted by the sample mixing module and the test results of the virus concentration test , and conduct mathematical modeling;

[0054] The infected person determination module obtains a model based on the mathematical modeling module, converts the detection problem into a sparse reconstruction problem, and solves the sparse solution. If the sparse solution meets the requirements, the mixed testing operation is terminated and the mixed testing results are output to locate the positive samples; otherwise, the sample mixing module is called to continue to the next mixed detection stage.

[0055] An embodiment of the present invention further provides an electronic device, including a memory and a processor;

[0056] The memory is used to store computer programs;

[0057] The processor is used to implement the above-mentioned large-scale sparse sample mixed testing method that significantly reduces the number of tests when executing the computer program.

[0058] An embodiment of the present invention further provides a computer-readable storage medium having a program stored thereon. When the program is executed by a processor, the above-mentioned mixed testing method for large-scale sparse samples that significantly reduces the number of tests is implemented.

[0059] The computer-readable storage medium may be an internal storage unit of any device with data processing capability described in any of the aforementioned embodiments, such as a hard disk or a memory. The computer-readable storage medium may also be an external storage device of any device with data processing capability, such as a plug-in hard disk, a smart media card (SMC), an SD card, a flash card, etc. equipped on the device. Furthermore, the computer-readable storage medium may also include both an internal storage unit and an external storage device of any device with data processing capability. The computer-readable storage medium is used to store the computer program and other programs and data required by any device with data processing capability, and may also be used to temporarily store data that has been output or is to be output.

[0060] Obviously, the above-described embodiments and drawings are only some examples of the present application. For those of ordinary skill in the art, the present application can also be applied to other similar situations based on these drawings without creative work. In addition, it is understandable that, although the work done in this development process may be complicated and lengthy, for those of ordinary skill in the art, certain changes in design, manufacturing or production based on the technical content disclosed in this application are only conventional technical means and should not be regarded as insufficient content disclosed in this application. Without departing from the concept of the present application, several variations and improvements can also be made, which all belong to the scope of protection of the present application. Therefore, the scope of protection of the present application shall be subject to the attached claims.

Claims

1. A method for mixed testing of large-scale sparse samples that significantly reduces the number of tests, characterized by: The following steps are involved: S1: For the collected The samples to be tested are numbered and an equal amount of share; S2: Set the number of mixing operations in each mixing detection stage , the number of samples mixed each time , the upper limit of the mixing times of each sample is , define the calibration reserve , used to verify the consistency of the sparse solution; define the count vector , used to record the number of times each sample has been mixed; define the mixing matrix Used to record the mixed detection process; define vector Used to record the virus concentration of mixed samples; S3: In the current mixed detection phase, execute Sub-sample mixing operation; Each time a sample is mixed, the samples whose current mixing times are less than A sample set is then randomly selected from the obtained sample set. non-repeated samples and mix them, and update the count vector; Detect the virus concentration of the mixed sample obtained from each mixing operation and update the vector ; S4: Update the mixing matrix based on the mixing operation used in S3 and the corresponding virus concentration detection results , and conduct mathematical modeling; S5: Based on the mathematical modeling results, the detection problem is converted into a sparse reconstruction problem, and a sparse solution is found. If the sparse solution meets the requirements, the mixed inspection operation is terminated and the mixed inspection result is output; If the requirements are not met, return to S3 and continue to the next mixed detection stage.

2. The method for mixed testing of large-scale sparse samples with significantly reduced testing times according to claim 1, characterized in that: The step S1 specifically includes: S11: The samples to be tested are numbered The virus concentration of all samples was ,in ; S12: Prepare an equal amount of each sample Each aliquot has the same virus concentration and is used for subsequent mixing operations.

3. The method for mixed testing of large-scale sparse samples with significantly reduced testing times according to claim 2, characterized in that: In step S1, the number of positive samples in the samples to be tested is , is an integer in the range .

4. The method for mixed testing of large-scale sparse samples with significantly reduced testing times according to claim 1, characterized in that: In step S2, the number of samples to be mixed each time is set. Less than , the number of mixed tests in each mixed test stage , check the reserve The value range is ; Count vector The initial value is all zero; the mixing matrix Initially, it is a matrix of all zeros; The initial value is all zeros.

