CN114496242A

CN114496242A - Myocardial infarction prediction method, system, equipment and medium

Info

Publication number: CN114496242A
Application number: CN202111590106.8A
Authority: CN
Inventors: 贾成友; 李宪凯; 吕中伟; 徐亚伟; 庄剑辉; 陈贺昌; 缪春建; 刘凡新; 曲伸; 杨建设
Original assignee: Shanghai Tenth Peoples Hospital
Current assignee: Shanghai Tenth Peoples Hospital
Priority date: 2021-12-23
Filing date: 2021-12-23
Publication date: 2022-05-13
Anticipated expiration: 2041-12-23
Also published as: CN114496242B

Abstract

The invention relates to a myocardial infarction prediction method, system, equipment and medium, the method includes obtaining the Raman detection data of the plasma sample; and inputting the Raman detection data into a pre-trained myocardial infarction prediction model to obtain a myocardial infarction prediction result. The method has the advantages that the detection time is greatly shortened by combining the Raman detection with the myocardial infarction prediction model, the myocardial infarction prediction result can be obtained within 8 minutes, the next step of revascularization is guided according to the detection time, and the detection time is far shorter than 15-20 minutes of the traditional method; the detection method is simple, various metabolites can be detected simultaneously by one-time test, and the myocardial infarction prediction result has high accuracy; the detection timeliness is good, and the whole abnormal change of the serum metabolite can be obtained immediately after the myocardial infarction occurs; the detection cost is low, a kit is not needed, a large amount of medical waste is not generated, and the environment pollution is avoided; the prediction method has high specificity, high sensitivity and high accuracy.

Description

Myocardial infarction prediction method, system, equipment and medium

Technical Field

The present invention relates to the field of medical data management technologies, and in particular, to a method, a system, a computer device, and a computer-readable storage medium for predicting myocardial infarction.

Background

At present, biochemical detection of acute myocardial infarction mainly depends on detecting two types of substances: one of the markers of myocardial injury, the most commonly used clinical markers are Myoglobin (Myoglobin), troponin (including troponin t (ctnt), troponin i (ctni), troponin c (ctnc), creatine kinase isoenzyme (CK-MB), brain natriuretic peptide precursor (pro-BNP) or N-terminal brain natriuretic peptide precursor (NT-proBNP), and Lactate Dehydrogenase (LDH), aspartate Aminotransferase (AST), and Creatine Kinase (CK) which have been used for the past, but these three types of specificity and sensitivity are poor and have been rarely used in recent years; and the second amino acid comprises alpha-ketoglutaric acid, alanine, glutamic acid, pyruvic acid and the like, and can reflect the metabolic abnormality of myocardial infarction to a certain extent. However, the most commonly used clinical biochemical indicators of myocardial infarction are cTnT, Myoglobin, CK-MB and pro-BNP/NT-proBNP, wherein cTnT has the highest specificity and Myoglobin has the highest sensitivity. For acute myocardial infarction patients, the detection of the myocardial injury marker needs to be completed within 20 minutes when the Chinese chest pain center requires to receive a diagnosis of chest pain patients.

At present, a kit is mainly used for detecting the myocardial injury marker, the principle of the kit is mainly an antigen-antibody sandwich compound, the compound is combined with streptavidin-coated magnetic particles and then transferred into a measuring cell and fixed on the surface of an electrode, unbound substances are washed away, an electrochemiluminescence reaction occurs after the electrode is electrified, an optical signal is detected by a photomultiplier tube and converted into an electrical signal, the electrical signal is processed by an instrument, and a sample is obtained by calculating a calibration curve. Generally, the detection takes about 15 to 20 minutes.

However, the prior art has the following drawbacks:

1) the detection process is complex: because the antigen-antibody reaction is needed, the primary antibody, the secondary antibody, the biotin label, the ruthenium label, the magnetic bead and other very complicated early-stage preparation and sample processing are needed;

2) the detection component is single: only one protein is detected in each detection, if three proteins are detected, three test tubes are required to be simultaneously detected, and a large amount of metabolites such as taurine (myocardial protection effect) are consumed in myocardial infarction, so that the detection consideration is out of the range;

3) the detection time is long: 15-20 minutes are needed, and valuable rescue time is wasted;

4) poor detection timeliness: the proteins can be released into blood after myocardial cell necrosis, so that the content of the proteins is not obviously increased in the early stage or the early stage of acute myocardial infarction, for example, troponin starts to increase 3-6 hours after the acute myocardial infarction occurs and reaches a peak in 10-24 hours, CK-MB increases in 6 hours after the myocardial infarction occurs and reaches a peak in 24 hours, in addition, the molecular weight of the troponin is smaller than that of CK-MB, the troponin enters blood earlier than that of CK-MB, and the myoglobin can increase within 30 minutes after myocardial injury, but the specificity is not high;

5) the detection cost is high: the general clinical acute myocardial infarction biochemical detection items comprise 4-5 items, the detection cost is about 500 yuan, the cost is high, and a large amount of medical insurance cost is occupied;

6) environmental pollution: the detection process can generate a large amount of waste of the kit, namely medical waste, the treatment is time-consuming, and the environment is easily polluted

At present, no effective solution is provided for the problems of complex detection process, single detection component, multiple detection items, high detection cost, long detection time, poor detection timeliness and environmental pollution in the related technology.

Disclosure of Invention

The present application aims to overcome the deficiencies in the prior art, and provides a myocardial infarction prediction method, system, device and storage medium, so as to solve at least the problems of complex detection process, single detection component, multiple detection items, high detection cost, long detection time, poor detection timeliness and environmental pollution existing in the related art.

In order to achieve the purpose, the technical scheme adopted by the application is as follows:

in a first aspect, the present invention provides a method for predicting myocardial infarction, comprising:

acquiring Raman detection data of the plasma sample;

inputting the Raman detection data into a pre-trained myocardial infarction prediction model to obtain a myocardial infarction prediction result;

wherein the accuracy rate of the myocardial infarction prediction result is 92.31%, the sensitivity is more than 97.22%, and the specificity is more than 99.41%.

