KR20170098025A - System and method for analyzing bio-signal based on big data - Google Patents
System and method for analyzing bio-signal based on big data Download PDFInfo
- Publication number
- KR20170098025A KR20170098025A KR1020160019883A KR20160019883A KR20170098025A KR 20170098025 A KR20170098025 A KR 20170098025A KR 1020160019883 A KR1020160019883 A KR 1020160019883A KR 20160019883 A KR20160019883 A KR 20160019883A KR 20170098025 A KR20170098025 A KR 20170098025A
- Authority
- KR
- South Korea
- Prior art keywords
- signal
- bio
- data
- feature value
- raw data
- Prior art date
Links
Images
Classifications
-
- G06F19/32—
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/024—Detecting, measuring or recording pulse rate or heart rate
-
- A61B5/0402—
-
- A61B5/0452—
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/08—Detecting, measuring or recording devices for evaluating the respiratory organs
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/145—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue
-
- G06F17/30194—
-
- G06F17/30318—
-
- G06F17/30339—
-
- G06F17/30592—
Abstract
BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a system and a method for analyzing a bio-signal based on a big data, and more particularly to a system and method for analyzing a bio-signal based on a big data, Receiving unit; And partitioning the characteristic values of the bio-signal extracted from the raw data of the received bio-signal into separate directories for each partition of the data warehouse Hadoop distributed file system (HDFS); And searching and extracting the corresponding biometric signal feature value using a query language corresponding to the feature value condition requested by the transceiver unit and providing the extracted biometric signal feature value as a result of the biometric signal analysis through the web service of the transceiver unit And a service providing unit.
Description
BACKGROUND OF THE
The wearable technology convergence industry has emerged as the core of strengthening competitiveness of IT and other industries as a fusion of human - technology combined human - centered emotion. As a result, the apparel sector is also attempting many new technologies into the realm of u-healthcare. In recent years, the demand for welfare has expanded, medical expenses have increased rapidly due to improvement of economic and living standards, population aging, and chronic degenerative diseases. As a result, the importance of u-healthcare centered on 'disease-centered treatment' and 'preventive and health-care centering' have increased, and u-healthcare has become more interested in managing individual health status anytime and anywhere.
According to this paradigm change, u-healthcare system should store data by using bio-signals and combine the stored data with technologies such as analysis and processing.
In recent years, wearable devices have emerged as a new growth engine as the growth of the smartphone market stagnated and the PC industry showed a stagnation. Wearable devices have evolved into integrated devices in the form of woven / apparel that can be directly attached to the skin, such as patches, in the form of accessories such as glasses, watches, bracelets and the like. In addition, wearable devices are being developed into various devices such as a living body implantable type in which they are directly implanted or taken into a human body. Therefore, as researches on wearable devices become more active, various healthcare products related to health are being developed.
However, in the case of a wearable device using the currently developed bio-signal, the bio-signal of the user obtained through the sensor attached to the device is measured to provide only simple services. Research on services that can prevent user 's health by collecting biometric data measured in wearable devices has not been done properly.
In some healthcare services, services are being used to detect abnormal signals in devices attached to the body of the elderly and suggest necessary measures. However, they are all services that can be backed up by data analysis technology that can quickly read human thoughts, movements, and vital signs.
Korea now has a very high potential for utilizing big data in the field of biohealth, such as world-class medical personnel, IT infrastructure and health insurance system for the general public. It is expected that the potential will be even greater if the data and social network service (SNS) data generated from wearable devices are integrated and utilized.
However, the bio-signals measured through the device are merely measured. Research on bio-signal analysis platform based on collecting, storing, analyzing and processing measured biological data and big data is not actively conducted.
Here, the biological signal refers to all information that can be measured through the device with respect to a signal generated in the human body. However, it is possible not only to measure the bio signal but also to check and manage the user's health condition through the measurement data collection. For example, an electrocardiogram (ECG) signal can be obtained from a heart of a person in a living body signal, and an acceleration (ACC) signal can be obtained from motion.
The collected biometric data can be classified into two types of data. First, it is a discontinuous data form with simple characteristic values such as body temperature, blood pressure, pulse, and oxygen saturation (SpO2). Second, it is continuous data such as electrocardiogram (ECG), acceleration (ACC), respiration and the like.
However, continuously input biological signals such as electrocardiogram and respiration are raw data which are not processed as atypical data, and their volume is large, and they should be classified into big data and processed. Consecutive data is huge and is not suitable for storage in a database. Non-sequential data also has simple feature values, but when measured with a short period of time, it also has a large amount of data and is not suitable for storing in a database.
Therefore, huge amounts of data that are not suitable for storing in the database should be classified and processed as big data. The biological signal itself is unprocessed irregular data. The characteristic value should be extracted through the analysis algorithm with the raw data of the biological signal (Law Data) and processed into analytical data.
On the other hand, if you look at Big Data, you can see big names like Google, Amazon and Wal-mart. However, as the amount of data to collect, store, process, and analyze from the moment became enormous, traditional systems could not handle huge amounts of data properly. The need for a new level of data processing and approach was further compromised by the inability to maintain services that could satisfy customers. They referred to this method as Big Data.
