CN113192627A - Patient and disease bipartite graph-based readmission prediction method and system - Google Patents
Patient and disease bipartite graph-based readmission prediction method and system Download PDFInfo
- Publication number
- CN113192627A CN113192627A CN202110396159.XA CN202110396159A CN113192627A CN 113192627 A CN113192627 A CN 113192627A CN 202110396159 A CN202110396159 A CN 202110396159A CN 113192627 A CN113192627 A CN 113192627A
- Authority
- CN
- China
- Prior art keywords
- patient
- patients
- disease
- bipartite graph
- medical
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 201000010099 disease Diseases 0.000 title claims abstract description 82
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 title claims abstract description 82
- 238000000034 method Methods 0.000 title claims description 23
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 21
- 238000012545 processing Methods 0.000 claims abstract description 8
- 230000006399 behavior Effects 0.000 claims description 21
- 230000002776 aggregation Effects 0.000 claims description 10
- 238000004220 aggregation Methods 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 6
- 230000004931 aggregating effect Effects 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 description 6
- 238000011160 research Methods 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 3
- 230000036541 health Effects 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 230000017105 transposition Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Data Mining & Analysis (AREA)
- Public Health (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Epidemiology (AREA)
- Pathology (AREA)
- Databases & Information Systems (AREA)
- Primary Health Care (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The present disclosure provides a patient and disease bipartite graph-based readmission prediction model, comprising: acquiring historical medical data of patients, and serializing the medical events of each patient by using the timestamps of diseases based on the historical medical data to obtain medical event sequence sets of all patients; constructing a bipartite graph of the patients and the medical events based on the medical event sequence sets of all the patients, expressing the historical illness information of the patients and establishing indirect connection between the patients suffering from the same illness; processing the bipartite graph of the patients and the medical events by using an embedded generation algorithm to obtain a final vector representation of each patient; predicting a readmission behavior of the patient based on the final vector representation.
Description
Technical Field
The disclosure belongs to the technical field of computers, and particularly relates to a patient and disease bipartite graph-based readmission prediction method and system.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
The study of a predictive model for readmission based on medical data is a current technique that is focused on a behavioral model for predicting whether a patient will be readmitted for a particular disease in a future period of time. Readmission events occur frequently and are costly, placing a significant burden on both the patient and the medical system. The rate of readmission, which is an important index for measuring the quality of medical services, is receiving increasing attention from governments, medical insurance companies, medical institutions, and patients. Therefore, dealing with the readmission prediction problem from the data analysis perspective is receiving increasing attention in the research field. Accurate prediction of readmission not only helps patients to know own physical health condition, guides healthy life behaviors. Therefore, how to build a model to accurately predict patient readmission becomes a critical issue.
The inventor finds that, in the research, most of the early researches use various regression techniques to establish a prediction model after reviewing the previous researches on the readmission prediction, and in recent decades, machine learning algorithms such as a decision tree and a support vector machine are increasingly widely applied to the readmission prediction. Recently, with the rise of deep learning, people use the recurrent neural network or its variants in deep learning to predict the readmission behavior using the serialized medical health record.
In summary, current approaches to dealing with readmission prediction problems typically use traditional machine learning classification algorithms, or utilize recurrent neural networks and their variants in deep learning. However, most of the existing methods only use the characteristic information of the patients to perform prediction, and the potential correlation relationship between the patients is rarely considered, so that the robustness of the prediction effect for similar people is not good, and especially, the patient with short hospitalization history, i.e. less case data, is difficult to perform the re-hospitalization prediction, so that the behavior prediction of how to perform re-hospitalization is the technical problem solved by the application for the situations of short hospitalization history data and less case data.
Disclosure of Invention
In order to overcome the defects of the prior art, the present disclosure provides a method and a system for predicting readmission based on a bipartite graph of a patient and a disease, wherein the bipartite graph of the patient and the disease is constructed according to medical data of the patient, and the readmission behavior is predicted based on an embedded generation algorithm.
In order to achieve the above object, one or more embodiments of the present disclosure provide the following technical solutions:
in one aspect, a patient and disease bipartite graph-based readmission prediction method is disclosed, comprising:
acquiring historical medical data of patients, and serializing the medical events of each patient by using the timestamps of diseases based on the historical medical data to obtain medical event sequence sets of all patients;
constructing a bipartite graph of the patients and the medical events based on the medical event sequence sets of all the patients, expressing the historical illness information of the patients and establishing indirect connection between the patients suffering from the same illness;
processing the bipartite graph of the patients and the medical events by using an embedded generation algorithm to obtain a final vector representation of each patient;
predicting a readmission behavior of the patient based on the final vector representation.