The Raman detection data is first Raman detection data and at least comprises the following components:

in some of these embodiments, the raman detection data is a second type of raman detection data that includes at least:

in some of these embodiments, prior to obtaining raman detection data for the plasma sample, the method further comprises:

obtaining a blood sample;

centrifuging the blood sample to obtain an initial plasma sample;

subjecting the initial plasma sample to a membrane filtration-based centrifugation process to obtain a plasma sample;

wherein the time for performing centrifugal treatment based on membrane filtration on the initial plasma sample is less than or equal to 5min, the volume of the initial plasma sample is less than or equal to 450 μ l, and the volume of the plasma sample is less than or equal to 5 μ l.

In some of these embodiments, the blood sample is centrifuged to obtain an initial plasma sample at 2000-3000 rpm for 1-3 min.

In some of these embodiments, the blood sample is centrifuged to obtain an initial plasma sample at 3000rpm for 2 min.

In some of these embodiments, the initial plasma sample is centrifuged based on membrane filtration to obtain a plasma sample with working parameters of 10000-15000 rpm for 3-6 min.

In some of these embodiments, the initial plasma sample is subjected to centrifugation based on membrane filtration to obtain a plasma sample with operating parameters of 12000rpm for 3 min.

In some of these embodiments, the plasma sample has a volume of 2 to 5 μ l.

In some of these embodiments, the membrane treatment is an ultrafiltration treatment.

In some of these embodiments, after obtaining the plasma sample, the method further comprises:

the plasma samples were placed on an aluminized slide and dried.

In some of these embodiments, the slide is pre-placed in an environment at 37 ℃.

In some of these examples, the plasma sample was dried for 30 seconds on an aluminized slide.

In some of these embodiments, the plasma sample is dried on an aluminized glass slide at 37 ℃.

In some of these embodiments, the raman detection parameters for obtaining raman detection data for a plasma sample are:

excitation wavelength: 532 nm;

power: 10-14 mw;

grating: 1200 g/mm;

single spectrum integration time: 16-20 s;

an objective lens: 100 x/0.9.

In some of these embodiments, acquiring raman detection parameters of raman detection data of the plasma sample further comprises:

5-8 profiles were collected for each of the plasma samples.

In some of these embodiments, prior to inputting the raman detection data into the pre-trained myocardial infarction prediction model, the method further comprises:

acquiring inquiry information;

inputting the Raman detection data and the inquiry information into a pre-trained myocardial infarction prediction model to obtain a myocardial infarction prediction result;

wherein the inquiry information comprises personal information, myocardial infarction related basic disease information and past operation history information;

wherein, in the myocardial infarction prediction model, the weight of the Raman detection data is 90%, and the weight of the inquiry information is 10%.

In some embodiments, the interrogation information includes coronary heart disease, past history of coronary stent implantation, hypertension, diabetes, hyperlipidemia, smoking history, and history of cerebral infarction.

In some embodiments, inputting the raman detection data into a pre-trained myocardial infarction prediction model to obtain a myocardial infarction prediction result comprises:

processing the Raman detection data, and mapping the Raman detection data into 1024-dimensional initial feature vectors;

inputting the 1024-dimensional initial feature vector into a pre-trained myocardial infarction prediction model to obtain a 4-dimensional final feature vector;

and inputting the 4-dimensional final feature vector into a classification function for processing so as to obtain a myocardial infarction prediction result.

In some of these embodiments, the classification function is a softmax classification function.

In some of these embodiments, the method of training a myocardial infarction prediction model comprises:

constructing a myocardial infarction prediction model, wherein the myocardial infarction prediction model comprises an input layer, a hidden layer and an output layer, the input of the input layer is 1024, the output of the input layer is 512, the input of the hidden layer is 512, the output of the hidden layer is 128, and the input of the output layer is 128, the output of the output layer is 4;

constructing a training set and a testing set according to Raman detection data, wherein the ratio of the training set to the testing set is 3: 1;

inputting the training set into the myocardial infarction prediction model for training to obtain an output result;

inputting the output result into a classification function for processing to obtain a myocardial infarction prediction result;

and testing the myocardial infarction prediction result to detect the effectiveness of training, and iterating the myocardial infarction prediction model according to the effectiveness detection result until the training is completed.

In some of these embodiments, the myocardial infarction prediction is tested as:

the test was performed every 10 rounds of training.

In some of these embodiments, the method for training a myocardial infarction prediction model further comprises:

calculating a loss value by using a cross entropy objective function;

and optimizing the weight of the myocardial infarction prediction model by using a random gradient descent method and a back propagation algorithm.

In a second aspect, the present invention provides a myocardial infarction prediction system comprising:

a plasma sample obtaining device for performing centrifugation on a blood sample to obtain an initial plasma sample, and performing centrifugation based on membrane filtration on the initial plasma sample to obtain a plasma sample; wherein the time for performing the membrane filtration-based centrifugation treatment on the initial plasma sample is less than or equal to 5 min;

the Raman detection device is used for carrying out Raman detection on the plasma sample so as to acquire Raman detection data;

the myocardial infarction prediction device is used for acquiring Raman detection data of the plasma sample, and inputting the Raman detection data into a pre-trained myocardial infarction prediction model to acquire a myocardial infarction prediction result;

In some of these embodiments, the plasma sample acquisition device comprises:

the anticoagulation blood taking unit is used for placing a blood sample;

a centrifugation unit for centrifuging the blood sample of the anticoagulation blood-taking unit to obtain a primary plasma sample, and centrifuging the primary plasma sample to obtain a plasma sample;

a membrane filtration unit for performing a membrane treatment on the initial plasma sample to obtain the plasma sample while the centrifugation unit performs a centrifugation treatment on the initial plasma sample.

In some embodiments, the anti-coagulant withdrawal unit is an EDTA-K2 anti-coagulant withdrawal unit.

In some of these embodiments, the membrane filtration unit is an ultrafiltration centrifuge tube.

In some of these embodiments, further comprising:

in some of these embodiments, the raman detection device employs confocal laser raman technology, surface enhanced raman scattering technology.

In a third aspect, the invention provides a computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method of predicting myocardial infarction as described above when executing the computer program.

