WO2021238092A1 - 基于心电与脑电信息结合的糖尿病前期检测系统和方法 - Google Patents

基于心电与脑电信息结合的糖尿病前期检测系统和方法 Download PDF

Info

Publication number
WO2021238092A1
WO2021238092A1 PCT/CN2020/128559 CN2020128559W WO2021238092A1 WO 2021238092 A1 WO2021238092 A1 WO 2021238092A1 CN 2020128559 W CN2020128559 W CN 2020128559W WO 2021238092 A1 WO2021238092 A1 WO 2021238092A1
Authority
WO
WIPO (PCT)
Prior art keywords
combined feature
eeg
ecg
feature set
dimensionality reduction
Prior art date
Application number
PCT/CN2020/128559
Other languages
English (en)
French (fr)
Inventor
聂泽东
李景振
刘宇航
Original Assignee
中国科学院深圳先进技术研究院
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 中国科学院深圳先进技术研究院 filed Critical 中国科学院深圳先进技术研究院
Priority to US17/312,946 priority Critical patent/US20220313172A1/en
Publication of WO2021238092A1 publication Critical patent/WO2021238092A1/zh

Links

Images

Classifications

    • 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
    • A61B5/14532—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 for measuring glucose, e.g. by tissue impedance measurement
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/25—Bioelectric electrodes therefor
    • A61B5/279—Bioelectric electrodes therefor specially adapted for particular uses
    • A61B5/28—Bioelectric electrodes therefor specially adapted for particular uses for electrocardiography [ECG]
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/25—Bioelectric electrodes therefor
    • A61B5/279—Bioelectric electrodes therefor specially adapted for particular uses
    • A61B5/291—Bioelectric electrodes therefor specially adapted for particular uses for electroencephalography [EEG]
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316—Modalities, i.e. specific diagnostic methods
    • A61B5/318—Heart-related electrical modalities, e.g. electrocardiography [ECG]
    • A61B5/33—Heart-related electrical modalities, e.g. electrocardiography [ECG] specially adapted for cooperation with other devices
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A61B5/7246—Details of waveform analysis using correlation, e.g. template matching or determination of similarity
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B2562/00—Details of sensors; Constructional details of sensor housings or probes; Accessories for sensors
    • A61B2562/04—Arrangements of multiple sensors of the same type
    • A61B2562/046—Arrangements of multiple sensors of the same type in a matrix array
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/25—Bioelectric electrodes therefor
    • A61B5/251—Means for maintaining electrode contact with the body
    • A61B5/256—Wearable electrodes, e.g. having straps or bands
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316—Modalities, i.e. specific diagnostic methods
    • A61B5/318—Heart-related electrical modalities, e.g. electrocardiography [ECG]
    • A61B5/346—Analysis of electrocardiograms
    • A61B5/349—Detecting specific parameters of the electrocardiograph cycle
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316—Modalities, i.e. specific diagnostic methods
    • A61B5/369—Electroencephalography [EEG]
    • A61B5/372—Analysis of electroencephalograms
    • A61B5/374—Detecting the frequency distribution of signals, e.g. detecting delta, theta, alpha, beta or gamma waves
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
    • A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
    • A61B5/6802—Sensor mounted on worn items
    • A61B5/6803—Head-worn items, e.g. helmets, masks, headphones or goggles
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7271—Specific aspects of physiological measurement analysis
    • A61B5/7275—Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor

Definitions

  • the present invention relates to the medical and health technical field, and more specifically, to a pre-diabetes detection system and method based on the combination of ECG and EEG information.
  • Pre-diabetes refers to a period of impaired blood glucose regulation, including impaired fasting blood glucose and impaired glucose tolerance, but has not yet reached the diagnostic criteria for diabetes. According to statistics, about 25% of young people and about 20% of young people have pre-diabetes.
  • pre-diabetes stage For people in the pre-diabetes stage, if there is no intervention, about 10% of them will progress to diabetes every year; and if the corresponding measures can be used to intervene in the pre-diabetes stage in time, such as using drugs, controlling their diet, strengthening exercise, etc., diabetes The risk can be reduced by 30%-75%, and the probability of returning to a normal blood sugar state can rise to about 70%. Therefore, the detection of pre-diabetes has very important significance.
  • the current methods for determining pre-diabetes include: 1) Obtain blood glucose concentration through blood sampling under fasting conditions. If the fasting blood glucose value is between 5.6 mmol/L and 7.0 mmol/L, it can be determined as pre-diabetes; 2) Carry out oral glucose Tolerance test, 2 hours after oral glucose, the blood glucose concentration is obtained by blood sampling. If the blood glucose value is between 7.8mmol/L and 11.1mmol/L, it is judged as pre-diabetes.
  • the existing technical solutions all need to collect venous blood or fingertip blood, which will bring greater pain and risk of infection to the patient, and the detection cost is relatively high.
  • the purpose of the present invention is to overcome the above-mentioned shortcomings of the prior art, and provide a pre-diabetes detection method based on the combination of ECG and EEG information, by carrying out an oral glucose tolerance test, and using a wearable device to obtain the ECG and EEG simultaneously Information, and then extract the relevant features of the ECG and EEG to realize the detection of pre-diabetes.
  • a pre-diabetes detection system based on the combination of ECG and EEG information includes:
  • Signal acquisition module use wearable devices to synchronously acquire the user's ECG signal and EEG signal in a non-invasive manner;
  • Feature extraction module Use multiple methods to perform dimensionality reduction processing on a combined feature set composed of ECG features and EEG features to obtain multiple dimensionality reduction combined feature sets, and analyze the multiple dimensionality reduction combined feature sets
  • the correlation with the blood glucose concentration value screens out the ECG characteristics and EEG characteristics that meet the set correlation standards, forming an optimized combined feature set
  • Multi-mode fusion module used to input the optimized combined feature set into the trained multiple types of neural network models, and obtain the user's pre-diabetes detection results by fusing the output results of the multiple types of neural networks .
  • the synchronously acquiring the user's ECG signal and EEG signal includes:
  • V1 is set in the fourth intercostal space on the right edge of the sternum
  • V2 is set in the fourth intercostal space on the left edge of the sternum
  • V3 is set in V2 and V4.
  • V4 is set at the intersection of the left mid-clavicular line and the fifth intercostal space
  • V5 is horizontal to the anterior axillary line
  • V6 is horizontal to the mid-axillary line;
  • the user wears an EEG electrode cap on the head, and six electrodes for monitoring EEG signals are distributed in the EEG electrode cap, corresponding to the frontal, occipital, and parietal lobes of the left and right hemispheres of the brain;
  • the feature extraction module performs the following process:
  • the Pearson correlation analysis method is used to analyze the correlation, and the correlation standard is set as correlation k>0.2 and P ⁇ 0.05, and P represents the probability of performing a hypothesis test on the correlation coefficient.
  • the using principal component analysis to perform dimensionality reduction processing on the combined feature set includes:
  • the eigenvectors of the eigenvalues sorted by size, the composition of the matrix u [u 1, u 2 , u 3, ..., u n], corresponding to descending eigenvalues are ⁇ 1, ⁇ 2, ⁇ 3 ,..., ⁇ n , intercept the value of the set ratio at the top of the sequence from the matrix u as the new feature point of each feature to achieve data dimensionality reduction.
  • the multiple types of neural network models include at least two types of support vector machines, random forests, convolutional neural networks, long- and short-term memory networks, and recurrent neural networks.
  • the multi-mode fusion module uses a voting method to fuse the output results of the multiple types of neural networks to obtain the detection result of whether the user belongs to the pre-diabetes stage.
