CN112768090A - Filtering system and method for chronic disease detection and risk assessment - Google Patents
Filtering system and method for chronic disease detection and risk assessment Download PDFInfo
- Publication number
- CN112768090A CN112768090A CN202110009976.5A CN202110009976A CN112768090A CN 112768090 A CN112768090 A CN 112768090A CN 202110009976 A CN202110009976 A CN 202110009976A CN 112768090 A CN112768090 A CN 112768090A
- Authority
- CN
- China
- Prior art keywords
- data
- weight
- chronic disease
- characteristic
- module
- 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
- 208000017667 Chronic Disease Diseases 0.000 title claims abstract description 60
- 238000001914 filtration Methods 0.000 title claims abstract description 39
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000001514 detection method Methods 0.000 title claims abstract description 22
- 238000012502 risk assessment Methods 0.000 title claims abstract description 19
- 230000036541 health Effects 0.000 claims abstract description 45
- 238000013135 deep learning Methods 0.000 claims abstract description 24
- 238000010276 construction Methods 0.000 claims abstract description 15
- 238000000605 extraction Methods 0.000 claims abstract description 14
- 201000010099 disease Diseases 0.000 claims description 15
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims description 15
- 230000002776 aggregation Effects 0.000 claims description 13
- 238000004220 aggregation Methods 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 13
- 238000012935 Averaging Methods 0.000 claims description 8
- 238000005457 optimization Methods 0.000 claims description 6
- 208000026350 Inborn Genetic disease Diseases 0.000 claims description 5
- 208000016361 genetic disease Diseases 0.000 claims description 5
- 238000012549 training Methods 0.000 abstract description 10
- 230000000694 effects Effects 0.000 abstract description 5
- 230000008569 process Effects 0.000 abstract description 5
- 238000007781 pre-processing Methods 0.000 abstract description 2
- 230000015654 memory Effects 0.000 description 15
- 238000010586 diagram Methods 0.000 description 6
- 239000013598 vector Substances 0.000 description 6
- 238000013473 artificial intelligence Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 238000003860 storage Methods 0.000 description 5
- 238000004590 computer program Methods 0.000 description 4
- 230000005291 magnetic effect Effects 0.000 description 4
- 238000004140 cleaning Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 208000024172 Cardiovascular disease Diseases 0.000 description 2
- 206010020772 Hypertension Diseases 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 206010012601 diabetes mellitus Diseases 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 208000031662 Noncommunicable disease Diseases 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000001684 chronic effect Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000037213 diet Effects 0.000 description 1
- 235000005911 diet Nutrition 0.000 description 1
- 230000006806 disease prevention Effects 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000004393 prognosis Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000003936 working memory 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
- G16H80/00—ICT specially adapted for facilitating communication between medical practitioners or patients, e.g. for collaborative diagnosis, therapy or health monitoring
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2457—Query processing with adaptation to user needs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/25—Integrating or interfacing systems involving database management systems
- G06F16/254—Extract, transform and load [ETL] procedures, e.g. ETL data flows in data warehouses
-
- 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
- G06N3/08—Learning methods
-
- 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)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Physics & Mathematics (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Biomedical Technology (AREA)
- Pathology (AREA)
- Primary Health Care (AREA)
- Epidemiology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Quality & Reliability (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The application aims to provide a filtering system and a filtering method for chronic disease detection and risk assessment, and the filtering system and the filtering method are used for obtaining multi-dimensional health information of a detected user; performing feature extraction and construction on the multi-dimensional health information to obtain standardized data; calculating the characteristic weight of the standardized data, and performing weighted summation on the obtained characteristic weight and the standardized data to obtain reconstructed input characteristic data so as to input the reconstructed input characteristic data into a deep learning frame model; by preprocessing the input data of the deep learning frame model, the weight of the weight information is enhanced, the training mode is more targeted, the training process of the model is faster, the effect is better, and the interpretability of the model is improved.
Description
Technical Field
The application relates to the field of chronic disease detection, in particular to a filtering system and method for chronic disease detection and risk assessment.
Background
Chronic non-infectious diseases (chronic diseases for short) are the main causes of death and disease burden of residents in China, and are important factors for restricting the improvement of health life expectancy. Practical experience at home and abroad proves that the prognosis of chronic diseases is closely related to the discovery of the diseases. The earlier the discovery and intervention, the better the treatment and management, and the key is early discovery and intervention. However, the current medical health system is not intended to take continuous care of the patients with chronic diseases, and people can only manage the chronic diseases by themselves under limited guidance. How to provide operable, sustainable feedback, and sustainable health surveillance is of paramount importance to provide personalized health advice to patients and reduced cost health services.
