CN117476214A - Data management method and system based on hospital information - Google Patents
Data management method and system based on hospital information Download PDFInfo
- Publication number
- CN117476214A CN117476214A CN202311500346.3A CN202311500346A CN117476214A CN 117476214 A CN117476214 A CN 117476214A CN 202311500346 A CN202311500346 A CN 202311500346A CN 117476214 A CN117476214 A CN 117476214A
- Authority
- CN
- China
- Prior art keywords
- vital sign
- data
- patient
- training
- sign data
- 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
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000013523 data management Methods 0.000 title claims abstract description 24
- 230000036772 blood pressure Effects 0.000 claims abstract description 13
- 230000036760 body temperature Effects 0.000 claims abstract description 13
- 230000036387 respiratory rate Effects 0.000 claims abstract description 13
- 230000005856 abnormality Effects 0.000 claims abstract description 12
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 claims abstract description 12
- 239000008280 blood Substances 0.000 claims abstract description 12
- 210000004369 blood Anatomy 0.000 claims abstract description 12
- 229910052760 oxygen Inorganic materials 0.000 claims abstract description 12
- 239000001301 oxygen Substances 0.000 claims abstract description 12
- 239000013598 vector Substances 0.000 claims description 74
- 238000012549 training Methods 0.000 claims description 62
- 238000012546 transfer Methods 0.000 claims description 55
- 239000011159 matrix material Substances 0.000 claims description 48
- 238000013527 convolutional neural network Methods 0.000 claims description 27
- 230000002159 abnormal effect Effects 0.000 claims description 15
- 238000000605 extraction Methods 0.000 claims description 8
- 238000013507 mapping Methods 0.000 claims description 7
- 238000005457 optimization Methods 0.000 claims description 7
- 230000002123 temporal effect Effects 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000013075 data extraction Methods 0.000 claims description 3
- 230000001419 dependent effect Effects 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 19
- 238000012545 processing Methods 0.000 abstract description 18
- 238000004458 analytical method Methods 0.000 abstract description 12
- 238000004422 calculation algorithm Methods 0.000 abstract description 9
- 230000006870 function Effects 0.000 description 14
- 239000000523 sample Substances 0.000 description 12
- 238000009826 distribution Methods 0.000 description 8
- 238000011176 pooling Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 6
- 230000004913 activation Effects 0.000 description 5
- 210000002569 neuron Anatomy 0.000 description 4
- 239000013610 patient sample Substances 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 238000005065 mining Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000010835 comparative analysis Methods 0.000 description 2
- 238000013136 deep learning model Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 230000036541 health Effects 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000003862 health status Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000013450 outlier detection Methods 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 238000012731 temporal analysis Methods 0.000 description 1
- 238000000700 time series analysis Methods 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 238000011269 treatment regimen Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- 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
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
-
- 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)
- Data Mining & Analysis (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Biomedical Technology (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Databases & Information Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Pathology (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- Biophysics (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The utility model discloses a data management method and system based on hospital information, it is through real-time supervision each patient's vital sign data, including blood pressure value, blood oxygen saturation, pulse value, body temperature value and respiratory rate value, and introduce data processing and analysis algorithm at the rear end and carry out the chronogenesis collaborative analysis of these patient's vital sign data, thereby catch chronogenesis association and semantic association between the patient, thereby carry out the medical data anomaly detection of this patient with the medical data of being convenient for to predetermined patient object's contrast analysis, in this way, can utilize the relation between each vital sign data of patient to realize the automated detection to patient medical data, thereby judge whether have the abnormality and take corresponding measure. Thus, the efficiency and the quality of data management can be improved, and reliable data support is provided for medical staff, so that the medical decision and the patient care quality are improved.
Description
Technical Field
The present application relates to the field of data governance, and more particularly, to a data governance method and system based on hospital information.
Background
Data governance refers to the process of managing and controlling data, aimed at ensuring the quality, availability, safety and compliance of the data. In hospital information systems, data management is particularly important because medical records contain a large amount of sensitive patient health information, and it is desirable to ensure the accuracy and integrity of the data to support medical decisions and provide high quality medical services.
Anomaly detection is an important aspect of data management that can help identify and correct anomalies in medical data. Abnormality detection is the identification of data that is significantly inconsistent with the normal mode or the expected mode by statistical analysis of the data. This may help to find anomalies in the data, such as outliers, etc. For example, in a medical record, if vital sign data of a patient is significantly different from most patients, further verification or correction thereof may be required.
Traditional data governance schemes typically require manual intervention and handling of anomalies. However, the amount of data in a hospital information system is enormous and complex, the way of manually processing medical information data for anomaly detection is time consuming and error prone, and manual processing may also be affected by subjective factors, leading to inconsistent results. Moreover, conventional schemes often detect anomalies in medical data based on experience of professionals and predefined rules and patterns, and cannot accommodate changes in data and new anomalies. In addition, in medical records, different vital sign data of a patient often have time sequence association relation, and data at a plurality of time points need to be comprehensively considered to accurately judge abnormal conditions. However, the conventional scheme cannot fully utilize such timing related information, resulting in insufficient accuracy of medical data detection for a patient object.
Accordingly, an optimized hospital information based data governance scheme is desired.
Disclosure of Invention
The present application has been made in order to solve the above technical problems. The embodiment of the application provides a data management method and a system based on hospital information, which monitor vital sign data of each patient in real time, including blood pressure value, blood oxygen saturation, pulse value, body temperature value and respiratory rate value, and introduce data processing and analysis algorithm at the back end to perform time sequence collaborative analysis of the vital sign data of the patients, so as to capture time sequence association and semantic association among the patients, thereby facilitating comparative analysis of medical data of a preset patient object to perform abnormal detection of the medical data of the patient, and further, can utilize association relation among the vital sign data of the patient to realize automatic detection of the medical data of the patient, so as to judge whether the patient is abnormal and take corresponding measures. Thus, the efficiency and the quality of data management can be improved, and reliable data support is provided for medical staff, so that the medical decision and the patient care quality are improved.
