Device for analyzing causes of sepsis based on matrix causal deentanglement

By constructing a hidden Markov chain structure causal graph and serialized variational automatic encoder, the hidden variables are decoupled, and the problem of sparse data and insufficient interpretability in the cause analysis of sepsis is solved, and the factors of sepsis are realized accurate analysis and early prediction.

CN118983108AActive Publication Date: 2024-11-19TIANJIN UNIV

Patent Information

Application Number
CN202411105211.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-13
Publication Date
2024-11-19
Estimated Expiration
2044-08-13

AI Technical Summary

Technical Problem

The prior art has problems with sparse data, structural causal graph construction problems, and insufficient interpretability of the model in the analysis of causes of sepsis, resulting in inaccurate analysis of causal relationships and difficulty in predicting sepsis early.

Method used

The sepsis cause factor analysis device based on matrix causal de-entanglement is adopted. Through the data acquisition module, the causal graph construction module, the model training module and the factor analysis module, the hidden Markov chain structure causal graph is constructed, and the serialized variable segmentation is used. The automatic encoder trains a serialized causal decoupling model to decouple hidden variables to improve analysis accuracy.

Benefits of technology

Accurate analysis and judgment of the causes of sepsis, improve the interpretability of the model and the understanding of sepsis-related indicators, can understand the mechanism of disease occurrence more clearly, reduce false correlation learning, and enhance the Ability to predict early stages of sepsis.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118983108A_ABST
    Figure CN118983108A_ABST
Patent Text Reader

Abstract

The invention discloses a sepsis cause factor analysis device based on matrix causal deentanglement, and relates to the field of intensive care medicine data processing and analysis, the device comprises a data acquisition module, a causal graph construction module, a model training module and a factor analysis module; the causal graph construction module is used for grouping historical electronic health record time sequence data and constructing a structural causal matrix based on a hidden Markov chain; the model training module is used for training a serialized causal decoupling model based on the structural causal matrix; the serialized causal decoupling model comprises a serialized variational automatic encoder, a priori network, a posteriori network and a generative network; the factor analysis module is used for grouping the current electronic health record time sequence data and determining the association strength of each group of data and sepsis based on a serialized causal decoupling model so as to analyze the cause factors of sepsis. According to the method, the analysis precision of the causes of sepsis is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of critical care medicine data processing and analysis, and in particular to a sepsis causal factor analysis device based on matrix-type causal disentanglement. Background Art

[0002] Sepsis, as one of the global health priorities, has always been a highly complex and diverse fatal disease. According to the World Health Organization, millions of people die from sepsis every year, and this number is still rising. The treatment of sepsis is highly time-dependent, and there is evidence that for every hour of delay in the use of antibiotics, the mortality rate of sepsis increases significantly. Early prediction of sepsis before the onset of the disease provides clinicians with valuable time and effectively reduces the mortality rate.

[0003] The causes of sepsis involve a variety of complex factors, including infection sources, host responses, and a variety of environmental and biological variables. The occurrence of sepsis is usually associated with severe infection, and the most common sources of infection include lung infection, abdominal infection, urinary tract infection, and skin and soft tissue infection. Infection triggers the host immune system to respond, leading to the massive release of inflammatory mediators. This excessive inflammatory response can cause extensive tissue damage and organ dysfunction.

[0004] In hospitals, the Sepsis-related Organ Failure Assessment (SOFA) is an effective method for assessing the severity of sepsis in patients. The SOFA score is based on several key indicators, including respiratory, cardiovascular, liver, coagulation, kidney, and central nervous system function. When the SOFA score is greater than or equal to 2, the patient is generally considered to have sepsis, the worst score in a day. Although the SOFA score can diagnose sepsis more accurately, it uses fewer indicators and is mainly used for diagnosis rather than early prediction.

[0005] In recent years, with the rapid development of machine learning and deep learning, these technologies have made some progress in analyzing and understanding the causes of sepsis. For example, Horng et al. used models such as support vector machines, generalized linear models, naive Bayes, and expectation maximization to conduct a retrospective analysis of sepsis patients in a tertiary academic teaching hospital, revealing the relationship between different causes and the occurrence of sepsis. Zhang et al. used Local Interpretable Model-Agnostic Explanations (LIME) to develop a model for real-time analysis of the causes of sepsis, which effectively supports clinical staff to understand the impact of different factors on sepsis. Liu et al. used text mining technology to analyze clinical texts to improve the understanding of the causes of sepsis. Gho et al. expanded the text mining algorithm for sepsis cause analysis and advocated modeling of common topics to improve the accurate analysis of the causes of sepsis.

[0006] However, deep learning methods face a series of challenges in the medical field, especially in the analysis of the causes of sepsis. First, due to its complexity, deep learning models lack interpretability. Although they can capture the correlation between data, they often have difficulty in parsing the true causal relationship, which may lead to the learning of false correlations and insufficient generalization ability.

[0007] Some research works have attempted to provide solutions to these problems. For example, Li et al. proposed a causal Markov model based on variational auto-encoder (VAE), which avoids the learning of false associations by distinguishing hidden variables that are causally related to the disease and those that are not directly related to the disease, and provides a robust method for analyzing the causes of sepsis. Nie et al. used the instrumental variable learning method to combine chest X-ray (CXR) images with electronic health record (EHR) data, giving instrumental variables medical semantics, effectively eliminating false correlations, and improving the ability to analyze the causes of sepsis. Nie et al. further enhanced the interpretability of chest X-ray images and the understanding of the causes of sepsis through structured causal models and backdoor adjustment methods.

[0008] To address the problem of insufficient medical data, Shang et al. used graph neural networks to understand the relationships between medical codes and pre-trained the model on patients' initial visit data. By simulating analysis tasks in actual scenarios, they improved the model's ability to understand the causes of sepsis and fine-tuned the model in subsequent patient visits to improve parsing accuracy.

[0009] In general, existing research has provided some ideas and methods for analyzing the causes of sepsis, but there are still some challenges, including how to make full use of EHR data to effectively solve the problem of data sparsity, how to construct a reasonable structural causal graph to analyze the correct causal relationship, and how to improve the interpretability of the model to clarify the relevant indicators of sepsis. Summary of the invention

[0010] The purpose of the present invention is to provide a sepsis causal factor analysis device based on matrix causal disentanglement, which can achieve accurate analysis and judgment of the cause of sepsis based on effective data decoupling analysis.