Big data refers to large data that is large in size, short in generation period, and includes not only numerical data but also text and image data, compared with data generated in the past analog environment. However, just because the capacity is large is not called big data. This means when the three-dimensional characteristics of the volume of data, the velocity of production of data, and the varieties of form are all considered. In recent years, it has become common to define values or complexities, and as the five elements are satisfied, they are suitable for big data.
Big Tables, Cassandra, Data Warehousing and Analysis Appliances, Distributed Systems, Google File System, Hadoop, Hbase, and Data Warehouse technologies, including Big Data's large volumes of data, collection and retrieval of unstructured data, data preprocessing and analysis, MapReduce, and non-relational databases.
Embodiments of the present invention define raw data of a vast amount of a living body signal (e.g., an electrocardiogram (ECG) signal, an acceleration signal, etc.) having a linear structure as big data, analyze the big data of the living body signal, To provide a big data-based bio-signal analysis system and method.
In the embodiments of the present invention, the characteristic values of the bio-signal extracted from the raw data of the bio-signal received through the web service are partitioned and stored in a partition for each partition of the data warehouse, The bio-signal characteristic value is extracted through the query language corresponding to the characteristic value condition and is provided as a result of the bio-signal analysis through the web service, so that only the file of the partitioned area can be retrieved, The present invention provides a system and method for analyzing bio-signals based on a large data.
According to a first aspect of the present invention, there is provided a bio-signal measuring apparatus for a user-wearing type, comprising: a transceiver for receiving raw data of bio-signals measured based on a big data through a web service; And partitioning the characteristic values of the bio-signal extracted from the raw data of the received bio-signal into separate directories for each partition of the data warehouse Hadoop distributed file system (HDFS); And searching and extracting the corresponding biometric signal feature value using a query language corresponding to the feature value condition requested by the transceiver unit and providing the extracted biometric signal feature value as a result of the biometric signal analysis through the web service of the transceiver unit A bio-signal analysis system based on a big data may be provided.
The transceiver can receive raw data of any one of an electrocardiogram signal, an acceleration signal, a breathing signal, and a blood coral saturation signal, which are atypical data based on big data.
The Hadoop distributed file system includes: a raw data storage unit for storing raw data of the received bio-signal to generate a repository for each bio-signal by analyzing raw data of the received bio-signal; A feature value extracting unit for extracting a feature value of a bio-signal by filtering raw data of the stored bio-signal; And a data warehouse having a data table divided into directories and partitioned, and partitioning the extracted feature values into separate directories for each partition of the data table.
Wherein the feature value extracting unit detects an R peak value among a P wave, a QRS wave, and a T wave when the raw data of the bio-signal is raw data of the electrocardiogram signal, (HRV), low frequency and high frequency band intensity (LF / HF) and standard deviation (SDNN: SDNN) of the predetermined normal RR interval during the recording time The feature value of at least one of the bio-signals can be extracted.
When the raw data of the bio-signal is raw data of the acceleration signal, the feature value extracting unit may extract a feature value of at least one of the exercise time, the calorie consumption, the movement distance, the number of steps and the behavior pattern.
Wherein the data warehouse stores a feature value of a divided bio-signal corresponding to a feature value condition of at least one of a user ID, a measurement year, a measurement month, a measurement date and a measurement time as a partition of the data warehouse corresponding to the feature value condition It can be stored separately in a separate directory.
The service provider can search for and extract the corresponding bio-signal feature value using the query language of the SQL-On-Hadoop corresponding to the requested feature value condition.
If the data warehouse is a Hive, the service provider may search for and extract a corresponding biometric signal feature value using a query language of HiveQL provided in the hive.
According to a second aspect of the present invention, there is provided a bio-signal measuring apparatus comprising: a raw data receiving step of receiving, through a web service, raw data of a bio-signal measured based on a big data in a user- A bio-signal storage step of partitioning the feature value of the bio-signal extracted from the raw data of the received bio-signal into a partition for each partition of the data warehouse and storing the partition; And a service providing step of searching for and extracting the corresponding biometric signal feature value using a query language corresponding to the requested characteristic value condition and providing the extracted biometric signal characteristic value as a result of a biometric signal analysis through a web service A big data-based bio-signal analysis method can be provided.
The raw data receiving step may receive raw data of at least one of an ECG signal, an acceleration signal, a breathing signal, and a blood coral saturation signal, which are atypical data based on a big data.
The bio-signal storing step may include storing raw data of the received bio-signal in a repository for each bio-signal by analyzing raw data of the received bio-signal to generate a repository for each bio-signal, Extracting a characteristic value of a biological signal by filtering raw data of the stored biological signal; And dividing the extracted feature value into a directory for each partition of the data table, and storing the separated feature value.