In a fourth aspect, the invention provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a method of predicting myocardial infarction as described above.

Compared with the related art, the myocardial infarction prediction method, the system, the equipment and the storage medium provided by the embodiment of the application greatly shorten the detection time by combining the Raman detection with the myocardial infarction prediction model, can obtain the myocardial infarction prediction result in about 8 minutes, and can determine whether to perform subsequent accurate detection according to the prediction result; the detection method is simple, multiple substances can be detected simultaneously by one-time test, and the myocardial infarction prediction result has high accuracy; the detection timeliness is good, and abnormal change of substances can be obtained before or after the myocardial infarction occurs; the detection cost is low, a kit is not needed, a large amount of medical waste is not generated, and the environment pollution is avoided; the prediction method has high specificity, high sensitivity and high accuracy.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:

fig. 1 is a flowchart of a myocardial infarction prediction method according to an embodiment of the present application (one);

fig. 2 is a flowchart of a myocardial infarction prediction method according to an embodiment of the present application (ii);

fig. 3 is a flowchart of a myocardial infarction prediction method according to an embodiment of the present application (iii);

fig. 4 is a flowchart of a myocardial infarction prediction method according to an embodiment of the present application (iv);

fig. 5 is a flow chart of a training method of an acute myocardial infarction model according to an embodiment of the present application;

fig. 6 is a frame diagram of a myocardial infarction prediction system according to an embodiment of the present application;

FIG. 7 is a block diagram of a plasma sample acquiring device according to an embodiment of the present application;

FIGS. 8 a-8 c are AUC curves for AUC > 0.8 for 58 metabolites compared to AMI and Con according to examples of the present application;

FIG. 9 is an AUC curve for 22 metabolites with AUC > 0.8 compared to (CAD + AF) for AMI according to an embodiment of the present application.

Wherein the reference numerals are: 600. a myocardial infarction prediction system;

610. a myocardial infarction prediction device;

620. a plasma sample acquiring device; 621. an anticoagulation blood taking unit; 622. a centrifugal unit; 623. a membrane filtration unit;

630. a Raman detection device.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be described and illustrated below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments provided in the present application without any inventive step are within the scope of protection of the present application.

It is obvious that the drawings in the following description are only examples or embodiments of the present application, and that it is also possible for a person skilled in the art to apply the present application to other similar contexts on the basis of these drawings without inventive effort. Moreover, it should be appreciated that in the development of any such actual implementation, as in any engineering or design project, numerous implementation-specific decisions must be made to achieve the developers' specific goals, such as compliance with system-related and business-related constraints, which may vary from one implementation to another.

Reference in the specification to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the specification. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of ordinary skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments without conflict.

Unless defined otherwise, technical or scientific terms referred to herein shall have the ordinary meaning as understood by those of ordinary skill in the art to which this application belongs. Reference to "a," "an," "the," and similar words throughout this application are not to be construed as limiting in number, and may refer to the singular or the plural. The present application is directed to the use of the terms "including," "comprising," "having," and any variations thereof, which are intended to cover non-exclusive inclusions; for example, a process, method, system, article, or apparatus that comprises a list of steps or elements (elements) is not limited to the listed steps or elements, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus. Reference to "connected," "coupled," and the like in this application is not intended to be limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The term "plurality" as referred to herein means two or more. "and/or" describes the association relationship of the associated object, indicating that there may be three relationships, for example, "a and/or B" may indicate: a exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. Reference herein to the terms "first," "second," "third," and the like, are merely to distinguish similar objects and do not denote a particular ordering for the objects.

Example 1

Fig. 1 is a flowchart (one) of a myocardial infarction prediction method according to an embodiment of the present invention. As shown in fig. 1, a myocardial infarction prediction method includes the following steps:

s102, acquiring Raman detection data of a plasma sample;

and S104, inputting the Raman detection data into a pre-trained myocardial infarction prediction model to obtain a myocardial infarction prediction result.

Wherein the myocardial infarction prediction result has the accuracy of 92.31 percent, the sensitivity of more than 97.22 percent and the specificity of more than 99.41 percent.

In step S102, the raman detection parameters of the raman detection data are:

excitation wavelength: 532 nm;

power: 10-14 mw;

grating: 1200 g/mm;

single spectrum integration time: 16-20 s;

an objective lens: 100 x/0.9.

Wherein the time of Raman detection data is 1-2 min.

Further, the acquiring raman detection parameters of the raman detection data of the plasma sample further comprises:

5-8 spectra were taken for each plasma sample.

In step S104, the time for obtaining the myocardial infarction prediction result is 0-2 min.

Preferably, the time to obtain a prediction of myocardial infarction is 30s-1 min.

In some of these embodiments, the raman detection data is a first type of raman detection data comprising at least:

by the 82 metabolites, whether the myocardial infarction risk exists can be predicted.

the correlation between myocardial infarction and etiology can be predicted through the 55 metabolites.

Fig. 2 is a flowchart of a myocardial infarction prediction method according to an embodiment of the present invention (ii). As shown in fig. 2, prior to obtaining raman detection data of a plasma sample, the method further comprises the steps of:

step S202, obtaining a blood sample;

step S204, carrying out centrifugal treatment on the blood sample to obtain an initial plasma sample;

and step S206, performing centrifugal processing based on membrane filtration on the initial plasma sample to obtain the plasma sample.

In step S202, obtaining a blood sample is to place collected blood into an anticoagulation blood vessel.

In step S204, the blood sample is centrifuged to obtain an initial plasma sample with working parameters of 2000-5000 rpm for 1-3 min.

Preferably, the blood sample is centrifuged to obtain an initial plasma sample with operating parameters of 3000rpm for 2 min.

In step S206, the initial plasma sample is placed in an ultrafiltration centrifuge tube, and the membrane treatment is an ultrafiltration treatment.

In step S206, membrane treatment and centrifugation treatment are simultaneously carried out on the initial plasma sample so as to obtain the plasma sample with working parameters of 10000-15000 rpm and 3-6 min.

Preferably, the initial plasma sample is subjected to both membrane treatment and centrifugation to obtain a plasma sample with working parameters of 12000rpm for 5 min.

