KR102661357B1 - Method and apparatus for analyzing medical data - Google Patents

Abstract

According to an embodiment of the present disclosure, a method for analyzing medical data of a medical institution is provided. The method includes obtaining a first medical dataset of the first medical institution and a second medical dataset of the second medical institution from a first medical institution and a second medical institution; generating a total synthetic dataset from the first medical dataset and the second medical dataset; And based on a target medical dataset of a target medical institution, obtaining an analysis result for the target medical dataset from the integrated synthetic dataset, and the first medical dataset and the second medical dataset. The data set and the integrated synthetic data set may include preset independent variables and preset dependent variables.

Description

Method and device for analyzing medical data {METHOD AND APPARATUS FOR ANALYZING MEDICAL DATA}

This disclosure relates to a method and apparatus for analyzing medical data, and specifically relates to a method and apparatus for analyzing a medical data set of a target medical institution from a synthetic data set.

A patient's medical data is considered personal information and sensitive information of the information subject under the Personal Information Protection Act, and when using a patient's medical data, there is a risk of personal information leakage or personal identification. As a result, medical data is less accessible outside of hospitals, and strict standards are followed for the use and application of medical data. Recently, in the medical field, the number of cases of applying medical synthetic data generated from medical data to research is increasing. Using synthetic medical data, research can be conducted without direct storage or export of medical data, and research results can be obtained without being restricted by accessibility due to security issues such as personal information leakage of existing personal information and sensitive information. In addition, synthetic data generation technology has recently been proposed as a solution to data input errors and data shortage problems, which can significantly reduce the time and economic costs associated with medical data analysis.

Republic of Korea registered patent 10-2403461 (2022.05.25)

This disclosure was developed in response to the above-described background technology, and analyzes medical data as a problem to be solved. For example, the present disclosure aims to solve the problem of obtaining analysis results for the target medical dataset from an integrated synthetic dataset based on the target medical dataset of the target medical institution.

Meanwhile, the technical problem to be achieved by the present disclosure is not limited to the technical problems mentioned above, and may include various technical problems within the scope of what is apparent to those skilled in the art from the contents described below.

The present disclosure was made in response to the above-described background technology, and seeks to provide a method and device for analyzing medical data from a medical institution.

According to one aspect of the present disclosure, a method of analyzing medical data of a medical institution performed by a computing device may be provided. The method includes obtaining a first medical dataset of the first medical institution and a second medical dataset of the second medical institution from a first medical institution and a second medical institution; generating a total synthetic dataset from the first medical dataset and the second medical dataset; And based on a target medical dataset of a target medical institution, obtaining an analysis result for the target medical dataset from the integrated synthetic dataset, and the first medical dataset and the second medical dataset. The data set and the integrated synthetic data set may include preset independent variables and preset dependent variables.

In one embodiment, generating the integrated synthetic data set includes obtaining clinical prior information including information about the correlation between the preset independent variable and the preset dependent variable; And it may include generating the integrated synthetic dataset from the clinical prior information, the first medical dataset, and the second medical dataset.

In one embodiment, the step of generating the integrated synthetic dataset includes generating a first synthetic dataset and a second medical data corresponding to the first medical dataset from the first medical dataset and the second medical dataset. Generating a second synthetic data set corresponding to the set; And it may include generating the integrated synthetic dataset by integrating the first synthetic dataset and the second synthetic dataset.

In one embodiment, the step of generating the first synthetic dataset and the second synthetic dataset is to generate the first medical dataset and the second medical data using a pre-trained GAN (Generative Adversarial Networks) model. It may include generating the first synthetic dataset and the second synthetic dataset from a set.

In one embodiment, the step of generating the first synthetic dataset and the second synthetic dataset is to generate data from the first medical dataset and the second medical dataset using a SMOTE (Synthetic Minority Oversampling Technique) algorithm. It may include generating the first synthetic dataset and the second synthetic dataset.

In one embodiment, the step of generating the first synthetic data set and the second synthetic data set includes obtaining clinical prior information including information about the correlation between the preset independent variable and the preset dependent variable. steps; And using a Bayesian algorithm based on Bayesian inference, the first synthetic data set and the second synthetic data set are created by reflecting the first medical data set and the second medical data set in the clinical prior information. It may include a creation step.

In one embodiment, from the first medical dataset and the second medical dataset, determining a first independent variable that is correlated with the dependent variable and is not included in the preset independent variable; , and the analysis result may include information about the correlation between the first independent variable and the preset dependent variable.

In one embodiment, the integrated synthetic dataset includes a first weight corresponding to the preset independent variable, and the step of obtaining an analysis result for the target medical dataset includes selecting the preset from the target medical dataset. Obtaining a second weight corresponding to the independent variable; Obtaining a third weight corresponding to the preset independent variable based on the first weight and the second weight; And it may include obtaining the analysis result from the integrated synthetic data set based on the preset independent variable, the preset dependent variable, and the third weight.

In one embodiment, the step of acquiring a third weight corresponding to the preset independent variable includes the number of first medical data in the first medical data set, the number of second medical data in the second medical data set, and the target. Determining a weighting degree of each of the first weight and the second weight based on the number of target medical data in the medical data set; and obtaining the third weight based on the weighting degree of each of the first weight and the second weight.

In one embodiment, the step of determining the weighting degree of each of the first weight and the second weight includes: acquiring clinical prior information including clinical statistical data in which clinical values are recorded; And based on the number of clinical medical data in the clinical statistical data, the number of first medical data in the first medical data set, the number of second medical data in the second medical data set, and the number of target medical data in the target medical data set. , may include determining a weighting degree of each of the first weight and the second weight.

In one embodiment, the step of obtaining an analysis result for the target medical dataset includes obtaining the target medical data from the integrated synthetic dataset based on the preset independent variable, the preset dependent variable, and the third weight. Generating a target synthetic dataset corresponding to the set; And it may include obtaining the analysis result from the target synthetic dataset.

In one embodiment, the step of obtaining an analysis result for the target medical data set is correlated with the dependent variable and the preset independent variable from the first medical data set and the second medical data set. determining a first independent variable not included; Generating a target synthetic dataset corresponding to the target medical dataset from the integrated synthetic dataset based on the preset independent variable, the first independent variable, the preset dependent variable, and the third weight; And it may include obtaining the analysis result from the target synthetic dataset.

In one embodiment, the analysis result of the target medical data set includes information about the preset independent variable, the preset dependent variable, the third weight, and the correlation between the preset independent variable and the preset dependent variable. may include.

In one embodiment, the first medical data set and the second medical data set have a data structure of a common data model (CDM) in the first raw medical data set and the second raw medical data set. It may be a converted medical dataset - the first medical dataset corresponds to the first raw medical dataset, and the second medical dataset corresponds to the second raw medical dataset.

In one embodiment, obtaining an analysis result for the target medical dataset includes generating a first integrated synthetic dataset from the first medical dataset and the second medical dataset; generating a second integrated synthetic dataset from the first medical dataset and the second medical dataset; And based on the target medical dataset, obtaining an analysis result for the target medical dataset from the first integrated synthetic dataset and the second integrated synthetic dataset, and the first integrated synthetic data The set includes a 1-1 weight corresponding to the preset independent variable, and the second integrated synthetic data set includes a 1-2 weight corresponding to the preset independent variable, and the 1-1 weight And the 1-2 weights may be different from each other.

In one embodiment, preset dependent variables included in each of the first integrated synthetic data set and the second integrated synthetic data set may have probability distributions based on different means and standard deviations.

In one embodiment, obtaining an analysis result for the target medical dataset includes obtaining a first analysis result for the target medical dataset from the first integrated synthetic dataset; Obtaining a second analysis result for the target medical dataset from the second integrated synthetic dataset; and obtaining an analysis result for the target medical dataset based on the first analysis result and the second analysis result.

In one embodiment, based on at least one of a preset effect size, preset power, and preset significance level, determining the required number of data of the medical dataset to generate the integrated synthetic dataset, further comprising: And each of the first medical data set and the second medical data set may include more medical data than the required number of data.

According to one aspect of the present disclosure, a computing device for analyzing medical data of a medical institution may be provided. The computing device includes at least one processor; and a memory, wherein the at least one processor acquires a first medical dataset of the first medical institution and a second medical dataset of the second medical institution from the first medical institution and the second medical institution, and Generating a total synthetic data set from the data set and the second medical data set, and analyzing the target medical data set from the total synthetic data set based on the target medical data set of the target medical institution. A result is obtained, and the first medical dataset, the second medical dataset, and the integrated synthetic dataset may include a preset independent variable and a preset dependent variable.

According to one aspect of the present disclosure, a computer program stored in a computer-readable storage medium may be provided. When the computer program is executed by one or more processors, it causes the one or more processors to perform operations for analyzing medical data of a medical institution, the operations comprising: receiving data from a first medical institution and a second medical institution; An operation of acquiring a first medical dataset and a second medical dataset of the second medical institution; An operation of generating a total synthetic data set from the first medical data set and the second medical data set; And based on a target medical dataset of a target medical institution, obtaining an analysis result for the target medical dataset from the integrated synthetic dataset, and the first medical dataset and the second medical dataset. The data set and the integrated synthetic data set may include preset independent variables and preset dependent variables.

According to some embodiments of the present disclosure, based on the target medical dataset of the target medical institution, analysis results for the target medical dataset may be obtained from the integrated synthetic dataset.

The effects that can be obtained from the present disclosure are not limited to the effects mentioned above, and other effects not mentioned can be clearly understood by those skilled in the art from the description below. .

Various aspects will now be described with reference to the drawings, where like reference numerals are used to collectively refer to like elements. In the following examples, for purposes of explanation, numerous specific details are set forth to provide a comprehensive understanding of one or more aspects. However, it will be clear that such aspect(s) may be practiced without these specific details. In other instances, well-known structures and devices are shown in block diagram form to facilitate describing one or more aspects.
1 is a block diagram of a computing device that obtains analysis results for a target medical dataset according to an embodiment of the present disclosure.
Figure 2 is a schematic diagram showing a network function according to an embodiment of the present disclosure.
Figure 3 is a flowchart showing a method by which a computing device obtains analysis results for a target medical dataset according to an embodiment of the present disclosure.
Figure 4 is a schematic diagram showing a process by which a computing device obtains analysis results for a target medical dataset according to an embodiment of the present disclosure.
Figure 5 is a schematic diagram showing a process for calculating the weight of a target medical institution according to an embodiment of the present disclosure.
6 is a brief, general schematic diagram of an example computing environment in which embodiments of the present disclosure may be implemented.