Wherein the extracting of the feature value of the bio-signal includes detecting an R peak value among the P wave, the QRS wave, and the T wave when the raw data of the bio-signal is the raw data of the electrocardiogram signal, Extracts feature values of at least one of the bio signal from the HRV, the low frequency and high frequency band intensity (LF / HF) and the standard deviation (SDNN) of the predetermined normal RR interval during a predetermined recording time can do.
Wherein the step of extracting the feature value of the bio-signal includes a step of, when the raw data of the bio-signal is the raw data of the acceleration signal, calculating a feature value of at least one of the exercise time, the calorie consumption, the movement distance, Can be extracted.
Wherein the step of separately storing the biometric signal in the partitioned directory includes a step of storing the feature value of the divided bio-signal corresponding to the at least one feature value condition among the user ID, the measurement year, the measurement month, the measurement date, Can be separately stored in a directory for each partition of the data warehouse.
The service providing step may retrieve and extract the corresponding bio-signal feature value using a query language of the SQL-On-Hadoop corresponding to the requested feature value condition.
In the case where the data warehouse is a Hive, the service providing step may search for and extract a corresponding bio-signal feature value using a query language of HiveQL provided in the hive.
Embodiments of the present invention define raw data of a vast amount of a living body signal (e.g., an electrocardiogram (ECG) signal, an acceleration signal, etc.) having a linear structure as big data, analyze the big data of the living body signal, .
In the embodiments of the present invention, the characteristic values of the bio-signal extracted from the raw data of the bio-signal received through the web service are partitioned and stored in a partition for each partition of the data warehouse, The bio-signal characteristic value is extracted through the query language corresponding to the characteristic value condition and is provided as a result of the bio-signal analysis through the web service, so that only the file of the partitioned area can be retrieved, have.
1 is a block diagram of a big data-based bio-signal analysis system according to an embodiment of the present invention.
2 is an explanatory diagram of various bio-signals generated from a human body applied to the embodiment of the present invention.
3 is an explanatory view of a smart garment applied to a bio-signal measuring device according to an embodiment of the present invention.
4 is an explanatory diagram of a smart clothing system applied to the embodiment of the present invention.
5 is an exemplary view of a continuously incoming electrocardiogram signal according to an embodiment of the present invention.
6 is an explanatory diagram of a web service applied to the embodiment of the present invention.
7 is an exemplary view of a SOAP message of a web service applied to an embodiment of the present invention.
8 is a configuration diagram of a hive system in the Hadoop distributed file system according to the embodiment of the present invention.
9 is a diagram illustrating an example of a file name and a storage format of raw data of a bio-signal according to an embodiment of the present invention.
10 is an exemplary view of an R peak value detected from an electrocardiogram signal according to an embodiment of the present invention.
11 is an explanatory diagram of characteristic values extractable from electrocardiogram data according to an embodiment of the present invention.
12 is an explanatory diagram of a process of creating a partitioned management table according to an embodiment of the present invention.
13 is an explanatory diagram of a process of creating a partitioned management table according to an embodiment of the present invention.
FIG. 14 is an explanatory diagram of a query processing process for extracting a feature value necessary for providing an analysis service according to the embodiment of the present invention.
15 is an exemplary view of information stored in a SOAP message according to an embodiment of the present invention;
16 is an exemplary view of a source code for storing an electrocardiogram signal according to an embodiment of the present invention.
17 is an explanatory diagram of a feedback process of feature values of an input bio-signal according to an embodiment of the present invention.
18 is an explanatory diagram of a main signal of an electrocardiogram signal according to an embodiment of the present invention.
19 and 20 are diagrams illustrating an example of a path where the raw data of the electrocardiogram signal is stored in the HDFS and raw data of the electrocardiogram signal in the bio-signal analysis system according to the embodiment of the present invention.
21 is an exemplary diagram of a feature value extraction source code of an electrocardiogram signal according to an embodiment of the present invention.
22 is an exemplary diagram of method source code for extracting feature values in an HRV class according to an embodiment of the present invention.
23 is an exemplary view of a feature value storage source code extracted from an electrocardiogram signal according to an embodiment of the present invention.
24 is an exemplary diagram of source code for loading a file storing a feature value according to an embodiment of the present invention into a hive.
25 is an exemplary view of a file loaded into a hive according to an embodiment of the present disclosure;
FIG. 26 is an exemplary diagram of a feature value stored in a hive according to an embodiment of the present invention. FIG.
Figure 27 is an illustration of an SOAP message requested for an autonomic nervous system service according to an embodiment of the present disclosure;
28 is an exemplary diagram of HiveQL used for extracting feature values of an autonomic nervous system service according to an embodiment of the present invention.
29 is an illustration of HiveQL for a momentum management service according to an embodiment of the present disclosure.
30 is an exemplary diagram of HiveQuels for a user behavior pattern analysis service according to an embodiment of the present invention.
31 is an explanatory diagram of the number of numbers corresponding to each action element of the user according to the embodiment of the present invention.
32 is an exemplary view of a soap message in which an autonomic nervous system management result according to an embodiment of the present invention is received.
