Diabetes complication risk prediction method, device, equipment and storage medium

Abstract

The application relates to the field of artificial intelligence, and discloses a diabetes complication risk prediction method, which comprises the following steps: acquiring clinical medical data and intestinal microorganism sequencing data of historical diabetes patients; calculating evolutionary distances between intestinal flora based on the sequencing data, and acquiring an abundance matrix according to the evolutionary distances; inputting the abundance matrix into a prediction model to obtain a feature vector; merging and convolving the feature vector and the clinical indexes of the historical diabetes patients to obtain a prediction probability; comparing the prediction probability with real labels of the historical diabetes patients, adjusting model parameters of the prediction model, and obtaining an optimized model; and predicting the complication risk of a current diabetes patient based on the optimized model according to a complication query demand of the current diabetes patient. The application further discloses a diabetes complication risk prediction device, an electronic device and a storage medium. The application can improve the accuracy of diabetes complication risk prediction.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of artificial intelligence, and in particular to a diabetes complication risk prediction method and device, an electronic device, and a computer readable storage medium. BACKGROUND

[0002] Diabetes is a chronic metabolic disease. During the long incubation period of diabetes, high blood sugar can damage many organs and tissues in the body of a patient, and often leads to various complications, such as stroke, heart disease, retinopathy, etc.

[0003] Intestinal microorganisms have a certain impact on the glucose metabolism of a patient. However, existing researches mostly predict the risk of diabetes complications based on clinical data, and the prediction results have limitations due to the single data. If the clinical data and microbial data are fused for analysis, the prediction performance can be greatly improved. However, the dimension of intestinal microorganisms is high, usually in the order of hundreds of thousands, and the sample size is very low, which reduces the prediction performance. In addition, the phylogenetic structure of the microbiome data, i.e., the evolutionary relationship between species, also has important value for disease prediction. Existing researches often ignore this information, which leads to ineffective utilization of microbial data. SUMMARY

[0004] The present application provides a diabetes complication risk prediction method, device and computer readable storage medium, which mainly aims to improve the accuracy of diabetes complication risk prediction.

[0005] To achieve the above-mentioned purpose, the present application provides a diabetes complication risk prediction method, which comprises:

[0006] obtaining clinical medical data and intestinal microbial sequencing data of a historical diabetes patient;

[0007] calculating evolutionary distances between intestinal flora based on the intestinal microbial sequencing data, and obtaining a plurality of flora abundance matrices according to the evolutionary distances;

[0008] sorting the plurality of flora abundance matrices, and inputting a preset complication risk prediction model to obtain a representation vector of the complication risk;

[0009] merging and convolving the representation vector and the clinical indicators of the historical diabetes patient to obtain a prediction probability of the historical diabetes patient suffering from complications;

[0010] comparing the prediction probability with a true label of the historical diabetes patient, adjusting model parameters of the complication risk prediction model to obtain an optimized model;

[0011] predicting a complication risk of the current diabetes patient based on the optimization model.

[0012] Optionally, the clinical medical data and the intestinal microorganism sequencing data of the historical diabetes patient are acquired, including:

[0013] inquiring identity information of the historical diabetes patient, and acquiring case data of the historical diabetes patient based on the identity information;

[0014] extracting the clinical medical data from the case data of the historical diabetes patient, and acquiring an intestinal microorganism sample of the historical diabetes patient;

[0015] extracting DNA from the intestinal microorganism sample to obtain intestinal flora DNA;

[0016] sequencing the intestinal flora DNA by a high-throughput sequencing method to obtain intestinal microorganism sequencing data.

[0017] Optionally, the calculating the evolutionary distance between intestinal floras based on the intestinal microorganism sequencing data includes:

[0018] acquiring a preset data analysis system, and inputting the intestinal microorganism sequencing data into the data analysis system to construct an intestinal flora evolutionary tree;

[0019] calculating the shortest path between floras by the intestinal flora evolutionary tree to obtain the evolutionary distance of the intestinal floras.