Preferably, the volume of the plasma sample is 2-5 μ l.

Preferably, the volume of the plasma sample is 2 μ l.

Further, after step S206, the method further comprises the steps of:

and step S208, placing the plasma sample on an aluminized glass slide for drying.

Wherein, the working parameter of placing the plasma sample on an aluminized glass slide for drying is 37 ℃.

Wherein the drying time is 30 s-2 min.

Through steps S202-S206, a plasma sample can be obtained within 4-9 min (generally 5-8 min, preferably 6-7 min) for subsequent detection, and the time for obtaining the sample is greatly shortened.

Fig. 3 is a flowchart (iii) of a myocardial infarction prediction method according to an embodiment of the present invention. As shown in fig. 3, before inputting the raman detection data into the pre-trained myocardial infarction prediction model, the method further comprises the following steps:

step S302, obtaining inquiry information;

and S304, inputting the Raman detection data and the inquiry information into a pre-trained myocardial infarction prediction model to obtain a myocardial infarction prediction result.

Wherein, the inquiry information comprises personal information, myocardial infarction basic disease information and past operation history information.

Preferably, the information for interrogation includes coronary heart disease, history of previous coronary stent implantation, hypertension, diabetes, hyperlipidemia, history of smoking, and history of cerebral infarction.

In the myocardial infarction prediction model, the weight of Raman detection data is 90%, and the weight of inquiry information is 10%.

Through steps S302 to S304, the accuracy of the myocardial infarction prediction result can be improved by adding the inquiry information.

Fig. 4 is a flowchart (iv) of a myocardial infarction predicting method according to an embodiment of the present invention. As shown in fig. 4, inputting the raman detection data into the pre-trained myocardial infarction prediction model to obtain a myocardial infarction prediction result includes:

s402, processing Raman detection data, and mapping the Raman detection data into 1024-dimensional initial characteristic vectors;

s404, inputting 1024-dimensional initial feature vectors into a pre-trained myocardial infarction prediction model to obtain 4-dimensional final feature vectors;

and step S406, inputting the 4-dimensional final feature vector into a classification function for processing so as to obtain a myocardial infarction prediction result.

In step S402, Raman detection data comprises wave number and intensity, wherein the wave number is in the range of 279-^-1A total of 1024 points.

In step S406, the classification function is a softmax classification function.

Fig. 5 is a flowchart of a training method of an acute myocardial infarction model according to an embodiment of the present invention. As shown in fig. 5, the method for training the myocardial infarction prediction model includes:

step S502, constructing a myocardial infarction prediction model, wherein the myocardial infarction prediction model comprises an input layer, a hidden layer and an output layer, the input of the input layer is 1024, the output of the input layer is 512, the input of the hidden layer is 512, the output of the hidden layer is 128, and the input of the output layer is 128, and the output of the output layer is 4;

step S504, a training set and a testing set are constructed according to the Raman detection data, wherein the proportion of the training set to the testing set is 3: 1;

step S506, inputting the training set into a myocardial infarction prediction model for training to obtain an output result;

step S508, inputting the output result into a classification function for processing to obtain a myocardial infarction prediction result;

and step S510, testing the myocardial infarction prediction result to detect the effectiveness of training, and iterating the myocardial infarction prediction model according to the effectiveness detection result until the training is finished.

In step S508, the classification function is a softmax classification function.

In step S510, a test is performed every 10 training rounds.

In step S510, the myocardial infarction prediction model is tested using a test set to detect the effectiveness of the training.

Further, the method for training the myocardial infarction prediction model further comprises the following steps:

step S512, calculating a loss value by using a cross entropy objective function;

and S514, optimizing the weight of the myocardial infarction prediction model by using a random gradient descent method and a back propagation algorithm.

Through the steps S502 to S514, the myocardial infarction prediction model with the accuracy rate of more than 92 percent, the sensitivity rate of more than 97 percent and the specificity of more than 99 percent can be obtained through training, and the accuracy rate, the sensitivity rate and the specificity are greatly improved under the condition of reducing the prediction time.

Fig. 6 is a frame diagram of a myocardial infarction prediction system in accordance with an embodiment of the present invention. As shown in fig. 6, the myocardial infarction prediction system 600 includes a myocardial infarction prediction device 610 for obtaining raman detection data of a plasma sample, and inputting the raman detection data into a pre-trained myocardial infarction prediction model to obtain a myocardial infarction prediction result.

In some embodiments, the myocardial infarction prediction device 610 includes, but is not limited to, a mobile terminal, a cloud server, a local server, a computer, a notebook, and the like.

Further, the myocardial infarction prediction system 600 further comprises a plasma sample acquiring device 620 and a raman detecting device 630. Wherein the plasma sample acquiring device 620 is used for performing centrifugal processing on the blood sample to acquire an initial plasma sample, and performing membrane processing and centrifugal processing on the initial plasma sample simultaneously to acquire the plasma sample; the raman detection device 630 is used for raman detection of the plasma sample to obtain raman detection data.

In some embodiments, the raman detection device 630 employs a confocal laser raman technique or a surface enhanced raman scattering technique.

Fig. 7 is a frame diagram of a plasma sample acquiring device according to an embodiment of the present invention. As shown in fig. 7, the plasma sample acquiring device 620 includes an anticoagulation unit 621, a centrifugation unit 622, and a membrane filtration unit 623. Wherein, the anticoagulation blood taking unit 621 is used for placing a blood sample; the centrifugation unit 622 is used for centrifuging the blood sample of the anticoagulation unit 621 to obtain a primary plasma sample, and centrifuging the primary plasma sample to obtain a plasma sample; the membrane filtration unit 623 is used for performing a membrane treatment on the initial plasma sample to obtain the plasma sample while the centrifugation unit 622 performs a centrifugation treatment on the initial plasma sample.

In some of these embodiments, anticoagulation blood sampling unit 621 is anticoagulation blood sampling vessel loaded with EDTA-K2 anticoagulation.

In some of these embodiments, the membrane filtration unit 623 is an ultrafiltration centrifuge tube.

Preferably, the membrane filtration unit 623 is a Millipore UFC503096Amicon Ultra-30K ultrafiltration centrifuge tube.