Further, the medical events of each patient are serialized using time stamps of the disease, each disease sequence snExpressed as:
wherein L isnIs a disease sequence snLength of (d)niThe ith disease, t, confirmed by the nth patient is shownniThe time stamp of when the nth patient confirmed the ith disease is indicated, and for all p<q,tnp≤tnq。
In a further technical scheme, when constructing the bipartite graph of the patient and the medical event, all the patients are used as two partsSet of patient nodes of graph, set of diseasesThe set of disease nodes after deduplication as a bipartite graph, the edges between the patient and the disease indicate that the patient has suffered from the disease.
In another aspect, a hospitalization behavior prediction system based on an embedded generation algorithm is disclosed, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the program, including:
acquiring historical medical data of patients, and serializing the medical events of each patient by using the timestamps of diseases based on the historical medical data to obtain medical event sequence sets of all patients;
constructing a bipartite graph of the patients and the medical events based on the medical event sequence sets of all the patients, expressing the historical illness information of the patients and establishing indirect connection between the patients suffering from the same illness;
processing the bipartite graph of the patients and the medical events by using an embedded generation algorithm to obtain a final vector representation of each patient;
predicting a readmission behavior of the patient based on the final vector representation.
In another aspect, a rehospitalization behavior prediction apparatus based on an embedded generation algorithm is disclosed, comprising:
a data serialization module configured to: acquiring historical medical data of patients, and serializing the medical events of each patient by using the timestamps of diseases based on the historical medical data to obtain medical event sequence sets of all patients;
a bipartite graph construction module configured to: constructing a bipartite graph of the patients and the medical events based on the medical event sequence sets of all the patients, expressing the historical illness information of the patients and establishing indirect connection between the patients suffering from the same illness;
a readmission behavior prediction module configured to: processing the bipartite graph of the patients and the medical events by using an embedded generation algorithm to obtain a final vector representation of each patient;
predicting a readmission behavior of the patient based on the final vector representation.
The above one or more technical solutions have the following beneficial effects:
the technical scheme includes that medical events of all patients are serialized by utilizing timestamps of diseases to obtain medical event sequence sets of all patients, a bipartite graph of the patients and the medical events is constructed based on the medical event sequence sets of all the patients, and original data are compressed in the representation form of the bipartite graph to enable the bipartite graph to be more compact, so that the problems of sparsity of medical data and short medical history data and few case data are effectively solved, the bipartite graph of the patients and the medical events is provided for scattered data of each patient, and medical history conditions of all the patients can be comprehensively displayed.
Most of the previous methods for predicting readmission are not good, but only use the characteristic information of the patient, and particularly, the readmission prediction is difficult for the patient with short hospitalization history, i.e. less case data. The invention can express the historical illness information of the patient and the indirect relation between the patients suffering from the same illness by constructing the bipartite graph of the patient and the illness, thereby utilizing the characteristic information of other potential related patients to assist in the prediction of readmission and improving the prediction effect.
Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure and are not to limit the disclosure.
FIG. 1 is a flow chart of a patient and disease bipartite graph based readmission prediction method.
Detailed Description
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the disclosure. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments according to the present disclosure. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
The embodiments and features of the embodiments in the present disclosure may be combined with each other without conflict.
Referring to fig. 1, the present embodiment discloses a hospital readmission prediction method based on a patient and disease bipartite graph, which constructs a patient and disease bipartite graph according to medical data of a patient, obtains a final vector representation of the patient through an embedded generation algorithm of the patient and disease bipartite graph, and performs prediction 5 by using the final vector representation of the patient.
The method specifically comprises the following steps:
the method comprises the following steps: constructing a patient and disease bipartite graph, first extracting all diseases from the patient's medical historyName of (1) Scale and time stampThe medical history of each patient is serialized according to the time stamp of the disease to obtain a disease sequence set, which is like { sn1,2,3, N, where N is the number of disease sequences. Each disease sequence snWatch capable of showingWherein L isnIs a disease sequence snLength of (d)niThe ith disease, t, confirmed by the nth patient is shownniThe time stamp of when the nth patient confirmed the ith disease is indicated, and for all p<q,tnp≤tnq. Then, a patient-to-disease bipartite graph is constructed based on the disease sequence sets of all patients. All patients are taken as a patient node set of the bipartite graph, anddisease collection And after the duplication removal, the disease node set is used as a bipartite graph. The border between the patient and the disease indicates that the patient has suffered from the disease. Then, the reciprocal of the time interval is normalized as the weight of the edge,the time interval here refers to the time difference between the patient's predicted time of illness and the patient's time of illness, the weighting of edges will be used later to indicate the degree of importance of a feature in a certain situation, the higher the final score, the easier it is to send And (4) generating.