[0020] Optionally, the merging and convolving the representation vector and the clinical indicators of the historical diabetes patient to obtain a prediction probability of the historical diabetes patient suffering from a complication includes:

[0021] vectorizing the clinical indicators to obtain a clinical indicator vector, and merging the representation vector and the clinical indicator vector to construct a joint vector;

[0022] extracting features of the joint vector by convolution to obtain a feature vector;

[0023] activating the feature vector and mapping it to a preset probability interval to obtain the prediction probability of the historical diabetes patient suffering from a complication.

[0024] Optionally, the comparing the prediction probability with a true label of the historical diabetes patient, adjusting model parameters of the complication risk prediction model to obtain an optimization model includes:

[0025] calculating a loss value of the prediction probability and the true label by cross-entropy;

[0026] The loss value is back-propagated by minimizing a loss function, and parameters of the complication risk prediction model are adjusted according to the loss value, so as to obtain an optimized model of the adjusted parameters.

[0027] Optionally, the feature of the joint vector is extracted by convolution to obtain a feature vector, including:

[0028] The joint vector is input into a first fully connected layer to extract preliminary features in the joint vector and obtain a preliminary feature vector;

[0029] The preliminary feature vector is input into a second fully connected layer to perform abstract feature extraction and mapping on the preliminary feature vector and obtain an abstract feature vector;

[0030] The abstract feature vector is input into a third fully connected layer to perform compression and integration on the abstract feature vector and obtain the feature vector.

[0031] Optionally, the plurality of bacterial community abundance matrices are obtained according to the evolutionary distances, including:

[0032] The evolutionary distances of the intestinal flora are sorted on a pre-constructed two-dimensional plane to construct an abundance matrix of the intestinal flora and obtain a plurality of bacterial community abundance matrices.

[0033] To solve the above problems, the application further provides a diabetes complication risk prediction device, the device comprising:

[0034] A data acquisition module is configured to acquire clinical medical data and intestinal microbial sequencing data of a historical diabetes patient;

[0035] A probability prediction module is configured to calculate evolutionary distances between intestinal bacterial communities based on the intestinal microbial sequencing data, obtain a plurality of bacterial community abundance matrices according to the evolutionary distances, sort the plurality of bacterial community abundance matrices, input the sorted plurality of bacterial community abundance matrices into a preset complication risk prediction model, and obtain a representation vector of the complication risk; and combine and convolve the representation vector and the clinical indicators of the historical diabetes patient to obtain a prediction probability of the historical diabetes patient suffering from a complication.

[0036] A model optimization module is configured to compare the prediction probability with a true label of the historical diabetes patient, adjust model parameters of the complication risk prediction model, and obtain an optimized model.

[0037] A risk prediction module is configured to predict a complication risk of a current diabetes patient based on a complication query demand of the current diabetes patient by using the optimized model.

[0038] To solve the above problems, the application further provides an electronic device, the electronic device comprising:

[0039] at least one processor; and

[0040] a memory connected with the at least one processor in communication; wherein

[0041] the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the above-mentioned diabetes complication risk prediction method.

[0042] To solve the above-mentioned problems, the present application also provides a computer readable storage medium, the computer readable storage medium stores at least one computer program, and the at least one computer program is executed by a processor in an electronic device to implement the above-mentioned diabetes complication risk prediction method.

[0043] The embodiment of the present application calculates the evolutionary distance between intestinal flora of historical diabetes patients by sequencing data of intestinal microorganisms of the historical diabetes patients, obtains a plurality of flora abundance matrices, sorts the flora abundance matrices, inputs the sorted flora abundance matrices into a complication risk prediction model to obtain a representation vector of the complication risk, and then combines the representation vector with clinical indicators of the historical diabetes patients to calculate a prediction probability of the historical diabetes patients suffering from complications, thereby realizing preliminary prediction of the complication incidence probability. The prediction probability is compared with a real label of the historical diabetes patients, parameters of the complication risk prediction model are adjusted, an optimized model is obtained, and the purpose of training and optimizing the complication risk prediction model is achieved. Finally, the complication risk of a current diabetes patient is predicted by the optimized model based on the complication demand of the current diabetes patient. Therefore, the diabetes complication risk prediction method, device, electronic device and computer readable storage medium provided by the present application can solve the problem of limitations in predicting diabetes patients due to single data. BRIEF DESCRIPTION OF DRAWINGS