[0011] To achieve the above object, the present invention provides a device for analyzing the causes of sepsis based on matrix causal disentanglement, comprising: A data acquisition module, used to acquire historical electronic health record time series data; the historical electronic health record time series data includes laboratory test indicators, personal attributes, clinical texts and disease labels at multiple times within a historical set period; the disease label is 0 or 1, wherein a disease label of 1 indicates sepsis, and a disease label of 0 indicates no sepsis; A causal graph construction module is connected to the data acquisition module and is used to group the historical electronic health record time series data to obtain multiple groups of historical factor data, and to construct a structural causal graph based on a hidden Markov chain based on each group of historical factor data to obtain a structural causal matrix; the structural causal graph is used to describe the historical factor data into a high-dimensional latent space; A model training module is connected to the causal graph construction module, and is used to train and optimize the serialized causal decoupling model based on the structural causal matrix according to each group of historical factor data, so as to obtain a trained serialized causal decoupling model; the serialized causal decoupling model includes a serialized variational autoencoder, a priori network, a posterior network and a generative network; the serialized variational autoencoder is used to encode the historical factor data; the priori network and the posterior network are used to parse the causal relationship in the structural causal graph according to the encoded historical factor data, and decouple to obtain multiple latent variables; the multiple latent variables are: latent variables related to sepsis, latent variables related to other diseases and latent variables unrelated to the disease; the generative network is used to analyze the correlation strength between the latent variables related to sepsis and sepsis; The factor analysis module is connected to the model training module and is used to group the current electronic health record time series data to obtain multiple groups of current factor data, and determine the correlation strength between each group of current factor data and sepsis based on the trained serialized causal decoupling model to analyze the causal factors of sepsis.

[0012] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects: the present invention provides a device for analyzing the causes of sepsis based on matrix causal disentanglement, constructs a structural causal graph based on hidden Markov chains, which can accurately describe the causal relationship of each variable over time, and proposes a framework based on serialized variational autoencoders to effectively learn the causal relationship described by the structural causal graph. Through this framework, the hidden variables related to sepsis are successfully decoupled from the confounding factors. The serialized causal decoupling model is trained and optimized using the time series data in the historical EHR, and the clinical data and physiological parameters of the patient are integrated, making full use of the time and space information in the EHR data to decouple complex causal relationships, which not only improves the interpretability of the cause analysis of sepsis, but also provides important support for in-depth understanding of the mechanism of sepsis, so that clinical researchers can understand the mechanism of sepsis more clearly. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0014] Figure 1 A schematic diagram of a module of a sepsis causal factor analysis device based on matrix causal disentanglement provided by the present invention.

[0015] Figure 2 A flow chart of a method for analyzing causal factors of sepsis based on matrix causal disentanglement provided by the present invention.

[0016] Figure 3 It is a structural causal graph based on hidden Markov chain.

[0017] Figure 4 Schematic diagram of a sequential variational autoencoder.

[0018] Figure 5 Schematic diagram of the network structure of the serialized causal decoupling model.

[0019] Figure 6 A schematic diagram of an implementation process of the method for analyzing causal factors of sepsis based on matrix causal disentanglement provided by the present invention.

[0020] Figure 7 A schematic diagram of another implementation process of the method for analyzing causal factors of sepsis based on matrix causal disentanglement provided by the present invention.

[0021] Figure 8Schematic diagram of the structure of grouping matrix decoupling analysis.

[0022] Fig. 9 Receiver operating characteristic curves of various models for early prediction of sepsis.

[0023] Fig.10 This is a diagram of the internal structure of a computer device. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] The purpose of the present invention is to provide a device for analyzing the causes of sepsis based on matrix causal disentanglement, combining causal disentanglement, hidden Markov chain, and serialized variational autoencoder technologies to establish a serialized causal decoupling model, make full use of the time and distribution information contained in the data, effectively avoid the influence of confounding factors, avoid false correlations, and improve the accuracy of sepsis cause analysis.

[0026] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0027] Example 1: Figure 1 As shown, this embodiment provides a device for analyzing causal factors of sepsis based on matrix causal disentanglement, including a data acquisition module 101, a causal graph construction module 102, a model training module 103 and a factor analysis module 104.

[0028] The data acquisition module 101 is used to acquire the historical electronic health record time series data.

[0029] The causal graph construction module 102 is connected to the data acquisition module 101. The causal graph construction module 102 is used to group the historical electronic health record time series data to obtain multiple groups of historical factor data, and construct a structural causal graph based on a hidden Markov chain based on each group of historical factor data to obtain a structural causal matrix.

[0030] The model training module 103 is connected to the causal graph construction module, and is used to train and optimize the serialized causal decoupling model based on each group of historical factor data and the structural causal matrix to obtain a trained serialized causal decoupling model.

[0031] The factor analysis module 104 is connected to the model training module 103. The factor analysis module 104 is used to group the current electronic health record time series data to obtain multiple groups of current factor data, and determine the correlation strength between each group of current factor data and sepsis based on the trained serialized causal decoupling model to analyze the causal factors of sepsis.

[0032] Furthermore, the model training module 104 includes: an exogenous variable determination submodule, a coding submodule, a priori and posterior submodule, an association strength determination submodule and a training submodule.

[0033] The exogenous variable determination submodule is used to extract the spatial distribution characteristics of laboratory test indicators and personal attributes at any time during the historical setting period to determine t exogenous variables at time; according to t -2 time exogenous variables, t -1 time exogenous variables and t The exogenous variable at the moment is encoded by the serialized variational autoencoder, and the t The time distribution information of the exogenous variables at the moment, where t For any time in the historical setting period, 1< t < T , T The final moment of the period set for history.

[0034] The encoding submodule is used to adopt the serialized variational autoencoder to t -2 laboratory test indicators, t -1 moment laboratory test indicators and t The laboratory test indicators at the time are coded to obtain t The time distribution information of laboratory test indicators at the moment. t -2 moments of personal attributes, t -1 moment of personal attributes and t The personal attributes of the moment are encoded to obtain t Time distribution information of personal attributes at each moment.