According to one aspect of the present application, there is provided a data governance method based on hospital information, comprising:
Extracting vital sign data of a plurality of patients at a plurality of predetermined time points within a predetermined time period from the medical records, wherein the vital sign data comprises a blood pressure value, a blood oxygen saturation, a pulse value, a body temperature value and a respiratory rate value;
carrying out vital sign mode association coding on vital sign data of a plurality of patients at a plurality of preset time points in a preset time period to obtain time sequence semantic association characteristics of the vital sign data of the patients;
acquiring medical data to be checked, wherein the medical data to be checked are vital sign data of a preset patient object at a plurality of preset time points in a preset time period;
extracting vital sign mode features of vital sign data of a plurality of preset time points of the preset patient object in a preset time period to obtain medical data time sequence correlation features to be checked;
mapping the medical data time sequence correlation characteristic to be checked into a high-dimensional space of the patient vital sign data time sequence semantic correlation characteristic to obtain a vital sign transfer correlation characteristic; and
based on the vital sign transfer-related features, it is determined whether there is an abnormality in the medical data to be tested.
According to another aspect of the present application, there is provided a data governance system based on hospital information, comprising:
A data extraction module for extracting vital sign data of a plurality of patients at a plurality of predetermined time points within a predetermined time period from a medical record, wherein the vital sign data comprises a blood pressure value, a blood oxygen saturation, a pulse value, a body temperature value and a respiratory rate value;
the association coding module is used for carrying out vital sign mode association coding on vital sign data of a plurality of patients at a plurality of preset time points in a preset time period so as to obtain time sequence semantic association characteristics of the vital sign data of the patients;
the medical data acquisition module to be checked is used for acquiring medical data to be checked, wherein the medical data to be checked is vital sign data of a preset patient object at a plurality of preset time points in a preset time period;
the feature extraction module is used for extracting vital sign mode features of vital sign data of a plurality of preset time points of the preset patient object in a preset time period to obtain medical data time sequence association features to be verified;
the vital sign transfer associated feature extraction module is used for mapping the medical data time sequence associated features to be verified into a high-dimensional space of the patient vital sign data time sequence semantic associated features to obtain vital sign transfer associated features; and
And the result generation module is used for determining whether the medical data to be tested is abnormal or not based on the vital sign transfer association characteristics.
Compared with the prior art, the data management method and system based on hospital information provided by the application monitor vital sign data of each patient in real time, including blood pressure value, blood oxygen saturation, pulse value, body temperature value and respiratory rate value, introduce data processing and analysis algorithm at the back end to perform time sequence collaborative analysis of the vital sign data of the patients, so as to capture time sequence association and semantic association among the patients, so that medical data of preset patient objects can be compared and analyzed to perform abnormal detection of the medical data of the patients, and in this way, the association relationship among the vital sign data of the patients can be utilized to realize automatic detection of the medical data of the patients, thereby judging whether the patients are abnormal and taking corresponding measures. Thus, the efficiency and the quality of data management can be improved, and reliable data support is provided for medical staff, so that the medical decision and the patient care quality are improved.
Drawings
The foregoing and other objects, features and advantages of the present application will become more apparent from the following more particular description of embodiments of the present application, as illustrated in the accompanying drawings. The accompanying drawings are included to provide a further understanding of embodiments of the application and are incorporated in and constitute a part of this specification, illustrate the application and not constitute a limitation to the application. In the drawings, like reference numerals generally refer to like parts or steps.
FIG. 1 is a flow chart of a data governance method based on hospital information according to an embodiment of the present application;
FIG. 2 is a system architecture diagram of a hospital information based data governance method in accordance with an embodiment of the present application;
FIG. 3 is a flow chart of a training phase of a hospital information based data governance method in accordance with an embodiment of the present application;
FIG. 4 is a flowchart of substep S2 of a hospital information based data governance method in accordance with an embodiment of the present application;
FIG. 5 is a block diagram of a hospital information based data management system according to an embodiment of the present application.
Detailed Description
Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application and not all of the embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein.
As used in this application and in the claims, the terms "a," "an," "the," and/or "the" are not specific to the singular, but may include the plural, unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that the steps and elements are explicitly identified, and they do not constitute an exclusive list, as other steps or elements may be included in a method or apparatus.
Although the present application makes various references to certain modules in a system according to embodiments of the present application, any number of different modules may be used and run on a user terminal and/or server. The modules are merely illustrative, and different aspects of the systems and methods may use different modules.
Flowcharts are used in this application to describe the operations performed by systems according to embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in order precisely. Rather, the various steps may be processed in reverse order or simultaneously, as desired. Also, other operations may be added to or removed from these processes.
Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application and not all of the embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein.
Traditional data governance schemes typically require manual intervention and handling of anomalies. However, the amount of data in a hospital information system is enormous and complex, the way of manually processing medical information data for anomaly detection is time consuming and error prone, and manual processing may also be affected by subjective factors, leading to inconsistent results. Moreover, conventional schemes often detect anomalies in medical data based on experience of professionals and predefined rules and patterns, and cannot accommodate changes in data and new anomalies. In addition, in medical records, different vital sign data of a patient often have time sequence association relation, and data at a plurality of time points need to be comprehensively considered to accurately judge abnormal conditions. However, the conventional scheme cannot fully utilize such timing related information, resulting in insufficient accuracy of medical data detection for a patient object. Accordingly, an optimized hospital information based data governance scheme is desired.