[0044] Figure 1 A flowchart of a diabetes complication risk prediction method provided by an embodiment of the present application is shown in the figure;

[0045] Figure 2 A detailed implementation flowchart of one step in the diabetes complication risk prediction method shown in the figure is shown in the figure; Figure 1

[0046] Figure 3 A detailed implementation flowchart of one step in the diabetes complication risk prediction method shown in the figure is shown in the figure; Figure 1

[0047] Figure 4 A functional module diagram of a diabetes complication risk prediction device provided by an embodiment of the present application is shown in the figure;

[0048] ​​Figure 5 A structural schematic diagram of an electronic device for implementing the diabetes complication risk prediction method according to an embodiment of the present application is shown in FIG. 1.

[0049] The implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0050] 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.

[0051] The embodiments of the present application provide a diabetes complication risk prediction method. The execution subject of the diabetes complication risk prediction method includes, but is not limited to, at least one of electronic devices such as a server and a terminal, which can be configured to execute the method provided by the embodiments of the present application. In other words, the diabetes complication risk prediction method can be executed by software or hardware installed in a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be a stand-alone server, or a cloud server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms, etc. basic cloud computing services.

[0052] Reference Figure 1 A flowchart of a diabetes complication risk prediction method according to an embodiment of the present application is shown in FIG. 1.

[0053] In the present embodiment, the diabetes complication risk prediction method includes:

[0054] S1, obtaining clinical medical data and intestinal microorganism sequencing data of historical diabetes patients.

[0055] In the embodiments of the present application, the clinical medical data is various data in the medical field for describing and recording information such as patient health status, medical history, diagnosis results and treatment plans, etc. The clinical medical data can come from various medical records, including doctor's handwritten paper medical records, text records in electronic medical record systems, etc. Among them, the clinical medical data is usually also closely related to the disease type, physiological state and medical history of the patient, for example, in the present scheme, the disease type is diabetes, and the clinical data will have rich blood glucose data of the patient.

[0056] Further, the intestinal microorganism sequencing data refers to data obtained by gene sequencing of intestinal microorganism community of the patient through high-throughput sequencing technology, and the intestinal microorganism sequencing data is usually DNA sequence data, and biological information extraction is usually performed on the intestinal microorganism sequencing data, so that the intestinal microorganism flora and quantity can be identified, and information about flora abundance, diversity, community structure and the like can be extracted.

[0057] In detail, the historical diabetes patient is a patient who has diabetes and has clear complication data records.

[0058] Referring to Figure 2 In the embodiment of the present application, the clinical medical data and the intestinal microorganism sequencing data of the historical diabetes patient are obtained, including:

[0059] S101, identity information of the historical diabetes patient is inquired, and case data of the historical diabetes patient is obtained based on the identity information;

[0060] S102, clinical medical data is extracted from the case data of the historical diabetes patient, and an intestinal microorganism sample of the historical diabetes patient is obtained;

[0061] S103, DNA extraction is performed on the intestinal microorganism sample to obtain intestinal flora DNA;

[0062] S104, the intestinal flora DNA is sequenced by high-throughput sequencing method to obtain intestinal microorganism sequencing data.

[0063] In detail, the identity information is used to represent the identity of the historical diabetes patient and has uniqueness, for example, the identity card number, serial number and the like of the diabetes patient. The high-throughput sequencing method is a method for rapidly and concurrently determining DNA sequence of a sample, and has the advantages of high speed, high precision and low cost compared with traditional low-throughput sequencing.

[0064] S2, evolutionary distance between intestinal flora is calculated based on the intestinal microorganism sequencing data, and a plurality of flora abundance matrices are obtained according to the evolutionary distance.

[0065] In the embodiment of the present application, the evolutionary distance is the shortest distance of the intestinal flora on the evolutionary tree constructed according to the intestinal microorganism sequencing data. The abundance matrix is a data matrix in the field of microbial ecology, and contains order of magnitude or abundance data of a microbial sample.