In the current market, for the prediction and prevention of chronic diseases, small and precise monitoring hardware is mostly focused, or an integrated management platform from data acquisition to data storage, data query and the like is adopted, and the usable mode comprises that the current situation of the chronic diseases is combined with AI (artificial intelligence) and the prediction is carried out on the total data; however, there are some problems in the conventional methods for combining chronic disease prevention and control with AI, the medical industry is concerned with life and has a strict supervision, and the problem of insufficient interpretability in deep learning is not completely solved, so that the chronic disease prediction based on the AI algorithm with insufficient interpretability needs time verification.
Disclosure of Invention
An object of the present application is to provide a filtering system and method for chronic disease detection and risk assessment, which solve the problems of insufficient interpretability and resource consumption of deep learning in the prior art using the current status of chronic disease in combination with AI.
According to one aspect of the present application, there is provided a filtering system for chronic disease detection and risk assessment, the system comprising:
a data acquisition module, a data aggregation module and a data filtering module,
the data acquisition module is used for acquiring multi-dimensional health information of a detected user;
the data aggregation module is used for carrying out feature extraction and construction on the multi-dimensional health information to obtain standardized data;
the data filtering module is used for calculating the characteristic weight of the standardized data, and carrying out weighted summation on the obtained characteristic weight and the standardized data to obtain reconstructed input characteristic data so as to input the reconstructed input characteristic data into the deep learning frame model.
Further, the system comprises: the health prediction module is used for predicting the health of the detected user according to the reconstructed input characteristic data to obtain a prediction result;
the feedback optimization module is used for determining a feedback result according to the prediction result and the multi-dimensional health information of the detected user and feeding the feedback result back to a database.
Further, the multi-dimensional health information includes: basic physical sign indexes, disease history, basic diseases, treatment records, living environment data, living habit data and family genetic disease history data.
Further, the data filtering module is used for dividing the historical data acquired from the database into different chronic disease data subsets according to the chronic disease types.
Further, the data filtering module is used for carrying out similarity calculation on the chronic disease data subset and the standardized data, and obtaining a weight value after averaging; and normalizing the weight to obtain a characteristic weight, and performing weighted summation on the obtained characteristic weight and the normalized data to obtain reconstructed input characteristic data.
According to another aspect of the present application, there is also provided a filtering method for chronic disease detection and risk assessment, the method comprising:
acquiring multi-dimensional health information of a detected user;
performing feature extraction and construction on the multi-dimensional health information to obtain standardized data;
and calculating the characteristic weight of the standardized data, and performing weighted summation on the obtained characteristic weight and the standardized data to obtain reconstructed input characteristic data so as to input the reconstructed input characteristic data into the deep learning frame model.
Further, before the calculating the feature weight of the normalized data, the method includes:
and dividing historical data acquired from the database into different chronic disease data subsets according to the chronic disease types.
Further, the calculating of the feature weight for the normalized data includes:
similarity calculation is carried out on the chronic disease data subset and the standardized data, and a weight is obtained after averaging;
and normalizing the weight to obtain the characteristic weight.
Further, the method comprises:
and carrying out weighted summation on the obtained characteristic weight and the standardized data to obtain reconstructed input characteristic data.
Further, before performing weighted summation on the obtained feature weight and the normalized data, the method includes:
and comparing all the obtained feature weights with a weight threshold, and if the current feature weight is smaller than the weight threshold, setting the current feature weight to be 0.
According to yet another aspect of the present application, there is also provided a computer readable medium having computer readable instructions stored thereon, the computer readable instructions being executable by a processor to implement the method as described above.
Compared with the prior art, the method and the device have the advantages that the multi-dimensional health information of the detected user is obtained; performing feature extraction and construction on the multi-dimensional health information to obtain standardized data; calculating the characteristic weight of the standardized data, and performing weighted summation on the obtained characteristic weight and the cleaned data to obtain reconstructed input characteristic data to be input into a deep learning frame model; by preprocessing the input data of the deep learning frame model, the weight of the weight information is enhanced, the training mode is more targeted, the training process of the model is faster, the effect is better, and the interpretability of the model is improved.