In the technical scheme of the application, a data management method based on hospital information is provided. Fig. 1 is a flow chart of a data governance method based on hospital information according to an embodiment of the present application. Fig. 2 is a system architecture diagram of a data management method based on hospital information according to an embodiment of the present application. As shown in fig. 1 and 2, a data management method based on hospital information according to an embodiment of the present application includes the steps of: s1, extracting vital sign data of a plurality of patients at a plurality of preset time points in a preset time period from medical records, wherein the vital sign data comprise blood pressure values, blood oxygen saturation, pulse values, body temperature values and respiratory rate values; s2, carrying out vital sign mode association coding on vital sign data of a plurality of patients at a plurality of preset time points in a preset time period to obtain time sequence semantic association characteristics of the vital sign data of the patients; s3, acquiring medical data to be checked, wherein the medical data to be checked are vital sign data of a preset patient object at a plurality of preset time points in a preset time period; s4, extracting vital sign mode features of vital sign data of a plurality of preset time points of the preset patient object in a preset time period to obtain medical data time sequence correlation features to be checked; s5, mapping the medical data time sequence correlation characteristic to be verified into a high-dimensional space of the patient vital sign data time sequence semantic correlation characteristic to obtain a vital sign transfer correlation characteristic; and S6, determining whether the medical data to be tested is abnormal or not based on the vital sign transfer association features.
In particular, the S1 extracts vital sign data of a plurality of patients at a plurality of predetermined time points within a predetermined time period from a medical record, wherein the vital sign data includes a blood pressure value, a blood oxygen saturation, a pulse value, a body temperature value, and a respiratory rate value. It should be appreciated that vital sign data is one of the common medical records in a hospital setting, including blood pressure values, blood oxygen saturation, pulse values, body temperature values, respiratory rate values, and the like. By monitoring and analyzing these vital sign data, medical personnel can discover changes in the patient's health condition in time and take appropriate intervention measures.
In particular, the step S2 is to perform vital sign pattern association coding on vital sign data of a plurality of predetermined time points of the plurality of patients in a predetermined time period to obtain time sequence semantic association features of the vital sign data of the patients. In particular, in one specific example of the present application, as shown in fig. 4, the S2 includes: s21, arranging vital sign data of a plurality of preset time points of each patient in a preset time period into a vital sign data time sequence correlation matrix according to a time dimension and a vital sign data sample dimension respectively, and obtaining vital sign data time sequence correlation feature vectors of a plurality of patients through a vital sign mode feature extractor based on a convolutional neural network model; and S22, enabling the patient vital sign data time sequence associated feature vectors to pass through a context encoder based on a converter to obtain global context patient vital sign data time sequence semantic associated feature vectors as the patient vital sign data time sequence semantic associated features.
Specifically, in S21, vital sign data of a plurality of predetermined time points of each patient in a predetermined time period are arranged as a vital sign data time sequence correlation matrix according to a time dimension and a vital sign data sample dimension, and then a vital sign pattern feature extractor based on a convolutional neural network model is used to obtain a plurality of vital sign data time sequence correlation feature vectors of the patient. The vital sign data of each patient not only has a time sequence dynamic change rule in the time dimension, but also has a time sequence collaborative hidden association relation. Therefore, in order to perform time sequence collaborative correlation analysis on different vital sign data of each patient, so as to improve accuracy of detecting abnormality of medical data, in the technical scheme of the application, vital sign data of a plurality of preset time points of each patient in a preset time period are required to be arranged into a vital sign data time sequence correlation matrix according to a time dimension and a vital sign data sample dimension, so that distribution information of different vital sign data of each patient in the time dimension and the sample dimension is integrated, so that characteristic analysis and characterization can be performed later. And then, carrying out feature mining on the vital sign data time sequence correlation matrix through a vital sign mode feature extractor based on a convolutional neural network model so as to extract time sequence collaborative correlation feature distribution information of vital sign data of each patient in a time dimension and a sample dimension, thereby obtaining time sequence correlation feature vectors of vital sign data of a plurality of patients. More specifically, each layer using the vital sign pattern feature extractor based on the convolutional neural network model performs respective processing on input data in forward transfer of the layer: carrying out convolution processing on input data to obtain a convolution characteristic diagram; pooling the convolution feature images based on a feature matrix to obtain pooled feature images; performing nonlinear activation on the pooled feature map to obtain an activated feature map; the output of the last layer of the vital sign mode feature extractor based on the convolutional neural network model is the vital sign data time sequence correlation feature vector of the patient, and the input of the first layer of the vital sign mode feature extractor based on the convolutional neural network model is the vital sign data time sequence correlation matrix.
Notably, convolutional neural networks (Convolutional Neural Network, CNN) are a type of deep learning model that is widely used in the fields of computer vision and image processing. It is excellent in processing data having a mesh structure such as images and audio. The following is the basic concept and gist of convolutional neural networks: the convolutional neural network takes a convolutional layer as a basic construction module. The convolution layer extracts features of the input data by applying a series of learnable filters (also called convolution kernels). The filter performs sliding operation on the input data and obtains a feature map through convolution operation. Each filter may capture a different characteristic of the input data; the pooling layer is used to reduce the size of the feature map and preserve the most important features. Common pooling operations include maximum pooling and average pooling. The pooling operation can reduce the parameter quantity of the model, reduce the calculation complexity and improve the translational invariance of the model to input; activation function: the convolution layer typically applies a nonlinear activation function, such as ReLU, after the convolution operation to introduce nonlinear features and increase the expressive power of the model; full tie layer: after passing through a series of convolution and pooling layers, a fully connected layer is typically used to perform the final classification or regression. The full-connection layer flattens the features of the previous layer into a vector, and is connected with the output layer through a weight matrix to generate a final prediction result; parameter sharing: an important feature of convolutional neural networks is parameter sharing. In the convolution layer, the parameters of each filter are used for processing different positions of input data, so that the number of parameters of a model is reduced, and the efficiency and generalization capability of the model are improved; depth structure: convolutional neural networks are typically composed of multiple convolutional layers and pooled layers, forming a depth structure. The deep architecture allows the network to learn more complex feature representations and to handle more abstract and advanced tasks.