[0066] The abundance matrix plays an important role in diversity analysis, and can be used to calculate diversity indexes of the microbial community in the ecosystem of the intestinal tract of the diabetes patient, such as species abundance, species evenness, Shannon diversity index and the like.

[0067] Further, the evolution distance between the intestinal flora is calculated based on the intestinal microorganism sequencing data, comprising:

[0068] A preset data analysis system is acquired, and the intestinal microorganism sequencing data is input into the data analysis system to construct an intestinal flora evolution tree;

[0069] The shortest path between the intestinal flora is calculated through the intestinal flora evolution tree to obtain the evolution distance between the intestinal flora.

[0070] In the embodiment of the present application, the data analysis system is a data analysis system constructed based on QIIME software (Quantitative Insights In Microbial Ecology), which is a software application for performing microbial community analysis, used for analyzing and interpreting nucleic acid sequence data from fungal, bacterial, viral and archaeal communities.

[0071] In the embodiment of the present application, the evolution tree can also be called a phylogenetic tree, which can use a tree branch graph to represent the genetic relationship between the intestinal flora, wherein the genetic relationship is estimated by the path between the communities.

[0072] Further, the plurality of flora abundance matrices are obtained according to the evolution distance, comprising:

[0073] The evolution distance of the intestinal flora is sorted on a pre-constructed two-dimensional plane to construct an abundance matrix of the intestinal flora, and a plurality of flora abundance matrices are obtained.

[0074] S3, the plurality of flora abundance matrices are sorted and input into a preset complication risk prediction model to obtain a representation vector of the complication risk.

[0075] In the embodiment of the present application, the complication risk prediction model is a risk prediction model constructed based on a convolutional neural network (Convolutional Neural Networks, CNN), and the representation vector is a vector representing the feature of the complication risk, containing all feature information that can reflect the health status of the diabetic patient, and the representation vector has a certain dimension number, which is convenient for input into a machine learning model for prediction.

[0076] In the embodiment of the present application, the plurality of flora abundance matrices are sorted and input into a preset complication risk prediction model to obtain a representation vector of the complication risk, comprising:

[0077] The plurality of flora abundance matrices are sorted to obtain a sorted matrix;

[0078] Features in the sorted matrix are extracted by the complication risk prediction model, and vectorization is performed to obtain the feature vector.

[0079] In detail, the sorted matrix is a set of plurality of flora abundance matrices.

[0080] S4, the feature vector and the clinical indicators of the historical diabetic patients are combined and convolved to obtain the prediction probability of the historical diabetic patients suffering from complications.

[0081] In the embodiment of the present application, the clinical indicators are parameters or measurement values for evaluating the physical health status of patients, and are usually associated with diseases or physiological functions in a specific medical field, such as blood pressure, blood glucose, pulse, body temperature, etc.

[0082] Further, the prediction probability is the complication incidence probability predicted based on the complication risk prediction model and the clinical indicators of the historical diabetic patients.

[0083] Referring to Figure 3 In the embodiment of the present application, the combination and convolution of the feature vector and the clinical indicators of the historical diabetic patients to obtain the prediction probability of the historical diabetic patients suffering from complications include:

[0084] S301, the clinical indicators are vectorized to obtain a clinical indicator vector, and the feature vector and the clinical indicator vector are combined to construct a joint vector;

[0085] S302, features of the joint vector are extracted by convolution to obtain a feature vector;

[0086] S303, the feature vector is activated and mapped to a preset probability interval to obtain the prediction probability of the historical diabetic patients suffering from complications.

[0087] In detail, the feature vector can be activated by a softmax activation function in the embodiment of the present application. The softmax activation function is a commonly used activation function, and is usually used in multi-classification problems. In a neural network model, the softmax activation function acts on the output of the last full connection layer to map it to a probability distribution.

[0088] In the embodiment of the present application, the extraction of features of the joint vector by convolution to obtain the feature vector includes:

[0089] The joint vector is passed through a first full connection layer to extract preliminary features in the joint vector to obtain a preliminary feature vector;

[0090] The preliminary feature vector is passed through a second fully connected layer to perform abstract feature extraction and mapping on the preliminary feature vector to obtain an abstract feature vector.