33 is an exemplary view of a soap message in which a momentum management result value according to an embodiment of the present invention is received.
34 is an explanatory diagram of an autonomic nervous system management service and a detailed view screen according to the embodiment of the present invention.
35 is an explanatory diagram of a momentum management service and a detailed view screen according to an embodiment of the present invention.
36 is an exemplary diagram of a SOAP message received as a result of a user behavior pattern according to an embodiment of the present disclosure;
37 is an explanatory diagram of a user behavior pattern analysis service and a detailed view screen according to an embodiment of the present invention.
38 is a flowchart illustrating a method of analyzing a big data based bio signal according to an embodiment of the present invention.
Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. Will be described in detail with reference to the portions necessary for understanding the operation and operation according to the present specification. In describing the embodiments of the present invention, description of technical contents which are well known in the technical field to which the present invention belongs and which are not directly related to the present specification will be omitted. This is for the sake of clarity without omitting the unnecessary explanation and without giving the gist of the present invention.
In describing the components of the present specification, the same reference numerals may be given to components having the same name, and the same reference numerals may be given to different drawings. However, even in such a case, it does not mean that the corresponding component has different functions according to the embodiment, or does not mean that it has the same function in different embodiments, and the function of each component is different from that of the corresponding embodiment Based on the description of each component in FIG.
1 is a block diagram of a big data-based bio-signal analysis system according to an embodiment of the present invention.
1, a big data-based
Hereinafter, the concrete configuration and operation of each component of the big data-based
The
The Hadoop distributed
The
The transmission and
First, the Hadoop distributed
The feature
The
For example, the feature
As another example, when the raw data of the biological signal is raw data of the acceleration signal, the feature
On the other hand, the
Then, the
For example, when the
2 is an explanatory diagram of various bio-signals generated from a human body applied to the embodiment of the present invention.
The types of the biological signals generated from the human body are shown in Fig. Biomedical signals are signals generated by the human body, and bio-signal measuring instruments can measure various kinds of bio-signals through devices capable of measuring bio-signals. They include electrocardiogram, breathing, body temperature, weight, blood pressure, pulse and body fat. In the case of an electrocardiogram or a pulse of the type of a biological signal, two or more electrodes are attached to the skin of a person and measured through a potential difference flowing therebetween. The sensor can be attached to the body and can be measured in various ways such as lifetime measurement and exercise measurement. The number of samples per second is called the sampling frequency. The higher the sampling frequency is, the more accurate the measurement is possible.
In addition, biometric data collected through these measurements can be systematically managed by database for disease prediction and health management
3 is an explanatory view of a smart garment applied to a bio-signal measuring device according to an embodiment of the present invention.
With the fusion of textile technology and IT technology in accordance with the paradigm of new digital society, smart apparel can measure and transmit biometric information as bio signal measurement, processing technology and information transmission technology can be made anytime and anywhere. A smart garment which is very easy to measure and transmit such a biological signal is a wearable wearable garment capable of collecting various bio-signals (for example, electrocardiogram, respiration, body temperature, blood pressure and exercise amount)
As shown in FIG. 3, the smart garment includes a signal transmission network for connecting the sensor and the communication module. A digital seal is used to connect the sensor and the module. For example, a digital yarn can be made of a metal fiber material of good rolling resistance and a fiber diameter of 100 mu m or less. As a result, the digital room can be a data transmission room suitable for smart clothes by eliminating the rigidity which is a characteristic inherent in the metal, and improving flexibility and flexing resistance.
4 is an explanatory diagram of a smart clothing system applied to the embodiment of the present invention.
As shown in FIG. 4, the smart clothing system measures biomedical signals using a biosensor in smart clothing. The smart apparel system then transmits the measured bio-signals to the device via the communication module. Then, the device is a system for transmitting a service to a server to provide a service.
5 is an exemplary view of a continuously incoming electrocardiogram signal according to an embodiment of the present invention.
Among all the information that can be measured through the bio-signal measurement device with respect to the signal generated from the human body, there is an electrocardiogram (ECG) signal as bio-signals that can be obtained from the heart. In addition, there is an acceleration (ACC) signal as a bio-signal that can be obtained from human motion.
Here, the raw data of the bio-signal collected through the bio-signal measuring device can be largely classified into discontinuous data and continuous data. Non-continuous data refers to data having simple characteristic values such as body temperature, blood pressure, pulse, and oxygen saturation. Continuous data refers to data that continuously flows, such as electrocardiogram, breathing, and acceleration.
As shown in FIG. 5, the raw data of the bio-signal having the linearly-introduced linear structure has a form and an amount of data that are not suitable for storing in the database. In the case of non-continuous data, the shorter the measurement period, the greater the amount of data, which is not suitable for database storage.
6 is an explanatory diagram of a web service applied to the embodiment of the present invention.
Web services are one or more functional components that can be accessed by using Internet standard protocols such as XML or HTTP.
As shown in FIG. 6, the
7 is an exemplary view of a SOAP message of a web service applied to an embodiment of the present invention.