Various embodiments are now described with reference to the drawings. In this specification, various descriptions are presented to provide an understanding of the disclosure. However, it is clear that these embodiments may be practiced without these specific descriptions.

As used herein, the terms “component,” “module,” “system,” and the like refer to a computer-related entity, hardware, firmware, software, a combination of software and hardware, or an implementation of software. For example, a component may be, but is not limited to, a process running on a processor, a processor, an object, a thread of execution, a program, and/or a computer. For example, both an application running on a computing device and the computing device can be a component. One or more components may reside within a processor and/or thread of execution. A component may be localized within one computer. A component may be distributed between two or more computers. Additionally, these components can execute from various computer-readable media having various data structures stored thereon. Components may transmit signals, for example, with one or more data packets (e.g., data and/or signals from one component interacting with other components in a local system, a distributed system, to other systems and over a network such as the Internet). Depending on the data being transmitted, they may communicate through local and/or remote processes.

Additionally, the term “or” is intended to mean an inclusive “or” and not an exclusive “or.” That is, unless otherwise specified or clear from context, “X utilizes A or B” is intended to mean one of the natural implicit substitutions. That is, either X uses A; X uses B; Or, if X uses both A and B, “X uses A or B” can apply to either of these cases. Additionally, the terms “or”, “and/or” as used herein should be understood to refer to and include all possible combinations of one or more of the related items listed.

Additionally, the terms “comprise” and/or “comprising” should be understood to mean that the corresponding feature and/or element is present. However, the terms “comprise” and/or “comprising” should be understood as not excluding the presence or addition of one or more other features, elements and/or groups thereof. Additionally, unless otherwise specified or the context is clear to indicate a singular form, the singular terms herein and in the claims should generally be construed to mean “one or more.”

And, the term “at least one of A or B” should be interpreted to mean “a case containing only A,” “a case containing only B,” and “a case of combining A and B.”

Those skilled in the art will additionally recognize that the various illustrative logical blocks, components, modules, circuits, means, logic, and algorithm steps described in connection with the embodiments disclosed herein may be implemented using electronic hardware, computer software, or a combination of both. It must be recognized that it can be implemented with To clearly illustrate the interchangeability of hardware and software, various illustrative components, blocks, configurations, means, logics, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented in hardware or software will depend on the specific application and design constraints imposed on the overall system. A skilled technician can implement the described functionality in a variety of ways for each specific application. However, such implementation decisions should not be construed as causing a departure from the scope of the present disclosure.

The description of the presented embodiments is provided to enable anyone skilled in the art to use or practice the present invention. Various modifications to these embodiments will be apparent to those skilled in the art. The general principles defined herein may be applied to other embodiments without departing from the scope of the disclosure. Therefore, the present invention is not limited to the embodiments presented herein. The present invention is to be interpreted in the broadest scope consistent with the principles and novel features presented herein.

In the present disclosure, terms represented by N, such as first, second, or third, are used to distinguish at least one entity. For example, the entities expressed as first and second may be the same or different from each other.

FIG. 1 is a diagram illustrating a computing device that obtains analysis results for a target medical dataset according to an embodiment of the present disclosure.

The configuration of the computing device 100 shown in FIG. 1 is only a simplified example. In one embodiment of the present disclosure, the computing device 100 may include different components for performing the computing environment of the computing device 100, and only some of the disclosed components may configure the computing device 100.

The computing device 100 according to some embodiments of the present disclosure may be a device for performing analysis on a target medical dataset. For example, the computing device 100 may be a device that generates a total synthetic data set and performs analysis on a target medical data set from the total synthetic data set. Computing device 100 may include any type of server or user terminal.

The computing device 100 may include a processor 110, a memory 130, and a network unit 150.

The processor 110 may consist of one or more cores, and the computing device 100 may include a central processing unit (CPU), a general purpose graphics processing unit (GPGPU), and a tensor processing unit (TPU). : tensor processing unit) may include a processor to perform operations related to data processing.

According to an embodiment of the present disclosure, the processor 110 may perform an operation for learning a neural network. For example, the processor 110 processes input data for learning in deep learning (DL), extracts features from input data, calculates errors, and updates neural network weights using backpropagation. Calculations for learning can be performed. At least one of the CPU, GPGPU, and TPU of the processor 110 may process learning of the network function. For example, CPU and GPGPU can work together to process learning of network functions and data classification using network functions. Additionally, in one embodiment of the present disclosure, processors of a plurality of computing devices may be used together to process learning of a network function and data classification using a network function.

The processor 110 may typically control the overall operation of the computing device 100. The processor 110 provides appropriate information or functions to the user by processing signals, data, information, etc. input or output through components included in the computing device 100 or running an application program stored in the memory 130. Or you can process it.

In one embodiment of the present disclosure, the memory 130 may store any type of information generated or determined by the processor 110 and/or any type of information received by the network unit 150. In one embodiment, memory 130 may store a database. A database may be a set of data stored in a form that the computing device 100 can process. For example, the memory 130 may include a medical data database. A medical data database may store a plurality of medical data.

In one embodiment of the present disclosure, the memory 130 is a flash memory type, hard disk type, multimedia card micro type, or card type memory (e.g. SD or It may include at least one type of storage medium among (Only Memory), magnetic memory, magnetic disk, and/or optical disk. The computing device 100 may operate in connection with web storage that performs a storage function of the memory 130 on the Internet. The description of the memory described above is merely an example, and the present disclosure is not limited thereto. Memory 130 may be operated by processor 110.

The network unit 150 according to an embodiment of the present disclosure may include any wired or wireless communication network capable of transmitting and receiving arbitrary types of data and signals, etc., as expressed in the present disclosure. The techniques described herein can be used in the networks mentioned above, as well as other networks.

Figure 2 is a diagram showing a network function according to an embodiment of the present disclosure.

Throughout this specification, the terms artificial intelligence-based model (e.g., summary information generation model, information classification model, etc.), computational model, neural network, network function, and neural network have the same meaning (interchangeable meaning). can be used

A neural network can generally consist of a set of interconnected computational units, which can be referred to as nodes. These nodes may also be referred to as neurons. A neural network consists of at least one node. Nodes (or neurons) that make up neural networks may be interconnected by one or more links.

Within a neural network, one or more nodes connected through a link may form a relative input node and output node relationship. The concepts of input node and output node are relative, and any node in an output node relationship with one node may be in an input node relationship with another node, and vice versa. As described above, input node to output node relationships can be created around links. One or more output nodes can be connected to one input node through a link, and vice versa.

In a relationship between an input node and an output node connected through one link, the value of the data of the output node may be determined based on the data input to the input node. Here, the link connecting the input node and the output node may have a weight. Weights may be variable and may be varied by the user or algorithm in order for the neural network to perform the desired function. For example, when one or more input nodes are connected to one output node by respective links, the output node is set to the values input to the input nodes connected to the output node and the links corresponding to each input node. The output node value can be determined based on the weight.

As described above, in a neural network, one or more nodes are interconnected through one or more links to form an input node and output node relationship within the neural network. The characteristics of the neural network may be determined according to the number of nodes and links within the neural network, the correlation between the nodes and links, and the value of the weight assigned to each link. For example, if the same number of nodes and links exist and two neural networks with different weight values of the links exist, the two neural networks may be recognized as different from each other.

A neural network may consist of a set of one or more nodes. A subset of nodes that make up a neural network can form a layer. Some of the nodes constituting the neural network may form one layer based on the distances from the first input node. For example, a set of nodes with a distance n from the initial input node may constitute n layers. The distance from the initial input node can be defined by the minimum number of links that must be passed to reach the node from the initial input node. However, this definition of a layer is arbitrary for explanation purposes, and the order of a layer within a neural network may be defined in a different way than described above. For example, a layer of nodes may be defined by distance from the final output node.

In one embodiment of the present disclosure, a set of neurons or nodes may be defined by the expression layer.

The initial input node may refer to one or more nodes in the neural network through which data is directly input without going through links in relationships with other nodes. Alternatively, in the relationship between nodes based on links within a neural network, it may refer to nodes that do not have other input nodes connected by links. Similarly, the final output node may refer to one or more nodes that do not have an output node in their relationship with other nodes among the nodes in the neural network. Additionally, hidden nodes may refer to nodes constituting a neural network other than the first input node and the last output node.

In a neural network according to an embodiment of the present disclosure, the number of nodes in the input layer may be the same as the number of nodes in the output layer, and as it progresses from the input layer to the hidden layer, the number of nodes decreases and then increases again. It could be an incremental neural network. In addition, the neural network according to another embodiment of the present disclosure may be a neural network in which the number of nodes in the input layer may be less than the number of nodes in the output layer, and the number of nodes decreases as it progresses from the input layer to the hidden layer. there is. In addition, the neural network according to another embodiment of the present disclosure may be a neural network in which the number of nodes in the input layer may be greater than the number of nodes in the output layer, and the number of nodes increases as it progresses from the input layer to the hidden layer. You can. A neural network according to another embodiment of the present disclosure may be a neural network that is a combination of the above-described neural networks.

A deep neural network (DNN) may refer to a neural network that includes multiple hidden layers in addition to the input layer and output layer. Deep neural networks allow you to identify latent structures in data. Deep neural networks include convolutional neural networks (CNNs), recurrent neural networks (RNNs), auto encoders, generative adversarial networks (GANs), and restricted Boltzmann machines ( It may include Restricted Boltzmann Machine (RBM), deep belief network (DBN), Q network, U network, Siamese network, etc. The description of the deep neural network described above is merely an example, and the present disclosure is not limited thereto.

Illustratively, the artificial intelligence-based model of the present disclosure may include a transformer, a generative pre-training transformer (GPT), and BERT (Bidirectional Encoder Representations from Transformers).