  • a method for detecting pre-diabetes based on the combination of ECG and EEG information includes the following steps:
  • the optimized combined feature set is input to the trained multiple types of neural network models, and the output results of the multiple types of neural networks are combined to obtain the user's pre-diabetes detection results.
  • the present invention has the advantages of non-invasive, no pain, convenience and comfort, low cost, real-time monitoring, etc., and can be widely used in the pre-diabetes detection of different groups such as children, adolescents, and the elderly.
  • Figure 1 is a schematic diagram of a pre-diabetes detection system based on a combination of ECG and EEG information according to an embodiment of the present invention
  • Fig. 2 is a schematic diagram of obtaining characteristic information of different segments from an ECG signal according to an embodiment of the present invention
  • Fig. 3 is a schematic diagram of selected ECG characteristics and EEG characteristics according to an embodiment of the present invention.
  • the pre-diabetes detection system based on the combination of ECG and EEG information provided by the embodiment of the present invention includes a signal acquisition module 110, a feature extraction module 120, a multi-mode fusion module 130 and a result output module 140.
  • the signal acquisition module 110 is used to collect the ECG signal and the EEG signal of the user (or called the subject). For example, the use of wearable devices to achieve simultaneous acquisition of the user's ECG and EEG signals during the oral glucose tolerance test.
  • the specific process of synchronously acquiring ECG and EEG signals is as follows: First, place the six electrode pads on the user's chest, and mark the electrode positions (or electrode identifiers) as V1-V6, which are the right sternum.
  • the user wears a disposable EEG electrode cap on the head.
  • the 6 electrodes are located in the frontal, occipital, and parietal lobes of the left and right hemispheres of the brain. .
  • the ECG acquisition equipment and the EEG acquisition equipment are turned on to realize the synchronous acquisition of the user's ECG and EEG signals. After 2 hours, stop collecting ECG and EEG signals, and export and save the data.
  • the feature extraction module 120 is used to reduce the dimensionality of ECG features and EEG features, and select some ECG features and EEG features most related to changes in blood glucose concentration for subsequent analysis, so as to improve the calculation speed.
  • the feature extraction module 120 can achieve dimensionality reduction in a variety of ways, such as using principal component analysis, independent component analysis, lasso regression analysis and other algorithms, and then perform correlation analysis on the features after multiple dimensionality reduction, and select those that are related to blood sugar. Some of the ECG characteristics and EEG characteristics that are most relevant to changes in concentration.
  • the specific implementation process of the feature extraction module 120 includes:
  • step S210 the ECG features are extracted from the ECG signal.
  • the ECG signal is analyzed, and the characteristic information of different segments is extracted from the ECG signal, including the characteristic information of 9 different segments such as RRI, RH, PH, QRS, PRQ, QT, QTC, ST, HR, etc.
  • the labeling of the fragment information is shown in Figure 2, and the detailed meaning is shown in Table 1.
  • Table 1 Feature information of different segments extracted from ECG signal
  • Step S220 extracting EEG features from EEG signals.
  • the EEG signal is analyzed, and the signal features of different frequency bands in the frontal, occipital, and parietal lobes of the left and right hemispheres of the brain are respectively extracted. For example, a total of 30 EEG feature information is extracted, as shown in Table 2 below.
  • Table 2 Feature information in different frequency bands extracted from EEG signals
  • step S230 multiple methods are used to reduce the dimensionality of the ECG feature and the EEG feature set.
  • the dimensionality reduction process based on principal component analysis is as follows:
  • the i-th feature can be expressed as:
  • x (i) (x 1 (i) ,x 2 (i) ,x 3 (i) ,...,x n (i) ) T (2)
  • the covariance matrix is a square matrix of size n ⁇ n with n feature points.
  • the dimensionality reduction process based on independent component analysis is as follows:
  • the i-th feature can be expressed as:
  • x (i) (x 1 (i) ,x 2 (i) ,x 3 (i) ,...,x n (i) ) T (6)
  • ⁇ , g are scale coefficients, and their values are determined by the maximum likelihood method.
  • the value of the W matrix is determined, the value of the Q matrix can be calculated, which is the feature set after dimensionality reduction through independent component analysis.
  • the dimensionality reduction process based on lasso regression analysis is as follows:
  • the i-th feature can be expressed as:
  • x (i) (x 1 (i) ,x 2 (i) ,x 3 (i) ,...,x n (i) ) T (10)
  • the lasso regression analysis process can be regarded as the solution process of convex optimization, namely:
  • J is the cost function and K is the correlation matrix to be solved.
  • K is the correlation matrix to be solved.
  • Step S240 Perform correlation analysis on the combined feature set after dimensionality reduction in multiple ways, and then select the ECG feature and the EEG feature that meet the set correlation standard to form an optimized combined feature set.
  • the ECG and EEG feature set Q after dimensionality reduction are obtained by independent component analysis, and the dimensionality reduction analysis by lasso regression ECG and EEG feature set Y, using the Pearson correlation analysis method to analyze the correlation between the above three ECG and EEG feature sets and blood glucose concentration after dimensionality reduction, that is, analyze the blood glucose concentration value and RRI, RH, PH, 9 ECG features such as QRS, PRQ, QT, QTC, ST, HR, and 30 such as F1 T , F1 D , F1 A , F1 B , F1 G , O1 T , O1 D , O1 A , O1 B , O1 G, etc.
  • the final selected features include RRI, QT, QRS three ECG features and O1 A , O1 D , P2 A , P2 B , P2 T five brain features Electrical characteristics.
  • the direction of the feature point projection with the largest variance can be selected through principal component analysis to obtain a low-dimensional feature set.
  • the principal component analysis is suitable for the case where the sample has a Gaussian distribution, while the independent component analysis does not require the sample to have a Gaussian distribution.
  • Lasso regression analysis can effectively reduce data dimensionality and accurately identify more important features. By adopting a variety of dimensionality reduction methods combined with correlation analysis for feature screening, while effectively reducing data dimensionality, it can accurately identify features that have a strong correlation with blood glucose concentration, which improves the subsequent processing speed and does not affect the detection accuracy. And it is not sensitive to the distribution of characteristic data, and the scope of application is wider.
  • the multi-mode fusion module 130 is used to obtain multiple detection results of pre-diabetes in different ways based on the selected ECG feature and EEG feature set.
  • support vector machines random forests, convolutional neural networks, long and short-term memory networks, recurrent neural networks, etc. can be used, and the extracted three ECG features of RRI, QT, QRS and O1 A , O1 D , P2
  • the five EEG features of A, P2 B , and P2 T are used as the input of the five network models, and appropriate parameters are selected.
  • the output results of each network model are obtained respectively, that is, whether it belongs to pre-diabetes . Further, the fusion result is obtained by using a voting method.
  • the result output module 140 is used to display the final judgment result to the user.
  • the detection result can be displayed by voice, text, etc.
  • the present invention uses a wearable device to obtain the ECG signal and the EEG signal synchronously, by reducing the dimensionality of the ECG feature and the EEG feature and selecting an optimized combination feature set, and then based on the fusion of multiple algorithms for diabetes
  • the pre-diabetes detection method can provide users with a non-invasive, pain-free, convenient and comfortable, low-cost pre-diabetes detection method, which can be widely used in the pre-diabetes detection of different groups such as children, adolescents, and the elderly.
  • the present invention may be a system, a method and/or a computer program product.
  • the computer program product may include a computer-readable storage medium loaded with computer-readable program instructions for enabling a processor to implement various aspects of the present invention.
  • the computer-readable storage medium may be a tangible device that can hold and store instructions used by the instruction execution device.
  • the computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing.
  • Non-exhaustive list of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM) Or flash memory), static random access memory (SRAM), portable compact disk read-only memory (CD-ROM), digital versatile disk (DVD), memory stick, floppy disk, mechanical encoding device, such as a printer with instructions stored thereon
  • RAM random access memory
  • ROM read-only memory
  • EPROM erasable programmable read-only memory
  • flash memory flash memory
  • SRAM static random access memory
  • CD-ROM compact disk read-only memory
  • DVD digital versatile disk
  • memory stick floppy disk
  • mechanical encoding device such as a printer with instructions stored thereon
