KR20170023382A - Disease forecast device based on concentration information of biomaterial and forecasting method thereof - Google Patents

Disease forecast device based on concentration information of biomaterial and forecasting method thereof Download PDF

Info

Publication number
KR20170023382A
KR20170023382A KR1020160004911A KR20160004911A KR20170023382A KR 20170023382 A KR20170023382 A KR 20170023382A KR 1020160004911 A KR1020160004911 A KR 1020160004911A KR 20160004911 A KR20160004911 A KR 20160004911A KR 20170023382 A KR20170023382 A KR 20170023382A
Authority
KR
South Korea
Prior art keywords
disease
disease prediction
concentration
prediction
client
Prior art date
Application number
KR1020160004911A
Other languages
Korean (ko)
Other versions
KR101903526B1 (en
Inventor
한영웅
최재훈
Original Assignee
한국전자통신연구원
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 한국전자통신연구원 filed Critical 한국전자통신연구원
Publication of KR20170023382A publication Critical patent/KR20170023382A/en
Application granted granted Critical
Publication of KR101903526B1 publication Critical patent/KR101903526B1/en

Links

Images

Classifications

    • G06F19/3431—
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
    • G01N21/59—Transmissivity
    • G01N21/5907—Densitometers
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K19/00—Record carriers for use with machines and with at least a part designed to carry digital markings
    • G06K19/06—Record carriers for use with machines and with at least a part designed to carry digital markings characterised by the kind of the digital marking, e.g. shape, nature, code
    • G06K19/06009—Record carriers for use with machines and with at least a part designed to carry digital markings characterised by the kind of the digital marking, e.g. shape, nature, code with optically detectable marking
    • G06K19/06037—Record carriers for use with machines and with at least a part designed to carry digital markings characterised by the kind of the digital marking, e.g. shape, nature, code with optically detectable marking multi-dimensional coding
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
    • G01N21/59—Transmissivity
    • G01N2021/5903—Transmissivity using surface plasmon resonance [SPR], e.g. extraordinary optical transmission [EOT]

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Pathology (AREA)
  • General Health & Medical Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Investigating Or Analysing Biological Materials (AREA)
  • Medical Informatics (AREA)
  • Immunology (AREA)
  • Biochemistry (AREA)
  • Public Health (AREA)
  • Theoretical Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Primary Health Care (AREA)
  • Epidemiology (AREA)

Abstract

A disease forecasting device based on the concentration of biomaterial according to an embodiment of the present invention includes a step of generating disease forecasting models according to the combination and concentration of biomaterials, a step of receiving the type information of biomaterial detected from a sample and the concentration information of the biomaterial from a disease forecasting client, a step of selecting one of the disease forecasting models having the highest degree of agreement with the type information of the biomaterial, a step of determining whether a disease corresponding to the selected disease forecasting model is diagnosed by referring to the concentration information of the biomaterial, or risk, and a step of transmitting the determination result to the disease forecasting client. So, it is possible to forecast various individual diseases.

Description

TECHNICAL FIELD The present invention relates to a device for predicting a disease based on the concentration of a biomolecule and a method for predicting the disease.

The present invention relates to a disease predicting device, and more particularly, to a disease predicting device capable of predicting an individual's disease by analyzing concentration data of a bio-material and an operation method thereof.

Recently, as interest in health has increased, there has been an increasing demand for health care technology that adds disease prevention concepts in addition to disease diagnosis and management. And the disease prediction device, which is one of the basic components of health care, is closely related to the sensor part and the analysis part. The sensor is able to measure a limited number of biomaterial concentrations, and the analyzer is specialized to this sensor and is configured to perform only predictions for specific diseases.

However, the development of a disease-specific and disease-specific disease-predicting device has great limitations in terms of usability and versatility because various diseases and bio-materials specific to each disease are continuously discovered. In addition, in order to predict the disease, an additional terminal device for visualizing the concentration of the measured bio-material and the disease prediction result is needed. In addition, individual health care through disease prediction can not be carried out personally because of the economic burden and the failure of motivation to purchase an additional terminal device. Thus, health management using these disease predictions is still being performed within a hospital-centered health care system.

This reality is contrary to the global trend in the healthcare field, where individuals manage their own medical data and provide various customized healthcare services based on the built up personal medical data.

It is an object of the present invention to provide a personal disease prediction apparatus based on a biomaterial concentration and capable of predicting diseases of various individuals without limitation to a specific disease using an individual smart terminal, And to provide a disease prediction method using the same.

A method for predicting a concentration based on a bio material according to an embodiment of the present invention includes the steps of generating disease prediction models based on combinations and concentrations of biomaterials, information on types of biomaterials detected from the samples, Selecting one of the disease prediction models from among the disease prediction models, the prediction accuracy of which is higher than the prediction accuracy of the disease prediction model; , And transmitting the determination result to the disease prediction client.

The apparatus for predicting a disease based on the concentration of a biomaterial according to an embodiment of the present invention includes a concentration detection sensor for sensing the type and concentration of a biomaterial from a sample and displaying the biomaterial in a QR code, A disease prediction client which converts the disease prediction client into a concentration information and a disease prediction request from the disease prediction client and information on the type and concentration of the biomaterial, And a disease prediction server for determining the presence or the risk of the disease by applying the concentration information of the bio material to the selected one of the models and transmitting the determined disease state or risk to the disease prediction client .

According to the disease predicting apparatus according to the embodiment of the present invention, since it is not limited to specific diseases and bio materials, various kinds of sensors can be used and various diseases can be predicted. In addition, it is possible to predict a variety of diseases using a personal smart terminal without additional cost, and the individual can lead the measurement and prediction history. In addition, the use of miniaturized sensors can reduce the personal burden of health care, and health care in the form of home care is possible. Sensors that can not be miniaturized can be operated in public health centers or hospitals, and individuals can directly receive the test results through their smart terminals. Therefore, when the disease prediction system of the present invention is used, it is expected that the risk of privacy protection and leak of personal medical information is low and it is possible to induce a strong motivation for improvement of health for an individual.