[0035] The a priori and posterior submodule is used to adopt the a priori network, according to t The time distribution information of the exogenous variables at the moment and t The posterior mean of each latent variable at time -1 is calculated t The prior mean and prior variance of each latent variable at the moment, and define the prior distribution; using the posterior network, according to t The time distribution information of the exogenous variables at the moment, t Time distribution information of laboratory test indicators at each moment,t Time distribution information of personal attributes at each moment and t -1 time, the posterior mean of each latent variable is determined t The posterior mean and posterior variance of the latent variable at time t, and define the posterior distribution.

[0036] The association strength determination submodule is used to adopt the generation network according to t The posterior mean and posterior variance of each latent variable at time t are determined t The hidden variables after reparameterization at each moment, and according to t After the hidden variables are reparameterized at each moment, reconstruction t Laboratory test indicators at all times and t The personal attributes of the moment, according to T The latent variables related to sepsis and other diseases after reparameterization at each moment are reconstructed T Clinical text of the moment, according to T The latent variables related to sepsis are reparameterized at each moment, and the strength of association between the latent variables related to sepsis and sepsis is analyzed.

[0037] The training submodule is used to reconstruct the laboratory test indicators at each moment within the historical set period, the personal attributes reconstructed at each moment, T Clinical texts that are constantly being reconstructed, T The predicted disease label at each moment, the association strength at each moment, the personal attributes at each moment, T Clinical texts at all times, T The diseased label at the moment, the prior distribution and the posterior distribution are optimized with the lower bound of evidence as the optimization target, so as to obtain a trained serialized causal decoupling model.

[0038] Furthermore, if Figure 2 As shown, the present invention is based on the above-mentioned sepsis causal factor analysis device based on matrix causal disentanglement, and also provides a sepsis causal factor analysis method based on matrix causal disentanglement, including steps 201 to 204.

[0039] Step 201: Obtain historical electronic health record time series data.

[0040] The historical electronic health record time series data includes laboratory test indicators at multiple times within a historical set period A , Personal attributes B , Clinical Texts Z (including antibiotic use) and disease labels Y .

[0041] Specifically, laboratory test indicators include platelet count, white blood cell count, blood pressure, etc. Personal attributes include height, weight, age, race, etc. Clinical texts include doctor's orders, surgical records, etc. In this embodiment, the EHR data of patients in the Intensive Care Unit (ICU) in the MIMIC-IV database are analyzed.

[0042] The present invention has a great impact on laboratory testing indicators. A , Personal attributes B , Clinical Texts Z Conduct group training, and keep clinical text in each group Z The data of antibiotics in the experiment were collected and then each group was analyzed separately. The following steps are the analysis process of each group.

[0043] Step 202: Group the historical electronic health record time series data to obtain multiple groups of historical factor data, and construct a structural causal graph based on a hidden Markov chain based on each group of historical factor data to obtain a structural causal matrix. The structural causal graph is used to describe the historical electronic health record time series data as a high-dimensional latent space.

[0044] The structural causal graph describes the known input data as an unknown high-dimensional latent space, then disentangles the high-dimensional latent space into different latent variables, reconstructs the input data according to the causal relationship, and finally uses only the latent variables related to sepsis to analyze the causal factors of sepsis.

[0045] like Figure 3 As shown, the structure of the structural causal graph is: and Co-generating exogenous variables , the exogenous variable The latent space described is decoupled into three latent variables: latent variables related to sepsis , hidden variables related to other diseases and hidden variables unrelated to the disease . At each time step, by Refactoring ,pass and Refactoring ,pass , and Refactoring . In the last time step, by The causative factors of sepsis were analyzed.

[0046] In a structural causal graph, a child node is determined only by the parent node pointing to it and an independent exogenous noise, as shown in the following formula.

[0047] .

[0048] .

[0049] .

[0050] .

[0051] in, Decide The distribution of for t Moment Impact The exogenous noise, , Represents a moment, represent The moment before the moment.

[0052] Right now t Exogenous variables at time Depend on t Laboratory testing indicators at all times , t Personal attributes of the moment and independent exogenous noise Decide; t Hidden variables associated with sepsis at all times Depend on t Hidden variables associated with sepsis at time -1 , t Exogenous variables at time and independent exogenous noise Decide; t Hidden variables associated with other diseases Depend on t -1 Hidden variables related to other diseases , t Exogenous variables at time and independent exogenous noise Decide; t Hidden variables that are not related to the disease at any time Depend on t- Hidden variables that are not related to the disease at time 1 , t Exogenous variables at time and independent exogenous noise Decide; t Laboratory testing indicators at all times Depend on t Hidden variables associated with sepsis at all times and independent exogenous noise Decide;t Personal attributes of the moment Depend on t Hidden variables associated with sepsis at all times , t Hidden variables associated with other diseases , t Hidden variables that are not related to the disease at any time and independent exogenous noise Decide; t Disease label at all times Depend on t Hidden variables associated with sepsis at all times and independent exogenous noise Decide; t Clinical text at the moment Depend on t Hidden variables associated with sepsis at all times , t Hidden variables associated with other diseases and independent exogenous noise Decide.

[0053] In the present invention, t Time refers to the continuous time in units of 1 hour. t hours.

[0054] Step 203: According to each set of historical factor data, based on the structural causal matrix, the serialized causal decoupling model is trained and optimized to obtain a trained serialized causal decoupling model.

[0055] The serialized causal decoupling model includes a serialized variational autoencoder, a priori network, a posterior network and a generative network. The serialized variational autoencoder is used to encode historical factor data. The priori network and the posterior network are used to parse the causal relationship in the structural causal graph according to the encoded historical factor data, and decouple to obtain multiple latent variables. The multiple latent variables are: latent variables related to sepsis, latent variables related to other diseases, and latent variables unrelated to the disease. The generative network is used to analyze the strength of association between the latent variables related to sepsis and sepsis.

[0056] The present invention adopts the causal disentanglement method of serialized variational autoencoder to study the causal relationship described by the hidden Markov chain and decouple the hidden variables. By designing a special encoder, the time and distribution characteristics in the input data are deeply mined and mapped to a high-dimensional latent space. The structure of the latent space is first defined and trained by a priori network to constrain the construction of the latent space, making it more reasonable and interpretable. This process not only helps to reveal the complex relationship between latent variables, but also identifies key factors related to sepsis.

[0057] Specifically, step 203 includes steps 301 to 308 .