  • the computer-readable storage medium used here is not interpreted as a transient signal itself, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (for example, light pulses through fiber optic cables), or through wires Transmission of electrical signals.
  • the computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to various computing/processing devices, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and/or a wireless network.
  • the network may include copper transmission cables, optical fiber transmission, wireless transmission, routers, firewalls, switches, gateway computers, and/or edge servers.
  • the network adapter card or network interface in each computing/processing device receives computer-readable program instructions from the network, and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing/processing device .
  • the computer program instructions used to perform the operations of the present invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, status setting data, or in one or more programming languages.
  • Source code or object code written in any combination, the programming language includes object-oriented programming languages such as Smalltalk, C++, etc., and conventional procedural programming languages such as "C" language or similar programming languages.
  • Computer-readable program instructions can be executed entirely on the user's computer, partly on the user's computer, executed as a stand-alone software package, partly on the user's computer and partly executed on a remote computer, or entirely on the remote computer or server implement.
  • the remote computer can be connected to the user's computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (for example, using an Internet service provider to connect to the user’s computer) connect).
  • LAN local area network
  • WAN wide area network
  • an electronic circuit such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), can be personalized by using the status information of the computer-readable program instructions.
  • the computer-readable program instructions are executed to realize various aspects of the present invention.
  • These computer-readable program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine that makes these instructions when executed by the processor of the computer or other programmable data processing device , A device that implements the functions/actions specified in one or more blocks in the flowchart and/or block diagram is produced. It is also possible to store these computer-readable program instructions in a computer-readable storage medium. These instructions make computers, programmable data processing apparatuses, and/or other devices work in a specific manner, so that the computer-readable medium storing the instructions includes An article of manufacture, which includes instructions for implementing various aspects of the functions/actions specified in one or more blocks in the flowcharts and/or block diagrams.
  • each block in the flowchart or block diagram can represent a module, program segment, or part of an instruction, and the module, program segment, or part of an instruction contains one or more components for realizing the specified logical function.
  • Executable instructions can be included in the blocks in the flowchart or block diagram.
  • the functions marked in the block may also occur in a different order from the order marked in the drawings. For example, two consecutive blocks can actually be executed in parallel, or they can sometimes be executed in the reverse order, depending on the functions involved.
  • each block in the block diagram and/or flowchart, and the combination of the blocks in the block diagram and/or flowchart can be implemented by a dedicated hardware-based system that performs the specified functions or actions Or it can be realized by a combination of dedicated hardware and computer instructions. It is well known to those skilled in the art that implementation through hardware, implementation through software, and implementation through a combination of software and hardware are all equivalent.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Public Health (AREA)
  • Medical Informatics (AREA)
  • Biomedical Technology (AREA)
  • General Health & Medical Sciences (AREA)
  • Pathology (AREA)
  • Molecular Biology (AREA)
  • Animal Behavior & Ethology (AREA)
  • Veterinary Medicine (AREA)
  • Surgery (AREA)
  • Biophysics (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Artificial Intelligence (AREA)
  • Psychiatry (AREA)
  • Physiology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Signal Processing (AREA)
  • Cardiology (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Physics (AREA)
  • Fuzzy Systems (AREA)
  • Evolutionary Computation (AREA)
  • Primary Health Care (AREA)
  • Databases & Information Systems (AREA)
  • Epidemiology (AREA)
  • Psychology (AREA)
  • Optics & Photonics (AREA)
  • Emergency Medicine (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
  • Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
  • Pulmonology (AREA)