In addition, the myocardial infarction prediction method of the embodiment of the present application may be implemented by a computer device. Components of the computer device may include, but are not limited to, a processor and a memory storing computer program instructions.

In some embodiments, the processor may include a Central Processing Unit (CPU), or A Specific Integrated Circuit (ASIC), or may be configured to implement one or more Integrated circuits of embodiments of the present Application.

In some embodiments, the memory may include mass storage for data or instructions. By way of example, and not limitation, memory may include a hard disk Drive (hard disk Drive, abbreviated HDD), a floppy disk Drive, a Solid State Drive (SSD), flash memory, an optical disc, a magneto-optical disc, tape, or a Universal Serial Bus (USB) Drive, or a combination of two or more of these. The memory may include removable or non-removable (or fixed) media, where appropriate. The memory may be internal or external to the data processing apparatus, where appropriate. In a particular embodiment, the memory is a Non-Volatile (Non-Volatile) memory. In particular embodiments, the Memory includes Read-Only Memory (ROM) and Random Access Memory (RAM). The ROM may be mask-programmed ROM, Programmable ROM (PROM), Erasable PROM (EPROM), Electrically Erasable PROM (EEPROM), Electrically rewritable ROM (EAROM), or FLASH Memory (FLASH), or a combination of two or more of these, where appropriate. The RAM may be a Static Random-Access Memory (SRAM) or a Dynamic Random-Access Memory (DRAM), where the DRAM may be a Fast Page Mode Dynamic Random-Access Memory (FPMDRAM), an Extended data output Dynamic Random-Access Memory (EDODRAM), a Synchronous Dynamic Random-Access Memory (SDRAM), and the like.

The memory may be used to store or cache various data files for processing and/or communication use, as well as possibly computer program instructions for execution by the processor.

The processor reads and executes the computer program instructions stored in the memory to implement any one of the myocardial infarction prediction methods in the above embodiments.

In some of these embodiments, the computer device may also include a communication interface and a bus. The processor, the memory and the communication interface are connected through a bus and complete mutual communication.

The communication interface is used for realizing communication among units, devices, units and/or equipment in the embodiment of the application. The communication interface may also be implemented with other components such as: the data communication is carried out among external equipment, image/data acquisition equipment, a database, external storage, an image/data processing workstation and the like.

A bus comprises hardware, software, or both that couple components of a computer device to one another. Buses include, but are not limited to, at least one of the following: data Bus (Data Bus), Address Bus (Address Bus), Control Bus (Control Bus), Expansion Bus (Expansion Bus), and Local Bus (Local Bus). By way of example, and not limitation, a Bus may include an Accelerated Graphics Port (AGP) or other Graphics Bus, an Enhanced Industry Standard Architecture (EISA) Bus, a Front-Side Bus (FSB), a Hyper Transport (HT) Interconnect, an ISA (ISA) Bus, an InfiniBand (InfiniBand) Interconnect, a Low Pin Count (LPC) Bus, a memory Bus, a microchannel Architecture (MCA) Bus, a PCI-Express (PCI-X) Bus, a Serial Advanced Technology Attachment (SATA) Bus, abbreviated VLB) bus or other suitable bus or a combination of two or more of these. A bus may include one or more buses, where appropriate. Although specific buses are described and shown in the embodiments of the application, any suitable buses or interconnects are contemplated by the application.

The computer device may execute the myocardial infarction prediction method in the embodiments of the present application.

In addition, in combination with the myocardial infarction prediction method in the foregoing embodiments, the present application embodiment may be implemented by providing a computer-readable storage medium. The computer readable storage medium having stored thereon computer program instructions; the computer program instructions, when executed by a processor, implement any of the myocardial infarction prediction methods of the above embodiments.

The myocardial infarction prediction method, the myocardial infarction prediction system, the computer equipment and the computer storage medium have the following advantages:

1) the sample pretreatment process is simple: compared with the conventional serum centrifugation method, the method only adds one membrane treatment process, can rapidly separate serum in the centrifugation process so as to carry out subsequent Raman detection, and has the centrifugation treatment time of only 6-8 minutes, which is shorter than that of the conventional serum centrifugation method;

2) the detection time is short: the Raman detection needs about 20 seconds at most, even if the detection is carried out on multiple points (such as three-point detection), the time consumption is not more than 1 hour at most;

3) the detection components are comprehensive: non-target metabolites, proteins and lipids in serum can be obtained, the components are comprehensive, and the data are reliable;

4) the detection timeliness is good: in the early stage or early stage of myocardial infarction, the change of metabolites can be rapidly analyzed through Raman detection, the early warning of acute myocardial infarction is carried out, and the subsequent accurate detection can be conveniently carried out by doctors in a targeted manner, for example, under the condition of high acute myocardial infarction, the doctors can carry out targeted detection;

5) the detection cost is low: the detection of the invention can be completed only by one sample processing tube and one detection slide, and the cost is very low;

6) low carbon and no pollution: a large amount of medical waste is not generated in the detection process, the environment is protected, the carbon is low, and the environment is not polluted.

Example 2

This embodiment is a specific application of the present invention.

1. Standard calibration

1.1 sample treatment

Obtaining blood samples, wherein the blood samples comprise blood samples of patients with acute myocardial infarction at the beginning of admission, blood samples of patients with acute myocardial infarction history and blood samples of normal persons;

respectively centrifuging the blood samples to obtain corresponding initial plasma samples, and transferring 1.5ml of the initial plasma samples into an EP tube, wherein the centrifugation conditions are 3000rpm and 2 min;

and (3) performing centrifugal treatment based on membrane filtration on the initial plasma sample to obtain a corresponding plasma sample, wherein 450 mu l of the initial plasma sample is put into an inner tube of a Millipore UFC503096Amicon Ultra-30K ultrafiltration centrifugal tube, and the centrifugal conditions are 12000rpm and 2.5-3 min.

Wherein, the number of the blood samples of the patients with acute myocardial infarction at the beginning of admission, the blood samples of the patients with acute myocardial infarction history and the blood samples of normal people is at least 4.