[0091] The abstract feature vector is passed through a third fully connected layer to perform compression and integration on the abstract feature vector to obtain the feature vector.

[0092] In the embodiment of the application, the first fully connected layer, the second fully connected layer and the third fully connected layer each contain hidden layer nodes and an activation function, which are used to extract and process features of the feature vector, wherein the hidden layer nodes can also be referred to as a hidden layer.

[0093] S5, comparing the prediction probability with the real label of the historical diabetic patient, adjusting the model parameters of the complication risk prediction model to obtain an optimized model.

[0094] In the embodiment of the application, the real label is complication data actually suffered by the historical diabetic patient, and the optimized model is a risk prediction model adjusted according to the prediction probability.

[0095] In the embodiment of the application, the comparison of the prediction probability with the real label of the historical diabetic patient and the adjustment of the model parameters of the complication risk prediction model to obtain an optimized model include:

[0096] The loss value of the prediction probability and the real label is calculated by cross-entropy.

[0097] The loss value is back-propagated by minimizing the loss function, the parameters of the complication risk prediction model are adjusted according to the loss value, and an optimized model with adjusted parameters is obtained.

[0098] In the embodiment of the application, the loss value is an index in the training process of a machine learning training model, which is used to measure the performance of the complication risk prediction model, the backpropagation is an algorithm for error backpropagation training in a neural network, which can calculate the gradient of the complication risk prediction model based on the chain rule, and update the parameters of the model by backpropagating the gradient.

[0099] S6, based on the complication query demand of the current diabetic patient, the complication risk of the current diabetic patient is predicted by the optimized model.

[0100] In the embodiment of the application, the intestinal microorganism sequencing data and the clinical medical data of the current diabetic patient are obtained, and the complication risk of the current diabetic patient is predicted by the optimized model according to the intestinal microorganism sequencing data and the clinical medical data.

[0101] Further, the current diabetes patient is a patient who has been determined to have diabetes but whose complications have not yet been determined.

[0102] The embodiment of the present application calculates the evolutionary distance between intestinal flora of historical diabetes patients by intestinal microbiota sequencing data, obtains a plurality of flora abundance matrices, then sorts the flora abundance matrices, inputs them into a complication risk prediction model to obtain a representation vector of the complication risk, and then combines and calculates the representation vector and the clinical indicators of the historical diabetes patients to obtain the prediction probability of the historical diabetes patients suffering from complications, thereby achieving preliminary prediction of the complication incidence probability. The prediction probability is compared with the true label of the historical diabetes patients to adjust the parameters of the complication risk prediction model to obtain an optimized model, so as to achieve the purpose of training and optimizing the complication risk prediction model. Finally, the complication risk of the current diabetes patient is predicted by the optimized model based on the complication demand of the current diabetes patient. Therefore, the diabetes complication risk prediction method, device, electronic equipment and computer readable storage medium provided by the present application can solve the problem of limitations in predicting diabetes patients in the case of single data.

[0103] As shown in Figure 4 FIG. 1 is a functional module diagram of a diabetes complication risk prediction device according to an embodiment of the present application.

[0104] The diabetes complication risk prediction device 100 according to the present application can be installed in an electronic device. According to the functions to be implemented, the diabetes complication risk prediction device 100 can include a data acquisition module 101, a probability prediction module 102, a model optimization module 103 and a risk prediction module 104. The modules according to the present application can also be referred to as units, which refer to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, and are stored in the memory of the electronic device.

[0105] In the present embodiment, the functions of each module / unit are as follows:

[0106] The data acquisition module 101 is configured to acquire clinical medical data and intestinal microbiota sequencing data of historical diabetes patients.

[0107] The probability prediction module 102 is configured to calculate the evolutionary distance between intestinal flora based on the intestinal microbiota sequencing data, and obtain a plurality of flora abundance matrices according to the evolutionary distance; sort the plurality of flora abundance matrices, and input them into a preset complication risk prediction model to obtain a representation vector of the complication risk; and combine and convolve the representation vector and the clinical indicators of the historical diabetes patients to obtain the prediction probability of the historical diabetes patients suffering from complications.