The artificial intelligence-based model of the present disclosure can be expressed by a network structure of any of the structures described above, including an input layer, a hidden layer, and an output layer.

Neural networks that can be used in the artificial intelligence-based model of the present disclosure include supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning, It can be learned using at least one of federated learning and incremental learning for distributed deep learning. Learning of a neural network may be a process of applying knowledge for the neural network to perform a specific operation to the neural network.

Neural networks can be trained to minimize output errors. In neural network learning, learning data is repeatedly input into the neural network, the output of the neural network and the error of the target for the learning data are calculated, and the error of the neural network is transferred from the output layer of the neural network to the input layer in the direction of reducing the error. This is the process of updating the weight of each node in the neural network through backpropagation. In the case of supervised learning, learning data in which the correct answer is labeled in each learning data is used (i.e., labeled learning data), and in the case of unsupervised learning, the correct answer may not be labeled in each learning data. That is, for example, in the case of supervised learning on data classification, the learning data may be data in which each training data is labeled with a category. Labeled training data is input to the neural network, and the error can be calculated by comparing the output (category) of the neural network with the label of the training data.

As another example, in the case of unsupervised learning on data classification, the error can be calculated by comparing the input training data with the neural network output. The calculated error is back-propagated in the neural network in the reverse direction (i.e., from the output layer to the input layer), and the connection weight of each node in each layer of the neural network may be updated according to back-propagation. The amount of change in the connection weight of each updated node may be determined according to the learning rate. The neural network's calculation of input data and backpropagation of errors can constitute a learning cycle (epoch). The learning rate may be applied differently depending on the number of repetitions of the learning cycle of the neural network. For example, in the early stages of neural network training, a high learning rate can be used to increase efficiency by allowing the neural network to quickly achieve a certain level of performance, and in the later stages of training, a low learning rate can be used to increase accuracy.

In the learning of neural networks, the training data can generally be a subset of real data (i.e., the data to be processed using the learned neural network), and thus the error for the training data is reduced, but the error for the real data is reduced. There may be an incremental learning cycle. Overfitting is a phenomenon in which errors in actual data increase due to excessive learning on training data. For example, a phenomenon in which a neural network that learned a cat by showing a yellow cat fails to recognize that it is a cat when it sees a non-yellow cat may be a type of overfitting. Overfitting can cause errors in machine learning algorithms to increase. To prevent such overfitting, various optimization methods can be used. To prevent overfitting, methods such as increasing the learning data, regularization, dropout to disable some of the network nodes during the learning process, and use of a batch normalization layer can be applied. You can.

In one embodiment, computing device 100 may utilize a pre-trained Generative Adversarial Networks (GAN) model. A pre-trained GAN model includes a generator neural network and a discriminator neural network, and the generator neural network generates synthetic data from real data, and the discriminator neural network is pre-trained to discriminate between real data and synthetic data. It can correspond to an artificial intelligence model. The generator neural network can be pre-trained through the above learning process to make it difficult for the discriminator neural network to distinguish between real data and generated synthetic data. The computing device 100 may use a pre-trained GAN model to generate a synthetic dataset having the probability distribution and characteristics of the actual medical dataset from the actual medical dataset.

Hereinafter, according to an embodiment of the present disclosure, the computing device 100 generates an integrated synthetic dataset from medical datasets of each of a plurality of medical institutions, and analyzes the target medical dataset of the target medical institution from the integrated synthetic dataset. A method for obtaining a result is disclosed.

Figure 3 is a flowchart showing a method by which a computing device obtains analysis results for a target medical dataset according to an embodiment of the present disclosure.

In step S301, the computing device 100 may acquire a first medical dataset of the first medical institution and a second medical dataset of the second medical institution from the first medical institution and the second medical institution. The computing device 100 may acquire medical data sets from each of a plurality of medical institutions. Hereinafter, for convenience of explanation, acquisition of medical data sets from a first medical institution and a second medical institution will be explained as an example. However, the method is not limited to this, and medical data sets can be acquired from each of a plurality of medical institutions, and an integrated synthetic data set can be generated from the plurality of medical data sets.

In one embodiment, a medical dataset may refer to a set of medical data. Medical data may include information related to the medical care of a specific person. For example, medical data may include variables related to at least one disease in the medical field. For example, medical data is data that includes a specific person's physical information, health information, and medical information, such as a specific person's CT image data, MRI test result data, electrocardiogram measurement result data, and electronic medical record (Electronic Medical Record, EMR) data, etc. may be included. Health information may include information about an individual's disease, such as the presence or absence of the disease, the name of the disease, and the staging of the disease. Medical information may include information about personal medical records, such as the presence or absence of a specific person's prescribed medication, the name of the prescribed medication, the period of taking the prescribed medication, the dosage of the prescribed medication, whether surgery was performed, the name of the surgery, and the details of the surgery. .

In one embodiment, medical data may include variables for a specific person's physical information, health information, medical information, etc. Additionally, medical data may include disease-related variables such as lifestyle habits (drinking, smoking, etc.), family history, age, gender, cholesterol level, and gene-related variables. The computing device 100 can set research subjects and purposes. In addition, independent variables and dependent variables among the variables included in the medical data can be set depending on the research object and purpose of the computing device 100. In one embodiment, the independent variable may refer to a causal variable, and the dependent variable may refer to a variable whose value is determined according to the independent variable. Each of the independent and dependent variables may include at least one variable. The first medical data set of the first medical institution and the second medical data set of the second medical institution may include preset independent variables and preset dependent variables.

For example, if the purpose of the study is to determine whether side effects occur with pioglitazone in patients with non-alcoholic fatty liver disease, the computing device 100 sets the research subject as a patient who took pioglitazone among patients with non-alcoholic fatty liver disease and synthesizes the To create the dataset, a medical dataset containing medical data of patients taking pioglitazone among non-alcoholic fatty liver patients can be obtained from the first and second medical institutions. In this case, the computing device 100 may set the presence or absence of non-alcoholic fatty liver disease, the stage of non-alcoholic fatty liver disease, the development period of non-alcoholic fatty liver disease, whether or not pioglitazone is being taken, the pioglitazone dose, and the pioglitazone taking period as independent variables. The computing device 100 may set variables related to side effects such as weight gain and edema (eg, body weight, presence of edema, etc.) as dependent variables.

In one embodiment, the first medical dataset and the second medical dataset are converted from the first raw medical dataset and the second raw medical dataset to the data structure of the Common Data Model (CDM). It could be a medical dataset. At this time, the first medical dataset may correspond to the first raw medical dataset, and the second medical dataset may correspond to the second raw medical dataset. In one embodiment, a raw medical dataset may be an unstructured medical dataset with a different data structure held by each of a plurality of medical institutions. In one embodiment, the common data model may mean a data model that sets the same data structure and standards to medical datasets with different data structures held by each of a plurality of medical institutions. For example, common data models may include OMOP-CDM, Sentinel-CDM, PCORnet CDM, etc. Variables included in raw medical data can be converted or classified into standardized data formats, labels, classes, etc., and converted into medical data with a data structure of a common data model. Through this, medical data sets from multiple medical institutions can be processed in batches.

In one embodiment, the computing device 100 may convert the first raw medical dataset and the second raw medical dataset into a first medical dataset and a second medical dataset having a data structure of a common data model. In another embodiment, the computing device 100 may acquire a first medical dataset and a second medical dataset having a data structure of a common data model from the first medical institution and the second medical institution.

In one embodiment, the computing device 100 may determine the required number of data in the medical dataset to generate an integrated synthetic dataset based on at least one of a preset effect size, preset power, and preset significance level. . Each of the first medical data set and the second medical data set may include more than the required number of medical data. For example, the required number of data may refer to the number of data required for a sample group to obtain meaningful results for the population. In one embodiment, the effect size may be the difference between the average values of the first and second medical datasets being studied divided by the standard deviation of the first medical dataset or the standard deviation of the second medical dataset. In one embodiment, power may be a value that represents a sufficient degree of probability to conclude that it is statistically significant. In one embodiment, the significance level is 1 minus the reliability. For example, if the reliability is 95%, the significance level may be 0.05, which is 1-0.95. The computing device 100 may set the effect size, power, and significance level, and determine the number of data required for each of the first and second medical datasets based on the set effect size, power, and significance level.

In one embodiment, the computing device 100 is a medical data set among a plurality of medical institutions that includes medical data including preset independent variables and preset dependent variables according to the research object and purpose. For , the corresponding medical dataset may not be used to create an integrated synthetic dataset.

In one embodiment, the computing device 100 selects the 1-1 medical data that does not include at least one of the preset dependent variable and the preset independent variable among the first medical data included in the first medical data set as the first medical data. It can be deleted from the dataset. Likewise, the computing device 100 selects the 2-1 medical data that does not include at least one of the preset dependent variable and the preset independent variable among the second medical data included in the second medical data set. It can be deleted.

In one embodiment, the computing device 100 includes 1-2 medical data including values that deviate from a preset threshold from clinical values included in clinical prior information among the first medical data included in the first medical data set. Can be deleted from the first medical dataset. Likewise, the computing device 100 selects the 2-2 medical data including a value that deviates from a preset threshold from the clinical value included in the clinical preliminary information among the second medical data included in the second medical data set as the second medical data. It can be deleted from the medical dataset. For example, medical data containing a value outside a preset threshold may be medical data containing an input error value.

In step S302, the computing device 100 may generate an integrated synthetic dataset from the first medical dataset and the second medical dataset. The integrated synthetic data set may include preset independent variables and preset dependent variables. In one embodiment, the integrated synthetic dataset may be a synthetic dataset created to follow a probability distribution of a preset dependent variable according to a preset independent variable calculated from the first medical dataset and the second medical dataset. For example, when a preset dependent variable includes a plurality of dependent variables, an integrated synthetic data set may be created to follow the probability distribution of each of the plurality of dependent variables according to the preset independent variable. In one embodiment, the computing device 100 may transmit the integrated synthetic dataset to an external device or external server. Through this, the technical effect can be achieved in that the research institution or researcher who requested the analysis can be provided with a synthetic dataset based on the actual medical dataset.