1.2 Raman detection

Taking 2 mul of the plasma sample, placing the plasma sample on an aluminized glass slide (which is heated at 37 ℃ in advance) for 30s-1min, adjusting the focusing of a Raman instrument in the period of time, and immediately performing Raman detection, wherein the Raman detection conditions are as follows: witech alpha300R laser confocal Raman instrument, excitation wavelength: 532nm, power: 10-14mw, grating: 1200g/mm, single spectrum integration time: 16-20s, objective: 100x/0.9, and 5-8 maps are collected for each plasma sample;

limit calibration and normalization processing were performed on all profiles, mean and PCA analysis was performed on all profiles for each plasma sample using the R language, and statistical analysis was performed on the difference peaks.

1.3 Mass spectrometric detection

And (3) taking the plasma sample for mass spectrum detection, and obtaining a mass spectrum detection result.

Wherein, the mass spectrometric detection mainly detects non-target metabolites, and performs PCA analysis and LDA analysis, and signal path analysis of differential metabolites on the whole plasma sample.

In the specific implementation verification process, the same sample is subjected to mass spectrometry detection and raman detection respectively, including two classifications (i.e., whether the sample is AMI) and four classifications (i.e., whether the sample is AMI, CAD, AF for coronary heart disease, and Con for a control group).

Based on the two classifications of raman detection (i.e., determining whether AMI or not), the prediction accuracy was 92.31%, and the sensitivity and specificity were as follows:

AMI：[1.0,0.8571]

and others: [0.8571,1.0].

Based on the two classifications of mass spectrometric detection (i.e., determination of AMI or not), the prediction accuracy was 92.31%, and the sensitivity and specificity were as follows:

AMI：[1.0,0.8889]

and others: [0.8889,1.0].

As shown above, in the case of the binary classification, the consistency of the judgment accuracy of the raman detection and the mass spectrometry on the same batch of samples is 100%.

Based on the four classifications of raman detection, the prediction accuracy was 76.92%, and the sensitivity and specificity were as follows:

AMI：[0.8333,0.8571]

AF：[0.0,1.0]

CAD:[0.5,1.0]

CON:[1.0,0.7778]。

based on the four classifications of mass spectrometric detection, the prediction accuracy was 76.92%, and the sensitivity and specificity were as follows:

AMI：[0.8,1.0]

AF：[1.0,0.8182]

CAD:[0.3333,1.0]

CON:[1.0,0.9]。

as shown above, in the case of the four-classification, the consistency of the judgment accuracy of the raman detection and the mass spectrometry detection on the same batch of samples is 100%.

The changes of metabolites are mainly the following aspects:

firstly, the method comprises the following steps: is structural damage and malfunction of the myocardium: the main performance is as follows:

(1) myocardial preservation sudden decrease: adenosine, an important nucleoside for myocardial protection, plays a role in protecting the myocardium by participating in at least 13 signaling pathways. In the experiment, the significant reduction in AMI patients reaches 0.042 times, which indicates that the myocardium obviously loses the protection; clinical studies have shown that: the large dose of intracoronary adenosine has a myocardial rescue effect on patients with acute ST-elevation myocardial infarction. Such a low fold reduction indicates that all 13 signaling pathways involved in adenosine are significantly reduced, and certainly have significant adverse effects on the structure and function of the myocardium.

(2) Increased myocardial damage: in this study, β -hydroxybutyrate (β -hydroxybutyrate) was increased 5.496 times as the most important metabolite in ketone bodies (80% content). The research shows that: the increase of beta hydroxybutyrate dehydrogenase (beta-HBD) can be used as a cardiac muscle injury marker, and the beta hydroxybutyrate dehydrogenase is closely related to myocardial infarction, so that HBD is very abundantly distributed in the heart, and the serum of mice and AMI patients is remarkably increased in literature reports. Another biological process is the significant reduction of Gap Junction (Gap Junction) related metabolites. In this study, three very important metabolites were involved in gap junctions, 1-octadecanoyl-2-arachidonoyl-cis-glycerol (0.047 fold) and L-glutamine (0.511 fold), Glutamic acid, 0.543 fold), respectively, which were involved in myocardial gap junctions. It can be seen that myocardial damage results in a significant increase in structural instability.

(3) Molecular transport abnormalities: ABC transporters (ABC transporters) signal pathways, which belong to a large and diverse family of proteins, each member of which contains two highly conserved ATP binding domains (ATPbinding cassettes), are known as ABC transporters, which dimerize by binding ATP, depolymerize after ATP hydrolysis, and transfer substrates bound thereto to the other side of the membrane by conformational changes. Although each ABC transporter transports only one substrate or class of substrates, there are members of the protein family that are capable of transporting ions, amino acids, nucleotides, polysaccharides, polypeptides, and even proteins. ABC transporters can also catalyze the turnover of lipids of the lipid bilayer between the two layers, which is of great significance in the generation and functional maintenance of membranes. Therefore, when the ABC transporters are abnormal in function, the transported molecules such as ions, amino acids, nucleotides, polysaccharides and polypeptides are abnormal, and the normal structure and function of the cardiac muscle are seriously affected. In this study, there are 9 substances involved in the ABC transporter signaling pathway, including L-glutamine (L-Glutamate, 0.511 fold), Glutamic acid, 0.543 fold), Uridine (Uridine, 0.873 fold), Taurine (Taurine, 0.849 fold), Choline (Choline, 0.495 fold), Histidine (Histinine, 0.399 fold), D-Mannose (D-Mannose, 1.82 fold), Betaine (Betaine, 0.8 fold), L-Histidine (L-Lysine, 0.768 fold), L-alanine (L-alanine0.733 fold). (some fold changes less than 2 fold, not shown in the table) thus visible molecular transport abnormalities are an important event and angle in the occurrence of AMI.