Specifically, the weight formula for calculating the edge is as follows:
eija weight representing the edge between patient i and disease j; t is tipRepresents the predicted time of patient i; t is tijIndicating the time when patient i suffers from disease j; n is a radical ofiRepresenting all neighbor nodes of patient i, i.e., all diseases patient i has.
Step two: information from the disease and potentially related patients is aggregated for each patient by an embedded generation algorithm of the patient and disease bipartite graph, resulting in a final vector representation for each patient. The algorithm aggregates neighborhood information of each node layer by layer, a patient node aggregates neighbor node characteristic information by using a patient aggregation module, a disease node aggregates neighbor node characteristic information by using a disease aggregation module, and the patient node and the disease node acquire more and more information from farther positions of a patient bipartite graph and a disease bipartite graph along with the continuation of iteration. The inputs of the two types of combination modules are the same, and are both two groups of node characteristics, hp={hp1,hp2,...,hpm},Is a patient node feature, where m is a patient nodeNumber of dots, FpIs the dimension of each patient node feature. h isd={hd1,hd2,...,hdn},Is a disease node characteristic, where n is the number of disease nodes, FdIs the dimension of each disease node feature. They will generate two sets of new node signatures h 'respectively'p={h′p1,h′p2,...,h′pm},And h'd={h′d1,h′d2,...,h′dm},As its output.
Specifically, the formula of the edge weight, the patient aggregation module and the disease aggregation module is as follows:
1) the weight update formula of the edge:
wherein · -TRepresenting transposition, | | representing join operation, NpiRepresenting the neighbors of the patient node i, hpiIs a patient node characteristic, hdjIs a disease node characteristic, eijIs the current attention coefficient, which is the attention coefficient of the previous layer, and if it is at the first layer, it is the initial weight of the edge.
2) A patient aggregation module:
3) a disease aggregation module:
whereinFeatures representing patient node i at level l-1;represents the characteristics of disease node j at layer l-1;is the weight of the edge connecting the node j and the node i on the l-1 layer; wl-1The weight matrix corresponding to the l-1 layer is used for transmitting information between different layers of the model; α is the previously-spoken attention coefficient calculation function; σ is a non-linear function.
Step three: the aggregated information is used to predict a patient's readmission. Aggregating information from the disease and related patients for each patient using a patient-to-disease bipartite graph embedding generation algorithm, wherein the aggregated information is used by a softmax layer to predict patient readmissionFor the last layer of the full connection layer, corresponding data is converted into a model Values enclosed between [ 0-1 ] and the sum being one, can be approximated as a probability, indicating the likelihood of a certain condition occurring。
The specific calculation process is as follows:
1) using the patient final vector zpAs an input to softmax, an output value between 0 and 1 is obtained.All knots The result sum is 1, resulting in similar probability results, the larger the value, the more likely.
y′=softmax(zp)
2) The loss function can be obtained according to the following formula:
the function of this function is to discriminate the difference between the calculated result and the true value, the larger the loss function is,the larger the difference from the true value is, the smaller the opposite. It can be used to amplify values close to the true value to reduce the loss function to approach the true value。
In summary, according to the embodiments of the present invention, the patient's readmission condition is predicted by the readmission prediction method based on the patient and the disease bipartite graph, the potential association relationship between the patients is considered, the health information of the patients is fully utilized, and the prediction accuracy is improved.
The object of the present embodiment is to provide a hospitalization behavior prediction system based on embedded generation algorithm, which includes a memory, a processor and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the specific steps of the above embodiment.
An object of the present embodiment is to provide a computer-readable storage medium.
A computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, performs the specific steps of the above-described implementation example.
The embodiment aims to provide a rehospitalization behavior prediction device based on an embedded generation algorithm, which comprises:
a data serialization module configured to: acquiring historical medical data of patients, and serializing the medical events of each patient by using the timestamps of diseases based on the historical medical data to obtain medical event sequence sets of all patients;
a bipartite graph construction module configured to: constructing a bipartite graph of the patients and the medical events based on the medical event sequence sets of all the patients, expressing the historical illness information of the patients and establishing indirect connection between the patients suffering from the same illness;
a readmission behavior prediction module configured to: processing the bipartite graph of the patients and the medical events by using an embedded generation algorithm to obtain a final vector representation of each patient;
predicting a readmission behavior of the patient based on the final vector representation.