[0058] Step 301: For any moment in the historical setting period, through a plurality of continuous fully connected layers Extract features from laboratory test indicators and personal attributes at the time , split it into two parts evenly along the last dimension of the tensor, the first part Directly defined as the mean , the latter part Through the activation function The variance is then defined as , and finally reparameterize to obtain an exogenous variable that conforms to the Gaussian distribution at the time .

[0059] Step 302: According to t -2 time exogenous variables, t -1 time exogenous variables and t The exogenous variable at the moment is encoded by the serialized variational autoencoder, and the t The time distribution information of the exogenous variables at the moment. Among them, t For any moment in the historical setting period, 1< t <T,T Set the last moment in the period for history.

[0060] In this embodiment, a point convolution based encoder is first used to t -2 time exogenous variables, t -1 time exogenous variables and t The exogenous variables at the time are encoded to obtain t Then, multiple preliminary feature maps at the moment t Multiple preliminary feature maps at the moment are concatenated, and a depth-wise convolution-based encoder is used to encode the concatenated preliminary feature maps to obtain t The final feature map at the moment. Finally, according to t The final feature map at the moment is calculated using the formula Sure t The time distribution information of the exogenous variables at the moment. Among them, for t Time distribution information of exogenous variables at the moment for t The final feature map at time, for t exogenous variables at time, represents a full connection operation, Represents the shape of the reshaped tensor, Represents the addition of corresponding points of tensors.

[0061] Step 303: According to t -2 laboratory test indicators at the moment, t -1 moment laboratory test indicators and t The laboratory test indicators at the time are encoded by the serialized variational autoencoder to obtain t Time distribution information of laboratory test indicators at each moment.

[0062] Step 304: According to t -2 moments of personal attributes, t -1 moment of personal attributes and t The personal attributes at the moment are encoded using the serialized variational autoencoder to obtain t Time distribution information of personal attributes at each moment.

[0063] like Figure 4 As shown, the serialized variational autoencoder will include three consecutive hours of data including the data at the moment to be encoded In parallel in the channel direction, first do one indivual Point convolution of convolution kernel: ;in, is the preliminary feature map, It is an encoder based on point convolution, where N is equal to the feature dimension of the encoded data, and the feature dimension refers to the number of indicators of laboratory test indicators.

[0064] Point convolution can weight the input feature map in the channel direction. Several convolution kernels will output several feature maps. Point convolution only changes the number of feature maps, not the size of the feature maps. preliminary feature maps, concatenating them together to form X The feature map is then X Depthwise convolution of convolution kernel: ;in, is the first preliminary feature map, is the Nth preliminary feature map, Indicates the concatenation of tensors. It is an encoder based on depthwise convolution. Each channel is convolved by only one convolution kernel. Depthwise convolution does not change the number of feature maps, but only the size of the feature maps. Finally, the shape of the feature map is reshaped, and together with the data to be encoded at this moment, the encoding results are added together through a fully connected layer to change the size and output, forming a residual sum structure: ;in, is the time distribution information of the input data, is a fully connected layer, Reshape a tensor.

[0065] Step 305: Using the prior network, according to t The time distribution information of the exogenous variables at the moment, t -1 time, the posterior mean of each latent variable is determined t The prior mean and prior variance of each latent variable at time , and define the prior distribution.

[0066] like Figure 5 In the prior network shown in the figure, spatial distribution features are first extracted from laboratory test indicators and personal attributes to jointly generate exogenous variables that can indicate sample distribution. The temporal distribution information is extracted through the serialized variational autoencoder and encoded as . Respectively t The hidden variables at the -1 moment are concatenated, and then the prior means and prior variances of the three hidden variables are parameterized through a layer of gated recurrent unit (GRU) and two different fully connected layers.

[0067] In this embodiment, the prior network is determined using the following formula t The prior mean and prior variance of each latent variable at time .

[0068] .

[0069] .

[0070] .

[0071] in, for t The prior mean of the latent variable associated with sepsis at each time point, for t The prior variance of the latent variable associated with sepsis at each moment, for t The prior mean of the latent variables related to other diseases at each moment, for t The prior variance of the latent variables related to other diseases at each moment, for t The prior mean of the latent variable that is not related to the disease at any time, for t The prior variance of the latent variable that is independent of the disease at each time step, for t The posterior mean of the latent variable associated with sepsis at time -1, for tThe posterior mean of the latent variables related to other diseases at time -1, for t The posterior mean of the latent variable that is not related to the disease at time -1, for t The time distribution information of the exogenous variables at the moment, represents a gated recurrent unit, , , , , , Denotes fully connected layers that are trained independently, and their superscripts indicate that the fully connected layers act on the generated variables { The prior mean of or prior variance .

[0072] Taking the latent variables related to sepsis as an example, and The size after splicing is 128, parameterized The GRU input size is 128, the output size is 256, the input size of the two fully connected layers is 256, and the output size is 64. This encoding method makes full use of the temporal and spatial information in the data, avoids learning false correlations in the data, and improves robustness.

[0073] The prior distribution is: .

[0074] in, is the prior distribution, are the parameters of the prior network, for t The set of hidden parameters at time, ={ , }, for t The hidden parameter set at time -1, for t exogenous variables at time, for t Hidden variables associated with sepsis at all times, for t Hidden variables related to sepsis at time -1, for t Hidden variables that are related to other diseases at all times, for t -1 time is related to other diseases. for t Hidden variables that are not related to the disease at any time, for t -1 time is a hidden variable that is not related to the disease.

[0075] Step 306: Using the posterior network, according to t The time distribution information of the exogenous variables at the moment, t Time distribution information of laboratory test indicators at each moment, t The time distribution information of personal attributes at each moment and t The posterior mean of each latent variable at time -1 is calculated t The posterior mean and posterior variance of the latent variable at time t, and define the posterior distribution.

[0076] The present invention sets a posterior distribution that is easy to learn to approximate the prior distribution, further optimizes the representation of the latent space through the posterior network, analyzes the causal relationship and decouples the low-dimensional latent variables. The latent variables are decoded into new samples in the original data space through the serialized variational autoencoder, and finally the causal factors of sepsis are analyzed by analyzing the latent variables related to sepsis.