1 is a block diagram illustrating a disease predicting apparatus according to an embodiment of the present invention.
FIG. 2 is a block diagram showing the configuration of the disease prediction client of FIG. 1; FIG.
FIG. 3 is a block diagram illustrating an exemplary configuration of the disease prediction server of FIG. 1;
FIG. 4 is a diagram showing the interactions between devices constituting the disease predicting apparatus of the present invention. FIG.
5 is a flowchart showing the sensing and data transmission method of the density detection sensor 100 of FIG.
6 is a flowchart briefly showing an operation method of the disease prediction client 200 of the present invention.
7 is a flowchart showing an operation method of the disease prediction server 400 for providing a disease prediction service.
FIG. 8 is a flowchart illustrating an exemplary method for generating the disease prediction model referred to in FIG. 7; FIG.
FIG. 9 is a table showing the biomaterial concentration learning data described in FIG.
FIG. 10 is a view illustrating an example of a method for providing a disease-predicting service based on concentration of a biomaterial performed in the disease predicting apparatus of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings, so that those skilled in the art can easily carry out the technical idea of the present invention. .

1 is a block diagram showing a disease prediction apparatus or system according to an embodiment of the present invention. Referring to FIG. 1, the disease prediction apparatus 10 of the present invention may include a concentration detection sensor 100, a disease prediction client 200, a network 300, and a disease prediction server 400.

The concentration detecting sensor 100 senses samples such as blood, body fluids, and urine. The concentration detection sensor 100 can measure the concentration value of each type of biomaterial (protein, metabolite, etc.) contained in the sample. The concentration detection sensor 100 can detect the kind of the bio-material measured from the sample and the concentration value for each type. The concentration detection sensor 100 can generate the QR code (Quick Response code) for the type of the detected biomaterial and the concentration value for each type. For example, the concentration detection sensor 100 may generate the type of the detected bio-material and the concentration value for each type using data in the form of JavaScript standard object notation (hereinafter referred to as JSON). Then, the density detection sensor 100 may convert the data created in the JSON format into a QR code. The converted QR code may be displayed on a display provided in the density detection sensor 100. [ It will be well understood that the transmission of the types of biomaterials detected and concentration values for each type is not limited to the QR code form. JSON format data may be transmitted to the disease prediction client 200 through a wired / wireless channel.

The disease prediction client 200 transmits the types of the detected biomaterials and concentration values for each type provided from the concentration detection sensor 100 to the disease prediction server 400. The disease prediction client 200 can visualize the disease prediction information provided from the disease prediction server 400 as data that can be recognized by the user. For example, the disease prediction client 200 can read the QR code displayed on the display module of the concentration detection sensor 100. [ Then, the QR code is parsed or decoded to transmit the kind of the effective biomaterial and the concentration value for each type to the disease prediction server 400. The disease prediction client 200 can display the disease prediction information provided from the disease prediction server 400 on the display as information of a form that can be confirmed by the user. In addition, the disease prediction client 200 stores the received disease prediction result and manages the disease prediction history of the user.

Here, the disease prediction client 200 may include functions and configurations for recognizing the types of biomaterials provided by the concentration detection sensor 100 and the concentration values for the respective types. That is, when the types of bio materials and the concentration values of the respective types are represented by QR codes, a camera module capable of recognizing such QR codes will be included. In the case where the type of the biomaterial provided by the concentration detection sensor 100 and the concentration value of each type are provided through a wired / wireless channel (for example, USB, WiFi, Blooth, etc.) Interface. Preferably, the disease prediction client 200 may be a portable communication terminal such as a smart phone, a tablet, a PDA, or the like, which is capable of wired / wireless communication and has a camera module.

The network 300 provides information exchange channels between the disease prediction client 200 and the disease prediction server 400. The network 300 may be, for example, a Code Division Multiple Access (CDMA), an Orthogonal Frequency Division Multiple Access (OFDMA), a 3G Standardization Project (3GPP), a Long Term Evolution (LTE, LTE- ) Or the like. The network 300 may include wireless communication channels such as satellite communication, Bluetooth, WiFi (Wireless Fidelity), and the like. Examples of wired communications for implementing network 300 may include Ethernet, gigabit Ethernet, InfiniBand (TM), digital subscriber line (DSL), fiber to the home (FTTH), and the like.

The disease prediction server 400 analyzes the types of biomaterials and concentration values provided from the disease prediction client 200 based on a disease prediction model in the form of Big data. The disease prediction server 400 can predict the disease status and the risk of the user by using the type of bio material and the concentration value of each type. The disease prediction server 400 will transmit the predicted disease status or risk to the disease prediction client 200. The disease prediction server 400 will include configurations for generating and storing a disease prediction model.

The disease predicting apparatus 10 based on the concentration of the biomaterial according to the embodiment of the present invention has been briefly described. The disease predicting apparatus 10 may be configured through a wired or wireless long-distance or near-field network 300. Here, the concentration detecting sensor 100 and the disease predicting client 200 may be configured as separate devices, but may be provided as a single device. That is, the concentration detection sensor 100 and the disease prediction client 200 may be implemented as a single device such as a wearable device worn by a user.

2 is a block diagram briefly showing the configuration of the disease prediction client 200 of FIG. Referring to FIG. 2, the disease prediction client 200 includes a processor 210, a working memory 220, a storage 230, a network interface 240, a camera module 250, a display device 260, (Not shown). It will be appreciated that the components of the disease prediction client 200 are not limited to the components shown. For example, the disease prediction client 200 may further include a hardware codec, a security block, and the like for processing image data.

The processor 210 may execute software (application programs, operating systems, device drivers) to be performed in the disease prediction client 200. The processor 210 will execute an operating system (OS) (not shown) loaded into the working memory 220. The processor 210 may execute various application programs to be run on an operating system (OS) basis. In particular, the processor 210 may execute a disease prediction client module 225 that is loaded into the working memory 220. The processor 210 may be provided as a homogeneous multi-core processor or a heterogeneous multi-core processor.