As shown in FIG. 7, the wearable bio-signal measurement instrument transmits the measured electrocardiogram signal to the
8 is a configuration diagram of a hive system in the Hadoop distributed file system according to the embodiment of the present invention.
Hadoop, a Java-based open source framework, is used for low-cost and cost-effective data processing for storing and distributing large amounts of biomedical data. Because Hadoop stores data copies, it has the advantage of being able to recover from data loss or failure. Hadoop's data processing method is a distributed computing system that stores data on multiple servers and simultaneously processes the data on each server where the data is stored. Map-Reduce is a sub-project of Hadoop that is divided into a Map stage for simple data division and a Reduce stage for collecting processed data. It is a computational programming model that parallelizes in a configured cluster.
Hive is a data warehouse infrastructure operating on Hadoop that provides data summarization, query and analysis capabilities. Hive has an important function that can easily perform data retrieval and analysis tasks that need to be done through Map-Reduce creation in general Hadoop ecosystem by using HiveQL similar to SQL . This allows the hive to query large data, allowing for easy access and analysis of Hadoop's data. HiveQL supports SQL type queries such as join, table creation, query and insert, so that users can easily query data.
The data in the hive is represented in the same table format as the relational database system. Table schema is stored and managed in Metastore as Metadata.
As shown in FIG. 8, the hive is a representative data warehouse using SQL-On-Hadoop. Hive can query and insert data using a SQL-like query language called HiveQL. The hive has components such as Java Database Connectivity (JDBC), MetaStore, and common language infrastructure (CLI). JDBC is a query compiler and execution engine that converts SQL query statements into MapReduce. MetaStore stores meta information table, partition, and column information for managing data on HDFS in DB table format. The CLI represents a console window for easy access to the hive.
For example, Apache Hive is a technology that can be analyzed in a query language called HiveQL. It is composed of map-ridus-based execution part, metadata information about data store, and query part from user or application program.
9 is a diagram illustrating an example of a file name and a storage format of raw data of a bio-signal according to an embodiment of the present invention.
As shown in Fig. 9, a wearable wearable device, a living body signal measuring device (e.g., smart clothes, etc.) measures an electrocardiogram signal. The electrocardiogram signal measured through the sensor attached to the bio-signal measuring instrument is stored in the Hadoop Distributed File System (HDFS) of Hadoop in the
10 is an exemplary view of an R peak value detected from an electrocardiogram signal according to an embodiment of the present invention.
As shown in FIG. 10, the
Here, the raw data of the bio signal having a linear structure is not meaningful only by the raw data itself. The electrocardiogram signal is a raw data having a linear structure. An electrocardiogram signal having a linear structure has an R-R interval as shown in FIG. The R-R interval is one of the important factors for analyzing the electrocardiogram signal, which is necessary for analyzing the electrocardiogram signal.
Thus, the
The transmission /
11 is an explanatory diagram of characteristic values extractable from electrocardiogram data according to an embodiment of the present invention.
As shown in FIG. 11, the
Hives are therefore more efficient than dealing with non-relational databases (NoSQL) -based map-reduction. A non-relational database (NoSQL) -based map-reshaping parses and processes a myriad of unspecified data, and consists of a pair of text-type keys and values, do. On the other hand, it is more efficient for a hive to use raw data where a schema exists and can be defined clearly. In addition, Hive has the advantage of easily analyzing and processing big data using HiveQL similar to SQL.
12 is an explanatory diagram of a process of creating a partitioned management table according to an embodiment of the present invention.
The
A typical file-based table needs to retrieve each record to see the results. On the other hand, if the feature values extracted in the partitioned data table in the
13 is an explanatory diagram of a process of creating a partitioned management table according to an embodiment of the present invention.
As shown in Fig. 13, the data file extracted from the raw data of the electrocardiogram signal by filtering is stored in accordance with the table format. Then, the
FIG. 14 is an explanatory diagram of a query processing process for extracting a feature value necessary for providing an analysis service according to the embodiment of the present invention.
The
Hereinafter, an embodiment in which a
15 is an exemplary view of information stored in a SOAP message according to an embodiment of the present invention;
As shown in FIG. 15, a living body signal measuring device, which is a wearable wearable device, measures an electrocardiogram signal. The bio-signal measuring device is connected to a user terminal (for example, a smart phone, etc.) 101 via Bluetooth, and the measured electrocardiogram signal is stored in the
The
16 is an exemplary view of a source code for storing an electrocardiogram signal according to an embodiment of the present invention.
As shown in FIG. 16, 'AttachementECG_Hadoop' has attachment information. The
17 is an explanatory diagram of a feedback process of feature values of an input bio-signal according to an embodiment of the present invention.
As shown in FIG. 17, the feature
18 is an explanatory diagram of a main signal of an electrocardiogram signal according to an embodiment of the present invention.
As shown in FIG. 18, the feature
19 and 20 are diagrams illustrating an example of a path where the raw data of the electrocardiogram signal is stored in the HDFS and raw data of the electrocardiogram signal in the bio-signal analysis system according to the embodiment of the present invention.