Drawings
Other features, objects and advantages of the present application will become more apparent upon reading of the following detailed description of non-limiting embodiments thereof, made with reference to the accompanying drawings in which:
FIG. 1 illustrates a schematic structural diagram of a filtering system for chronic disease detection and risk assessment provided in accordance with an aspect of the present application;
FIG. 2 is a schematic structural diagram of a filtering system under a deep learning framework for chronic disease detection and risk assessment according to an embodiment of the present application;
fig. 3 shows a schematic flow diagram of a filtering method for chronic disease detection and risk assessment according to another aspect of the present application.
The same or similar reference numbers in the drawings identify the same or similar elements.
Detailed Description
The present application is described in further detail below with reference to the attached figures.
In a typical configuration of the present application, the terminal, the device serving the network, and the trusted party each include one or more processors (e.g., Central Processing Units (CPUs)), input/output interfaces, network interfaces, and memory.
The Memory may include volatile Memory in a computer readable medium, Random Access Memory (RAM), and/or nonvolatile Memory such as Read Only Memory (ROM) or flash Memory (flash RAM). Memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, Phase-Change RAM (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), flash Memory or other Memory technology, Compact Disc Read-Only Memory (CD-ROM), Digital Versatile Disc (DVD) or other optical storage, magnetic cassette tape, magnetic tape storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include non-transitory computer readable media (transient media), such as modulated data signals and carrier waves.
Fig. 1 shows a schematic structural diagram of a filtering system for chronic disease detection and risk assessment according to an aspect of the present application, the system comprising: the system comprises a data acquisition module 11, a data aggregation module 12 and a data filtering module 13, wherein the data acquisition module 11 is used for acquiring multi-dimensional health information of a detected user; the data aggregation module 12 is configured to perform feature extraction and construction on the multi-dimensional health information to obtain standardized data; the data filtering module 13 is configured to perform feature weight calculation on the normalized data, perform weighted summation on the obtained feature weight and the normalized data, and obtain reconstructed input feature data to be input into the deep learning framework model. The system is applied to a filter under a deep learning framework of chronic disease detection and risk assessment, the chronic disease detection is detection of chronic diseases, a data filter layer is added before deep learning training, data are pre-trained in a filter mode, multi-dimensional data of detected people are screened in the pre-training process, a certain subset of input features are concentrated, weight is introduced through an input sequence, a position set with related information is considered preferentially to generate next output, and the interpretability of a deep learning framework model is improved by paying attention to the position set of the related information.
Specifically, the data obtaining module 11 is configured to obtain multi-dimensional health information of a detected user; here, the multi-dimensional health information includes: basic physical sign indexes, disease history, basic diseases, treatment records, living environment data, living habit data and family genetic disease history data. The data obtaining module 11 obtains multi-dimensional user information, and mainly includes: basic physical sign indexes, disease history, basic diseases, treatment records, living environment data, living habit data, family genetic disease history data, user behavior data and other multidimensional data, so that a data set can be constructed by a subsequent module.
Specifically, the data aggregation module 12 is configured to perform feature extraction and construction on the multi-dimensional health information to obtain standardized data; here, extraction and construction are performed according to the data acquired by the data acquisition module 11, so that exchange and aggregation of distributed, heterogeneous and network-spanning multi-source information resources are realized, and sharing of a unified platform and multi-source data resources is realized. The multi-party data processed according to the platform standard are concentrated to a central database, and then data services are provided to the outside in a unified standard, so that the data become reusable information resource services according to certain business rules. The extraction and construction can be realized by using an ETL tool, such as an OWB of Oracle, an SSIS service of SQL Server2005, information and the like.
Specifically, the data filtering module 13 is configured to perform feature weight calculation on the normalized data, and perform weighted summation on the obtained feature weight and the normalized data to obtain reconstructed input feature data, so as to input the reconstructed input feature data into the deep learning framework model. Here, the data filtering module 13 performs data filtering on the normalized input data, performs feature reconstruction and screening, introduces feature weights, preferentially considers the key feature information set, performs weighted summation on the obtained feature weights and the data of the cleaned normalized person under test to obtain reconstructed input feature data, for example, obtains feature weights W1, W2, and W3, obtains data of three chronic diseases corresponding to the cleaned person under test a1, a2, and A3, and performs weighted summation: w1 a1+ W2 a2+ W3 A3. The calculation of the characteristic weight can be iterated for a plurality of times so as to highlight the effect of the key information subset; and inputting the obtained reconstructed input characteristic data into a deep learning framework model, wherein the deep learning framework model is used for chronic disease detection and risk assessment, and the output result is a prediction result of the detected personnel.