An operating system (OS) or basic application programs (Application Programs) will be loaded into the working memory 220 at boot time. For example, the OS image stored in the storage 230 at the boot time of the disease prediction client 200 is loaded into the working memory 220 based on the boot sequence. All the input / output operations of the disease prediction client 200 can be supported by the operating system (OS). Likewise, applications may be loaded into the working memory 220 for selection by the user or provision of basic services. The working memory 220 may be used as a buffer memory for storing image data provided from an image sensor such as a camera. The working memory 220 may be a volatile memory such as a static random access memory (SRAM) or a dynamic random access memory (DRAM), or a nonvolatile memory such as a PRAM, an MRAM, a ReRAM, a FRAM, and a NOR flash memory.

In particular, the disease prediction client module 225 is loaded into the working memory 220. The disease predicting client module 225 has a function of verifying and parsing the bio material and concentration information provided in the form of a QR code through the camera module 250 and transmitting the parsed information to the disease prediction server 400. In addition, the disease prediction client module 225 controls to display the prediction result provided from the disease prediction server 400 on the display device 260.

The storage 230 is provided as a storage medium of the disease prediction client 200. The storage 230 may store an application program, an OS image, and various data. In particular, the storage 230 may store disease prediction result data provided from the disease prediction server 400. The accumulated disease prediction result data can be managed by the disease prediction client module 225 as a disease history for the user. The storage 230 may be provided as a memory card (MMC, eMMC, SD, MicroSD, etc.). The storage 230 may include a NAND-type flash memory having a large storage capacity. Alternatively, the storage 230 may include a next generation non-volatile memory such as a PRAM, an MRAM, a ReRAM, and a FRAM, or a NOR flash memory.

The network interface 240 provides interfacing between the disease prediction client 200 and the network 300. That is, the network interface 240 provides a signal path between the disease prediction client 200 and the network 300 for transmitting the detected bio material and concentration information to the disease prediction server 400. In addition, the network interface 240 will also provide a signal path for receiving a disease prediction result from the disease prediction server 400. The network interface 240 may be implemented in a mobile terminal such as CDMA, OFDMA, 3GPP, Long Term Evolution (LTE), and the like, when the disease prediction client 200 is provided as a portable personal terminal. , LTE-A), or a fourth generation standard (4G), satellite communication, Bluetooth, and wireless fidelity (WiFi). Alternatively, the network interface 240 may be a wired communication interface such as Ethernet, gigabit Ethernet, InfiniBand (TM), digital subscriber line (DSL), fiber to the home (FTTH) It will be appreciated that the communication scheme of the network interface 240 is not limited to the schemes described above.

The camera module 250 senses moving images or images to the disease prediction client 200 and provides the sensed images. In particular, the camera module 250 can capture the QR code-type biomaterials and their concentration information displayed on the display of the concentration detection sensor 100. The QR code captured by the image information can be converted into a JSON format or a general data format by decoding or parsing.

The display device 260 provides various display functions of the disease prediction client 200. The display device 260 may display a user authentication or a verification result of the QR code for receiving the disease prediction service. In addition, the display device 260 may display the disease prediction result transmitted from the disease prediction server 400 in various tables or graphs. The display device 260 may include, for example, a display controller and a display panel that convert image information according to the specification of the display. In addition, the display device 260 may be provided as a touch pad capable of recognizing a user's touch.

The system bus 270 provides a network between the functional blocks that constitute the disease prediction client 200. The system bus 270 may include, for example, a data bus, an address bus, and a control bus. The data bus is the path through which the data travels. Mainly, a memory access path to the working memory 220 or the storage 230 will be provided. An address bus provides an address exchange path between functional blocks (IPs). The control bus provides a path for transferring control signals between functional blocks (IPs). However, the configuration of the system bus 270 is not limited to the above description, and may further include arbitration means for efficient management.

Although not shown, the disease prediction client 200 may further include various user interfaces. The user interfaces may communicate user input data or output data to other configurations via the system bus 270. For example, the user interfaces may display a keyboard screen, interactive information, or the like for inputting data to the display device 260 under the control of the processor 210. [ The user can input data such as authentication information to the disease prediction client 200 through the user interface.

According to the above description, the disease prediction client 200 can transmit the bio-material provided from the concentration detection sensor 100 and the concentration information of each bio-material to the disease prediction server 400. In addition, the disease prediction client 200 displays the disease prediction result provided from the disease prediction server 400 to the user. In addition, the disease prediction client 200 can store the disease prediction result in the storage 230 and manage it as a history of the user's disease information.

Here, it is described that the disease prediction client module 225 is provided as one software module to the disease prediction client 200, but the present invention is not limited thereto. The disease prediction client module 225 may be installed in the disease prediction client 200 in the form of a hardware module composed of one functional block.

FIG. 3 is a block diagram illustrating an exemplary configuration of the disease prediction server of FIG. 1; 3, the disease prediction server 400 includes a processor 410, a working memory 420, a storage 430, a network interface 440, a user interface 450, and a system bus 460 .

The processor 410 drives software of various algorithms that the disease prediction server 400 performs. For example, the processor 410 may call an application program, an operating system, or device drivers from the working memory 420 or the storage 430 for the main function of the disease prediction server 400 and execute the program. In particular, the processor 410 may execute a disease prediction server module 425 to provide a concentration-based disease predicting function of the biomaterial of the present invention.

An operating system (OS), application programs, and various process data may be loaded into the working memory 420. In particular, the working memory 420 may load the disease prediction server module 425 for processing the biomaterial and concentration information provided from the disease prediction client 200 to predict the disease. The disease prediction server module 425 may be stored in the storage 430 and loaded into the working memory 420 upon receiving an execution instruction by the user. The disease prediction server module 425 loaded in the working memory 420 is called and executed by the processor 410. [ In accordance with the execution of the disease prediction server module 425, the received bio material and concentration information based on the disease prediction model 435 of the big data type generated in advance will be processed. That is, the biomaterial and concentration information of the user is compared with similar items in the disease prediction model 435, and the disease condition and the risk of the user can be predicted according to the comparison result. The disease prediction result is transmitted to the disease prediction client 200 again via the network interface 440.