[0108] The model optimization module 103 is configured to compare the predicted probability with the real label of the historical diabetic patient, adjust the model parameter of the complication risk prediction model, and obtain an optimized model.

[0109] The risk prediction module 104 is configured to predict the complication risk of the current diabetic patient by the optimized model based on the complication query demand of the current diabetic patient.

[0110] In detail, each module in the diabetes complication risk prediction device 100 in the embodiment of the present application adopts the same technical means as the diabetes complication risk prediction method in the above Figures 1 to 3 , and can produce the same technical effects, which will not be described here.

[0111] As shown in Figure 5 , it is a structural schematic diagram of an electronic device for implementing the diabetes complication risk prediction method according to an embodiment of the present application.

[0112] The electronic device 1 can include a processor 10, a memory 11, a communication bus 12, and a communication interface 13, and can further include a computer program stored in the memory 11 and executable on the processor 10, such as a diabetes complication risk prediction program.

[0113] In some embodiments, the processor 10 can be composed of an integrated circuit, for example, a single packaged integrated circuit, or a plurality of packaged integrated circuits with the same function or different functions, including one or more combinations of central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control core of the electronic device, which connects all components of the electronic device through various interfaces and lines, executes or runs the programs or modules stored in the memory 11 (such as executing the diabetes complication risk prediction program), and calls the data stored in the memory 11 to perform various functions and process data of the electronic device.

[0114] The memory 11 includes at least one type of readable storage medium, such as a flash memory, a mobile hard disk, a multimedia card, a card-type memory (e.g., an SD or DX memory, etc.), a magnetic memory, a disk, an optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of an electronic device, such as a mobile hard disk of the electronic device. In other embodiments, the memory 11 can also be an external storage device of the electronic device, such as a plug-in mobile hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Further, the memory 11 can include both an internal storage unit and an external storage device of the electronic device. The memory 11 can be used not only to store application software and various data installed in the electronic device, such as the code of the diabetes complication risk prediction program, but also to temporarily store data that has been output or will be output.

[0115] The communication bus 12 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. The bus is configured to enable connection and communication between the memory 11 and at least one processor 10, etc.

[0116] The communication interface 13 is used for communication between the electronic device and other devices, including a network interface and a user interface. Optionally, the network interface can include a wired interface and / or a wireless interface (e.g., a WI-FI interface, a Bluetooth interface, etc.), which is typically used to establish a communication connection between the electronic device and other electronic devices. The user interface can be a display (Display), an input unit (such as a keyboard (Keyboard)), and optionally, the user interface can also be a standard wired interface, a wireless interface. Optionally, in some embodiments, the display can be an LED display, a liquid crystal display, a touch liquid crystal display, and an OLED (Organic Light-Emitting Diode) touch, etc. The display can also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device and to display a visualized user interface.

[0117] Figure 5 Only the electronic device with components is shown, and those skilled in the art can understand that, Figure 5The illustrated structure does not constitute a limitation on the electronic device 1, and can include fewer or more components than illustrated, or combine certain components, or arrange different components.

[0118] For example, although not shown, the electronic device can further include a power supply (such as a battery) for powering individual components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management device, so that functions such as charge management, discharge management, and power consumption management can be achieved through the power management device. The power supply can also include one or more direct current or alternating current power sources, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, and any other components. The electronic device can also include various sensors, Bluetooth modules, Wi-Fi modules, and the like, which will not be described here.

[0119] It should be understood that the embodiments are only for illustration and are not limited in the scope of the patent application by this structure.