As shown in Fig. 19, the raw data of the electrocardiogram signal measured from the wearable wearable device and stored in the HDFS of the
The transmission /
As shown in Fig. 20, the raw data of the electrocardiogram signal is irregular data which is meaningless in itself. Therefore, the feature
21 is an exemplary diagram of a feature value extraction source code of an electrocardiogram signal according to an embodiment of the present invention.
As shown in Fig. 21, an ECG parser (ECG parser) inheriting a bio-signal parser (VitalSignParser) is created in the feature value extraction source code. The ECG parser (ECGParser) passes the HRV class in the startParsing method as a parameter to the file path and the sampling rate.
22 is an exemplary diagram of method source code for extracting feature values in an HRV class according to an embodiment of the present invention.
As shown in FIG. 22, the feature
23 is an exemplary view of a feature value storage source code extracted from an electrocardiogram signal according to an embodiment of the present invention.
As shown in FIG. 23, the raw data of the electrocardiogram signal is stored in a file in a form suitable for the schema information stored in the meta store to store the feature values obtained through filtering in the hive.
24 is an exemplary diagram of source code for loading a file storing a feature value according to an embodiment of the present invention into a hive.
As shown in FIG. 24, the
25 is an exemplary view of a file loaded into a hive according to an embodiment of the present disclosure;
As shown in FIG. 25, a partitioned table in a hive enhances performance and stores data in a logical form such as a hierarchical structure.
At this time, if the file (row unit) is read and processed, if all the data are loaded and processed, the map-reducing operation becomes more frequent. Therefore, the hive according to the embodiment of the present invention can improve the processing speed by reducing the data to be read from the file through the partitioning. The user ID, the measurement year, and the measurement month are generated, and the generated feature value data is partitioned and stored in the hive to be loaded.
FIG. 26 is an exemplary diagram of a feature value stored in a hive according to an embodiment of the present invention. FIG.
As shown in FIG. 26, HRV (array), LF / HF (array) by the measurement date, hour, minute, second and delimiter (:) by a comma (SDNN), a sampling-rate (int), and the like are stored in the memory. For example, the type of feature value data may be all of the data types used in JAVA.
Hereinafter, an embodiment of a bio-signal analysis service (for example, an autonomic nervous system service, a momentum management, and a user behavior pattern analysis service) served by the
The
Figure 27 is an illustration of an SOAP message requested for an autonomic nervous system service according to an embodiment of the present disclosure;
The raw data of the bio-signal collected through the measurement is transmitted to the
As shown in FIG. 27, three parameter information about the user ID, the start date and the end date are expressed in the body element of the Soap message.
28 is an exemplary diagram of HiveQL used for extracting feature values of an autonomic nervous system service according to an embodiment of the present invention.
As shown in FIG. 28, when the user requests the Soap message from the
The
29 is an illustration of HiveQL for a momentum management service according to an embodiment of the present disclosure.
As shown in FIG. 29, HiveQL for the momentum management service is shown. The
30 is an exemplary diagram of HiveQuels for a user behavior pattern analysis service according to an embodiment of the present invention.
As shown in FIG. 30, HiveQL for the user behavior pattern analysis service is shown. For example, the
31 is an explanatory diagram of the number of numbers corresponding to each action element of the user according to the embodiment of the present invention.
As shown in FIG. 31, in the case of the behavior pattern of the user, the number is expressed as 1 to 5 for five behavior factors. The number of numbers corresponding to each behavior pattern element is shown.
32 is an exemplary view of a soap message in which an autonomic nervous system management result according to an embodiment of the present invention is received.
The
As shown in Fig. 32, the
As described above, in the embodiment of the present invention, the
The
33 is an exemplary view of a soap message in which a momentum management result value according to an embodiment of the present invention is received.
As shown in Fig. 33, a Soap message for the momentum management is shown. The
Then, the
34 is an explanatory diagram of an autonomic nervous system management service and a detailed view screen according to the embodiment of the present invention.
As shown in FIG. 34, the
In Fig. 34, the blue dot indicates stability and the red dot indicates unstable. In addition, graphs of the values of the autonomic nervous system balance and the average pulse of the user are graphically displayed through the detailed view screen. The
35 is an explanatory diagram of a momentum management service and a detailed view screen according to an embodiment of the present invention.
35, the
In FIG. 35, the
The
36 is an exemplary diagram of a SOAP message received as a result of a user behavior pattern according to an embodiment of the present disclosure;
As shown in FIG. 36, the
37 is an explanatory diagram of a user behavior pattern analysis service and a detailed view screen according to an embodiment of the present invention.
As shown in FIG. 37, when a measurement date is selected through the
38 is a flowchart illustrating a method of analyzing a big data based bio signal according to an embodiment of the present invention.