The data of the disease prediction model 435 may be loaded into the working memory 420 while the disease prediction calculation is being performed. In addition, the data may be loaded into the working memory 420 and collected or merged during the learning operation to generate the disease prediction model 435 for disease prediction.

The storage 430 is provided as a storage medium of the disease prediction server 400. Various databases provided for the tasks performed by the disease prediction server 400 may be stored in the storage 430. The storage 430 may provide a non-volatile memory function for storing the results of the operations performed by the disease prediction server 400. For example, the storage 430 may store a disease prediction model 435 for comparing biomaterial and concentration information provided in the disease prediction client 200. [ The disease prediction model 435 may be updated periodically or as needed by the user with information on new biomaterials, signs thereof, and new diseases. The storage 430 may also store concentration learning data of the bio material for disease prediction. The disease prediction server 400 may update and adjust the disease prediction model as necessary according to the concentration learning data of the bio material. The storage 430 may be provided as a hard disk (HDD), a solid state drive (SSD), or the like capable of storing a large amount of data as an example.

The network interface 440 provides a communication channel for data exchange between the disease prediction server 400 and the disease prediction client 200. The disease prediction server 400 may receive the bio material and concentration information from the disease prediction client 200 through the network interface 440. [ The disease prediction server 400 may transmit the disease prediction result to the disease prediction client 200 through the network interface 440. It will be appreciated that the network interface may be provided according to various communication standards of wired or wireless.

The user interface 450 provides interfacing between the disease prediction server 400 and the user. The user interface 450 may include various input and output devices. Material concentration learning data for generating the disease prediction model 435 to the disease prediction server 400 through the user interface. The construction method of the disease prediction model 435 of the big data type using the biomaterial concentration learning data will be described in detail in the following drawings.

The system bus 460 provides a network between the functional blocks constituting the disease prediction server 400.

According to the above description, the disease prediction server 400 can predict disease and risk based on the bio material and concentration information provided from the disease prediction client 200. The disease prediction server 400 may transmit the disease prediction result to the disease prediction client 200. [ In addition, the disease prediction server 400 may periodically or, if necessary, update the disease prediction model provided for the disease prediction calculation. Therefore, it is possible to expand the versatility of the concentration-based disease predicting apparatus 10 of a biomaterial by updating a new disease or a symptom caused by a new biomaterial.

FIG. 4 is a view showing the interactions between devices constituting the disease prediction system of the present invention. FIG. Referring to FIG. 4, transmission of biomaterial and concentration information between the concentration detection sensor 100, the disease prediction client 200, and the disease prediction server 400 and the transmission relation of the disease prediction result will be described.

In step S10, sensing of the user's biomaterial by the concentration detection sensor 100 will be performed. For example, the concentration detection sensor 100 may sense a user's blood sample, urine sample, or the like to detect the type and concentration of metabolites contained in the sample. The thus-sensed bio-material and concentration information may be reprocessed by the concentration detection sensor 100 and displayed in various formats of information.

In step S12, the bio-material and concentration information may be transmitted to the disease prediction client 200. [ For example, the bio-material and concentration information may be converted into a QR code form in the concentration detection sensor 100 and displayed. The biomaterial and concentration information in the form of a QR code displayed on the display can be captured and recognized by the camera module of the disease prediction client 200.

In step S20, the disease prediction client 200 can check the validity of the bio-material and concentration information provided from the concentration detection sensor 100. [ Here, the disease prediction client 200 may verify whether the biomaterial and the concentration information are meaningful information capable of predicting the disease and verify the validity.

In step S22, when the disease prediction client 200 determines that the bio material and the concentration information are valid, the disease prediction client 200 transmits the bio material and the concentration information to the disease prediction server 400. [

In step S30, the disease prediction server 400 performs a concentration-based disease prediction calculation on the received bio-material and concentration information. The disease prediction calculation may be performed with reference to the disease prediction model built in the disease prediction server 400. [

In step S32, the disease prediction server 400 transmits the disease prediction result to the disease prediction client 200. [

In step S34, the disease prediction client 200 informs the user of the received disease prediction result through an information display device such as a display. The disease prediction client 200 may store the received disease prediction result as a disease prediction history of the user.

In the foregoing, the transfer of information carried out in the disease predicting device 10 has been briefly described. However, it will be understood that the transmission method of the bio material and the concentration information and the transmission method of the disease prediction result are not limited to the above-mentioned examples. The concentration detecting sensor 100 and the disease predicting client 200 may be constituted by a single device such as a wearable device. In this case, the conversion process of the biomaterial and concentration information into the QR code may be omitted.

5 is a flowchart showing the sensing and data transmission method of the density detection sensor 100 of FIG. Referring to FIG. 5, when a sample such as blood or urine is input to the concentration detection sensor 100 and a request is made to measure the concentration of the bio material and the bio materials, the measurement and display procedure of the concentration of the bio material of the present invention starts do.

In step S110, the concentration detection sensor 100 senses the concentration of each type of metabolites from the provided sample. The limits of the types and concentration of bio-materials that can be measured according to the performance of the concentration detection sensor 100 may vary. In order to measure the type and concentration of a more precise biomaterial, a large concentration detection sensor 100 capable of sensing with a high resolution may be applied.

In step S120, the concentration detection sensor 100 may convert the concentration information of each type of the measured bio-materials into data of JSON format (JavaScript Object Notation Format). Here, an example has been described in which concentration information for each type of biomaterial is converted into an open standard JSON format. However, the present invention is not limited thereto, and may be transformed into data in various formats other than the syntax of JavaScript.

In step S130, the concentration detection sensor 100 converts the concentration information of each type of biomaterial converted into the JSON format into a QR code. At this time, the data size of the concentration information for each type of biomaterial should be a size acceptable to the QR code. It will be appreciated that, in another embodiment, in the case of concentration information for each type of relatively small-sized biomaterial, it may be converted into a bar code form. However, it will be understood that the display method of the concentration information for each type of the biomaterial is not limited to the above-described examples, and various modifications are possible.