[0120] The diabetes complication risk prediction program stored in the memory 11 in the electronic device 1 is a combination of multiple instructions that, when executed in the processor 10, can achieve:

[0121] Obtaining clinical medical data and gut microbiome sequencing data of historical diabetes patients;

[0122] Calculating evolutionary distances between gut microbiota based on the gut microbiome sequencing data, and obtaining multiple microbiota abundance matrices according to the evolutionary distances;

[0123] Sorting the multiple microbiota abundance matrices and inputting them into a preset complication risk prediction model to obtain a representation vector of the complication risk;

[0124] Merging and convolving the representation vector and the clinical indicators of the historical diabetes patients to obtain a prediction probability of the historical diabetes patients suffering from complications;

[0125] Comparing the prediction probability with the true label of the historical diabetes patients to adjust the model parameters of the complication risk prediction model to obtain an optimized model;

[0126] Based on the complication query demand of a current diabetes patient, predicting the complication risk of the current diabetes patient through the optimized model.

[0127] Specifically, the specific implementation method of the processor 10 for the above instructions can refer to the description of the related steps in the corresponding embodiments of the accompanying drawings, which will not be described here.

[0128] Further, the modules / units integrated in the electronic device 1 are stored in a computer readable storage medium if they are realized in the form of software function units and sold or used as independent products. The computer readable storage medium can be volatile or non-volatile. For example, the computer readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM).

[0129] The application further provides a computer readable storage medium, which stores a computer program, and the computer program can realize the following when executed by a processor of an electronic device:

[0130] Obtaining clinical medical data and intestinal microorganism sequencing data of a historical diabetes patient;

[0131] Calculating evolutionary distances between intestinal flora based on the intestinal microorganism sequencing data, and obtaining a plurality of flora abundance matrices according to the evolutionary distances;

[0132] Sorting the plurality of flora abundance matrices, and inputting a preset complication risk prediction model to obtain a representation vector of the complication risk;

[0133] Merging and convolving the representation vector and the clinical indicators of the historical diabetes patient to obtain a prediction probability of the historical diabetes patient suffering from complications;

[0134] Comparing the prediction probability with a true label of the historical diabetes patient, adjusting model parameters of the complication risk prediction model to obtain an optimized model;

[0135] Based on a complication query demand of a current diabetes patient, predicting a complication risk of the current diabetes patient through the optimized model.

[0136] In several embodiments provided in the application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other manners. For example, the above described device embodiments are merely illustrative, and the division of the modules can be different from the above. For example, the division of the modules can be different from the above.

[0137] The modules illustrated as separate components can or can not be physically separated, and the components illustrated as modules can or can not be physical units, i.e., can be located in one place, or can be distributed on a plurality of network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the embodiments.

[0138] In addition, each functional module in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware, or in the form of hardware plus software function module.

[0139] It is obvious for those skilled in the art that the present application is not limited to the details of the above exemplary embodiments, and the present application can be realized in other specific forms without departing from the spirit or essential characteristics of the present application.

[0140] Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting, and the scope of the present application is defined by the appended claims rather than the above description, and therefore all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present application. Any additional reference signs in the claims should not be considered as limiting the claims involved.

[0141] The blockchain referred to in the present application is a new application mode of computer technologies such as distributed data storage, peer-to-peer transmission, verification mechanism, and encryption algorithm. Blockchain, in essence, is a decentralized database, which is a series of data blocks associated using cryptographic methods, each data block containing information of a batch of network transactions, for verifying the validity (anti-fake) of the information and generating the next block. The blockchain can include a blockchain underlying platform, a platform product service layer, and an application service layer, etc.

[0142] The embodiments of the present application can acquire and process related data based on artificial intelligence technology. Among them, artificial intelligence (Artificial Intelligence, AI) is to use digital computers or digital computer controlled machines to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0143] In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. The plurality of units or devices stated in the system claims can also be realized by one unit or device through software or hardware. The words first, second, etc. are used to indicate names, not any particular order.

[0144] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the technical solutions of the present application.