In one embodiment of the present application, the system comprises: the health prediction module is used for predicting the health of the detected user according to the reconstructed input characteristic data to obtain a prediction result; the feedback optimization module is used for determining a feedback result according to the prediction result and the multi-dimensional health information of the detected user and feeding the feedback result back to a database. The health prediction module carries out monitoring prediction on the detected personnel according to the input reconstructed input characteristic data to obtain a prediction result. The data reconstructed based on the data filtering module is subjected to model construction based on a deep learning algorithm, input data of the model is preprocessed, weight of the gravity information is enhanced, a training mode is changed from a normal large-scale data searching general rule into targeted model training, and therefore the training process is faster in speed and better in effect. The feedback optimization module gives intervention measures such as exercise and diet suggestions of corresponding diseases to the prediction result of the health prediction module, compares the prediction result with the real health result of the detected person, gives a positive feedback result if the prediction result is consistent with the real health result of the detected person, and feeds the feedback result back to a database to update sample data in the database, wherein the database is a standard sample database and stores the multi-dimensional data information of the user acquired in history.
In an embodiment of the present application, the data filtering module 13 is configured to divide historical data acquired from a database into different chronic disease data subsets according to chronic disease categories; and cleaning and standardizing the different chronic disease data subsets to obtain a standard data subset. Here, the historical data acquired from the database is divided into different chronic disease data subsets according to the type of the chronic disease, such as a hypertension set, a diabetes set, a cardiovascular disease set and the like, wherein the historical data is sample data obtained by multi-dimensional user data information acquired in history. And carrying out data cleaning and standardization on the data of the detected users divided into different data subsets, thereby ensuring that the data dimension is consistent with the dimension in the database and the standard is consistent.
In connection with the above embodiment, the data filtering module 13 is configured to perform similarity calculation on the chronic disease data subset and the standardized data, and obtain a weight value after averaging; and normalizing the weight to obtain the characteristic weight. And carrying out similarity calculation on the data of the detected user and the obtained chronic disease data subset, and averaging to obtain a weight value, wherein the weight value is the similarity between the data of the detected user and the data of various diseases, and the higher the similarity is, the larger the weight value is, otherwise, the smaller the weight value is. The calculation of the specific similarity satisfies the following conditions:
wherein, P { (k)1,W(F1)),(k2,W(F2)0,…,(kn,W(Fn) Denotes the data feature vector in the chronic disease subset, U { (U) }1,w(1)),(u2,w(2)),...,(unW (n)) } the feature vector of the detected user.
Normalizing the weight to obtain a directly usable weight, and normalizing by using a softmax () normalization exponential function:
the denominator is the sum of natural base e indexes of all dimensions of all k-dimensional vectors z, and the numerator is the index of the natural base e of the dimension to be solved.
Fig. 2 is a schematic structural diagram of a filtering system under a deep learning framework for chronic disease detection and risk assessment in an embodiment of the present application, including a data acquisition module, a data aggregation module, a data filtering module, a health prediction module, and a feedback optimization module; the data acquisition module is used for acquiring multi-dimensional user information, inputting the acquired information into the data aggregation module, extracting and constructing multi-dimensional and multi-feature data by the data aggregation module, cleaning and standardizing the data, inputting the standardized data into the data filtering module, and calculating feature weights by the data filtering module according to historical data acquired from a sample library and the standardized data input by the data aggregation module, so that the acquired feature weights and data of detected personnel are subjected to weighted summation to obtain reconstructed input feature data; and the health prediction module performs health prediction according to the reconstructed input characteristic data based on the deep learning framework model to obtain a prediction result, the feedback optimization module intervenes in the prediction result to obtain a feedback result, and the feedback result is fed back to the sample library. The multi-dimensional data of the detected personnel are screened, a certain subset of input features is focused on, the position set with related information is preferentially considered by introducing weight on the input sequence to generate the next output, and the interpretability of the model is improved by focusing attention on the position set with the related information.
Fig. 3 shows a schematic flow diagram of a filtering method for chronic disease detection and risk assessment according to another aspect of the present application, the method comprising: step S11 to step S13, wherein,
in step S11, multi-dimensional health information of the detected user is acquired; here, the multi-dimensional health information includes: basic physical sign indexes, disease history, basic diseases, treatment records, living environment data, living habit data, family genetic disease history data, user behavior data and other multidimensional data, so that a data set is constructed and a training set is preprocessed in the subsequent process.