In one embodiment, the computing device 100 may acquire clinical prior information including information about the correlation between a preset independent variable and a preset dependent variable. In one embodiment, clinical prior information may include prior clinical studies, clinical statistical data, etc. for pre-established research subjects and purposes. For example, clinical preliminary information includes information on the study period, disease (or drug) code information for the target disease for the purpose of the study, information on the prevalence and incidence of the target disease by year/medical institution, pre-set independent variables, and It may include statistical data (e.g., statistical quantity, statistical method, statistical data, statistical range, statistical unit, etc.) for the preset dependent variable. In one embodiment, clinical preliminary information may include clinical statistics in which clinical values are recorded. For example, clinical statistical data may include information about the number of clinical medical data.

In one embodiment, information about the correlation between a preset independent variable and a preset dependent variable may be information about a numerical change in a preset dependent variable according to a preset independent variable. For example, a preset independent variable may include a plurality of independent variables, and a corresponding weight may be assigned to each of the plurality of independent variables. The weight corresponding to each of the plurality of independent variables may reflect the correlation between the preset independent variable and the preset dependent variable. For example, when a preset independent variable and a preset dependent variable have a linear proportional relationship, the weight corresponding to each of the plurality of independent variables may be a proportionality constant. Based on the weights for the preset independent variables, the probability distribution of the preset dependent variable according to the preset independent variable may be obtained.

In one embodiment, the computing device 100 may generate an integrated synthetic dataset from the acquired clinical prior information, the first medical dataset, and the second medical dataset. For example, the computing device 100 calculates the probability distribution of a preset dependent variable according to a preset independent variable in clinical statistical data included in clinical preliminary information and the probability distribution of the preset dependent variable in the first and second medical datasets. You can create an integrated synthetic data set that follows the probability distribution of the preset dependent variable according to the set independent variable. Through this, it is possible to achieve the technical effect of generating a synthetic dataset that follows a probability distribution based on clinical values using clinical prior information. In addition, it is possible to achieve the technical effect of generating a synthetic dataset that follows a probability distribution based on actual data using medical data from actual medical institutions rather than simply based on clinical values.

In one embodiment, when the preset independent variable and the preset dependent variable are numeric variables having numerical values, the computing device 100 determines each numerical interval of the preset independent variable from the first medical data set and the second medical data set. The probability distribution of preset dependent variable values can be calculated. Additionally, the computing device 100 may calculate a probability distribution of a preset dependent variable value for each numerical interval of a preset independent variable from clinical prior information, a first medical data set, and a second medical data set. The computing device 100 may generate an integrated synthetic data set that follows the probability distribution of the preset dependent variable values for each numerical interval of the calculated preset independent variables. In this case, the integrated synthetic data set may have a weight corresponding to each numerical interval of the preset independent variable according to the probability distribution of the preset dependent variable value. For example, the weight may correspond to the average value of the preset dependent variable according to the numerical interval of the preset independent variable.

For example, if the research purpose is to study the correlation between obesity and diabetes, the independent variable can be set as the BMI (Body Mass Index) level and the dependent variable can be set as the blood sugar level. The computing device 100 may calculate a probability distribution of blood sugar levels for each BMI index value section from the first and second medical data sets in which each medical data includes the BMI index and blood sugar level. The computing device 100 may generate an integrated synthetic dataset having a calculated probability distribution.

In one embodiment, when the preset independent variable is a numeric variable and the preset dependent variable is a logical variable with a value of 1 (True) or 0 (False) depending on whether a disease (or side effect) occurs, etc., computing The device 100 may calculate a probability distribution for the probability of occurrence of a preset dependent variable according to the value of a preset independent variable from the first medical data set and the second medical data set. Additionally, the computing device 100 may calculate a probability distribution for the probability of occurrence of a preset dependent variable according to the value of the preset independent variable from clinical prior information, the first medical data set, and the second medical data set. For example, if the preset dependent variable is 1 (True), a disease (or side effect) corresponding to the preset dependent variable occurs, and if the preset dependent variable is 0 (False), the preset dependent variable This may mean that the corresponding disease (or side effect) has not occurred. The computing device 100 may generate an integrated synthetic data set that follows a probability distribution for the probability of occurrence of a preset dependent variable according to the calculated values of the preset independent variable. In this case, the integrated synthetic data set may have weights corresponding to the preset independent variables according to the probability distribution for the probability of occurrence of the preset dependent variable. For example, the weight may correspond to the probability of occurrence of a preset dependent variable according to the preset value of the independent variable.

For example, if the research purpose is to study the correlation between obesity and diabetes, the independent variable can be set as the BMI (Body Mass Index) value and the dependent variable can be set as the presence or absence of diabetes. The computing device 100 may calculate a probability distribution for the probability of developing diabetes according to the BMI index from the first and second medical datasets in which each medical data includes the BMI index and the presence or absence of diabetes. The computing device 100 may generate an integrated synthetic dataset having a calculated probability distribution.

In one embodiment, when the preset independent variable is a logical variable with a value of 1 (True) or 0 (False) depending on whether a disease occurs or the presence or absence of a prescription drug, and the preset dependent variable is a numeric variable, The computing device 100 may calculate a probability distribution of a preset dependent variable value according to a preset independent variable with a value of 1 (True) from the first medical data set and the second medical data set. In addition, the computing device 100 calculates the probability distribution of a preset dependent variable value according to a preset independent variable with a value of 1 (True) from clinical prior information, the first medical data set, and the second medical data set. You can. The computing device 100 may generate an integrated synthetic data set that follows a probability distribution of preset dependent variable values according to the calculated preset independent variables. In this case, the integrated synthetic data set may have weights corresponding to the preset independent variables according to the probability distribution of the preset dependent variable values. For example, the weight may correspond to the average value of a preset dependent variable according to a preset independent variable.

For example, when studying the side effects of pioglitazone on patients with non-alcoholic fatty liver disease, the independent variables can be set as the presence or absence of non-alcoholic fatty liver disease and the use of pioglitazone, and the dependent variable can be set as the amount of weight gain. The computing device 100 calculates a probability distribution of the amount of weight gain in non-alcoholic fatty liver patients due to taking pioglitazone from the first medical dataset and the second medical dataset including medical data on non-alcoholic fatty liver patients taking pioglitazone. can do. The computing device 100 may generate an integrated synthetic dataset having a calculated probability distribution.

In one embodiment, when the preset independent variable and the preset dependent variable are logical variables, the computing device 100 determines the preset dependent variable according to the preset independent variable from the first medical dataset and the second medical dataset. The probability of occurrence can be calculated. Additionally, the computing device 100 may calculate the probability of occurrence of a preset dependent variable according to a preset independent variable from clinical prior information, the first medical data set, and the second medical data set. The computing device 100 may generate an integrated synthetic data set having a probability of occurrence of a preset dependent variable according to the calculated preset independent variable. In this case, the integrated synthetic data set may have weights corresponding to the preset independent variables according to the probability of occurrence of the preset dependent variable. For example, the weight may correspond to the probability of occurrence of a preset dependent variable according to a preset independent variable.

For example, if the research purpose is to study the correlation between obesity and diabetes, the independent variable can be set as obesity and the dependent variable can be set as the presence or absence of diabetes. The computing device 100 may calculate a probability distribution for the probability of developing diabetes according to obesity from the first and second medical datasets, which include obesity and diabetes in each medical data. The computing device 100 may generate an integrated synthetic dataset having a calculated probability distribution.

In one embodiment, each of the preset independent variable and the preset dependent variable may include at least one of a numeric variable and a logical variable. For example, the computing device 100 may calculate a weight corresponding to each of a plurality of independent variables according to a method for calculating the probability distribution (or probability of occurrence) of a preset dependent variable according to the preset independent variable. In one embodiment, the average value of each of the plurality of dependent variables may be calculated as a weighted average obtained by multiplying each of the plurality of independent variables by a plurality of weights corresponding to each of the plurality of independent variables. For example, W1,… ,Wn is the weight corresponding to each independent variable, When Xn is a plurality of independent variables, the average value of each dependent variable can be obtained as (W1*X1 + W2*X2 +…+ Wn*Xn)/(W1 + W2 +…+ Wn).

In one embodiment, the computing device 100 generates a first synthetic dataset corresponding to the first medical dataset and a second synthetic dataset corresponding to the second medical dataset from the first medical dataset and the second medical dataset. can be created. A detailed description of how the computing device 100 generates a synthetic data set from each of a plurality of medical institutions and an integrated synthetic data set from the synthetic data set corresponding to each of the plurality of medical institutions will be described later with reference to FIG. 4. Let's do it.

In step S303, the computing device 100 may obtain an analysis result for the target medical dataset from the integrated synthetic dataset based on the target medical dataset of the target medical institution. In one embodiment, the target medical institution may be a medical institution for which analysis results are to be obtained. For example, the computing device 100 may obtain information about the target medical institution from a research institution or researcher trying to obtain analysis results about the target medical institution. The computing device 100 may obtain information about the target medical institution from an external device or external server. In one embodiment, the target medical institution may be one of a first medical institution and a second medical institution. In another embodiment, the target medical institution may be a medical institution that is not included in the first medical institution and the second medical institution. The analysis results for the target medical dataset may include information about the correlation between preset independent variables and preset dependent variables in the target medical dataset. In one embodiment, the computing device 100 may transmit analysis results for the target medical dataset to an external device or external server. Through this, the technical effect can be achieved in that the research institution or researcher who requested the analysis can receive analysis results that reflect the target medical dataset in the integrated synthetic dataset.

For example, the computing device 100 obtains analysis results for the target medical dataset by reflecting the probability distribution of the preset dependent variable in the target medical dataset to the probability distribution of the preset dependent variable in the integrated synthetic dataset. can do. For example, the computing device 100 may apply target medical information to a first weight corresponding to a preset independent variable in an integrated synthetic data set generated based on clinical prior information, a first medical data set, and a second medical data set. Analysis results for the target medical dataset can be obtained by reflecting the second weight corresponding to the preset independent variable in the dataset. Through this, it is possible to obtain analysis results that reflect the probability distribution of the target medical dataset and are not over-fitted to the target medical dataset.