[0077] like Figure 5 In the posterior network shown in the figure, spatial distribution features are also extracted from laboratory test indicators and personal attributes to jointly generate exogenous variables that can indicate sample distribution. The laboratory test indicators, personal attributes and exogenous variables are encoded into , and .Will , and Respectively t The hidden variables at the -1 moment are concatenated, and then the means and variances of the three hidden variables are parameterized through a layer of long short-term memory network (LSTM) and two different fully connected layers.

[0078] In this embodiment, the posterior network is determined by the following formula t The posterior mean and posterior variance of each latent variable at time t.

[0079] .

[0080] .

[0081] .

[0082] in, for t The posterior mean of the latent variable associated with sepsis at time t, for t The posterior variance of the latent variable associated with sepsis at time t, for tThe posterior mean of the latent variables related to other diseases at each moment, for t The posterior variance of the latent variables related to other diseases at each moment, for t The posterior mean of the latent variable that is not related to the disease at any given moment, for t The posterior variance of the latent variable that is independent of the disease at each time step, for t The posterior mean of the latent variable associated with sepsis at time -1, for t The posterior mean of the latent variables related to other diseases at time -1, for t The posterior mean of the latent variable that is not related to the disease at time -1, for t The time distribution information of the exogenous variables at the moment, for t Time distribution information of laboratory test indicators at each moment, for t Time distribution information of personal attributes at each moment, represents the long short-term memory network, , , , , , Represents different fully connected layers.

[0083] Taking the latent variables related to sepsis as an example, , , and The size after splicing is 320, parameterized The LSTM layer has an input size of 320 and an output size of 128, and the two fully connected layers have an input size of 128 and an output size of 64.

[0084] The posterior distribution is: .

[0085] in, is the posterior distribution, are the parameters of the posterior network, for t The first intermediate parameter set at time, , is the second intermediate parameter set, , for t Laboratory test indicators at all times, for t The personal attributes of the moment, forT Clinical texts of the moment, for T The disease label at all times, for T Hidden variables associated with sepsis at all times, for T Hidden variables that are related to other diseases at all times, Represents a continuous multiplication operation.

[0086] Step 307: Using the generated network, according to t The posterior mean and posterior variance of each latent variable at time t are determined t The hidden variables after reparameterization at each moment, and according to t After the hidden variables are reparameterized at each moment, reconstruction t Laboratory test indicators and t The personal attributes of the moment. T The latent variables after reparameterization are constantly reconstructed to reconstruct clinical texts, and based on the latent variables related to sepsis, the strength of association between each group of historical factor data and sepsis is confirmed.

[0087] like Figure 5 In the generative network shown in FIG, after obtaining the posterior mean and posterior variance, the latent variables following the Gaussian distribution can be reparameterized based on the posterior mean and posterior variance.

[0088] In this embodiment, the generation network is determined by the following formula: t The hidden variables after reparameterization at each moment.

[0089] .

[0090] .

[0091] .

[0092] in, for t The latent variables related to sepsis after reparameterization at each moment, for t Hidden variables related to other diseases after constant reparameterization, for t The disease-independent latent variables after reparameterization at all times, for t The posterior mean of the latent variable associated with sepsis at time t, for t The posterior variance of the latent variable associated with sepsis at time t, for t The posterior mean of the latent variables related to other diseases at each moment, for t The posterior variance of the latent variables related to other diseases at each moment, for t The posterior mean of the latent variable that is not related to the disease at any given moment, for t The posterior variance of the latent variable that is independent of the disease at each time step, For one with A random matrix of the same size that follows a normal distribution with mean 0 and variance 1, For one with A random matrix of the same size that follows a normal distribution with mean 0 and variance 1, For one with A random matrix of the same size that follows a normal distribution with mean 0 and variance 1, Represents the multiplication of corresponding points of two matrices.

[0093] The generative network is reconstructed using the following formula t Laboratory test indicators at all times, t Personal attributes of the moment and T Clinical text of the moment.

[0094] .

[0095] .

[0096] .

[0097] in, for t Laboratory testing indicators are constantly reconstructed. for t Personal attributes that are constantly reconstructed, for T Clinical texts that are constantly being reconstructed. for t The latent variables related to sepsis after reparameterization at each moment, for t Hidden variables related to other diseases after constant reparameterization, for t The disease-independent latent variables after reparameterization at all times, Represents a generative model consisting of multiple consecutive fully connected layers.

[0098] For each moment, the generative model is a Gaussian distribution parameterized by a fully connected layer. For example, in the generation process of personal attributes, the three re-parameterized latent variables are concatenated and passed through multiple consecutive fully connected layers as a generative network with an input size of 128 and an output size of 7.

[0099] In this embodiment, laboratory test indicators and personal attributes are reconstructed at each moment, and clinical text is generated only at the last moment. At the same time, after fully learning to extract features from multiple consecutive hours of data, at the last moment, the decoupled latent variables related to sepsis are used for classification to determine T Predicted disease label at the moment: ,in, It is a classification model. It is a low-dimensional latent variable that has been extracted, so there is no need for a complex classification model. Multiple continuous fully connected layers are used, with an input size of 64 and an output size of 2. Finally, the classification result is output through a sigmoid layer. The classification result is represented by the score of the disease label. The higher the score, the stronger the association strength of sepsis in the historical factor data set.

[0100] The distribution of laboratory test indicators is , the distribution of personal attributes is , the distribution of clinical text and disease labels is: .

[0101] The goal of optimizing the network structure is the Evidence Lower Bound (ELBO): .

[0102] According to the causal hidden Markov condition, the joint distribution probability at each moment is: .

[0103] Substituting the joint distribution probability into ELBO, we obtain the optimization objective of the serialized causal decoupling model as described in the following formula.

[0104] .

[0105] .

[0106] .

[0107] .

[0108] in, is the optimization target value of the serialized causal decoupling model, for t The time distribution information of the exogenous variables at the moment, for t -1 time distribution information of exogenous variables, Express expectations, is the first intermediate variable, is the second intermediate variable, describes the divergence between the prior and posterior distributions, is the third intermediate variable, exist did not exist before the moment The posterior distribution at the moment is approximately equal to the prior distribution, which is equal to 0, so Can be omitted.

[0109] Step 308: Reconstruct the laboratory test indicators and personal attributes at each moment in the historical setting period, and T Clinical texts and predicted disease labels reconstructed at each moment are combined with laboratory test indicators, personal attributes, T The clinical text, actual disease label, prior distribution and posterior distribution at each moment are used to optimize the parameters of the prior network and the posterior network by maximizing ELBO to obtain a trained serialized causal decoupling model.