 In step S140, the density detection sensor 100 displays the generated QR code on the provided display. In another embodiment, the concentration detection sensor 100 may display concentration information for each type of biomaterial produced by the barcode on a display (not shown).

In the above, a method has been described in which the concentration detection sensor 100 senses the concentration of each kind of biomaterials contained in blood or urine samples and displays them as information. However, the concentration information of each kind of the detected biomaterial may be transmitted to the disease prediction client 200 without being displayed on the display. In addition, the types of samples for sensing biomaterials are not limited to blood or urine. For example, various bio-secretions such as saliva, tears, and sweat can be provided as a sample for detecting the biomaterial of the present invention.

6 is a flowchart briefly showing an operation method of the disease prediction client 200 of the present invention. Referring to FIG. 6, the disease prediction client 200 receives and transmits concentration information for each type of bio material according to the disease prediction client module 225 (see FIG. 2), and receives and displays disease prediction results.

In step S210, the disease prediction client module 225 of the disease prediction client 200 will be executed. The disease prediction client module 225 can be activated by an execution command of an application provided to the terminal by the user.

In step S220, the user authentication requesting execution of the application will be performed. Information on the concentration of biomaterials in the individual's body or disease prediction results are personal medical information. Therefore, access to private personal information should be provided with appropriate level of security through authentication process. Therefore, in order to drive the disease prediction client module 225 for disease prediction, a user authentication process must be performed first. Various methods such as ID / pass (ID / Pass) and fingerprint recognition may be applied to the user authentication method.

In step S230, the disease prediction client 200 receives concentration information for each type of biomaterial provided from the concentration detection sensor 100. [ For example, the disease prediction client 200 can receive the QR code displayed on the display of the concentration detection sensor 100 through the camera module 250 (see FIG. 2). For this operation, the disease prediction client module 225 may be set such that the user selects the QR code recognition menu after the user authentication or automatically moves to the QR code recognition screen. It will be appreciated that the concentration information for each kind of bio material may be transmitted to the disease prediction client 200 in various ways besides QR code or bar code.

In step S240, the disease prediction client 200 will verify the validity of the concentration information for each kind of bio material provided in the form of a QR code. The camera module 250 is used for recognition of the QR code by the disease prediction client 200. [ It will be determined whether or not the QR code displayed on the display of the density detection sensor 100 is normally recognized. Then, the recognized QR code is analyzed and it will be checked whether the disease type, type of biomaterial, and concentration information for each type of biomaterial exist in the QR code in the JSON format. The name of the biomaterial and the concentration information for each type of biomaterial are essential elements to be included in the biomaterial and concentration information. The type of disease to be predicted may be inputted by the user through the input interface mounted on the concentration detection sensor 100 and may be sent from the concentration detection sensor 100 to the disease prediction client 200. Or the type of the disease may be directly input by the user at the time of the disease prediction client module 225. [ If there is a problem in the QR code through the above-described check (No direction), the procedure moves to step S230 and the re-recognition of the QR code will proceed. On the other hand, if it is determined that the QR code is valid (Yes direction), the procedure moves to step S250.

In step S250, a valid QR code is parsed, and the type of disease, type of biomaterial, and concentration information for each type of biomaterial are extracted. In addition, at this stage, the type of disease, type of biomaterial, and concentration information for each type of biomaterial extracted in a transmission format suitable for the network 300 may be modulated.

In step S260, the disease prediction client 200 transmits the type of the extracted disease, the type of the biomaterial, and the concentration information of the type of the biomaterial to the disease prediction server 400. In addition, the disease prediction client 200 may request the disease prediction server 400 to predict the risk of the user's disease or disease based on the transmitted information.

In step S270, the disease prediction client 200 waits until a response of the disease prediction server 400 is received. That is, the disease prediction client 200 will monitor whether the disease prediction result is received from the disease prediction server 400. [ If the disease prediction result is not received (No direction), the atmosphere is maintained, and if reception of the disease prediction result is confirmed (Yes direction), the procedure goes to step S280.

In step S280, the disease prediction client 200 displays the disease prediction result to the user as image information of a recognizable type. For example, the disease prediction client 200 may display the disease prediction result on a display as an image in the form of a character or a graph.

In step S290, the disease prediction client 200 will store the received disease prediction result in the storage 230. [ The disease prediction client 200 will manage accumulated disease prediction results as an illness prediction history of the authenticated user.

The method for implementing the disease prediction service in the disease prediction client 200, which can be implemented as a personal terminal, has been described above.

7 is a flowchart showing an operation method of the disease prediction server 400 for providing a disease prediction service. Referring to FIG. 7, the disease prediction method of the present invention using concentration information of each type of biomaterial will be described.

In step S310, the disease prediction server 400 generates a disease prediction model. As a precondition of the disease prediction calculation, the disease prediction server 400 should be provided with a previously generated disease prediction model in the form of big data. The disease prediction model may be stored in storage 430 and loaded into working memory 420. The disease prediction server 400 can sort various bio material combinations corresponding to various diseases by using learning data in advance. Based on the selected biomaterial combination, a disease prediction model can be created and a disease prediction model big data can be constructed. The storage 430 may include a disease prediction model repository 436 for storing a large-capacity disease prediction model.

In step S320, the disease prediction server module 425 will be executed in the disease prediction server 400. [ The disease prediction server module 425 may be provided as a daemon type program. It will be appreciated that the disease prediction server module 425 may be activated upon startup of the disease prediction server 400 and may be executed in response to operation of a network or other program. When the disease prediction server module 425 is activated, it is possible to receive a disease prediction request from the disease prediction client 200. [

In step S330, the disease prediction server 400 will monitor whether the disease prediction request and the concentration information of the bio material and type are received from the disease prediction client 200. [ If there is no disease prediction request from the disease prediction client 200 (No direction), the disease prediction server 400 will continuously monitor whether or not the disease prediction request is received. If the disease prediction request and the bio-material concentration information are received from the disease predicting client 200 (Yes direction), the procedure will move to step S340.