In one embodiment, the computing device 100 may determine, from the first medical dataset and the second medical dataset, a first independent variable that is correlated with a preset dependent variable and is not included in the preset independent variable. . The computing device 100 may calculate a relational expression for each of the plurality of variables included in the first medical data set and the second medical data set, and determine the variable that is correlated with the preset dependent variable as the first independent variable. there is. At this time, the analysis result of the target medical dataset may include information about the correlation between the first independent variable and the preset dependent variable. Through this, a technical effect can be achieved that can identify independent variables that the research institution or researcher has not considered among a plurality of variables that are correlated with a preset dependent variable.

In one embodiment, the computing device 100 may obtain weights to be reflected in analysis results for the target medical dataset from clinical prior information, an integrated synthetic dataset, and the target medical dataset. A detailed description of the method for obtaining weights to be reflected in the analysis results of the target medical dataset will be described later with reference to FIG. 5.

In one embodiment, the computing device 100 may generate a plurality of integrated synthetic datasets from the first medical dataset and the second medical dataset. Hereinafter, for convenience of explanation, a method by which the computing device 100 generates a first integrated synthetic dataset and a second integrated synthetic dataset from the first medical dataset and the second medical dataset will be described as an example. .

In one embodiment, the computing device 100 may generate a first integrated synthetic dataset from the first medical dataset and the second medical dataset. Additionally, the computing device 100 may generate a second integrated synthetic dataset from the first medical dataset and the second medical dataset. The computing device 100 may obtain analysis results for the target medical dataset from the first integrated synthetic dataset and the second integrated synthetic dataset, based on the target medical dataset. The first integrated synthetic data set may include a 1-1 weight corresponding to a preset independent variable, and the second integrated synthetic data set may include a 1-2 weight corresponding to a preset independent variable. At this time, the 1-1 weight and the 1-2 weight may be different from each other. For example, when creating an integrated synthetic data set from the first medical data set and the second medical data set, the values of the preset independent variable and the preset dependent variable included in each synthetic data generated are within the set probability distribution. may be different from each other. Accordingly, each time an integrated synthetic dataset is created from the first medical dataset and the second medical dataset, the value of the synthetic data included in the integrated synthetic dataset and the first weight of the integrated synthetic dataset may be different. .

In one embodiment, preset dependent variables included in each of the first integrated synthetic data set and the second integrated synthetic data set may have probability distributions based on different means and standard deviations. For example, when the 1-1 weight and the 1-2 weight corresponding to each of the first integrated synthetic data set and the second integrated synthetic data set are different from each other, the first integrated synthetic data set and the second integrated synthetic data set The probability distribution of each of the three may be different.

In one embodiment, the computing device 100 may obtain a first analysis result for the target medical dataset from the first integrated synthetic dataset. Additionally, the computing device 100 may obtain a second analysis result for the target medical dataset from the second integrated synthetic dataset. The first analysis result and the second analysis result for the target medical dataset may have different third weights and probability distributions for the preset dependent variable. That is, the first analysis result and the second analysis result for the target medical dataset may be different analysis results. The computing device 100 may obtain an analysis result for the target medical dataset based on the first analysis result and the second analysis result. For example, the analysis result for the target medical dataset may be an analysis result including a first analysis result and a second analysis result. In one embodiment, the computing device 100 may transmit the analysis result of the target medical dataset including the first analysis result and the second analysis result to an external device or an external server. Through this, multiple synthetic datasets are created from the first and second medical datasets, multiple simulations are performed on the target medical dataset based on the multiple synthetic datasets, and the research institution that requested the analysis Alternatively, the technical effect of providing researchers with highly reliable analysis results can be achieved.

Figure 4 is a schematic diagram showing a process by which a computing device obtains analysis results for a target medical dataset according to an embodiment of the present disclosure.

Referring to FIG. 4, the computing device 100 according to an embodiment receives first medical data set 401a, second medical data set 401b, and third medical data from the first medical institution, the second medical institution, and the third medical institution. A dataset 401c can be obtained. The computing device 100 may generate a first synthetic dataset 402a from the first medical dataset 401a. The computing device 100 may generate a second synthetic dataset 402b from the second medical dataset 401b. The computing device 100 may generate a third synthetic dataset 402c from the third medical dataset 401c. The computing device 100 may generate an integrated synthetic dataset 403 by integrating the first synthetic dataset 402a, the second synthetic dataset 402b, and the third synthetic dataset 402c.

In one embodiment, the computing device 100 obtains an analysis result 405 for the target medical dataset 404 from the integrated synthetic dataset 403 based on the target medical dataset 404 of the target medical institution. You can. For example, the computing device 100 reflects the probability distribution of the preset dependent variable in the target medical dataset 404 to the probability distribution of the preset dependent variable in the integrated synthetic dataset 403, thereby generating target medical data. The analysis result 405 for the set 404 can be obtained. Hereinafter, for convenience of explanation, a method by which the computing device 100 generates the integrated synthetic data set 403 from the first medical data set 401a and the second medical data set 401b will be described as an example. . However, it is not limited to this, and the computing device 100 generates an integrated synthetic dataset 403 from the first medical dataset 401a, the second medical dataset 401b, and the third medical dataset 401c in the same manner. can be created.

In one embodiment, the computing device 100 generates a first synthetic dataset 402a and a first synthetic dataset corresponding to the first medical dataset 401a from the first medical dataset 401a and the second medical dataset 401b. 2 A second synthetic data set (402b) corresponding to the medical data set (401b) can be created. The computing device 100 may generate an integrated synthetic dataset 403 by integrating the first synthetic dataset 402a and the second synthetic dataset 402b.

In one embodiment, the computing device 100 uses a pre-trained GAN model to generate a first synthetic dataset 402a and a second synthesis from the first medical dataset 401a and the second medical dataset 401b. A dataset 402b can be created. For example, the computing device 100 may use a pre-trained GAN model to generate a first synthetic dataset 402a having a probability distribution of a preset dependent variable in the first medical dataset 401a. there is. Additionally, the computing device 100 may use a pre-trained GAN model to generate a second synthetic dataset 402b having a probability distribution of a preset dependent variable in the second medical dataset 401b.

In one embodiment, the computing device 100 may generate a clinical medical data set that follows a probability distribution of a preset dependent variable in clinical prior information. The computing device 100 may generate a first synthetic dataset 402a from the first medical dataset 401a and the clinical medical dataset using a pre-trained GAN model. The computing device 100 may determine the number of clinical medical datasets to be created based on the preset first reflection ratio between the clinical prior information and the first medical dataset 401a. For example, the number of clinical medical datasets to be generated may be determined based on the number of first medical datasets 401a and a preset first reflection ratio. The computing device 100 generates a second synthetic dataset 402b and a third synthetic dataset (402b) from the second medical dataset 401b and the third medical dataset 401c using a GAN model pre-trained in the same manner. 402c) can be created.

In one embodiment, the computing device 100 uses the SMOTE (Synthetic Minority Oversampling Technique) algorithm to generate a first synthetic dataset 402a and a first synthetic dataset 402a from the first medical dataset 401a and the second medical dataset 401b. A second synthetic data set 402b can be created. In one embodiment, the SMOTE algorithm calculates K nearest neighbor data belonging to the same class using the K-NN (K-Nearest Neighbor) algorithm for data sampled from a dataset, and combines the sampled data with K nearest neighbors. It may refer to an algorithm that generates data corresponding to an arbitrary point located on a line segment connected to each of neighboring data as synthetic data.

For example, the computing device 100 operates on a line segment connecting each of the first medical data included in the first medical data set 401a and the nearest neighboring medical data on a plane or space with a plurality of independent variables as axes. You can select any point located in . The computing device 100 may generate first synthetic data having numerical values of each of a plurality of independent variables corresponding to the selected point. At this time, the dependent variable of the first synthetic data may have the same class as the dependent variable class of the first medical data and the nearest neighbor medical data. For example, the dependent variable class may be a class that indicates whether a disease (or side effect) occurs. The computing device 100 may generate a first synthetic data set 402a from a plurality of first synthetic data. The computing device 100 generates the second synthetic dataset 402b and the third synthetic dataset 402c from the second medical dataset 401b and the third medical dataset 401c using the SMOTE algorithm in the same manner. can be created.

In one embodiment, the computing device 100 uses a Bayesian algorithm based on Bayesian inference to reflect the first medical data set 401a and the second medical data set 401b in the clinical prior information. 1 A synthetic dataset 402a and a second medical dataset 402b can be created. For example, the computing device 100 uses a Bayesian algorithm based on Bayesian inference to reflect the first medical data set 401a in the clinical prior information to generate the first synthetic data set 402a. You can. Likewise, the computing device 100 may generate the second medical data set 402b by reflecting the second medical data set 401b in the clinical prior information using a Bayesian algorithm based on Bayesian inference. . In one embodiment, the Bayesian algorithm may refer to an algorithm that infers a posterior probability distribution by applying medical data to a prior probability distribution.

For example, the computing device 100 may obtain a prior probability distribution of a preset dependent variable according to a preset independent variable, based on clinical prior information. The computing device 100 sets the probability distribution to beta distribution, gamma distribution, normal distribution, etc. for the first medical data included in the first medical data set 401a, obtains the likelihood, and applies it to the prior probability distribution. can do. The computing device 100 may obtain a posterior probability distribution of a preset dependent variable according to a preset independent variable by applying the first medical dataset 401a to the prior probability distribution. The computing device 100 may generate a first synthetic data set 402a based on the mean value and standard deviation of the obtained posterior probability distribution. The computing device 100 may generate a first synthetic data set 402a that follows a probability distribution based on a mean value and standard deviation that are the same as or similar to the mean value and standard deviation of the obtained posterior probability distribution. The computing device 100 uses a Bayesian algorithm based on Bayesian inference in the same manner to obtain the second synthetic data set 402b and the third medical data set 401c from the second medical data set 401b and the third medical data set 401c. A synthetic data set 402c can be created.