The steps involved in the apparatus of the above embodiment correspond to the first embodiment of the method, and the detailed description thereof can be found in the relevant description of the first embodiment. The term "computer-readable storage medium" should be taken to include a single medium or multiple media containing one or more sets of instructions; it should also be understood to include any medium that is capable of storing, encoding or carrying a set of instructions for execution by a processor and that cause the processor to perform any of the methods of the present disclosure.
Those skilled in the art will appreciate that the modules or steps of the present disclosure described above can be implemented using general purpose computer means, or alternatively, they can be implemented using program code executable by computing means, whereby the modules or steps may be stored in memory means for execution by the computing means, or separately fabricated into individual integrated circuit modules, or multiple modules or steps thereof may be fabricated into a single integrated circuit module. The present disclosure is not limited to any specific combination of hardware and software.
The above description is only a preferred embodiment of the present disclosure and is not intended to limit the present disclosure, and various modifications and changes may be made to the present disclosure by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present disclosure should be included in the protection scope of the present disclosure.
Although the present disclosure has been described with reference to specific embodiments, it should be understood that the scope of the present disclosure is not limited thereto, and those skilled in the art will appreciate that various modifications and changes can be made without departing from the spirit and scope of the present disclosure.
Claims (10)
1. A patient and disease bipartite graph-based readmission prediction method, comprising:
acquiring historical medical data of patients, and serializing the medical events of each patient by using the timestamps of diseases based on the historical medical data to obtain medical event sequence sets of all patients;
constructing a bipartite graph of the patients and the medical events based on the medical event sequence sets of all the patients, expressing the historical illness information of the patients and establishing indirect connection between the patients suffering from the same illness;
processing the bipartite graph of the patients and the medical events by using an embedded generation algorithm to obtain a final vector representation of each patient;
predicting a readmission behavior of the patient based on the final vector representation.
2. The bi-graph patient-to-disease based readmission prediction method of claim 1, wherein the time stamps of the disease are used to sequence the medical events of each patient, and each disease sequence snExpressed as:
3. The method of claim 1, wherein constructing the bipartite graph of patients and medical events comprises constructing the bipartite graph of patients and medical events by using all patients as a set of patient nodes in the bipartite graph and using the set of diseases as a set of nodes in the bipartite graphThe set of disease nodes after deduplication as a bipartite graph, the edges between the patient and the disease indicate that the patient has suffered from the disease.
4. A patient-to-disease bipartite graph-based readmission prediction method as claimed in claim 1, wherein the final vector is used as input for softmax, resulting in an output value between 0 and 1.
5. The method of claim 1, wherein the reciprocal of the time interval is normalized as a weight of the patient-to-medical event bipartite graph edge.
6. A rehospitalization behavior prediction device based on an embedded generation algorithm is characterized by comprising:
a data serialization module configured to: acquiring historical medical data of patients, and serializing the medical events of each patient by using the timestamps of diseases based on the historical medical data to obtain medical event sequence sets of all patients;
a bipartite graph construction module configured to: constructing a bipartite graph of the patients and the medical events based on the medical event sequence sets of all the patients, expressing the historical illness information of the patients and establishing indirect connection between the patients suffering from the same illness;
a readmission behavior prediction module configured to: processing the bipartite graph of the patients and the medical events by using an embedded generation algorithm to obtain a final vector representation of each patient;
predicting a readmission behavior of the patient based on the final vector representation.
7. The embedded generation algorithm based rehospitalization behavior prediction device of claim 6, wherein said bipartite graph building module constructs a bipartite graph of patients and medical events, takes all patients as a patient node set of the bipartite graph, and takes a disease set as a disease setThe set of disease nodes after deduplication as a bipartite graph, the edges between the patient and the disease indicate that the patient has suffered from the disease.
8. The apparatus of claim 7, further comprising: a patient aggregation module and a disease aggregation module;
the patient aggregation module is configured to: aggregating the characteristic information of the neighbor nodes of the patient nodes;
the disease aggregation module is configured to: and aggregating the disease nodes with the characteristic information of the neighbor nodes.
9. A rehospitalization behavior prediction system based on an embedded generation algorithm, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the following steps when executing the program, comprising:
acquiring historical medical data of patients, and serializing the medical events of each patient by using the timestamps of diseases based on the historical medical data to obtain medical event sequence sets of all patients;
constructing a bipartite graph of the patients and the medical events based on the medical event sequence sets of all the patients, expressing the historical illness information of the patients and establishing indirect connection between the patients suffering from the same illness;
processing the bipartite graph of the patients and the medical events by using an embedded generation algorithm to obtain a final vector representation of each patient;
predicting a readmission behavior of the patient based on the final vector representation.