Specifically, the S22 uses the plurality of patient vital sign data time-series associated feature vectors through a context encoder based on a converter to obtain global context patient vital sign data time-series semantic associated feature vectors as the patient vital sign data time-series semantic associated features. In consideration of the fact that in the data management of hospital information, implicit correlation features exist among time sequence cooperative features of vital sign data of each patient, namely, when abnormality detection of medical data of a patient object is actually carried out, the time sequence cooperative correlation features among the vital sign data of all patients need to be integrated, so that feature comparison analysis is carried out, and therefore the accuracy of abnormality detection is improved. Therefore, in the technical scheme of the application, the plurality of patient vital sign data time sequence associated feature vectors are further encoded in a context encoder based on a converter, so that global context associated feature information based on patient samples among the vital sign data time sequence collaborative associated features of each patient is extracted, and therefore global context patient vital sign data time sequence semantic associated feature vectors are obtained. More specifically, the plurality of patient vital sign data time sequence correlation feature vectors are arranged in one dimension to obtain global patient vital sign data time sequence correlation feature vectors; calculating the product between the global patient vital sign data time sequence correlation feature vector and the transpose vector of each patient vital sign data time sequence correlation feature vector in the plurality of patient vital sign data time sequence correlation feature vectors to obtain a plurality of self-attention correlation matrices; respectively carrying out standardization processing on each self-attention correlation matrix in the plurality of self-attention correlation matrices to obtain a plurality of standardized self-attention correlation matrices; obtaining a plurality of probability values by using a Softmax classification function through each normalized self-attention correlation matrix in the normalized self-attention correlation matrices; weighting each patient vital sign data time sequence associated feature vector in the patient vital sign data time sequence associated feature vectors by taking each probability value in the probability values as a weight so as to obtain the context semantic patient vital sign data time sequence associated feature vectors; and cascading the plurality of context semantic patient vital sign data time sequence associated feature vectors to obtain the global context patient vital sign data time sequence semantic associated feature vector.
Notably, the converter-based context encoder is a deep learning model for processing sequence data, with the most commonly used model being the transducer model. Wherein the transducer model is composed of a plurality of encoder layers, each encoder layer containing a multi-headed self-attention mechanism and a feed-forward neural network.
It should be noted that, in other specific examples of the present application, vital sign pattern association encoding may be performed on vital sign data of the plurality of patients at a plurality of predetermined time points within a predetermined period of time in other manners to obtain time-sequence semantic association features of the vital sign data of the patient, for example: vital sign data is collected for a plurality of patients over a predetermined period of time. Such data may include vital sign indicators of blood pressure, heart rate, body temperature, respiratory rate, etc.; the collected vital sign data is preprocessed. This includes data cleaning, missing value processing, outlier detection, denoising, and the like. The quality and the accuracy of the data are ensured; the vital sign data is mode-dependent encoded using a suitable encoding method. This may include extracting features and patterns of vital sign data using machine learning algorithms, time series analysis methods, or domain knowledge; and extracting time sequence semantic association features from the encoded vital sign data. This may include extracting features such as trends, periodicity, mutation points, etc. in the time series, and calculating statistical indicators such as mean, variance, maximum, minimum, etc.; data analysis and modeling is performed using the extracted features. Machine learning algorithms or statistical methods can be applied to explore temporal semantic association features of vital sign data, such as cluster analysis, time series prediction, association rule mining, etc.; the analysis results are interpreted and applied in the monitoring, diagnosis or prediction of patient vital sign data. Based on the analysis results, personalized treatment regimens may be formulated or pre-warning systems provided to monitor the health status of the patient.
In particular, the step S3 is to acquire medical data to be checked, wherein the medical data to be checked is vital sign data of a predetermined patient object at a plurality of predetermined time points within a predetermined time period. It should be understood that, in the actual detection of medical data for a predetermined patient object, firstly medical data to be examined is acquired, wherein the medical data to be examined is vital sign data of the predetermined patient object at a plurality of predetermined time points within a predetermined period of time.
In particular, the step S4 is to extract vital sign pattern features of vital sign data of the predetermined patient object at a plurality of predetermined time points within a predetermined period of time to obtain time sequence correlation features of medical data to be verified. Particularly, in one specific example of the present application, the medical data to be tested is arranged into a vital sign data time sequence correlation matrix according to a time dimension and a vital sign data sample dimension, so that the time sequence collaborative correlation distribution information of each vital sign data in the medical data to be tested in the time and sample dimension is integrated, and then the time sequence collaborative correlation distribution information of each vital sign data in the medical data to be tested is subjected to feature mining through the vital sign pattern feature extractor based on the convolutional neural network model, so that the time sequence collaborative correlation feature distribution information of each vital sign data of the predetermined patient in the time and sample dimension is extracted, and the time sequence collaborative correlation feature vector of the medical data to be tested is obtained. More specifically, each layer using the vital sign pattern feature extractor based on the convolutional neural network model performs respective processing on input data in forward transfer of the layer: carrying out convolution processing on input data to obtain a convolution characteristic diagram; pooling the convolution feature images based on a feature matrix to obtain pooled feature images; performing nonlinear activation on the pooled feature map to obtain an activated feature map; the output of the last layer of the vital sign mode feature extractor based on the convolutional neural network model is the medical data time sequence correlation feature vector to be checked, and the input of the first layer of the vital sign mode feature extractor based on the convolutional neural network model is the vital sign data time sequence correlation matrix.