For example, if the research purpose is to study the blood pressure control effect of furosemide on patients with high blood pressure, the computing device 100 may use independent variables such as presence of high blood pressure, whether furosemide is taken, period of taking furosemide, dosage, etc. and the dependent variable can be set to systolic blood pressure (SBP). The computing device 100 may obtain a prior probability distribution of systolic blood pressure according to taking furosemide based on prior clinical studies or clinical statistical data included in clinical prior information. The computing device 100 may obtain a posterior probability distribution of systolic blood pressure according to taking furosemide by applying the first medical dataset and the second medical dataset to the prior probability distribution. The computing device 100 may generate a first synthetic data set 402a that follows a probability distribution based on the mean value and standard deviation of the obtained posterior probability distribution.

Figure 5 is a schematic diagram showing a process for calculating the weight of a target medical institution according to an embodiment of the present disclosure.

In one embodiment, the computing device 100 provides analysis results for the target medical dataset 503 based on the clinical prior information 501, the integrated synthetic dataset 502, and the target medical dataset 503. A third weight 504 corresponding to the preset independent variable to be applied can be obtained. In one embodiment, when the clinical prior information 501 is reflected in generating the integrated synthetic dataset 502, the computing device 100 generates the integrated synthetic dataset 502 and the target medical dataset 503 of the target medical institution. Based on this, a third weight 504 corresponding to a preset independent variable to be applied to the analysis result for the target medical dataset 503 can be obtained.

In one embodiment, the integrated synthetic dataset 502 may include a first weight corresponding to a preset independent variable. The computing device 100 may obtain a second weight corresponding to a preset independent variable from the target medical dataset 503. The computing device 100 may obtain a third weight 504 corresponding to a preset independent variable based on the first weight and the second weight. The computing device 100 may obtain an analysis result for the target medical dataset 503 from the integrated synthetic dataset 502 based on the preset independent variable, the preset dependent variable, and the third weight 504. .

In one embodiment, the computing device 100 displays statistics, ranges, and units of preset independent variables (and preset dependent variables) in the clinical prior information 501, the number of data, and probability of the target medical dataset 503. The third weight ( 504) can be obtained. For example, the computing device 100 may determine the third weight 504 to be larger as the likelihood ratio of the target medical data set 503 relative to the clinical prior information 501 increases, the number of data increases, and the variance decreases. . That is, the clinical prior information can serve as a global parameter for calculating the third weight 504, and the target medical dataset can serve as a local parameter for calculating the third weight 504.

In one embodiment, the computing device 100 based on the number of first medical data in the first medical dataset, the number of second medical data in the second medical dataset, and the number of target medical data in the target medical dataset 503. , the weighting degree of each of the first weight and the second weight can be determined. For example, the weighting degree of each of the first weight and the second weight may mean a reflection ratio to be reflected in the third weight 504. Based on the number of synthetic data in the integrated synthetic data set 502 and the number of target medical data in the target medical data set 503, the computing device 100 determines the probability distribution of the dataset with a larger number of data by using the third weight 504 ) can be determined to determine the degree of weighting of each of the first weight and the second weight. At this time, the number of synthetic data in the integrated synthetic data set 502 may be equal to the sum of the number of first medical data and the number of second medical data. The computing device 100 may obtain the third weight 504 based on the weighting degree of each of the first weight and the second weight.

In one embodiment, the computing device 100 may acquire clinical prior information 501 including a clinical medical dataset in which clinical values are recorded. The computing device 100 is based on the number of clinical medical data in the clinical medical dataset, the number of first medical data in the first medical dataset, the number of second medical data in the second medical dataset, and the number of target data in the target medical dataset. Thus, the weighting degree of each of the first weight and the second weight can be determined. The computing device 100 compares the number of synthetic data of the integrated synthetic data set 502 and the number of target medical data of the target medical data set 503, so that the probability distribution of the dataset with a larger number of data is more reflected. The weighting degree of each weight and the second weight can be determined. At this time, the number of synthetic data in the integrated synthetic data set 502 may be equal to the sum of the number of clinical medical data, the number of first medical data, and the number of second medical data. The computing device 100 may obtain the third weight 504 based on the weighting degree of each of the first weight and the second weight.

In one embodiment, the computing device 100 selects a target corresponding to the target medical dataset 503 from the integrated synthetic dataset 502 based on the preset independent variable, the preset dependent variable, and the third weight 504. You can create synthetic datasets. The probability distribution of a preset dependent variable in the target synthetic data set may be determined based on the third weight 504. The computing device 100 may obtain an analysis result for the target medical dataset 503 from the target synthetic dataset.

In one embodiment, the computing device 100 may determine, from the first medical dataset and the second medical dataset, a first independent variable that is correlated with the dependent variable and is not included in the preset independent variable. The computing device 100 generates a target synthetic dataset corresponding to the target medical dataset from the integrated synthetic dataset 502 based on the preset independent variable, the first independent variable, the preset dependent variable, and the third weight 504. can be created. The computing device 100 may obtain an analysis result for the target medical dataset 503 from the target synthetic dataset. Through this, the technical effect of being able to create a synthetic data set including independent variables that the research institution or researcher has not considered among a plurality of variables that are correlated with a preset dependent variable can be achieved.

In one embodiment, the analysis results for the target medical data set 503 include information about a preset independent variable, a preset dependent variable, a third weight 504, and the correlation between the preset independent variable and the preset dependent variable. may include. For example, the correlation between independent variables and dependent variables included in the analysis results may be obtained based on the third weight 504.

According to an embodiment of the present disclosure, a computer-readable medium storing a data structure is disclosed. Data structure can refer to the organization, management, and storage of data to enable efficient access and modification of data. Data structure can refer to the organization of data to solve a specific problem (e.g., retrieving data, storing data, or modifying data in the shortest possible time). A data structure may be defined as a physical or logical relationship between data elements designed to support a specific data processing function. Logical relationships between data elements may include connection relationships between user-defined data elements. Physical relationships between data elements may include actual relationships between data elements that are physically stored in a computer-readable storage medium (e.g., a persistent storage device). A data structure may specifically include a set of data, relationships between data, and functions or instructions applicable to the data. Through an effectively designed data structure, the computing device 100 can perform calculations while minimally using the resources of the computing device 100. Specifically, the computing device 100 can increase the efficiency of operations, reading, insertion, deletion, comparison, exchange, and search through an effectively designed data structure.

Data structures can be divided into linear data structures and non-linear data structures depending on the type of data structure. A linear data structure may be a structure in which only one piece of data is connected to another piece of data. Linear data structures may include List, Stack, Queue, and Deque. A list can refer to a set of data that has an internal order. The list may include a linked list. A linked list may be a data structure in which data is connected in such a way that each data is connected in a single line with a pointer. In a linked list, a pointer may contain connection information to the next or previous data. Depending on its form, a linked list can be expressed as a singly linked list, a doubly linked list, or a circularly linked list. A stack may be a data listing structure that allows limited access to data. A stack can be a linear data structure in which data can be processed (for example, inserted or deleted) at only one end of the data structure. Data stored in the stack may have a data structure (LIFO-Last in First Out) where the later it enters, the sooner it comes out. A queue is a data listing structure that allows limited access to data. Unlike the stack, it can be a data structure (FIFO-First in First Out) where data stored later is released later. A deck can be a data structure that can process data at both ends of the data structure.

A non-linear data structure may be a structure in which multiple pieces of data are connected behind one piece of data. Nonlinear data structures may include graph data structures. A graph data structure can be defined by vertices and edges, and an edge can include a line connecting two different vertices. Graph data structure may include a tree data structure. A tree data structure may be a data structure in which there is only one path connecting two different vertices among a plurality of vertices included in the tree. In other words, it may be a data structure that does not form a loop in the graph data structure.

Throughout this specification, computational model, neural network, network function, and neural network may be used interchangeably. Below, it is described in a unified manner as a neural network. Data structures may include neural networks. And the data structure including the neural network may be stored in a computer-readable medium. Data structures including neural networks also include data preprocessed for processing by a neural network, data input to the neural network, weights of the neural network, hyperparameters of the neural network, data acquired from the neural network, activation functions associated with each node or layer of the neural network, neural network It may include a loss function for learning. A data structure containing a neural network may include any of the components disclosed above. In other words, the data structure including the neural network includes data preprocessed for processing by the neural network, data input to the neural network, weights of the neural network, hyperparameters of the neural network, data acquired from the neural network, activation functions associated with each node or layer of the neural network, neural network It may be configured to include all or any combination of loss functions for learning. In addition to the configurations described above, a data structure containing a neural network may include any other information that determines the characteristics of the neural network. Additionally, the data structure may include all types of data used or generated in the computational process of a neural network and is not limited to the above. Computer-readable media may include computer-readable recording media and/or computer-readable transmission media. A neural network can generally consist of a set of interconnected computational units, which can be referred to as nodes. These nodes may also be referred to as neurons. A neural network consists of at least one node.

The data structure may include data input to the neural network. A data structure containing data input to a neural network may be stored in a computer-readable medium. Data input to the neural network may include learning data input during the neural network learning process and/or input data input to the neural network on which training has been completed. Data input to the neural network may include data that has undergone pre-processing and/or data subject to pre-processing. Preprocessing may include a data processing process to input data into a neural network. Therefore, the data structure may include data subject to preprocessing and data generated by preprocessing. The above-described data structure is only an example and the present disclosure is not limited thereto.

The data structure may include the weights of the neural network. (In this specification, weights and parameters may be used with the same meaning.) And the data structure including the weights of the neural network may be stored in a computer-readable medium. A neural network may include multiple weights. Weights may be variable and may be varied by the user or algorithm in order for the neural network to perform the desired function. For example, when one or more input nodes are connected to one output node by respective links, the output node is set to the values input to the input nodes connected to the output node and the links corresponding to each input node. Based on the weight, the data value output from the output node can be determined. The above-described data structure is only an example and the present disclosure is not limited thereto.