In step S340, the disease prediction server 400 performs a search for a disease prediction model constructed in advance using the received bio-material and concentration information. The disease prediction server 400 will search the disease prediction model repository 436 for disease prediction models corresponding to biomaterial combinations including at least one or both of the received biomaterials.

In step S350, the disease prediction server 400 selects one disease prediction model having the highest prediction accuracy of the disease among the plurality of searched disease models. The prediction accuracy is a value determined in the prediction model generation process. That is, in the prediction model generation process, the prediction accuracy for each of the disease prediction models constructed by combining one bio-material or a plurality of bio-materials will be calculated. And the prediction accuracy for each of the computed disease prediction models is stored in the disease prediction model repository 436. [ The evaluation process of the prediction accuracy will be described in detail in FIG. 8 to be described later.

In step S360, the disease prediction server 400 performs a concentration-based prediction operation of the bio material in the disease prediction model condition selected in step S350. That is, the disease prediction server 400 can input the concentration information of each of the received biomaterials into the selected disease prediction model to determine the disease status, the progress of the disease, and the risk. For example, the predicted degree of disease progression or risk may be determined by referring to the concentration of a specific biomaterial which is highly correlated with the degree of disease progression.

In step S370, the disease prediction server 400 transmits the disease prediction result to the disease prediction client 200. [ When the disease prediction result is transmitted, the requested operation from the disease prediction client 200 is completed.

In the above description, the type and concentration-based disease prediction method of the bio-material in the disease prediction server 400 has been described. The disease type, disease-specific prediction models, prediction accuracy of each prediction model, type of biomaterial, and the like, which are constructed in the disease prediction model repository 436, can be updated periodically or by learning when necessary. Therefore, according to the disease prediction server 400 of the present invention, it is not limited to a specific type of biomaterial or disease. It is also possible to predict the various diseases and to ensure the generality of services that adaptively provide disease risk.

FIG. 8 is a flowchart illustrating an exemplary method for generating the disease prediction model referred to in FIG. 7; FIG. Referring to Fig. 8, a method for generating a disease prediction model based on a biomaterial will be exemplarily described.

In step S311, the disease prediction server 400 reads the biomass concentration learning data from the biomass concentration learning data repository 438. [ Biomaterial concentration learning data is data obtained by extracting concentration values of a plurality of biomaterials by analyzing samples of body fluid or urine extracted from each of a normal control group and a disease group (Diesease) for a specific disease. An example of such biomaterial concentration learning data is shown in the table of FIG.

In step S312, the disease prediction server 400 reads the biomaterial concentration learning data and then evaluates the disease prediction accuracy for each individual biomaterial. Disease prediction accuracy is the process of generating a disease prediction model using a single bio-material and evaluating how accurately the disease prediction model generated based on the cross-validation method distinguishes between normal and disease. In this process, various kinds of classification methodologies such as K-Nearest Neighbors (KNN), SVM (Support Vector Machine), and RF (Random Forest) can be applied to the algorithm for generating a disease prediction model. By way of example, cross-validation methods can be applied to various cross-validation methodologies such as Leave One Out Cross-Validation (LOOCV) and N-fold Cross-Validation.

In step S313, when the disease prediction accuracy evaluation is completed for all the bio materials, the disease prediction server 400 selects only biomaterials that satisfy the accuracy of the reference value or more. The selection criterion can be an Accuracy value or a P-value obtained through the cross-validation method.

In step S314, the disease prediction server 400 generates a bio-material combination using all the selected bio-materials. Combinations of biomaterials can be combined to include all biomaterials selected from one biomaterial.

In step S315, the disease prediction server 400 performs a prediction accuracy evaluation of the disease for all the bio-material combinations generated. The prediction accuracy evaluation may be the same as the evaluation of the disease prediction accuracy for each bio material performed in step S312.

In step S316, the disease prediction server 400 determines a prediction model of the disease for the combination of the substances whose prediction accuracy is equal to or higher than the reference value, and stores it in the disease prediction model repository 436. [ In the disease prediction model, data generated according to the driving method of the prediction algorithm can be stored in various forms such as a file and a database.

In the above, a method of generating a disease prediction model using biomass concentration learning data has been described. Biomaterial concentration learning data can be updated when new diseases or biomaterials are generated. Thus, a disease prediction model can also be updated at any time.

FIG. 9 is a table showing the biomaterial concentration learning data described in FIG. Referring to FIG. 9, the bio-material concentration learning data may have a data structure in the form of a two-dimensional matrix. The rows and columns of the bio-material concentration learning data represent a single sample and a single bio-material, respectively, and the information of rows and columns can also be expressed in a form in which they are mutually replaced. Each single sample is tagged with class information (normal or diseased person). And there can be multiple diseases in one dataset. The biomaterial concentration learning data may be input in a file format such as a text file or an excel file, or may be input from a database.

FIG. 10 is a diagram illustrating an example of a method for providing a disease-based disease prediction service of a bio material, which is performed in the disease prediction apparatus 10 of the present invention. Referring to Figure 10, methods and apparatus for predicting cardiovascular disease based on metabolite concentration will be illustratively illustrated. It is to be understood, however, that these examples are not intended to limit the invention to the particular embodiments, but are to be understood to include all modifications, equivalents, and alternatives falling within the spirit and scope of the invention. The disease prediction system for predicting cardiovascular disease may include a concentration detection sensor 100, a disease prediction client 200, and a disease prediction server 400. The concentration detection sensor 100, the disease prediction client 200, and the disease prediction server 400 may be configured substantially the same as those of FIG.

The concentration detection sensor 100 can measure the concentration of specific metabolites in the blood using the blood sample. The concentration detection sensor 300 may be, for example, a high-sensitivity biosensor using LSPR (Local Surface Plasmon Resonance), which is a kind of optical method. The concentration detection sensor 300 can measure the permeability of the medium that changes in proportion to the concentration of the target metabolite after reacting using a specific enzyme of the metabolite in the blood sample. The concentration value of the metabolite can be measured by converting the measured value to the concentration. The concentration detecting sensor 300 may include a reaction part including a microfluid chip, a pump for moving a microfluid and a valve, a measuring part composed of a laser diode (LD) and a photodiode (PD) A touch-type LCD display module for inputting disease information and displaying a QR code, a control unit for control, and the like.