Claims

1. A method of predicting the risk of diabetic complications, characterized by, The method comprises: obtaining clinical medical data and intestinal microorganism sequencing data of historical diabetes patients; calculating evolutionary distances between intestinal flora based on the intestinal microorganism sequencing data, and obtaining a plurality of flora abundance matrices according to the evolutionary distances; sorting the plurality of flora abundance matrices and inputting them into a preset complication risk prediction model to obtain a representation vector of the complication risk; vectorizing clinical indicators of the historical diabetes patients to obtain a clinical indicator vector, combining the representation vector and the clinical indicator vector to construct a joint vector, extracting features of the joint vector by convolution to obtain a feature vector, activating and mapping the feature vector to a preset probability interval to obtain a prediction probability of the historical diabetes patients suffering from complications; comparing the prediction probability with a true label of the historical diabetes patients, adjusting model parameters of the complication risk prediction model to obtain an optimized model; predicting a complication risk of a current diabetes patient based on a complication query demand of the current diabetes patient through the optimized model; wherein the extracting features of the joint vector by convolution to obtain a feature vector comprises: extracting preliminary features in the joint vector through a first fully connected layer to obtain a preliminary feature vector; performing abstract feature extraction and mapping on the preliminary feature vector through a second fully connected layer to obtain an abstract feature vector; performing compression and integration on the abstract feature vector through a third fully connected layer to obtain the feature vector.

2. The method for predicting the risk of diabetic complications as described in claim 1, characterized in that, The obtaining of the clinical medical data and the intestinal microorganism sequencing data of the historical diabetes patients comprises: querying identity information of the historical diabetes patients and obtaining case data of the historical diabetes patients based on the identity information; extracting clinical medical data from the case data of the historical diabetes patients and obtaining intestinal microorganism samples of the historical diabetes patients; extracting DNA from the intestinal microorganism samples to obtain intestinal flora DNA; sequencing the intestinal flora DNA by high-throughput sequencing to obtain intestinal microorganism sequencing data.

3. The method for predicting the risk of diabetic complications as described in claim 1, characterized in that, The calculation of the evolutionary distances between the intestinal flora based on the intestinal microorganism sequencing data comprises: obtaining a preset data analysis system and inputting the intestinal microorganism sequencing data into the data analysis system to construct an intestinal flora evolutionary tree; calculating the shortest path between the flora through the intestinal flora evolutionary tree to obtain the evolutionary distances of the intestinal flora.

4. The method for predicting the risk of diabetic complications as described in claim 1, characterized in that, The comparison of the prediction probability with the true label of the historical diabetes patients and the adjustment of the model parameters of the complication risk prediction model to obtain the optimized model comprise: calculating a loss value of the prediction probability and the true label by cross-entropy; performing back propagation on the loss value by minimizing a loss function, adjusting parameters of the complication risk prediction model according to the loss value to obtain an optimized model with adjusted parameters.

5. The method for predicting the risk of diabetic complications as described in claim 1, characterized in that, The obtaining of the plurality of flora abundance matrices according to the evolutionary distances comprises: The evolutionary distance of the intestinal flora is sorted on a pre-constructed two-dimensional plane, an abundance matrix of the intestinal flora is constructed, and a plurality of flora abundance matrices are obtained.

6. A diabetes complication risk prediction device for implementing the diabetes complication risk prediction method according to any one of claims 1 to 5, characterized by, The device comprises: a data acquisition module configured to acquire clinical medical data and intestinal microbial sequencing data of a historical diabetes patient; a probability prediction module configured to calculate evolutionary distances between intestinal floras based on the intestinal microbial sequencing data, and obtain a plurality of flora abundance matrices according to the evolutionary distances; sort the plurality of flora abundance matrices, and input a preset complication risk prediction model to obtain a representation vector of the complication risk; and merge and convolve the representation vector and the clinical indicators of the historical diabetes patient to obtain a prediction probability of the historical diabetes patient suffering from complications; a model optimization module configured to compare the prediction probability with a true label of the historical diabetes patient, adjust model parameters of the complication risk prediction model, and obtain an optimized model; a risk prediction module configured to predict a complication risk of a current diabetes patient based on a complication query demand of the current diabetes patient by using the optimized model.

7. An electronic device, comprising: The electronic device comprises: at least one processor; and a memory connected in communication with the at least one processor; wherein the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to perform the diabetes complication risk prediction method according to any one of claims 1 to 5.

8. A computer readable storage medium storing a computer program, characterized in that, The computer program is executed by the processor to implement the diabetes complication risk prediction method according to any one of claims 1 to 5.

Images