In particular, the step S5 is to map the medical data time sequence correlation characteristic to be verified into a high-dimensional space of the patient vital sign data time sequence semantic correlation characteristic so as to obtain a vital sign transfer correlation characteristic. That is, in the technical solution of the present application, the time-series association feature vector of the medical data to be checked is used as a query feature vector, and a transfer matrix of the time-series semantic association feature vector of the medical data to be checked relative to the global context patient vital sign data is calculated, so that the time-series collaborative association feature information among the vital sign data of the predetermined patient object is mapped into a high-dimensional space of the time-series pattern collaborative association feature information of the vital signs of the plurality of patients, thereby facilitating anomaly detection of the medical data to be checked of the predetermined patient object.
In particular, the S6, based on the vital sign transfer correlation feature, determines whether there is an abnormality in the medical data to be tested. That is, in the technical solution of the present application, the transfer matrix is passed through a classifier to obtain a classification result, where the classification result is used to indicate whether there is an abnormality in the medical data to be tested. Specifically, in the technical scheme of the application, the classification label of the classifier is an evaluation label for judging whether the medical data to be checked is abnormal or not, so after the classification result is obtained, whether the medical data to be checked is normally detected or not can be based on the classification result. Specifically, the transfer matrix is unfolded into classification feature vectors based on row vectors or column vectors; performing full-connection coding on the classification feature vectors by using a plurality of full-connection layers of the classifier to obtain coded classification feature vectors; and passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
A Classifier (Classifier) refers to a machine learning model or algorithm that is used to classify input data into different categories or labels. The classifier is part of supervised learning, which performs classification tasks by learning mappings from input data to output categories.
The fully connected layer (Fully Connected Layer) is one type of layer commonly found in neural networks. In the fully connected layer, each neuron is connected to all neurons of the upper layer, and each connection has a weight. This means that each neuron in the fully connected layer receives inputs from all neurons in the upper layer, and weights these inputs together, and then passes the result to the next layer.
The Softmax classification function is a commonly used activation function for multi-classification problems. It converts each element of the input vector into a probability value between 0 and 1, and the sum of these probability values equals 1. The Softmax function is commonly used at the output layer of a neural network, and is particularly suited for multi-classification problems, because it can map the network output into probability distributions for individual classes. During the training process, the output of the Softmax function may be used to calculate the loss function and update the network parameters through a back propagation algorithm. Notably, the output of the Softmax function does not change the relative magnitude relationship between elements, but rather normalizes them. Thus, the Softmax function does not change the characteristics of the input vector, but simply converts it into a probability distribution form.
It should be appreciated that training of the convolutional neural network model-based vital sign pattern feature extractor, the converter-based context encoder, and the classifier is required prior to inference using the neural network model described above. That is, in the hospital information-based data governance method of the present application, a training stage is further included for training the vital sign pattern feature extractor based on the convolutional neural network model, the converter-based context encoder, and the classifier.
Fig. 3 is a flowchart of a training phase of a hospital information based data governance method in accordance with an embodiment of the present application. As shown in fig. 3, a data management method based on hospital information according to an embodiment of the present application includes: a training phase comprising: s110, training data are acquired, wherein the training data comprise training vital sign data of a plurality of patients at a plurality of preset time points in a preset time period, medical data to be tested are trained, and whether abnormal true values exist in the medical data to be tested; s120, training vital sign data of a plurality of preset time points of each patient in a preset time period are respectively arranged into a training vital sign data time sequence association matrix according to a time dimension and a vital sign data sample dimension, and then a vital sign mode feature extractor based on a convolutional neural network model is used for obtaining a plurality of training patient vital sign data time sequence association feature vectors; s130, enabling the plurality of time sequence associated feature vectors of vital sign data of the training patients to pass through the context encoder based on the converter to obtain time sequence semantic associated feature vectors of vital sign data of the training global context patients; s140, arranging the medical data to be tested in a training vital sign data time sequence correlation matrix according to a time dimension and a vital sign data sample dimension, and then obtaining a training medical data time sequence correlation feature vector to be tested through the vital sign mode feature extractor based on the convolutional neural network model; s150, taking the time sequence associated feature vector of the medical data to be checked in training as a query feature vector, and calculating a training transfer matrix of the time sequence semantic associated feature vector of the vital sign data of the patient relative to the training global context; s160, passing the training transfer matrix through the classifier to obtain a classification loss function value; and S170, training the vital sign mode feature extractor, the context encoder and the classifier based on the convolutional neural network model based on the classification loss function value and through gradient descent direction propagation, wherein the training transfer feature vector obtained after the training transfer matrix is unfolded is subjected to weight space iterative recursive directional proposal optimization at each iteration of the training.