Secondly, the method comprises the following steps: significant abnormalities in phospholipid metabolic pathways

Phospholipid materials are an important part of the biological membrane and are a mixture of similar components. Is one of the components of bile and biomembrane surfactant, and is involved in the biological processes of protein recognition and signal transduction by cell membranes. The phospholipid mainly comprises Phosphatidylcholine (PC), Phosphatidylinositol (PI), Phosphatidylethanolamine (PE), Phosphatidylserine (PS), Phosphatidylglycerol (PG), Phosphatidic Acid (PA), etc. Glycerol and choline are important components of phospholipids, and the results of this study show that: at AMI, phospholipid metabolic pathways are markedly abnormal. In this study, the decrease in phosphatidylcholine LPC (18:2) in AMI group was 0.086-fold compared to the control group. The research in the literature finds that: the serum LPC 14: 0-LPC 18:0 of AMI group patients are lower than that of the control group subjects in all five LPCs substances (P < 0.05). In addition, the present study also found that other metabolites of phospholipids, including phosphatidylcholine (18:1e/9-hode, 0.053 fold), 1-stearoyl-2-linoleoyl-sn-glycero-3-phosphoethanolamine (0.11 fold), phosphatidylethanolamine (18:1e/10-hdohe, 0.134 fold), phosphatidylinositol 36:2(0.156 fold), phosphatidylinositol 38:4(0.1 fold), and phosphatidylcholine (16:1e/17-hdohe, 0.157 fold), were all significantly reduced. A total of 28 phospholipid metabolites were significantly altered in different classes when AMI was compared to normal and other disease groups! Upon KEGG analysis of these metabolites, a significant majority of them were found to be reduced, with few significant increases. Effect of phospholipids on cardiac function: the mechanisms by which these changes, such as the reduction in LPC species levels, may be associated with the calcium mobilisation effect of the phosphatidic acid species causing extra calcium influx, which in turn affects the contractile capacity of the heart.

It follows that phospholipid metabolism abnormalities have a very important significance in the diagnosis of AMI. The specific metabolite changes and AUC are shown in tables 1 and 2 below and in FIGS. 8 a-8 c, FIG. 9:

TABLE 1 Change in 82 metabolites and AUC from first Raman measurement data

TABLE 2 Change in 55 metabolites and AUC for the second type of Raman data

Thirdly, the method comprises the following steps: significant change in energy supply

In glycolysis related intermediates, including citric acid (Citrate, 4.726 fold increase) concentration was significantly increased, intermediate isocitric acid (Isocitrate, 0.683 fold), lactic acid (Dl-lactate, 0.763 fold-not mentioned in the table) were all slightly decreased. Citric acid, the first species of sugar to be oxidized, increases significantly suggesting that the aerobic oxidation of sugar also increases significantly. Lactic acid as an emergency energy source should be first energized during glycolysis and should also accumulate significantly and rise, but in fact decrease, indicating that lactic acid must be converted to another form to energize! Meanwhile, the beta-hydroxybutyrate is also an important component (accounting for 80 percent of the ketone body) in the ketone body and is also obviously enhanced, which indicates that the mode of energy metabolism is completely changed. That is, because AMI consumes a large amount of energy, current aerobic oxidation has not been able to meet the needs of the body.

Fourthly: significant alterations in amino acid metabolism

The main changes are that the alanine metabolic pathway, the aspartic acid metabolic pathway, the glutamine metabolic pathway and the amino acid synthesis (8 substances) are obviously changed, and no matter the three-carbon or six-carbon amino acid can also be used as energy supply to participate in the energy metabolism process.

These changes are a very comprehensive and systematic reflection of the full picture of AMI. The detection of raman in this process can detect these changes in the whole aspect very comprehensively, and by performing raman and mass spectrometric detection on the same substance respectively, and comparing with a neural network model, it is found that: the accuracy of the result is completely consistent with that of the mass spectrum detection. Fully indicates the accuracy and reliability of Raman direct detection of serum.

1.4 Standard calibration

And performing Raman detection on the standard substance of the 24 metabolites, and forming a standard Raman peak spectrum collection for judging AMI.

2. Myocardial infarction prediction model

2.1 construction of myocardial infarction prediction model

The myocardial infarction prediction model is constructed by adopting a BP neural network and comprises an input layer, a hidden layer and an output layer, wherein the input of the input layer is 1024, the output of the input layer is 512, the input of the hidden layer is 512, the output of the hidden layer is 128, and the input of the output layer is 128 and the output of the output layer is 4.

2.2 construction of data sets

Uniformly mapping Raman detection data of different plasma samples and standard samples into 1024-dimensional initial characteristic vectors, and storing the initial characteristic vectors into a txt file, wherein the initial characteristic vectors comprise two columns of data, one column is wave number, and the other column is intensityThe wave numbers of the data are consistent and are 279 and 2186cm^-1A total of 1024 points;

dividing the Raman detection data into a training set and a testing set, wherein the proportion of the training set to the testing set is 3: 1.

the raman detection data can be classified into 4 categories, which are AF (atrial fibrillation group), AMI (acute myocardial infarction group), CAD (coronary heart disease group), and Con (control group).

2.3 training myocardial infarction prediction model

Inputting the training set into a myocardial infarction prediction model to obtain an output result;

and inputting the output result into a softmax classification function for processing so as to obtain a myocardial infarction prediction result.

2.4 iterative myocardial infarction prediction model

Inputting the test set into the trained myocardial infarction prediction model to obtain an output result;

inputting the output result into a softmax classification function for processing so as to obtain a myocardial infarction prediction result;

the predicted myocardial infarction result is compared with the original result of the test set to test the effectiveness of the training.

The myocardial infarction prediction model was trained for 300 rounds using the training set, and every 10 rounds were tested to verify the effectiveness of the training.

2.5 optimized myocardial infarction prediction model

And calculating a loss value by using a cross entropy objective function, and optimizing a weight parameter by using a random gradient descent method and a back propagation algorithm.

Wherein the weight parameters include:

the raman detection data was weighted 90% and the interrogation information (clinical information) was weighted 10%.

Among the clinical information, the time from onset to admission (2 points), the history of coronary heart disease (2 points), the past history of PCI, the later history of hypertension, the history of diabetes, the history of hyperlipidemia, the history of cerebral infarction and the history of smoking are each divided into one point.

Wherein, the first disease onset time is also graded, 2 points within 12 hours, 1.5 points within 12-24 hours, 1 point within 24-72 hours, and 0.5 point within 72 hours.