In this example, the process of predicting cardiovascular disease using four metabolic substances, hypoxanthine, inosine, linoleic acid, and oleic acid will be described. The concentration values of the four metabolites measured by the concentration detection sensor 100 will be converted into QR codes together with the disease information and displayed on the display of the concentration detection sensor 100. [ When a user runs a disease prediction client application in a disease prediction client 200 such as a smart phone and performs user authentication using a previously registered fingerprint, the corresponding QR code can be recognized. The QR code can be captured and recognized through the camera module of the disease prediction client 200. This recognition process may be represented by an image 262 on the display of the disease prediction client 200. The recognized QR codes are automatically analyzed and extracted as the disease type, the name of the four biomaterials, and the biomass concentration value. The extracted information will be transmitted to the disease prediction server 400.

The disease prediction server 400 searches all prediction models including one or more of the above four metabolites. The disease prediction server 400 selects one prediction model having the highest accuracy among the prediction models. In this embodiment, the disease prediction model may be generated by KNN (K-Nearest Neighbors) and SVM (Support Vector Machine). Assume that the SVM-based disease prediction model using three metabolites, hypoxanthine, inosine, and linoleic acid, has the highest accuracy. Then, the disease prediction is performed by inputting the concentration values of hypoxanthine, inosine, and linoleic acid into the corresponding model. After the disease prediction, the disease prediction server 400 will transmit the biomaterial type, disease state, and prediction accuracy information used in the prediction to the disease prediction client 200.

The disease prediction client 200 can visualize information received from the disease prediction server 400 as an image 264 on a display. The disease prediction client 200 stores the received disease prediction result in the storage and manages it as the disease measurement history of the authenticated user.

The procedure for predicting cardiovascular disease by the disease prediction system of the present invention has been briefly described above. However, it should be understood that the diseases predictable by the disease prediction system of the present invention are not limited to these, and that all diseases capable of being detected through the concentration of the biomaterial can be predicted. Updates using biomass concentration learning data can also be used to predict all existing and future diseases.

While the invention has been shown and described with reference to certain preferred embodiments thereof, it will be understood by those skilled in the art that various changes and modifications may be made therein without departing from the spirit and scope of the invention. Therefore, the scope of the present invention should not be limited to the above-described embodiments, but should be determined by the equivalents of the claims of the present invention as well as the claims of the following.

Claims (19)

Generating disease prediction models according to combinations and concentrations of biomaterials;
Receiving type information of the biomaterial detected from the sample and concentration information of the biomaterial from the disease prediction client;
Selecting one of the disease prediction models having the highest prediction accuracy;
Determining whether or not the disease corresponding to the selected disease prediction model is diagnosed by referring to the concentration information of the bio material; And
And transmitting the diagnosis or risk of the determined disease to the disease prediction client. The method according to claim 1,
Wherein the disease prediction client parses the QR code to provide the type information of the bio material and the concentration information of the bio material. 3. The method of claim 2,
Converting the type information of the bio material detected from the sample and the concentration information of the bio material into data of JSON format and converting the data of the JSON format into the QR code. The method of claim 3,
Wherein the disease prediction client recognizes the QR code using a camera module. The method according to claim 1,
Wherein generating the disease prediction models comprises:
Reading biomaterial concentration learning data;
Evaluating prediction accuracy of each of the biomaterials registered in the biomaterial concentration learning data;
Selecting biomaterials having a prediction accuracy higher than a reference value by referring to the evaluation result;
Generating combinations of the selected biomaterials;
Evaluating a prediction accuracy for each of the combinations; And
Selecting a biomaterial combination having the prediction accuracy equal to or higher than a reference value, and storing the biomaterial combination as a disease prediction model. 6. The method of claim 5,
Wherein the biomaterial concentration learning data includes concentration data of biomaterials corresponding to each of a normal group and a disease group. 6. The method of claim 5,
Evaluating the prediction accuracy of each of the biomaterials, or the prediction accuracy for each of the combinations, comprising:
Generating a disease prediction model; And
And applying a cross validation method for evaluating the ability of the disease prediction model to distinguish between normal and disease. 8. The method of claim 7,
Wherein at least one algorithm of KNN (K-Nearest Neighbors), SVM (Support Vector Machine), and RF (Random Forest) is applied in the step of generating the disease prediction model. 8. The method of claim 7,
Wherein the cross validation method comprises Leave One Out Cross-Validation (LOOCV) or N-fold Cross-Validation (NCV). The method according to claim 1,
Displaying the determination result on the display in the disease prediction client; And
And storing the determination result as a disease prediction history. The method according to claim 1,
Further comprising receiving information on the type of disease from the disease prediction client. A disease-predicting device based on the concentration of a bio-material comprising:
A concentration detecting sensor for sensing the type and concentration of the biomolecule from the sample, displaying the always-sensed information and disease information in a QR code;
A disease prediction client that recognizes the QR code and converts the QR code into the type and concentration information of the bio material; And
The method comprising: receiving a disease prediction request from the disease prediction client and the type and concentration information of the bio material; selecting one of the disease prediction models having the highest prediction accuracy for the disease information; And a disease prediction server for determining whether the disease is a disease or a risk by applying concentration information of the bio-material, and transmitting the determined disease or risk to the disease prediction client. 13. The method of claim 12,
Wherein the concentration detection sensor converts the type and concentration of the bio material and the disease information into data of JSON format and then converts the data into the QR code. 13. The method of claim 12,
Wherein the concentration detection sensor includes a display module for displaying the QR code. 13. The method of claim 12,
Wherein the disease prediction client includes a camera module for receiving the QR code as image information. 13. The method of claim 12,
Wherein the disease predicting client comprises a display device for indicating the presence or the risk of the disease transmitted from the disease predicting server. 17. The method of claim 16,
Wherein the disease prediction client includes a storage for storing the presence or the degree of the disease. 13. The method of claim 12,
Wherein the disease prediction client performs a user authentication operation for authenticating a user who is able to access the disease or the risk. 13. The method of claim 12,
Wherein the disease prediction server generates the disease prediction models using biomass concentration learning data.
KR1020160004911A 2015-08-19 2016-01-14 Disease forecast device based on concentration information of biomaterial and forecasting method thereof KR101903526B1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
KR20150116807 2015-08-19
KR1020150116807 2015-08-19