The
The transmission /
Then, the
Then, the
The transmitting and receiving
Meanwhile, the
Then, the
As described above, the
The
The
It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit or essential characteristics thereof. Therefore, the embodiments disclosed in the present invention are intended to illustrate rather than limit the scope of the present invention, and the scope of the technical idea of the present invention is not limited by these embodiments. The scope of protection of the present invention should be construed according to the following claims, and all technical ideas within the scope of equivalents should be construed as falling within the scope of the present invention.
100: Biological signal analysis system
101: User terminal
110: Transmitting /
120: Hadoop Distributed File System (HDFS)
121: raw data storage unit
122: Feature value extraction unit
123: Data Warehouse
130: Service Offering
610; Web server
Claims (16)
And partitioning the characteristic values of the bio-signal extracted from the raw data of the received bio-signal into separate directories for each partition of the data warehouse Hadoop distributed file system (HDFS); And
The biometric signal feature value is retrieved and extracted using a query language corresponding to the feature value condition requested by the transceiver unit and the extracted biometric signal feature value is provided as a biometric signal analysis result through the web service of the transceiver unit Service Offering
Based biological signal analysis system.
The transmitting /
A big-data-based bio-signal analysis system that receives raw data of a bio-signal of at least one of an ECG signal, an acceleration signal, a breathing signal, and a blood coral saturation signal, which are atypical data based on a big data.
The Hadoop distributed file system
A raw data storage unit for analyzing raw data of the received bio-signal to generate a repository for each bio-signal, and storing raw data of the received bio-signal in a repository for each generated bio-signal;
A feature value extracting unit for extracting a feature value of a bio-signal by filtering raw data of the stored bio-signal; And
Wherein the data warehouse comprises a data table divided into directories according to partitions, and the extracted feature values are partitioned and stored in a directory for each partition of the data table,
Based biological signal analysis system.
The feature value extracting unit
A R peak value of a P wave, a QRS wave, and a T wave when the raw data of the bio-signal is the raw data of the electrocardiogram signal; At least one of heart rate variability (HRV), low frequency and high frequency band intensity (LF / HF), and standard deviation (SDNN: SDNN) Big-data-based bio-signal analysis system that extracts feature values of bio-signals.
The feature value extracting unit
Wherein the feature value of at least one of the exercise time, the calorie consumption, the movement distance, the number of steps and the behavior pattern is extracted when the raw data of the bio-signal is the raw data of the acceleration signal.
The data warehouse
The feature values of the divided bio-signals corresponding to the feature value condition of at least one of the user ID, the measurement year, the measurement month, the measurement date, and the measurement time are divided into a directory for each partition of the data warehouse corresponding to the feature value condition Big data based bio signal analysis system to store.
The service providing unit
Wherein the bio-signal characteristic value is retrieved and extracted using a query language of the SQL-On-Hadoop corresponding to the requested feature value condition.
The service providing unit
Wherein the bio-signal characteristic value is retrieved and extracted using a query language of HiveQL provided in the hive when the data warehouse is a hive.
A bio-signal storage step of partitioning the feature value of the bio-signal extracted from the raw data of the received bio-signal into a partition for each partition of the data warehouse and storing the partition; And
A service providing step of searching and extracting the corresponding bio-signal feature value using a query language corresponding to the requested feature value condition, and providing the extracted bio-signal feature value as a result of a bio-signal analysis through a web service
Wherein the bio-signal analysis method comprises:
The raw data receiving step
A method for analyzing a bio-signal based on a big data, the method comprising: receiving raw data of at least one of an ECG signal, an acceleration signal, a respiration signal, and a blood coral saturation signal, which are atypical data based on Big Data.
The bio-signal storage step
Analyzing raw data of the received bio-signal to generate a repository for each bio-signal, and storing raw data of the received bio-signal in a repository for each generated bio-signal;
Extracting a characteristic value of a biological signal by filtering raw data of the stored biological signal; And
Partitioning the extracted feature values into separate directories for each partition of the data table and storing them
Wherein the bio-signal analysis method comprises:
The step of extracting the feature value of the bio-
A R peak value of a P wave, a QRS wave, and a T wave when the raw data of the bio-signal is the raw data of the electrocardiogram signal; Wherein the characteristic value of at least one of the biological signals is extracted from HRV, low frequency and high frequency band intensities (LF / HF) and standard deviation (SDNN) of predetermined normal RR intervals.
The step of extracting the feature value of the bio-
Wherein the feature value of at least one of the exercise time, the calorie consumption, the movement distance, the number of steps and the behavior pattern is extracted when the raw data of the bio-signal is the raw data of the acceleration signal.
The step of separately storing in the partitioned directory
The feature values of the divided bio-signals corresponding to the feature value condition of at least one of the user ID, the measurement year, the measurement month, the measurement date, and the measurement time are divided into a directory for each partition of the data warehouse corresponding to the feature value condition A method for analyzing a bio-signal based on a big data to be stored.
The service providing step
Wherein the biometric signal feature value is retrieved and extracted using a query language of the SQL-On-Hadoop corresponding to the requested feature value condition.