In particular, in the technical solution of the present application, each patient vital sign data time-sequence associated feature vector expresses a time-sequence-sample cross dimension associated feature of vital sign data of a corresponding patient, so that after the plurality of patient vital sign data time-sequence associated feature vectors pass through a context encoder based on a converter, a context associated representation of vital sign data features in a whole patient sample space can be obtained, and thus, when calculating a transfer matrix of the to-be-verified medical data time-sequence associated feature vector corresponding to a specific patient relative to the global context patient vital sign data time-sequence semantic associated feature vector, the transfer matrix includes domain transfer features of vital sign data features in a single patient sample space-a whole patient sample space, so that when the transfer matrix is classified by a classifier, the weight matrix of the classifier has difficulty in training an effect of converging relative to a class label belonging to a predetermined sample space domain, and the classifier is influenced. Therefore, when classifying the transfer matrix through a classifier, the applicant of the present application performs a weighted space iterative recursive directional proposed optimization on the transfer feature vector obtained after the expansion of the transfer matrix at each iteration, specifically expressed as:
Wherein M is 1 And M 2 The weight matrix of the last iteration and the current iteration are respectively V c Is the training transfer characteristic vector obtained after the training transfer matrix is unfolded,represents matrix multiplication, +. the exponential operation of the vector represents the calculation of a natural exponential function value raised to the power of the eigenvalue of each position in the vector, V' c And representing the optimized training transfer characteristic vector obtained after the optimized training transfer matrix is unfolded. Here, the weighted spatial iterative recursive directed proposed optimization may be performed by transforming the initial transferred feature vector V to be classified c As anchor points, to iterate in weight space based on the weight matrix corresponding to the transfer feature vector V c Anchor footprints (anchor footprints) under different spatial domain scale feature distribution dimensions are obtained as directional proposals (oriented proposal) recursively iterating in the weight space, so that class confidence and local accuracy of weight matrix convergence are improved based on prediction proposals, and training effect of the transfer matrix through a classifier is improved. Therefore, the automatic detection of the medical data of the patient can be realized, whether the medical data of the patient is abnormal or not can be judged, and corresponding measures are taken, so that the efficiency and the quality of data management are improved, and reliable data support is provided for medical staff.
In summary, the data management method based on hospital information according to the embodiment of the application is explained, by monitoring vital sign data of each patient in real time, including blood pressure value, blood oxygen saturation, pulse value, body temperature value and respiratory rate value, and introducing a data processing and analyzing algorithm at the back end to perform time sequence collaborative analysis of the vital sign data of the patients, so as to capture time sequence association and semantic association between the patients, so as to facilitate comparative analysis of medical data of a predetermined patient object to perform abnormal detection of the medical data of the patient, and thus, the association relationship between each vital sign data of the patient can be utilized to realize automatic detection of the medical data of the patient, thereby judging whether the patient is abnormal and taking corresponding measures. Thus, the efficiency and the quality of data management can be improved, and reliable data support is provided for medical staff, so that the medical decision and the patient care quality are improved.
Further, a data management system based on hospital information is also provided.
FIG. 5 is a block diagram of a hospital information based data management system according to an embodiment of the present application. As shown in fig. 5, a hospital information-based data governance system 300 according to an embodiment of the present application includes: a data extraction module 310 for extracting vital sign data of a plurality of patients at a plurality of predetermined time points within a predetermined time period from medical records, wherein the vital sign data includes a blood pressure value, a blood oxygen saturation, a pulse value, a body temperature value, and a respiratory rate value; the association encoding module 320 is configured to perform vital sign pattern association encoding on vital sign data of the plurality of patients at a plurality of predetermined time points within a predetermined time period to obtain time sequence semantic association features of the vital sign data of the patients; a medical data to be checked acquisition module 330, configured to acquire medical data to be checked, where the medical data to be checked is vital sign data of a predetermined patient object at a plurality of predetermined time points within a predetermined time period; the feature extraction module 340 is configured to perform vital sign mode feature extraction on vital sign data of the predetermined patient object at a plurality of predetermined time points within a predetermined time period to obtain a time sequence correlation feature of the medical data to be verified; the vital sign transfer associated feature extraction module 350 is configured to map the to-be-verified medical data time sequence associated feature to a high-dimensional space of the patient vital sign data time sequence semantic associated feature to obtain a vital sign transfer associated feature; and a result generation module 360 for determining whether there is an abnormality in the medical data to be tested based on the vital sign transfer correlation characteristics.
As described above, the hospital information-based data governance system 300 according to the embodiment of the present application may be implemented in various wireless terminals, such as a server or the like having a hospital information-based data governance algorithm. In one possible implementation, hospital information based data management system 300 according to embodiments of the present application may be integrated into a wireless terminal as a software module and/or hardware module. For example, the hospital information-based data governance system 300 may be a software module in the operating system of the wireless terminal or may be an application developed for the wireless terminal; of course, the hospital information based data management system 300 could equally be one of the many hardware modules of the wireless terminal.
Alternatively, in another example, the hospital information-based data governance system 300 and the wireless terminal may also be separate devices, and the hospital information-based data governance system 300 may be connected to the wireless terminal via a wired and/or wireless network and communicate interactive information in accordance with a agreed data format.
The foregoing description of the embodiments of the present disclosure has been presented for purposes of illustration and description, and is not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the various embodiments described. The terminology used herein was chosen in order to best explain the principles of the embodiments, the practical application, or the improvement of technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
Claims (9)
1. A data governance method based on hospital information, comprising:
extracting vital sign data of a plurality of patients at a plurality of predetermined time points within a predetermined time period from the medical records, wherein the vital sign data comprises a blood pressure value, a blood oxygen saturation, a pulse value, a body temperature value and a respiratory rate value;
carrying out vital sign mode association coding on vital sign data of a plurality of patients at a plurality of preset time points in a preset time period to obtain time sequence semantic association characteristics of the vital sign data of the patients;
acquiring medical data to be checked, wherein the medical data to be checked are vital sign data of a preset patient object at a plurality of preset time points in a preset time period;
extracting vital sign mode features of vital sign data of a plurality of preset time points of the preset patient object in a preset time period to obtain medical data time sequence correlation features to be checked;
mapping the medical data time sequence correlation characteristic to be checked into a high-dimensional space of the patient vital sign data time sequence semantic correlation characteristic to obtain a vital sign transfer correlation characteristic; and
based on the vital sign transfer-related features, it is determined whether there is an abnormality in the medical data to be tested.