[0110] like Figure 6 The figure shows a schematic diagram of an implementation process of the method for analyzing the causes of sepsis based on matrix causal disentanglement provided by the present invention, firstly obtaining initial data, then preprocessing the initial data, and then performing sepsis cause analysis based on the prior network, the posterior network and the generative network.

[0111] Step 204: Group the current electronic health record time series data to obtain multiple groups of current factor data, and determine the correlation strength between each group of current factor data and sepsis based on the trained serialized causal decoupling model to analyze the causal factors of sepsis.

[0112] It should be noted that the execution subjects of the above steps 201 to 204 are all computers.

[0113] like Figure 7As shown, the present invention extracts laboratory test indicators, personal attributes and clinical texts of ICU patients to establish a data set. The samples in the data set are a complex and entangled high-dimensional space. The present invention decouples the high-dimensional space into low-dimensional hidden variables, and classifies the hidden variables related to sepsis, effectively eliminating the influence of confounding factors on model analysis. A structural causal graph describing the relationship between hidden variables and observed data is constructed using a hidden Markov chain, and a serialized variational autoencoder is used to learn causal features and decouple hidden variables. The samples in the data set are cleaned and normalized, and a training set, a test set and a validation set are established. The appropriate time step and feature dimension are set, and the training set is used to train and optimize the serialized causal decoupling model. After the training is completed, the trained model is evaluated according to the validation set, and the parameters are adjusted again and trained repeatedly to achieve the best effect. Finally, the trained serialized causal decoupling model is applied to the current electronic health record data, and the causal factors of sepsis are analyzed by constructing a causal structure matrix and analyzing the correlation strength between different data and sepsis. The present invention integrates the patient's clinical data and physiological parameters, fully utilizes the temporal and spatial information in ICU patient data, decouples complex causal relationships, improves the interpretability of the results, and achieves accurate prediction of sepsis.

[0114] like Figure 8 As shown, the method for analyzing the causal factors of sepsis based on matrix causal disentanglement provided by the present invention includes the following key steps. First, the initial data is preprocessed to extract key information such as laboratory test indicators, personal attributes, clinical texts, etc. to ensure the quality and consistency of the data. Next, the preprocessed data set is grouped, such as into multiple groups of data such as {G1, G2}, {G1, G3}, etc., to obtain a group set. A causal decoupling analysis is performed on each group of data, and a structural causal graph based on a hidden Markov chain is constructed, and the data is encoded using a serialized variational autoencoder to analyze the causal relationship. Among them, the hidden variables are decoupled into hidden variables related to sepsis, hidden variables related to other diseases, and hidden variables unrelated to the disease. After that, the sepsis prediction score of each group set is obtained, and the size of the score is used to determine which group has factors that are more relevant to sepsis, and then further grouping is performed, and the above operation is repeated to obtain the final factor judgment result. The present invention first constructs a grouping matrix to obtain multiple grouping sets at one time. After multiple group judgments, superposition judgments are performed based on the analyzed results to obtain the final cause analysis results.

[0115] In order to improve the analysis speed, the present invention adopts parallel computing, and at the same time, takes into account the importance of antibiotic use, and distributes it as a key factor in all groups to ensure that it fully considers the relevance of sepsis. Through causal decoupling analysis, the sepsis prediction score of each group is calculated, and the factors that are more relevant to sepsis in each group are judged according to the size of the prediction score. Through repeated screening and grouping, the analysis is gradually refined, and the key causal factors of sepsis are finally determined. After multiple iterative analyses, the analysis results of each group are combined by the method of superposition judgment to form a final causal analysis report. This process can not only effectively avoid the influence of confounding factors and avoid false correlations, but also significantly improve the accuracy of sepsis causal analysis, providing an important reference for clinical diagnosis and treatment. In particular, through parallel computing and comprehensive consideration of antibiotic factors, the present invention can quickly and accurately identify the key factors related to sepsis, thereby improving the efficiency and reliability of causal analysis.

[0116] The feasibility of the sepsis causal factor analysis method based on matrix causal disentanglement provided by the present invention is verified in combination with specific data sets and calculation formulas.

[0117] The data set used to evaluate the early prediction method of sepsis of the present invention is the Medical Information Mart for Intensive Care (MIMIC-IV v2.2) database, which is an EHR format data set, including various types of data such as personal attributes, laboratory test results, vital signs, disease diagnosis, and drug use. In order to save time and cost in clinical practice, when the doctor believes that a certain indicator of the patient is within the normal range, no further testing is performed, so there are a large number of missing values ​​in the database. At the same time, due to irregular operation of the recorder or some unknown problems, there will also be some abnormal values ​​in the database. The present invention cleans the data based on the normal value range provided by Harutyunyan et al., merges all observations in units of one hour, and takes the latest one if the same indicator is tested multiple times within one hour. Arrange according to the size of the missing rate, discard the indicators with too high missing rates, and select the method of filling up and filling down for the missing values. If a patient has no measured value for a certain indicator, it means that the doctor believes that the patient's indicator is normal, so a value within the normal range is filled.

[0118] After data cleaning and filling, select the maximum and minimum standardization method for normalization: .

[0119] in, is the normalized index, For an indicator, For an indicator in the sample The minimum value of For an indicator in the sample The normalized data set is more standardized and free from the influence of data anomalies, thus improving the accuracy of prediction.

[0120] After data processing, we get 3511 positive samples and 17538 negative samples, which are divided into training set, test set and validation set in a ratio of 3:1:1. The maximum number of training times is set to 200 and the batch size is set to 128. The Adam optimizer is used in training, and the initial learning rate is 1e-3, which is reduced to one-fifth of the original rate every 10 training times.

[0121] The sepsis causal factor analysis method based on matrix causal disentanglement proposed in this invention was evaluated on MIMIC-IV. Fig. 9 As shown, the performance is measured by the area under the receiver operator characteristic (ROC) curve (Area Under Curve, AUC) and the F1 score, as shown in the following formula.

[0122] .

[0123] .

[0124] .

[0125] .