The service providing step
Wherein the bio-signal feature value is retrieved and extracted using a query language of HiveQL provided in the hive when the data warehouse is a hive.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020160019883A KR101827087B1 (en) | 2016-02-19 | 2016-02-19 | System and method for analyzing bio-signal based on big data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020160019883A KR101827087B1 (en) | 2016-02-19 | 2016-02-19 | System and method for analyzing bio-signal based on big data |
Publications (2)
Publication Number | Publication Date |
---|---|
KR20170098025A true KR20170098025A (en) | 2017-08-29 |
KR101827087B1 KR101827087B1 (en) | 2018-02-08 |
Family
ID=59760342
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
KR1020160019883A KR101827087B1 (en) | 2016-02-19 | 2016-02-19 | System and method for analyzing bio-signal based on big data |
Country Status (1)
Country | Link |
---|---|
KR (1) | KR101827087B1 (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108334845A (en) * | 2018-02-06 | 2018-07-27 | 北京华睿集成科技有限公司 | Data positioning method and data positioning system |
KR20190054741A (en) * | 2017-11-14 | 2019-05-22 | 주식회사 케이티 | Method and Apparatus for Quality Management of Data |
KR20220032662A (en) * | 2020-09-08 | 2022-03-15 | 닥터애니케어 주식회사 | System for providing emergency alarming service using voice message |
WO2023080288A1 (en) * | 2021-11-03 | 2023-05-11 | 홍석우 | Biosignal visualization service system using pan cancer cell diagnosis kit based on glucose metabolism gene microarray chip |
-
2016
- 2016-02-19 KR KR1020160019883A patent/KR101827087B1/en active IP Right Grant
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20190054741A (en) * | 2017-11-14 | 2019-05-22 | 주식회사 케이티 | Method and Apparatus for Quality Management of Data |
CN108334845A (en) * | 2018-02-06 | 2018-07-27 | 北京华睿集成科技有限公司 | Data positioning method and data positioning system |
KR20220032662A (en) * | 2020-09-08 | 2022-03-15 | 닥터애니케어 주식회사 | System for providing emergency alarming service using voice message |
WO2023080288A1 (en) * | 2021-11-03 | 2023-05-11 | 홍석우 | Biosignal visualization service system using pan cancer cell diagnosis kit based on glucose metabolism gene microarray chip |
Also Published As
Publication number | Publication date |
---|---|
KR101827087B1 (en) | 2018-02-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Pham et al. | Heart rate variability in psychology: A review of HRV indices and an analysis tutorial | |
Kikhia et al. | Utilizing a wristband sensor to measure the stress level for people with dementia | |
Salehizadeh et al. | A novel time-varying spectral filtering algorithm for reconstruction of motion artifact corrupted heart rate signals during intense physical activities using a wearable photoplethysmogram sensor | |
KR101827087B1 (en) | System and method for analyzing bio-signal based on big data | |
Ahmed et al. | Early detection of atrial fibrillation based on ECG signals | |
CN105808946B (en) | Mobile telemedicine system based on cloud computing | |
Nurmi et al. | Systematic review on machine-learning algorithms used in wearable-based eHealth data analysis | |
Peternel et al. | A presence-based context-aware chronic stress recognition system | |
CN101785702A (en) | Health information system, method, corresponding device, equipment and reagent carrier | |
Jovanović et al. | A mobile crowd sensing application for hypertensive patients | |
Park et al. | Design of wearable EEG devices specialized for passive brain–computer interface applications | |
Page et al. | Visualization of health monitoring data acquired from distributed sensors for multiple patients | |
KR20160125543A (en) | User-oriented healthcare big data service method, computer program and system | |
CN107092799A (en) | A kind of vital sign data management system and method based on cloud computing | |
Tervonen et al. | Ultra-short window length and feature importance analysis for cognitive load detection from wearable sensors | |
Harezlak et al. | Application of time-scale decomposition of entropy for eye movement analysis | |
Choi et al. | Advantage of vital sign monitoring using a wireless wearable device for predicting septic shock in febrile patients in the emergency department: a machine learning-based analysis | |
Rossi et al. | Error estimation of ultra-short heart rate variability parameters: effect of missing data caused by motion artifacts | |
Benchekroun et al. | Preprocessing methods for ambulatory HRV analysis based on HRV distribution, variability and characteristics (DVC) | |
Baldini et al. | A Real-Time, Open-Source, IoT-like, Wearable Monitoring Platform | |
Shao et al. | Research of HRV as a measure of mental workload in human and dual-arm robot interaction | |
KR20210004058A (en) | A Novel Healthcare Monitoring Method and Apparatus Using Wearable Sensors and Social Networking Data | |
Bae et al. | Vital block and vital sign server for ECG and vital sign monitoring in a portable u-vital system | |
CN108962395A (en) | One kind is acquired in real time based on parallel score rank physiological signal and analysis method | |
Taoum et al. | Validity of ultra-short-term hrv analysis using ppg—a preliminary study |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
A201 | Request for examination | ||
E902 | Notification of reason for refusal | ||
E701 | Decision to grant or registration of patent right | ||
GRNT | Written decision to grant |