In step S12, performing feature extraction and construction on the multi-dimensional health information to obtain standardized data; in this case, extraction and construction are performed according to the acquired data, so that exchange and aggregation of distributed, heterogeneous and network-crossing multi-source information resources are realized, and sharing of a unified platform and multi-source data resources is realized. The multi-party data processed according to the platform standard are concentrated to a central database, and then data services are provided to the outside in a unified standard, so that the data become reusable information resource services according to certain business rules. The extraction and construction can be realized by using an ETL tool, such as an OWB of Oracle, an SSIS service of SQL Server2005, information and the like.
In step S13, feature weights are calculated for the normalized data, and the obtained feature weights are weighted and summed with the normalized data to obtain reconstructed input feature data for input into the deep learning framework model. Here, feature weights are introduced, a key feature information set is considered preferentially, the obtained feature weights and data of the cleaned and normalized person under examination are subjected to weighted summation to obtain reconstructed input feature data, for example, if the obtained feature weights are W1, W2 and W3, and the data of three chronic diseases corresponding to the cleaned person under examination are a1, a2 and A3, the weighted summation is performed: w1 a1+ W2 a2+ W3 A3. The calculation of the feature weight can be iterated for multiple times to achieve the effect of highlighting the key point information subset, and therefore the key point information subset is input into the deep learning frame model for prediction.
In an embodiment of the application, historical data acquired from a database is divided into different chronic disease data subsets according to chronic disease types. Here, the historical data acquired from the database is divided into different chronic disease data subsets according to the type of the chronic disease, such as a hypertension set, a diabetes set, a cardiovascular disease set and the like, wherein the historical data is sample data obtained by multi-dimensional user data information acquired in history.
In step S13, similarity between the chronic disease data subset and the normalized data is calculated, and a weight is obtained after averaging; normalizing the weight to obtain a characteristic weight; and carrying out weighted summation on the obtained characteristic weight and the standardized data to obtain reconstructed input characteristic data. And performing similarity calculation on the cleaned and standardized data of the detected user and the chronic disease data subset, and averaging to obtain a weight, wherein the weight is the similarity between the data of the detected user and various disease data, and the higher the similarity is, the larger the weight is, otherwise, the smaller the weight is. The calculation of the specific similarity satisfies the following conditions:
wherein, P { (k)1,W(F1)),(k2,W(F2)),…,(kn,W(Fn) Denotes the data feature vector in the chronic disease subset, U { (U) }1,w(1)),(u2,w(2)),...,(unW (n)) } the feature vector of the detected user.
Normalizing the weight to obtain a directly usable weight, and normalizing by using a softmax () normalization exponential function:
the denominator is the sum of natural base e indexes of all dimensions of all k-dimensional vectors z, and the numerator is the index of the natural base e of the dimension to be solved.
In a preferred embodiment of the present application, before performing weighted summation on the obtained feature weights and the cleaned data, all the obtained feature weights may be compared with a weight threshold, and if the current feature weight is smaller than the weight threshold, the current feature weight is set to 0. Here, a threshold is set for the feature weight, and a weight smaller than the threshold is set to be 0, so that interference of noise data is reduced, feature dimensions are reduced, a focus information subset is highlighted, and prediction data is made to be more interpretable.
In addition, a computer readable medium is provided, on which computer readable instructions are stored, the computer readable instructions being executable by a processor to implement the aforementioned filtering method for chronic disease detection and risk assessment.
In an embodiment of the present application, there is also provided a filtering apparatus for chronic disease detection and risk assessment, the apparatus including:
one or more processors; and
a memory storing computer readable instructions that, when executed, cause the processor to perform the operations of the method as previously described.
For example, the computer readable instructions, when executed, cause the one or more processors to:
acquiring multi-dimensional health information of a detected user;
performing feature extraction and construction on the multi-dimensional health information to obtain standardized data;
and calculating the characteristic weight of the standardized data, and performing weighted summation on the obtained characteristic weight and the cleaned data to obtain reconstructed input characteristic data so as to input the reconstructed input characteristic data into a deep learning frame model.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the spirit and scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include such modifications and variations as well.