2. The hospital information-based data governance method of claim 1, wherein vital sign pattern-dependent encoding of vital sign data of the plurality of patients at a plurality of predetermined time points within a predetermined time period to obtain patient vital sign data temporal semantic-dependent features, comprises:
the vital sign data of a plurality of preset time points of each patient in a preset time period are respectively arranged into a vital sign data time sequence correlation matrix according to a time dimension and a vital sign data sample dimension, and then a vital sign mode feature extractor based on a convolutional neural network model is used for obtaining vital sign data time sequence correlation feature vectors of a plurality of patients; and
and the patient vital sign data time sequence associated feature vectors are used as the patient vital sign data time sequence associated feature by a context encoder based on a converter to obtain global context patient vital sign data time sequence associated feature vectors.
3. The hospital information-based data governance method of claim 2, wherein vital sign pattern feature extraction of vital sign data of the predetermined patient subject at a plurality of predetermined time points within a predetermined time period to obtain a medical data timing correlation feature to be verified comprises:
And after the medical data to be checked is arranged into a vital sign data time sequence correlation matrix according to the time dimension and the vital sign data sample dimension, the medical data time sequence correlation feature vector to be checked is obtained through the vital sign mode feature extractor based on the convolutional neural network model and is used as the medical data time sequence correlation feature to be checked.
4. A hospital information based data administration method according to claim 3, wherein mapping the medical data temporal correlation feature to be verified into a high-dimensional space of the patient vital sign data temporal semantic correlation feature to obtain a vital sign transfer correlation feature comprises:
and taking the time sequence associated feature vector of the medical data to be checked as a query feature vector, and calculating a transfer matrix of the time sequence semantic associated feature vector of the medical data relative to the vital sign data of the global context patient as the vital sign transfer associated feature.
5. The hospital information-based data governance method of claim 4, wherein determining whether there is an abnormality in medical data to be tested based on the vital sign transfer-related characteristics comprises:
and the transfer matrix passes through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the medical data to be tested has abnormality or not.
6. The hospital information-based data governance method of claim 5, further comprising the step of training: for training the convolutional neural network model-based vital sign pattern feature extractor, the converter-based context encoder, and the classifier.
7. The hospital information-based data governance method of claim 6, wherein said training step comprises:
acquiring training data, wherein the training data comprises training vital sign data of a plurality of patients at a plurality of preset time points in a preset time period, training medical data to be tested, and whether the medical data to be tested has an abnormal true value or not;
training vital sign data of a plurality of preset time points of each patient in a preset time period are respectively arranged into a training vital sign data time sequence correlation matrix according to a time dimension and a vital sign data sample dimension, and then a vital sign mode feature extractor based on a convolutional neural network model is used for obtaining vital sign data time sequence correlation feature vectors of a plurality of training patients;
passing the plurality of training patient vital sign data time-series associated feature vectors through the converter-based context encoder to obtain a training global context patient vital sign data time-series semantic associated feature vector;
The medical data to be tested are arranged into a training vital sign data time sequence correlation matrix according to the time dimension and the physical sign data sample dimension, and then the training vital sign data time sequence correlation feature vector is obtained through the vital sign mode feature extractor based on the convolutional neural network model;
taking the time sequence associated feature vector of the training medical data to be checked as a query feature vector, and calculating a training transfer matrix of the time sequence semantic associated feature vector of the training medical data to be checked relative to the vital sign data of the training global context patient;
passing the training transfer matrix through the classifier to obtain a classification loss function value;
training the vital sign mode feature extractor, the converter-based context encoder and the classifier based on the classification loss function value and through gradient descent direction propagation, wherein at each iteration of the training, a weight-space iterative recursive directed proposal optimization is performed on training transfer feature vectors obtained after the training transfer matrix expansion.
8. The hospital information-based data governance method according to claim 7, wherein at each iteration of the training, the training transfer feature vector obtained after the training transfer matrix is developed is subjected to weight space iterative recursive directed proposed optimization with the following optimization formula to obtain an optimized training transfer matrix;
Wherein, the optimization formula is:
wherein M is 1 And M 2 The weight matrix of the last iteration and the current iteration are respectively V c Is the training transfer characteristic vector obtained after the training transfer matrix is unfolded,represents matrix multiplication, +. the exponential operation of the vector represents the calculation of a natural exponential function value raised to the power of the eigenvalue of each position in the vector, V' c Representation ofAnd the optimized training transfer characteristic vector is obtained after the optimized training transfer matrix is unfolded.