As an example and not a limitation, the weights may include weights that are changed during the neural network learning process and/or weights for which neural network learning has been completed. Weights that change during the neural network learning process may include weights that change at the start of the learning cycle and/or weights that change during the learning cycle. Weights for which neural network training has been completed may include weights for which a learning cycle has been completed. Therefore, the data structure including the weights of the neural network may include weights that are changed during the neural network learning process and/or the data structure including the weights for which neural network learning has been completed. Therefore, the above-mentioned weights and/or combinations of each weight are included in the data structure including the weights of the neural network. The above-described data structure is only an example and the present disclosure is not limited thereto.

The data structure including the weights of the neural network may be stored in a computer-readable storage medium (e.g., memory, hard disk) after going through a serialization process. Serialization may be the process of converting a data structure into a form that can be stored on the same or a different computing device 100 and later reconstructed and used. The computing device 100 can transmit and receive data over a network by serializing the data structure. The data structure including the weights of the serialized neural network can be reconstructed in the same computing device 100 or another computing device 100 through deserialization. The data structure including the weights of the neural network is not limited to serialization. Furthermore, the data structure including the weights of the neural network is a data structure to increase computational efficiency while minimizing the use of resources of the computing device 100 (e.g., in a non-linear data structure, B-Tree, Trie, m-way search tree, AVL tree, Red-Black Tree) may be included. The foregoing is merely an example and the present disclosure is not limited thereto.

The data structure may include hyper-parameters of a neural network. And the data structure including the hyperparameters of the neural network can be stored in a computer-readable medium. A hyperparameter may be a variable that can be changed by the user. Hyperparameters include, for example, learning rate, cost function, number of learning cycle repetitions, weight initialization (e.g., setting the range of weight values subject to weight initialization), Hidden Unit. It may include a number (e.g., number of hidden layers, number of nodes in hidden layers). The above-described data structure is only an example and the present disclosure is not limited thereto.

6 is a brief, general schematic diagram of an example computing environment in which embodiments of the present disclosure may be implemented.

Although the disclosure has been described above as being capable of being implemented generally by computing device 100, those skilled in the art will recognize that the disclosure can be combined with computer-executable instructions and/or other program modules that can be executed on one or more computers and/ Alternatively, it will be well known that it can be implemented as a combination of hardware and software.

Typically, program modules include routines, programs, components, data structures, etc. that perform specific tasks or implement specific abstract data types. Additionally, those skilled in the art will understand that the methods of the present disclosure can be used on single-processor or multiprocessor computer systems, minicomputers, mainframe computers, as well as personal computers, handheld computing devices, microprocessor-based or programmable consumer electronics, etc. It will be appreciated that other computer system configurations may be implemented, including, but not limited to, each of which may operate in connection with one or more associated devices.

The described embodiments of the disclosure can also be practiced in distributed computing environments where certain tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote memory storage devices.

Computers typically include a variety of computer-readable media. Computer-readable media can be any medium that can be accessed by a computer, and such computer-readable media includes volatile and non-volatile media, transitory and non-transitory media, removable and non-transitory media. Includes removable media. By way of example, and not limitation, computer-readable media may include computer-readable storage media and computer-readable transmission media. Computer-readable storage media refers to volatile and non-volatile media, transient and non-transitory media, removable and non-removable, implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Includes media. Computer readable storage media may include RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital video disk (DVD) or other optical disk storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage. This includes, but is not limited to, a device, or any other medium that can be accessed by a computer and used to store desired information.

A computer-readable transmission medium typically implements computer-readable instructions, data structures, program modules, or other data on a modulated data signal, such as a carrier wave or other transport mechanism. Includes all information delivery media. The term modulated data signal refers to a signal in which one or more of the characteristics of the signal have been set or changed to encode information within the signal. By way of example, and not limitation, computer-readable transmission media includes wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, RF, infrared, and other wireless media. Combinations of any of the above are also intended to be included within the scope of computer-readable transmission media.

An example environment 1100 is shown that implements various aspects of the present disclosure, including a computer 1102, which includes a processing unit 1104, a system memory 1106, and a system bus 1108. do. System bus 1108 couples system components, including but not limited to system memory 1106, to processing unit 1104. Processing unit 1104 may be any of a variety of commercially available processors. Dual processors and other multiprocessor architectures may also be used as processing unit 1104.

System bus 1108 may be any of several types of bus structures that may further be interconnected to a memory bus, peripheral bus, and local bus using any of a variety of commercial bus architectures. System memory 1106 includes read only memory (ROM) 1110 and random access memory (RAM) 1112. The basic input/output system (BIOS) is stored in non-volatile memory 1110, such as ROM, EPROM, and EEPROM, and is a basic input/output system that helps transfer information between components within the computer 1102, such as during startup. Contains routines. RAM 1112 may also include high-speed RAM, such as static RAM, for caching data.

Computer 1102 may also include an internal hard disk drive (HDD) 1114 (e.g., EIDE, SATA)—the internal hard disk drive 1114 may also be configured for external use within a suitable chassis (not shown). Yes - a magnetic floppy disk drive (FDD) 1116 (e.g., for reading from or writing to a removable diskette 1118), and an optical disk drive 1120 (e.g., a CD-ROM for reading the disk 1122 or reading from or writing to other high-capacity optical media such as DVDs). Hard disk drive 1114, magnetic disk drive 1116, and optical disk drive 1120 are connected to system bus 1108 by hard disk drive interface 1124, magnetic disk drive interface 1126, and optical drive interface 1128, respectively. ) can be connected to. The interface 1124 for implementing an external drive includes at least one or both of Universal Serial Bus (USB) and IEEE 1394 interface technologies.

These drives and their associated computer-readable media provide non-volatile storage of data, data structures, computer-executable instructions, and the like. For computer 1102, drive and media correspond to storing any data in a suitable digital format. Although the description of computer-readable media above refers to removable optical media such as HDDs, removable magnetic disks, and CDs or DVDs, those of ordinary skill in the art would also recognize zip drives, magnetic cassettes, flash memory cards, and cartridges. It will be appreciated that other types of computer-readable media may also be used in the exemplary operating environment, and that any such media may contain computer-executable instructions for performing the methods of the present disclosure. .

A number of program modules may be stored in the drive and RAM 1112, including an operating system 1130, one or more application programs 1132, other program modules 1134, and program data 1136. All or portions of the operating system, applications, modules and/or data may also be cached in RAM 1112. It will be appreciated that the present disclosure may be implemented on various commercially available operating systems or combinations of operating systems.

A user may enter commands and information into computer 1102 through one or more wired/wireless input devices, such as a keyboard 1138 and a pointing device such as mouse 1140. Other input devices (not shown) may include microphones, IR remote controls, joysticks, game pads, stylus pens, touch screens, etc. These and other input devices are connected to the processing unit 1104 through an input device interface 1142, which is often connected to the system bus 1108, but may also include a parallel port, an IEEE 1394 serial port, a game port, a USB port, an IR interface, It can be connected by other interfaces, etc.

A monitor 1144 or other type of display device is also connected to system bus 1108 through an interface, such as a video adapter 1146. In addition to monitor 1144, computers typically include other peripheral output devices (not shown) such as speakers, printers, etc.

Computer 1102 may operate in a networked environment using logical connections to one or more remote computers, such as remote computer(s) 1148, via wired and/or wireless communications. Remote computer(s) 1148 may be a workstation, computing device computer, router, personal computer, portable computer, microprocessor-based entertainment device, peer device, or other conventional network node, and is generally connected to computer 1102. For simplicity, only memory storage device 1150 is shown, although it includes many or all of the components described. The logical connections depicted include wired/wireless connections to a local area network (LAN) 1152 and/or a larger network, such as a wide area network (WAN) 1154. These LAN and WAN networking environments are common in offices and companies and facilitate enterprise-wide computer networks, such as intranets, all of which can be connected to a worldwide computer network, such as the Internet.

When used in a LAN networking environment, computer 1102 is connected to local network 1152 through wired and/or wireless communications network interfaces or adapters 1156. Adapter 1156 may facilitate wired or wireless communication to LAN 1152, which also includes a wireless access point installed thereon for communicating with wireless adapter 1156. When used in a WAN networking environment, the computer 1102 may include a modem 1158 or be connected to a communicating computing device on the WAN 1154 or to establish communications over the WAN 1154, such as over the Internet. have other means. Modem 1158, which may be internal or external and a wired or wireless device, is coupled to system bus 1108 via serial port interface 1142. In a networked environment, program modules described for computer 1102, or portions thereof, may be stored in remote memory/storage device 1150. It will be appreciated that the network connections shown are exemplary and that other means of establishing a communications link between computers may be used.

Computer 1102 may be associated with any wireless device or object deployed and operating in wireless communications, such as a printer, scanner, desktop and/or portable computer, portable data assistant (PDA), communications satellite, wirelessly detectable tag. Performs actions to communicate with any device or location and telephone. This includes at least Wi-Fi and Bluetooth wireless technologies. Accordingly, communication may be a predefined structure as in a conventional network or may simply be ad hoc communication between at least two devices.

Wi-Fi (Wireless Fidelity) allows connection to the Internet, etc. without wires. Wi-Fi is a wireless technology, like cell phones, that allows these devices, such as computers, to send and receive data indoors and outdoors, anywhere within the coverage area of a cell tower. Wi-Fi networks use wireless technology called IEEE 802.11 (a, b, g, etc.) to provide secure, reliable, and high-speed wireless connections. Wi-Fi can be used to connect computers to each other, to the Internet, and to wired networks (using IEEE 802.3 or Ethernet). Wi-Fi networks can operate in the unlicensed 2.4 and 5 GHz wireless bands, for example, at data rates of 11 Mbps (802.11a) or 54 Mbps (802.11b), or in products that include both bands (dual band). .

Those skilled in the art will understand that information and signals may be represented using any of a variety of different technologies and techniques. For example, data, instructions, commands, information, signals, bits, symbols and chips that may be referenced in the above description include voltages, currents, electromagnetic waves, magnetic fields or particles, optical fields. It can be expressed by particles or particles, or any combination thereof.