5. The method for mixed testing of large-scale sparse samples with significantly reduced testing times according to claim 1, characterized in that: The step S3 specifically includes: S31: A hybrid detection phase is performed Sub-sample mixed detection operation; S32: For Sample mixing operation, where , filter out from all samples those with mixing times less than A sample collection of: ； Where: Indicates The candidate sample set available for mixing in the sample mixing operation. Indicates The number of times samples have been mixed; From the candidate sample set Random selection non-repeating samples and mix them to get a mixed sample, and update the count vector; where, ; S33: Repeat step S32 until the a composite sample; S34: Yes The mixed samples were tested one by one and the virus concentration was recorded as ,in, Representation vector No. to elements.

6. The method for mixed testing of large-scale sparse samples with significantly reduced testing times according to claim 5, characterized in that: The step S4 comprises the following specific steps: Update the mixing matrix according to the sample mixing scheme in step S3 No. to Row, mixing matrix Each row of corresponds to a mixing operation, and each row includes elements, each element position corresponds to a sample number to be tested, and the element value , where the element value is It means that the sample to be tested corresponding to the element position participates in the mixing, and the mixed test sample is expressed as: ； Where: Represents the mixing matrix No. to OK, represents the virus concentration of all samples, Representation vector No. to elements.

7. The method for mixed testing of large-scale sparse samples with significantly reduced testing times according to claim 6, characterized in that: The step S5 specifically includes: S51: Convert the detection problem into a linear sparse reconstruction problem, the goal is to find a sparse solution At the same time: ； ； Where: Represents the mixing matrix Before rows, with dimensions , Represents vector Before elements, with dimensions , Represents the mixing matrix Before Row, Dimension , Represents vector Before elements, with dimensions ; S52: Use the sparse signal recovery algorithm to solve the two optimization problems in step S51 and obtain sparse solutions respectively. and ; S53: Comparatively sparse solution and If the consistency ,in is the set threshold, then it is determined to be accurately located Virus infected people, The position of the non-zero element in corresponds to the number of the positive sample to be tested, and the mixed test operation is terminated; if , then return to step S3 and continue to the next mixed detection stage.

8. A large-scale sparse sample mixed testing system implementing the method according to any one of claims 1 to 7, characterized in that: include: The sample preprocessing module is used to number the collected samples and prepare equal amounts of each sample. share; Parameter setting module, set the number of mixed tests in each mixed test stage , the number of samples mixed each time , the upper limit of the number of mixing times for each sample , define the calibration reserve , count vector, mixing matrix , mixed sample virus concentration vector ; The sample mixing module performs Sample mixing operations are performed each time; in each mixing operation, samples with current mixing times less than 1 are selected from all samples. A sample set is then randomly selected from the obtained sample set. non-repeated samples and mix them, and update the count vector; Virus concentration detection, used to detect the virus concentration of the mixed sample obtained from each mixing operation and update the vector ; Mathematical modeling module, which updates the mixing matrix based on the mixing operation adopted by the sample mixing module and the test results of the virus concentration test , and conduct mathematical modeling; The infected person determination module obtains a model based on the mathematical modeling module, converts the detection problem into a sparse reconstruction problem, and solves the sparse solution. If the sparse solution meets the requirements, the mixed testing operation is terminated to locate the positive sample; otherwise, the sample mixing module is called to continue to the next mixed detection stage.

9. An electronic device, characterized in that: It includes a processor and a memory, wherein the memory stores machine executable instructions that can be executed by the processor, and the processor executes the machine executable instructions to implement the mixed testing method for large-scale sparse samples that significantly reduces the number of tests as described in any one of claims 1 to 7.

10. A machine-readable storage medium, characterized in that: The machine-readable storage medium stores machine-executable instructions, which, when called and executed by a processor, are used to implement the mixed testing method for large-scale sparse samples that significantly reduces the number of tests as described in any one of claims 1 to 7.

Images