[0126] in, is the area under the ROC curve, It is the proportion of samples predicted to be positive that are actually positive, reflecting the accuracy of the model in predicting positive classes. is the proportion of all actual positive classes that are correctly predicted as positive, reflecting the ability of the model to capture positive classes. The F1 score represents the harmonic mean of precision and recall, and is used to measure the balance between precision and recall. , is the prediction score of the positive sample, is the prediction score of the negative sample, is the number of positive samples, is the number of negative samples, TP is the number of true positive samples in the prediction results, FP is the number of false positive samples in the prediction results, and FN is the number of false negative samples in the prediction results.

[0127] The present invention uses Grad-CAM to generate a grayscale image to represent the results of the causal factor analysis. The horizontal axis in the grayscale image represents the number of hours from the time when sepsis occurs, and the vertical axis represents the variable name. The lighter the color, the lower the contribution of the variable to the model prediction of sepsis at this time, and the darker the color, the higher the contribution. When a variable is black at all time points, it means that the variable is clinically important for indicating sepsis.

[0128] Embodiment 2: A computer device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a method for analyzing causal factors of sepsis based on matrix-based causal disentanglement.

[0129] Embodiment 3: A computer-readable storage medium stores a computer program, which, when executed by a processor, implements a method for analyzing causal factors of sepsis based on matrix-based causal disentanglement.

[0130] Embodiment 4: A computer program product, comprising a computer program, which, when executed by a processor, implements a method for analyzing causal factors of sepsis based on matrix-based causal disentanglement.

[0131] Embodiment 5: A computer device, which may be a database, and its internal structure diagram may be as follows: Fig.10 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store pending transactions. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for analyzing the causes of sepsis based on matrix causal disentanglement is implemented.

[0132] It should be noted that the object information (including but not limited to object device information, object personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in the present invention are all information and data authorized by the object or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0133] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided by the present invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided by the present invention may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided by the present invention may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0134] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0135] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A sepsis causal factor analysis device based on matrix causal disentanglement, It is characterized in that The sepsis causal factor analysis device based on matrix-type causal disentanglement includes: A data acquisition module, used to acquire historical electronic health record time series data; the historical electronic health record time series data includes laboratory test indicators, personal attributes, clinical texts and disease labels at multiple times within a historical set period; the disease label is 0 or 1, wherein a disease label of 1 indicates sepsis, and a disease label of 0 indicates no sepsis; A causal graph construction module is connected to the data acquisition module and is used to group the historical electronic health record time series data to obtain multiple groups of historical factor data, and to construct a structural causal graph based on a hidden Markov chain based on each group of historical factor data to obtain a structural causal matrix; the structural causal graph is used to describe the historical factor data into a high-dimensional latent space; A model training module is connected to the causal graph construction module, and is used to train and optimize the serialized causal decoupling model based on the structural causal matrix according to each group of historical factor data, so as to obtain a trained serialized causal decoupling model; the serialized causal decoupling model includes a serialized variational autoencoder, a priori network, a posterior network and a generative network; the serialized variational autoencoder is used to encode the historical factor data; the priori network and the posterior network are used to parse the causal relationship in the structural causal graph according to the encoded historical factor data, and decouple to obtain multiple latent variables; the multiple latent variables are: latent variables related to sepsis, latent variables related to other diseases and latent variables unrelated to the disease; the generative network is used to analyze the correlation strength between the latent variables related to sepsis and sepsis; The factor analysis module is connected to the model training module and is used to group the current electronic health record time series data to obtain multiple groups of current factor data, and determine the correlation strength between each group of current factor data and sepsis based on the trained serialized causal decoupling model to analyze the causal factors of sepsis.

2. The device for analyzing the causes of sepsis based on matrix causal disentanglement according to claim 1, It is characterized in that The model training module includes: The exogenous variable determination submodule is used to extract the spatial distribution characteristics of laboratory test indicators and personal attributes at any time during the historical setting period to determine t exogenous variables at time; according to t -2 time exogenous variables, t -1 time exogenous variables and t The exogenous variable at the moment is encoded by the serialized variational autoencoder, and the t The time distribution information of the exogenous variables at the moment, where t For any time in the historical setting period, 1< t < T , T The final moments of the period set for history; The encoding submodule is used to adopt the serialized variational autoencoder to t -2 laboratory test indicators, t -1 moment laboratory test indicators and t The laboratory test indicators at the time are coded to obtain t The time distribution information of laboratory test indicators at the moment; the serialized variational automatic encoder is used to t -2 moments of personal attributes, t -1 moment of personal attributes and t The personal attributes of the moment are encoded to obtain t Time distribution information of personal attributes at each moment; A priori and posterior submodule, used to adopt the prior network, according to t The time distribution information of the exogenous variables at the moment and t The posterior mean of each latent variable at time -1 is calculated t The prior mean and prior variance of each latent variable at the moment, and define the prior distribution; using the posterior network, according to t The time distribution information of the exogenous variables at the moment, t Time distribution information of laboratory test indicators at each moment, t Time distribution information of personal attributes at each moment and t -1 time, the posterior mean of each latent variable is determined t The posterior mean and posterior variance of the hidden variables at the moment, and define the posterior distribution; The association strength determination submodule is used to adopt the generation network according to t The posterior mean and posterior variance of each latent variable at time t are determined t The hidden variables after reparameterization at each moment, and according to t After the hidden variables are reparameterized at each moment, reconstruction t Laboratory test indicators at all times and t The personal attributes of the moment, according to T The latent variables related to sepsis and other diseases after reparameterization at each moment are reconstructed T Clinical text of the moment, according to T The latent variables related to sepsis after reparameterization at each moment were used to analyze the strength of association between the latent variables related to sepsis and sepsis; The training submodule is used to reconstruct the laboratory test indicators at each moment within the historical set period, the personal attributes reconstructed at each moment, T Clinical texts that are constantly being reconstructed, T The predicted disease label at each moment, the association strength at each moment, the personal attributes at each moment, T Clinical texts at all times, T The diseased label at the moment, the prior distribution and the posterior distribution are optimized with the lower bound of evidence as the optimization target, so as to obtain a trained serialized causal decoupling model.