9. A data management system based on hospital information, comprising:
a data extraction module for extracting vital sign data of a plurality of patients at a plurality of predetermined time points within a predetermined time period from a medical record, wherein the vital sign data comprises a blood pressure value, a blood oxygen saturation, a pulse value, a body temperature value and a respiratory rate value;
the association coding module is used for carrying out vital sign mode association coding on vital sign data of a plurality of patients at a plurality of preset time points in a preset time period so as to obtain time sequence semantic association characteristics of the vital sign data of the patients;
the medical data acquisition module to be checked is used for acquiring medical data to be checked, wherein the medical data to be checked is vital sign data of a preset patient object at a plurality of preset time points in a preset time period;
The feature extraction module is used for extracting vital sign mode features of vital sign data of a plurality of preset time points of the preset patient object in a preset time period to obtain medical data time sequence association features to be verified;
the vital sign transfer associated feature extraction module is used for mapping the medical data time sequence associated features to be verified into a high-dimensional space of the patient vital sign data time sequence semantic associated features to obtain vital sign transfer associated features; and
and the result generation module is used for determining whether the medical data to be tested is abnormal or not based on the vital sign transfer association characteristics.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311500346.3A CN117476214A (en) | 2023-11-10 | 2023-11-10 | Data management method and system based on hospital information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311500346.3A CN117476214A (en) | 2023-11-10 | 2023-11-10 | Data management method and system based on hospital information |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117476214A true CN117476214A (en) | 2024-01-30 |
Family
ID=89632738
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311500346.3A Pending CN117476214A (en) | 2023-11-10 | 2023-11-10 | Data management method and system based on hospital information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117476214A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117672532A (en) * | 2024-01-31 | 2024-03-08 | 吉林大学 | Hospitalized patient nursing risk assessment early warning monitoring system and method |
| CN117850601A (en) * | 2024-03-08 | 2024-04-09 | 南昌大学第二附属医院 | System and method for automatically detecting vital signs of handheld PDA |
| CN118098603A (en) * | 2024-04-26 | 2024-05-28 | 吉林大学 | Intensive care clinical information system and method for eicu |
| CN118395135A (en) * | 2024-02-02 | 2024-07-26 | 简帷(杭州)软件有限公司 | Data preprocessing system and method for supply chain planning system |
| CN118447991A (en) * | 2024-07-08 | 2024-08-06 | 吉林大学 | Hemodialysis nursing quality index data reporting management system and terminal equipment |
| CN118507074A (en) * | 2024-07-16 | 2024-08-16 | 吉林大学 | Intelligent nursing system and method for tumor patients in radiotherapy and chemotherapy period |
-
2023
- 2023-11-10 CN CN202311500346.3A patent/CN117476214A/en active Pending
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117672532A (en) * | 2024-01-31 | 2024-03-08 | 吉林大学 | Hospitalized patient nursing risk assessment early warning monitoring system and method |
| CN117672532B (en) * | 2024-01-31 | 2024-04-05 | 吉林大学 | Hospitalized patient nursing risk assessment early warning monitoring system and method |
| CN118395135A (en) * | 2024-02-02 | 2024-07-26 | 简帷(杭州)软件有限公司 | Data preprocessing system and method for supply chain planning system |
| CN117850601A (en) * | 2024-03-08 | 2024-04-09 | 南昌大学第二附属医院 | System and method for automatically detecting vital signs of handheld PDA |
| CN117850601B (en) * | 2024-03-08 | 2024-05-14 | 南昌大学第二附属医院 | System and method for automatically detecting vital signs of handheld PDA |
| CN118098603A (en) * | 2024-04-26 | 2024-05-28 | 吉林大学 | Intensive care clinical information system and method for eicu |
| CN118447991A (en) * | 2024-07-08 | 2024-08-06 | 吉林大学 | Hemodialysis nursing quality index data reporting management system and terminal equipment |
| CN118507074A (en) * | 2024-07-16 | 2024-08-16 | 吉林大学 | Intelligent nursing system and method for tumor patients in radiotherapy and chemotherapy period |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117476214A (en) | Data management method and system based on hospital information | |
| JP2022037241A (en) | Abnormality detection system, abnormality detection method, abnormality detection program, and method for generating learned model | |
| WO2021257128A2 (en) | Quantum computing based deep learning for detection, diagnosis and other applications | |
| CN110136842A (en) | Morbidity prediction technique, device and the computer readable storage medium of acute infectious disease | |
| CN111009321A (en) | Application method of machine learning classification model in juvenile autism auxiliary diagnosis | |
| CN117598700B (en) | Intelligent blood oxygen saturation detection system and method | |
| CN118098603B (en) | Intensive care clinical information system and method for eicu | |
| CN117542524A (en) | Intelligent nursing information processing system and method | |
| CN117542474A (en) | Remote nursing monitoring system and method based on big data | |
| CN117598674B (en) | Multi-parameter heart function monitoring system and method | |
| CN117643461B (en) | Heart rate intelligent monitoring system and method based on artificial intelligence | |
| CN117648643A (en) | Rigging predictive diagnosis method and device based on artificial intelligence | |
| CN116929815A (en) | Equipment working state monitoring system and method based on Internet of things | |
| WO2023108418A1 (en) | Brain atlas construction and neural circuit detection method and related product | |
| CN110378394A (en) | More physiological data convergence analysis methods neural network based | |
| CN117257302B (en) | Personnel mental health state assessment method and system | |
| CN117393153B (en) | Shock real-time risk early warning and monitoring method and system based on medical internet of things time sequence data and deep learning algorithm | |
| CN118016279A (en) | Analysis diagnosis and treatment platform based on artificial intelligence multi-mode technology in breast cancer field | |
| CN117831743A (en) | Household care health monitoring management system and method based on Internet of things technology | |
| JP2021192155A (en) | Program, method and system for supporting abnormality detection | |
| CN117530684B (en) | Blood glucose abnormality detection and early warning system and method based on health big data | |
| CN117079017A (en) | Credible small sample image identification and classification method | |
| CN114611556B (en) | Multi-category motor imagery task identification method based on graph neural network | |
| Rajmohan et al. | G-Sep: A deep learning algorithm for detection of long-term sepsis using bidirectional gated recurrent unit | |
| CN117122286B (en) | Intelligent rehabilitation equipment |
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 |