Specifically, the past medical history is taken as an example for explanation:

	heart peduncle	Coronary heart disease	Atrial fibrillation	Control group
					History of hypertension	a1	a2	a3	a4(＝0)
History of diabetes	b1	b2	b3	b4(＝0)

According to the statistical result of the clinical data, the disease incidence probability of the disease corresponding to the disease can be obtained from the total number of the diseases of the certain disease and the number of the corresponding cases of the disease history, as shown in the table above. Such as: if the patient has a history of hypertension, then he/she has a probability of having myocardial infarction of a 1%, a probability of having coronary heart disease of a 2%, and so on.

Therefore, a mode of combining intelligent Raman data classification and clinical data is designed to analyze the final diagnosis result, and the Raman classification accounts for 90% and the clinical information accounts for 10%. Assuming that the patient # 1 has a history of hypertension, if the probability of the disease of the patient # 1 predicted by the raman classification model after passing through the Sigmoid activation function is c1, c2, c3, and c4 (corresponding to the four types of diseases), the new probability of the disease after weighted summation is d1 ═ c1 ═ 90% + a1 × 10%, d2 ═ c2 = 90% + a2 × 10%, d3 ═ c 3% + a3 × 10%, d4 ═ c 4%90 + a4 × 10%, where the sum of d1, d2, d3, and d4 is not 1. And then d1, d2, d3 and d4 are subjected to a Softmax activation function to obtain final prediction probabilities e1, e2, e3 and e4, and a final prediction result of a disease corresponding to a value with the highest probability is selected, wherein the sum of e1, e2, e3 and e4 is 1. That is, if the value of e3 is the largest here, the diagnosis of patient # 1 is atrial fibrillation.

2.6 trained myocardial infarction prediction model

Through continuous training, the experimental data of the myocardial infarction prediction model are as follows:

if only Raman detection is used, the experimental data are as follows:

on the basis of only using Raman detection, the accuracy of a test set can reach 92.23% by applying a model obtained by a neural network algorithm, the sensitivity to AMI (acute myocardial infarction) can reach 97.22%, and the specificity can reach 99.41%. The neural network algorithm can be used for analyzing the myocardial infarction patients through Raman spectrum data, and the obtained result is very high in precision.

The main research process of the invention is as follows:

performing mass spectrum and Raman detection on each sample, combining clinical information as weight, improving the quality through artificial intelligence algorithm fusion, fusing Raman, mass spectrum and clinical information together, establishing an AMI detection model by using 75% of data, finally integrating the model into a Raman detection process in an algorithm form, and finally directly detecting serum by Raman and combining the algorithm fusion rapidly to achieve the effect of rapidly judging AMI based on Raman detection.

The invention takes a normal control group as a diagnosis group, takes other diseases of the heart (including stable coronary heart disease and arrhythmia) as a differential diagnosis group, introduces the Raman detection data after the serum detection of a patient or the normal control group into a pre-trained myocardial infarction risk prediction model, namely obtains the myocardial infarction risk prediction result rapidly (within 2 minutes); in a total of 52 samples, the accuracy of the raman and mass spectral analysis results were completely consistent with 25% (i.e., 13 cases) of data, 92.31% (12/13).

Therefore, there was little difference between direct detection of metabolites in serum using raman and detection of gold standard Mass spectrometry (Mass spectrometry), confirming the reliability of the myocardial infarction risk prediction method of the present invention.

According to the invention, an AMI model is established by clinically diagnosed AMI patient data, the clinically diagnosed AMI model can be used as a total of 82 metabolites for predicting significant differential expression (including increase and decrease) of AMI by Raman detection, and the accuracy rate of the myocardial infarction risk prediction result is more than 92.31%, the sensitivity is 97.22%, and the specificity is 99%.

The most important advantages of the invention are:

firstly, the consistency of the method and a gold standard mass spectrometry method is very good in the aspect of the accuracy of detecting AMI, which shows that the reliability of a Raman-based AMI detection methodology is very high;

secondly, the time for detecting the myocardial damage marker based on the enzyme method is greatly shortened, the myocardial infarction risk prediction result can be obtained within 8 minutes, and the myocardium is protected from necrosis as much as possible before the next treatment;

thirdly, the detection method is simple, and a plurality of substances can be detected simultaneously by one-time test;

fourthly, the detection timeliness is good, and abnormal change of substances can be obtained before or after the myocardial infarction occurs;

fifthly, the detection cost is low, and the medical insurance cost can be greatly reduced;

and sixthly, a large amount of medical waste is not generated, and the environment pollution is avoided.

All possible combinations of the technical features of the above embodiments may not be described for the sake of brevity, but should be considered as within the scope of the present disclosure as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims

1. A method of predicting myocardial infarction, comprising:

acquiring Raman detection data of a plasma sample;

2. A method of predicting myocardial infarction as defined in claim 1, wherein the raman detected data is a first type of raman detected data comprising at least:

(ii) a Or

The Raman detection data is second type Raman detection data, and at least comprises the following data:

3. a myocardial infarction prediction method according to claim 1 or claim 2, wherein prior to obtaining raman detection data of a plasma sample, the method further comprises:

obtaining a blood sample;

centrifuging the blood sample to obtain an initial plasma sample;

4. A method of predicting myocardial infarction according to any one of claims 1 to 3, wherein before inputting the raman detection data into a pre-trained model for predicting myocardial infarction, the method further comprises:

acquiring inquiry information;

5. The myocardial infarction prediction method according to any one of claims 1 to 4, wherein inputting the Raman detection data into a pre-trained myocardial infarction prediction model to obtain a myocardial infarction prediction result comprises:

6. A myocardial infarction prediction method according to any one of claims 1 to 8, wherein the training method of the myocardial infarction prediction model comprises:

7. The myocardial infarction prediction method of claim 6, wherein the training method of the myocardial infarction prediction model further comprises:

calculating a loss value by using a cross entropy objective function;

8. A myocardial infarction prediction system comprising:

9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of predicting myocardial infarction as defined in any one of claims 1 to 6 when executing the computer program.

10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out a method of predicting myocardial infarction according to any one of claims 1 to 6.