10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method steps of any one of the preceding claims 1-6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110396159.XA CN113192627A (en) | 2021-04-13 | 2021-04-13 | Patient and disease bipartite graph-based readmission prediction method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110396159.XA CN113192627A (en) | 2021-04-13 | 2021-04-13 | Patient and disease bipartite graph-based readmission prediction method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113192627A true CN113192627A (en) | 2021-07-30 |
Family
ID=76975640
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110396159.XA Pending CN113192627A (en) | 2021-04-13 | 2021-04-13 | Patient and disease bipartite graph-based readmission prediction method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113192627A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114496283A (en) * | 2022-02-15 | 2022-05-13 | 山东大学 | Disease prediction system based on path reasoning, storage medium and equipment |
WO2022136978A1 (en) * | 2020-12-22 | 2022-06-30 | International Business Machines Corporation | Medical event prediction by health record monitoring |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107591204A (en) * | 2017-08-01 | 2018-01-16 | 山东大学 | Be in hospital again Forecasting Methodology and system based on sequential portrait figure |
CN110310740A (en) * | 2019-04-15 | 2019-10-08 | 山东大学 | Based on see a doctor again information forecasting method and the system for intersecting attention neural network |
-
2021
- 2021-04-13 CN CN202110396159.XA patent/CN113192627A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107591204A (en) * | 2017-08-01 | 2018-01-16 | 山东大学 | Be in hospital again Forecasting Methodology and system based on sequential portrait figure |
CN110310740A (en) * | 2019-04-15 | 2019-10-08 | 山东大学 | Based on see a doctor again information forecasting method and the system for intersecting attention neural network |
Non-Patent Citations (2)
Title |
---|
刘志奇: "基于病人疾病二部图的再入院预测研究", 《中国优秀博硕士学位论文全文数据库(硕士)基础科学辑》 * |
藤田一弥等 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2022136978A1 (en) * | 2020-12-22 | 2022-06-30 | International Business Machines Corporation | Medical event prediction by health record monitoring |
GB2617947A (en) * | 2020-12-22 | 2023-10-25 | Ibm | Medical event prediction by health record monitoring |
CN114496283A (en) * | 2022-02-15 | 2022-05-13 | 山东大学 | Disease prediction system based on path reasoning, storage medium and equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Sun et al. | Disease prediction via graph neural networks | |
CN109659033B (en) | Chronic disease state of an illness change event prediction device based on recurrent neural network | |
WO2018171533A1 (en) | Review machine learning system | |
CN113196314B (en) | Adapting a predictive model | |
Zheng et al. | Resolving the bias in electronic medical records | |
CN104572583B (en) | Method and system for data densification | |
US20110112380A1 (en) | Method and System for Optimal Estimation in Medical Diagnosis | |
Zheng et al. | Capturing feature-level irregularity in disease progression modeling | |
CN113192627A (en) | Patient and disease bipartite graph-based readmission prediction method and system | |
TWI814154B (en) | Method for predicting disease based on medical image | |
CN109213871A (en) | Patient information knowledge mapping construction method, readable storage medium storing program for executing and terminal | |
CN112925857A (en) | Digital information driven system and method for predicting associations based on predicate type | |
WO2023025956A1 (en) | Method and system for deconvolution of bulk rna-sequencing data | |
CN110335160B (en) | Medical care migration behavior prediction method and system based on grouping and attention improvement Bi-GRU | |
WO2014130287A1 (en) | Method and system for propagating labels to patient encounter data | |
Teo et al. | Current trends in readmission prediction: an overview of approaches | |
Ranjini et al. | An ensemble of heterogeneous incremental classifiers for assisted reproductive technology outcome prediction | |
Sampath et al. | Ensemble Nonlinear Machine Learning Model for Chronic Kidney Diseases Prediction | |
CN116403728B (en) | Data processing device for medical treatment data and related equipment | |
Bhat et al. | An efficient prediction model for diabetic database using soft computing techniques | |
Andersson | Predicting Patient Length Of Stay at Time of Admission Using Machine Learning | |
Sumathi et al. | Machine learning based pattern detection technique for diabetes mellitus prediction | |
Sisimayi et al. | AI-enabled case detection model for infectious disease outbreaks in resource-limited settings | |
CN115240843A (en) | Fairness prediction system based on structure causal model | |
JP2024500470A (en) | Lesion analysis methods in medical images |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210730 |
|
RJ01 | Rejection of invention patent application after publication |