Those skilled in the art will understand that the various illustrative logical blocks, modules, processors, means, circuits and algorithm steps described in connection with the embodiments disclosed herein may be used in electronic hardware, (for convenience) It will be understood that it may be implemented by various forms of program or design code (referred to herein as software) or a combination of both. To clearly illustrate this interoperability of hardware and software, various illustrative components, blocks, modules, circuits and steps have been described above generally with respect to their functionality. Whether this functionality is implemented as hardware or software depends on the specific application and design constraints imposed on the overall system. A person skilled in the art of this disclosure may implement the described functionality in various ways for each specific application, but such implementation decisions should not be construed as departing from the scope of this disclosure.

The various embodiments presented herein may be implemented as a method, apparatus, or article of manufacture using standard programming and/or engineering techniques. The term article of manufacture includes a computer program, carrier, or media accessible from any computer-readable storage device. For example, computer-readable storage media include magnetic storage devices (e.g., hard disks, floppy disks, magnetic strips, etc.), optical disks (e.g., CDs, DVDs, etc.), smart cards, and flash. Includes, but is not limited to, memory devices (e.g., EEPROM, cards, sticks, key drives, etc.). Additionally, various storage media presented herein include one or more devices and/or other machine-readable media for storing information.

It is to be understood that the specific order or hierarchy of steps in the processes presented is an example of illustrative approaches. It is to be understood that the specific order or hierarchy of steps in processes may be rearranged within the scope of the present disclosure, based on design priorities. The appended method claims present elements of the various steps in a sample order but are not meant to be limited to the particular order or hierarchy presented.

The description of the presented embodiments is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be applied to other embodiments without departing from the scope of the disclosure. Thus, the present disclosure is not limited to the embodiments presented herein but is to be interpreted in the broadest scope consistent with the principles and novel features presented herein.

Claims (20)

In a method of analyzing medical data from a medical institution performed by a computing device,
Obtaining a first medical dataset of the first medical institution and a second medical dataset of the second medical institution from a first medical institution and a second medical institution;
generating a total synthetic dataset from the first medical dataset and the second medical dataset; and
Based on a target medical dataset of a target medical institution, obtaining an analysis result for the target medical dataset from the integrated synthetic dataset;
Includes,
The first medical data set, the second medical data set, and the integrated synthetic data set include preset independent variables and preset dependent variables,
The integrated synthetic data set includes a first weight corresponding to the preset independent variable, and
The steps for obtaining analysis results for the target medical dataset are:
Obtaining a second weight corresponding to the preset independent variable from the target medical data set;
Obtaining a third weight corresponding to the preset independent variable based on the first weight and the second weight; and
Obtaining the analysis result from the integrated synthetic data set based on the preset independent variable, the preset dependent variable, and the third weight;
Including,
method.
According to claim 1,
The step of generating the integrated synthetic data set is,
Obtaining clinical prior information including information about the correlation between the preset independent variable and the preset dependent variable; and
generating the integrated synthetic data set from the clinical prior information, the first medical data set, and the second medical data set;
Including,
method.
According to claim 1,
The step of generating the integrated synthetic data set is,
Generating a first synthetic dataset corresponding to the first medical dataset and a second synthetic dataset corresponding to the second medical dataset from the first medical dataset and the second medical dataset; and
Integrating the first synthetic data set and the second synthetic data set to generate the integrated synthetic data set;
Including,
method.
According to claim 3,
The step of generating the first synthetic dataset and the second synthetic dataset is,
Generating the first synthetic dataset and the second synthetic dataset from the first medical dataset and the second medical dataset using a pre-trained GAN (Generative Adversarial Networks) model;
Including,
method.
According to claim 3,
The step of generating the first synthetic dataset and the second synthetic dataset is,
Generating the first synthetic dataset and the second synthetic dataset from the first medical dataset and the second medical dataset using a SMOTE (Synthetic Minority Oversampling Technique) algorithm;
Including,
method.
According to claim 3,
The step of generating the first synthetic dataset and the second synthetic dataset is,
Obtaining clinical prior information including information about the correlation between the preset independent variable and the preset dependent variable; and
Using a Bayesian algorithm based on Bayesian inference, the first synthetic data set and the second synthetic data set are generated by reflecting the first medical data set and the second medical data set in the clinical prior information. steps;
Including,
method.
According to claim 1,
determining, from the first medical data set and the second medical data set, a first independent variable that is correlated with the dependent variable and is not included in the preset independent variable;
It further includes, and
The analysis result includes information about the correlation between the first independent variable and the preset dependent variable,
method.
delete According to claim 1,
The step of obtaining a third weight corresponding to the preset independent variable is:
Based on the number of first medical data in the first medical data set, the number of second medical data in the second medical data set, and the number of target medical data in the target medical data set, the first weight and the second weight are respectively determining the degree of weighting; and
Obtaining the third weight based on the weighting degree of each of the first weight and the second weight;
Including,
method.
According to clause 9,
The step of determining the weighting degree of each of the first weight and the second weight is:
Obtaining clinical preliminary information including clinical statistics in which clinical values are recorded; and
Based on the number of clinical medical data in the clinical statistical data, the number of first medical data in the first medical data set, the number of second medical data in the second medical data set, and the number of target medical data in the target medical data set, determining a weighting degree of each of the first weight and the second weight;
Including,
method.
According to claim 1,
The steps for obtaining analysis results for the target medical dataset are:
Generating a target synthetic dataset corresponding to the target medical dataset from the integrated synthetic dataset based on the preset independent variable, the preset dependent variable, and the third weight; and
Obtaining the analysis result from the target synthetic dataset;
Including,
method.
According to claim 1,
The steps for obtaining analysis results for the target medical dataset are:
determining, from the first medical data set and the second medical data set, a first independent variable that is correlated with the dependent variable and is not included in the preset independent variable;
Generating a target synthetic dataset corresponding to the target medical dataset from the integrated synthetic dataset based on the preset independent variable, the first independent variable, the preset dependent variable, and the third weight; and
Obtaining the analysis result from the target synthetic dataset;
Including,
method.
According to claim 1,
The analysis results for the target medical data set are:
Containing information about the preset independent variable, the preset dependent variable, the third weight, and the correlation between the preset independent variable and the preset dependent variable,
method.
According to claim 1,
The first medical data set and the second medical data set are medical data sets converted from the first raw medical data set and the second raw medical data set to the data structure of the Common Data Model (CDM). in - the first medical dataset corresponds to the first raw medical dataset, and the second medical dataset corresponds to the second raw medical dataset -,
method.
In a method of analyzing medical data from a medical institution performed by a computing device,
Obtaining a first medical dataset of the first medical institution and a second medical dataset of the second medical institution from a first medical institution and a second medical institution;
generating a total synthetic dataset from the first medical dataset and the second medical dataset; and
Based on a target medical dataset of a target medical institution, obtaining an analysis result for the target medical dataset from the integrated synthetic dataset;
Includes,
The first medical data set, the second medical data set, and the integrated synthetic data set include preset independent variables and preset dependent variables,
The steps for obtaining analysis results for the target medical dataset are:
generating a first integrated synthetic dataset from the first medical dataset and the second medical dataset;
generating a second integrated synthetic dataset from the first medical dataset and the second medical dataset; and
Based on the target medical dataset, obtaining an analysis result for the target medical dataset from the first integrated synthetic dataset and the second integrated synthetic dataset;
contains, and
The first integrated synthetic data set includes a 1-1 weight corresponding to the preset independent variable,
The second integrated synthetic data set includes 1-2 weights corresponding to the preset independent variables,
The 1-1 weight and the 1-2 weight are different from each other,
method.
According to claim 15,
The preset dependent variables included in each of the first integrated synthetic data set and the second integrated synthetic data set have probability distributions based on different means and standard deviations,
method.
According to claim 15,
The step of obtaining analysis results for the target medical dataset is,
Obtaining a first analysis result for the target medical dataset from the first integrated synthetic dataset;
Obtaining a second analysis result for the target medical dataset from the second integrated synthetic dataset; and
Obtaining an analysis result for the target medical dataset based on the first analysis result and the second analysis result;
Including,
method.
According to claim 1,
Based on at least one of a preset effect size, preset power, and preset significance level, determining the required number of data of the medical dataset for generating the integrated synthetic dataset;
It further includes, and
Each of the first medical data set and the second medical data set includes medical data greater than the required number of data,
method.
A computing device for analyzing medical data from a medical institution,
at least one processor; and
Memory;
Includes,
The at least one processor,
Obtaining a first medical data set of the first medical institution and a second medical data set of the second medical institution from the first medical institution and the second medical institution,
Generate a total synthetic dataset from the first medical dataset and the second medical dataset, and
Based on the target medical dataset of the target medical institution, an analysis result for the target medical dataset is obtained from the integrated synthetic dataset,
The first medical data set, the second medical data set, and the integrated synthetic data set include preset independent variables and preset dependent variables,
The integrated synthetic data set includes a first weight corresponding to the preset independent variable, and
Obtaining analysis results for the target medical dataset is,
Obtaining a second weight corresponding to the preset independent variable from the target medical data set,
Based on the first weight and the second weight, a third weight corresponding to the preset independent variable is obtained, and
Obtaining the analysis result from the integrated synthetic data set based on the preset independent variable, the preset dependent variable, and the third weight,
Computing device.
A computer program stored in a computer-readable storage medium, wherein the computer program, when executed by one or more processors, causes the one or more processors to perform operations for analyzing medical data of a medical institution, the operations comprising:
Obtaining a first medical data set of the first medical institution and a second medical data set of the second medical institution from a first medical institution and a second medical institution;
An operation of generating a total synthetic data set from the first medical data set and the second medical data set; and
An operation of obtaining an analysis result for the target medical dataset from the integrated synthetic dataset based on the target medical dataset of a target medical institution;
Includes,
The first medical data set, the second medical data set, and the integrated synthetic data set include preset independent variables and preset dependent variables,
The integrated synthetic data set includes a first weight corresponding to the preset independent variable, and
The operation of obtaining analysis results for the target medical dataset is:
Obtaining a second weight corresponding to the preset independent variable from the target medical data set;
Obtaining a third weight corresponding to the preset independent variable based on the first weight and the second weight; and
Obtaining the analysis result from the integrated synthetic data set based on the preset independent variable, the preset dependent variable, and the third weight;
Including,
A computer program stored on a computer-readable storage medium.

Images