Publications (2)

Publication Number Publication Date
KR20170023382A true KR20170023382A (en) 2017-03-03
KR101903526B1 KR101903526B1 (en) 2018-10-05

Family

ID=58410595

Family Applications (1)

Application Number Title Priority Date Filing Date
KR1020160004911A KR101903526B1 (en) 2015-08-19 2016-01-14 Disease forecast device based on concentration information of biomaterial and forecasting method thereof

Country Status (1)

Country Link
KR (1) KR101903526B1 (en)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20180099185A (en) * 2017-02-28 2018-09-05 연세대학교 산학협력단 Method, server and system for generating disease prediction models
KR20180120469A (en) * 2017-04-27 2018-11-06 권오일 System for analyzing and predecting disease
WO2018236188A1 (en) * 2017-06-23 2018-12-27 한국기초과학지원연구원 Method for diagnosing myocardial infarction or predicting prognosis of same
KR20200114975A (en) * 2019-03-29 2020-10-07 연세대학교 산학협력단 Method for predicting of bacteremia risk and device for predicting of bacteremia risk using the same
KR20210016912A (en) * 2019-08-06 2021-02-17 울산과학기술원 Method and system to predict the progression of periodontitis
KR20220156681A (en) 2021-05-18 2022-11-28 주식회사 웨이센 Disease occurrence prediction system and method using three-dimensional multichannel data
KR20220168930A (en) * 2021-06-17 2022-12-26 주식회사 휴이노 Method, system and non-transitory computer-readable recording medium for managing analysis model for bio-signal

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR102190682B1 (en) * 2018-10-31 2020-12-14 한국 한의학 연구원 Method and system for screening diabetes using multiple biological signals

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007101482A (en) * 2005-10-07 2007-04-19 Matsushita Electric Ind Co Ltd Measuring tip and analytical method therefor
CA2783536A1 (en) * 2009-12-09 2011-06-16 Aviir, Inc. Biomarker assay for diagnosis and classification of cardiovascular disease
JP5603639B2 (en) * 2010-04-23 2014-10-08 国立大学法人京都大学 Learning device for prediction device and computer program therefor

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20180099185A (en) * 2017-02-28 2018-09-05 연세대학교 산학협력단 Method, server and system for generating disease prediction models
KR20180120469A (en) * 2017-04-27 2018-11-06 권오일 System for analyzing and predecting disease
WO2018236188A1 (en) * 2017-06-23 2018-12-27 한국기초과학지원연구원 Method for diagnosing myocardial infarction or predicting prognosis of same
KR20200114975A (en) * 2019-03-29 2020-10-07 연세대학교 산학협력단 Method for predicting of bacteremia risk and device for predicting of bacteremia risk using the same
KR20210016912A (en) * 2019-08-06 2021-02-17 울산과학기술원 Method and system to predict the progression of periodontitis
KR20220156681A (en) 2021-05-18 2022-11-28 주식회사 웨이센 Disease occurrence prediction system and method using three-dimensional multichannel data
KR20220168930A (en) * 2021-06-17 2022-12-26 주식회사 휴이노 Method, system and non-transitory computer-readable recording medium for managing analysis model for bio-signal

Also Published As

Publication number Publication date
KR101903526B1 (en) 2018-10-05

Similar Documents

Publication Publication Date Title
KR101903526B1 (en) Disease forecast device based on concentration information of biomaterial and forecasting method thereof
US10032114B2 (en) Predicting application performance on hardware accelerators
US10713716B2 (en) Verification model using neural networks
JP2017538195A (en) Hierarchical deep convolutional neural network
US11120369B2 (en) Communication generation using sparse indicators and sensor data
KR102460442B1 (en) Time series data processing device, health predicting system including the same, and method for operating time series data processing device
US20240086383A1 (en) Search engine optimization by selective indexing
US20170124523A1 (en) System and method for providing service for appointment at doctor's based on health status
CN105228531A (en) Menstruation starts pre-settled date accountant, program and organism analytical equipment
Matthews et al. Rapid dengue and outbreak detection with mobile systems and social networks
Mavrogiorgou et al. A pluggable IoT middleware for integrating data of wearable medical devices
KR20190000681A (en) A Method for diagnosis or progrosis prediction of myocardial infarction
Aybey et al. SENSDeep: an ensemble deep learning method for protein–protein interaction sites prediction
Ammunét et al. Deep learning tools are top performers in long non-coding RNA prediction
Chierici et al. Machine learning models for predicting endocrine disruption potential of environmental chemicals
JP2019003408A (en) Evaluation method, computer, and program for hyperparameter
Zhang et al. An intelligent method of cancer prediction based on mobile cloud computing
Chakraborty et al. Smart healthcare systems using big data
KR102284440B1 (en) Method to broker deep learning model transactions perfomed by deep learning model transaction brokerage servers
CN114664458A (en) Patient classification device, computer device and storage medium
Choudhury A framework for safeguarding artificial intelligence systems within healthcare
CN113990514A (en) Abnormality detection device for doctor diagnosis and treatment behavior, computer device and storage medium
JP2023518251A (en) Systems and methods for predicting expression levels
Kowall et al. Use of areas under the receiver operating curve (AROCs) and some caveats
Wu et al. Supervised variable selection in randomised controlled trials prior to exploration of treatment effect heterogeneity: an example from severe malaria

Legal Events

Date Code Title Description
A201 Request for examination
E902 Notification of reason for refusal
AMND Amendment
E601 Decision to refuse application
AMND Amendment
X701 Decision to grant (after re-examination)
GRNT Written decision to grant