It should be noted that the present application may be implemented in software and/or a combination of software and hardware, for example, implemented using Application Specific Integrated Circuits (ASICs), general purpose computers or any other similar hardware devices. In one embodiment, the software programs of the present application may be executed by a processor to implement the steps or functions described above. Likewise, the software programs (including associated data structures) of the present application may be stored in a computer readable recording medium, such as RAM memory, magnetic or optical drive or diskette and the like. Additionally, some of the steps or functions of the present application may be implemented in hardware, for example, as circuitry that cooperates with the processor to perform various steps or functions.
In addition, some of the present application may be implemented as a computer program product, such as computer program instructions, which when executed by a computer, may invoke or provide methods and/or techniques in accordance with the present application through the operation of the computer. Program instructions which invoke the methods of the present application may be stored on a fixed or removable recording medium and/or transmitted via a data stream on a broadcast or other signal-bearing medium and/or stored within a working memory of a computer device operating in accordance with the program instructions. An embodiment according to the present application comprises an apparatus comprising a memory for storing computer program instructions and a processor for executing the program instructions, wherein the computer program instructions, when executed by the processor, trigger the apparatus to perform a method and/or a solution according to the aforementioned embodiments of the present application.
It will be evident to those skilled in the art that the present application is not limited to the details of the foregoing illustrative embodiments, and that the present application may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned. Furthermore, it is obvious that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. A plurality of units or means recited in the apparatus claims may also be implemented by one unit or means in software or hardware. The terms first, second, etc. are used to denote names, but not any particular order.
Claims (11)
1. A filtering system for chronic disease detection and risk assessment, the system comprising:
a data acquisition module, a data aggregation module and a data filtering module,
the data acquisition module is used for acquiring multi-dimensional health information of a detected user;
the data aggregation module is used for carrying out feature extraction and construction on the multi-dimensional health information to obtain standardized data;
the data filtering module is used for calculating the characteristic weight of the standardized data, and carrying out weighted summation on the obtained characteristic weight and the standardized data to obtain reconstructed input characteristic data so as to input the reconstructed input characteristic data into the deep learning frame model.
2. The system of claim 1, wherein the system comprises: the health prediction module is used for predicting the health of the detected user according to the reconstructed input characteristic data to obtain a prediction result;
the feedback optimization module is used for determining a feedback result according to the prediction result and the multi-dimensional health information of the detected user and feeding the feedback result back to a database.
3. The system of claim 1, wherein the multi-dimensional health information comprises: basic physical sign indexes, disease history, basic diseases, treatment records, living environment data, living habit data and family genetic disease history data.
4. The system of claim 1, wherein the data filtering module is configured to classify the historical data obtained from the database into different chronic disease data subsets according to chronic disease categories.
5. The system of claim 4, wherein the data filtering module is configured to perform similarity calculation on the chronic disease data subset and the normalized data, and obtain a weight value after averaging; and normalizing the weight to obtain a characteristic weight, and performing weighted summation on the obtained characteristic weight and the normalized data to obtain reconstructed input characteristic data.
6. A filtering method for chronic disease detection and risk assessment, the method comprising:
acquiring multi-dimensional health information of a detected user;
performing feature extraction and construction on the multi-dimensional health information to obtain standardized data;
and calculating the characteristic weight of the standardized data, and performing weighted summation on the obtained characteristic weight and the standardized data to obtain reconstructed input characteristic data so as to input the reconstructed input characteristic data into the deep learning frame model.
7. The method of claim 6, wherein the computing of the feature weights for the normalized data is preceded by:
and dividing historical data acquired from the database into different chronic disease data subsets according to the chronic disease types.
8. The method of claim 7, wherein computing the feature weights for the normalized data comprises:
similarity calculation is carried out on the chronic disease data subset and the standardized data, and a weight is obtained after averaging;
and normalizing the weight to obtain the characteristic weight.
9. The method of claim 8, wherein the method comprises:
and carrying out weighted summation on the obtained characteristic weight and the standardized data to obtain reconstructed input characteristic data.
10. The method of any one of claims 6 to 8, wherein prior to the weighted summation of the derived feature weights and the normalized data, comprising:
and comparing all the obtained feature weights with a weight threshold, and if the current feature weight is smaller than the weight threshold, setting the current feature weight to be 0.