3. The device for analyzing the causes of sepsis based on matrix causal disentanglement according to claim 2, It is characterized in that The exogenous variable determination submodule is based on t -2 time exogenous variables, t -1 time exogenous variables and t The exogenous variable at the moment is encoded by the serialized variational autoencoder, and the t The time distribution information of exogenous variables at the moment, including: Using a point convolution based encoder, t -2 time exogenous variables, t -1 time exogenous variables and t The exogenous variables at the time are encoded to obtain t Multiple preliminary feature maps at the moment; Will t Multiple preliminary feature maps at the moment are concatenated, and a depth-wise convolution-based encoder is used to encode the concatenated preliminary feature maps to obtain t Final feature map at moment; according to t The final feature map at the moment is calculated using the formula Sure t The time distribution information of the exogenous variables at the moment; among them, for t Time distribution information of exogenous variables at the moment for t The final feature map at time, for t The exogenous variable at time, represents a full connection operation, Represents the shape of the reshaped tensor, Represents the addition of corresponding points of tensors.

4. The device for analyzing the causes of sepsis based on matrix causal disentanglement according to claim 2, It is characterized in that The prior network is determined by the following formula t The prior mean and prior variance of each latent variable at time: ; ; ; in, for t The prior mean of the latent variable associated with sepsis at each time point, for t The prior variance of the latent variable associated with sepsis at each moment, for t The prior mean of the latent variables related to other diseases at each moment, for t The prior variance of the latent variables related to other diseases at each moment, for t The prior mean of the latent variable that is not related to the disease at any time, for t The prior variance of the latent variable that is independent of the disease at each time step, for t The posterior mean of the latent variable associated with sepsis at time -1, for t The posterior mean of the latent variables related to other diseases at time -1, for t The posterior mean of the latent variable that is not related to the disease at time -1, for t The time distribution information of the exogenous variables at the moment, represents a gated recurrent unit, , , , , , Represents different fully connected operations.

5. The device for analyzing the causes of sepsis based on matrix causal disentanglement according to claim 2, It is characterized in that The posterior network is determined by the following formula t The posterior mean and posterior variance of each latent variable at time: ; ; ; in, for t The posterior mean of the latent variable associated with sepsis at time t, for t The posterior variance of the latent variable associated with sepsis at time t, for t The posterior mean of the latent variables related to other diseases at each moment, for t The posterior variance of the latent variables related to other diseases at each moment, for t The posterior mean of the latent variable that is not related to the disease at any given moment, for t The posterior variance of the latent variable that is independent of the disease at each time step, for t The posterior mean of the latent variable associated with sepsis at time -1, for t The posterior mean of the latent variables related to other diseases at time -1, for t The posterior mean of the latent variable that is not related to the disease at time -1, for t The time distribution information of the exogenous variables at the moment, for t Time distribution information of laboratory test indicators at each moment, for t Time distribution information of personal attributes at each moment, represents the long short-term memory network, , , , , , Represents different fully connected operations.

6. The device for analyzing the causes of sepsis based on matrix causal disentanglement according to claim 2, It is characterized in that The generating network is determined by the following formula t The hidden variables after reparameterization at each moment: ; ; ; in, for t The latent variables related to sepsis after reparameterization at each moment, for t Hidden variables related to other diseases after constant reparameterization, for t The disease-independent latent variables after reparameterization at all times, for t The posterior mean of the latent variable associated with sepsis at time t, for t The posterior variance of the latent variable associated with sepsis at time t, for t The posterior mean of the latent variables related to other diseases at each moment, for t The posterior variance of the latent variables related to other diseases at each moment, for t The posterior mean of the latent variable that is not related to the disease at any given moment, for t The posterior variance of the latent variable that is independent of the disease at each time step, For one with A random matrix of the same size that follows a normal distribution with mean 0 and variance 1, For one with A random matrix of the same size that follows a normal distribution with mean 0 and variance 1, For one with A random matrix of the same size that follows a normal distribution with mean 0 and variance 1, It means multiplying the corresponding points of two matrices. Represents the addition of corresponding points of tensors.

7. The device for analyzing the causes of sepsis based on matrix causal disentanglement according to claim 2, It is characterized in that The generative network is reconstructed using the following formula t Laboratory test indicators at all times, t Personal attributes of the moment and T Clinical text of the moment: ; ; ; in, for t Laboratory testing indicators are constantly reconstructed. for t Personal attributes that are constantly reconstructed, for T Clinical texts that are constantly being reconstructed. for t The latent variables related to sepsis after reparameterization at each moment, for t Hidden variables related to other diseases after constant reparameterization, for t The disease-independent latent variables after reparameterization at all times, Represents a generative model consisting of multiple consecutive fully connected layers.

8. The device for analyzing the causes of sepsis based on matrix causal disentanglement according to claim 2, It is characterized in that The prior distribution is: ; in, is the prior distribution, are the parameters of the prior network, for t The set of hidden parameters at time, ={ , }, for t The hidden parameter set at time -1, for t The exogenous variable at time, for t Hidden variables associated with sepsis at all times, for t Hidden variables related to sepsis at time -1, for t Hidden variables that are related to other diseases at all times, for t -1 time is related to other diseases, for t Hidden variables that are not related to the disease at any time, for t -1 hidden variables that are not related to the disease; The posterior distribution is: ; in, is the posterior distribution, are the parameters of the posterior network, for t The first intermediate parameter set at time, , is the second intermediate parameter set, , for t Laboratory test indicators at all times, for t The personal attributes of the moment, for T Clinical texts of the moment, for T The disease label at all times, for T Hidden variables associated with sepsis at all times, for T Hidden variables that are related to other diseases at all times, Represents a continuous multiplication operation.

9. The device for analyzing the causes of sepsis based on matrix causal disentanglement according to claim 8, It is characterized in that The optimization goal of the serialized causal decoupling model is: ; ; ; ; in, is the optimization target value of the serialized causal decoupling model, for t The time distribution information of the exogenous variables at the moment, for t -1 time distribution information of exogenous variables, is the first intermediate variable, is the second intermediate variable, describes the divergence between the prior and posterior distributions, is the third intermediate variable, Express expectations.

Citation Information

Patent Citations

  • Video compressed sensing recovery method and system based on generative network and low-rank regularization

    CN116645280A

  • Deep learning and ChatGPT-based sepsis early detection device

    CN117153422A

  • A computer-implemented method of analysing genetic data about an organism

    US20200402614A1

Cited By

  • Medical time series data anti-fact prediction method and device based on causal decoupling

    CN119742083A

  • A Counterfactual Prediction Method and Device for Medical Time-Series Data Based on Causal Disentanglement

    CN119742083B