11. A computer readable medium having computer readable instructions stored thereon which are executable by a processor to implement the method of any one of claims 6 to 10.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110009976.5A CN112768090A (en) | 2021-01-05 | 2021-01-05 | Filtering system and method for chronic disease detection and risk assessment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110009976.5A CN112768090A (en) | 2021-01-05 | 2021-01-05 | Filtering system and method for chronic disease detection and risk assessment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112768090A true CN112768090A (en) | 2021-05-07 |
Family
ID=75699816
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110009976.5A Pending CN112768090A (en) | 2021-01-05 | 2021-01-05 | Filtering system and method for chronic disease detection and risk assessment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112768090A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113393934A (en) * | 2021-06-07 | 2021-09-14 | 义金(杭州)健康科技有限公司 | Health trend estimation method and prediction system based on vital sign big data |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109935338A (en) * | 2019-03-07 | 2019-06-25 | 平安科技（深圳）有限公司 | Data prediction processing method, device and computer equipment based on machine learning |
| CN111248913A (en) * | 2020-01-21 | 2020-06-09 | 山东师范大学 | Chronic obstructive pulmonary disease prediction system, equipment and medium based on transfer learning |
| CN112002410A (en) * | 2020-08-20 | 2020-11-27 | 医渡云（北京）技术有限公司 | Infectious disease state prediction method and device, storage medium, and electronic device |
-
2021
- 2021-01-05 CN CN202110009976.5A patent/CN112768090A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109935338A (en) * | 2019-03-07 | 2019-06-25 | 平安科技（深圳）有限公司 | Data prediction processing method, device and computer equipment based on machine learning |
| CN111248913A (en) * | 2020-01-21 | 2020-06-09 | 山东师范大学 | Chronic obstructive pulmonary disease prediction system, equipment and medium based on transfer learning |
| CN112002410A (en) * | 2020-08-20 | 2020-11-27 | 医渡云（北京）技术有限公司 | Infectious disease state prediction method and device, storage medium, and electronic device |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113393934A (en) * | 2021-06-07 | 2021-09-14 | 义金(杭州)健康科技有限公司 | Health trend estimation method and prediction system based on vital sign big data |
| CN113393934B (en) * | 2021-06-07 | 2022-07-12 | 义金(杭州)健康科技有限公司 | Health trend estimation method and prediction system based on vital sign big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10127477B2 (en) | Distributed event prediction and machine learning object recognition system | |
| US10354204B2 (en) | Machine learning predictive labeling system | |
| US10275690B2 (en) | Machine learning predictive labeling system | |
| US11631032B2 (en) | Failure feedback system for enhancing machine learning accuracy by synthetic data generation | |
| Chen et al. | Entity embedding-based anomaly detection for heterogeneous categorical events | |
| US10521734B2 (en) | Machine learning predictive labeling system | |
| CN111612041B (en) | Abnormal user identification method and device, storage medium and electronic equipment | |
| CN112732690B (en) | Stabilizing system and method for chronic disease detection and risk assessment | |
| US10956825B1 (en) | Distributable event prediction and machine learning recognition system | |
| US11010691B1 (en) | Distributable event prediction and machine learning recognition system | |
| US11379685B2 (en) | Machine learning classification system | |
| Ding et al. | The effectiveness of multitask learning for phenotyping with electronic health records data | |
| Balasubramaniam et al. | Feature selection and dwarf mongoose optimization enabled deep learning for heart disease detection | |
| CN111612038A (en) | Abnormal user detection method and device, storage medium and electronic equipment | |
| US20220391760A1 (en) | Combining model outputs into a combined model output | |
| Akundi et al. | Big Data analytics in healthcare using Machine Learning algorithms: a comparative study | |
| CN118468061B (en) | Automatic algorithm matching and parameter optimizing method and system | |
| US20200151603A1 (en) | Distributable event prediction and machine learning recognition system | |
| Qudsi et al. | Predictive data mining of chronic diseases using decision tree: A case study of health insurance company in Indonesia | |
| Sudharson et al. | Performance analysis of enhanced adaboost framework in multifacet medical dataset | |
| CN108304568B (en) | Real estate public expectation big data processing method and system | |
| Swetha et al. | Leveraging Scalable Classifier Mining for Improved Heart Disease Diagnosis | |
| CN112768090A (en) | Filtering system and method for chronic disease detection and risk assessment | |
| CN110957046A (en) | Medical health case knowledge matching method and system | |
| CN116524284A (en) | Nuclear magnetic resonance image analysis method, system, device and storage medium |
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: 20210507 |
|
| RJ01 | Rejection of invention patent application after publication |