Abstract

一种基于心电与脑电信息结合的糖尿病前期检测系统和方法，包括：信号获取模块(110)，其利用可穿戴设备以无创方式同步获取用户的心电信号和脑电信号；特征提取模块(120)，其利用多种方式对由心电特征和脑电特征构成的组合特征集进行降维处理，获得多个降维的组合特征集，并通过分析多个降维的组合特征集与血糖浓度值的相关性选出满足设定相关性标准的心电特征和脑电特征，构成优化的组合特征集；多模融合模块(130)，用于将优化的组合特征集分别输入至经训练的多种类型的神经网络模型，通过融合多种类型的神经网络的输出结果，获得用户糖尿病前期的检测结果。这一方法和系统能够提供无创、无疼痛、便捷舒适、低成本的糖尿病前期检测方案。

Description

基于心电与脑电信息结合的糖尿病前期检测系统和方法 技术领域
本发明涉及医疗健康技术领域，更具体地，涉及一种基于心电与脑电信息结合的糖尿病前期检测系统和方法。
背景技术
在由血糖正常的健康群体，逐渐发展成糖尿病人群的过程中存在一个时期，即糖尿病前期。糖尿病前期是指血糖调节功能受损,包括空腹血糖受损以及糖耐量受损，但是尚未达到糖尿病诊断标准的一段时期。据统计，有约25％的年轻人,以及约20％的青少年都患有糖尿病前期。对处于糖尿病前期的人，如果不进行干预，每年会有10％左右进展为糖尿病；而如果在糖尿病前期能及时采用相应措施进行干预，例如使用药物、控制其饮食、加强运动等，发生糖尿病的危险能够下降30％-75％，恢复为正常血糖状况的几率最高可升至70％左右。因此，对糖尿病前期的检测具有非常重要的意义。
目前糖尿病前期的判定方法包括：1)在空腹情况下通过采血获取血糖浓度值，如果空腹血糖值在5.6mmol/L至7.0mmol/L之间，则可判定为糖尿病前期；2)开展口服葡萄糖耐量试验，口服葡萄糖2小时后，通过采血获取血糖浓度，如果血糖值在7.8mmol/L至11.1mmol/L，则判定为糖尿病前期。然而，现有技术方案均需采集静脉血或者指尖血，会给患者带来较大疼痛和感染的风险，并且检测成本较高。
发明内容
本发明的目的是克服上述现有技术的缺陷，提供一种基于心电和脑电信息相结合的糖尿病前期检测方法，通过开展口服葡萄糖耐量试验，并利用可穿戴设备同步获取心电和脑电信息，进而提取心电和脑电的相关特征， 实现糖尿病前期的检测。
根据本发明的第一方面，提供一种基于心电与脑电信息结合的糖尿病前期检测系统。该系统包括：
信号获取模块：利用可穿戴设备以无创方式同步获取用户的心电信号和脑电信号；
特征提取模块：利用多种方式对由心电特征和脑电特征构成的组合特征集进行降维处理，获得多个降维的组合特征集，并通过分析所述多个降维的组合特征集与血糖浓度值的相关性筛选出满足设定相关性标准的心电特征和脑电特征，构成优化的组合特征集；
多模融合模块：用于将所述优化的组合特征集分别输入至经训练的多种类型的神经网络模型，通过融合所述多种类型的神经网络的输出结果，获得用户糖尿病前期的检测结果。
在一个实施例中，所述同步获取用户的心电信号和脑电信号包括：
将用于监测心电信号的六个电极片V1至V6分别放置在用户胸部，其中V1设置在胸骨右缘第四肋间，V2设置在胸骨左缘第四肋间，V3设置在V2与V4之间连线的中点，V4设置在左锁骨中线与第五肋间交叉处，V5水平于腋前线，V6水平于腋中线；
用户在头部佩戴脑电电极帽，该脑电电极帽中分布设置用于监测脑电信号的六个电极，分别对应大脑左半球和右半球的额叶、枕叶、顶叶；
开展葡萄糖耐量试验，并开启心电采集设备和脑电采集设备，以同步获取用户的心电信号和脑电信号。
在一个实施例中，所述特征提取模块执行以下过程：
从所述心电信号中提取多个不同片段的特征信息，从所述脑电信号中分别提取对应大脑不同位置的不同频段的多个脑电特征信息，构成所述组合特征集；
利用主成分分析对所述组合特征集进行降维处理，获得第一组合特征集；
利用独立成分分析对所述组合特征集进行降维处理，获得第二组合特征集；
利用套索回归分析对所述组合特征集进行降维处理，获得第三组合特征集；
分别分析所述第一组合特征集、所述第二组合特征集和所述第三组合特征集与血糖浓度的相关性，进而筛选出满足设定相关性标准的心电特征和脑电特征，构成所述优化的组合特征集。
在一个实施例中，利用Pearson相关分析法分析相关性，所述相关性标准设置为相关性k＞0.2且P≤0.05，P表示对相关性系数进行假设检验的概率。
在一个实施例中，所述利用主成分分析对所述组合特征集进行降维处理包括：
计算所述组合特征集中各特征的特征点的协方差矩阵；
计算所述协方差矩阵的特征向量和对应的特征值：
将特征向量按特征值的大小排序，组成矩阵u＝[u 1,u 2,u 3,...,u n]，对应的特征值由大到小分别是λ 1,λ 2,λ 3,...,λ n，从矩阵u中截取排序靠前的设定比例的值作为每个特征的新特征点，以实现数据降维。
在一个实施例中，所述多种类型的神经网络模型包括支持向量机、随机森林、卷积神经网络、长短期记忆网络、递归神经网络中的至少两种类型。
在一个实施例中，所述多模融合模块采用投票法融合所述多种类型的神经网络的输出结果，获得用户是否属于糖尿病前期的检测结果。
根据本发明的第二方面，提供一种基于心电与脑电信息结合的糖尿病前期检测方法。该方法包括以下步骤：
利用可穿戴设备以无创方式同步获取用户的心电信号和脑电信号；
利用多种方式对由心电特征和脑电特征构成的组合特征集进行降维处理，获得多个降维的组合特征集，并通过分析所述多个降维的组合特征集与血糖浓度值的相关性筛选出满足设定相关性标准的心电特征和脑电特征，构成优化的组合特征集；
将所述优化的组合特征集分别输入至经训练的多种类型的神经网络模型，通过融合所述多种类型的神经网络的输出结果，获得用户糖尿病前 期的检测结果。
与现有技术相比，本发明的优点在于，具有无创、无疼痛、便捷舒适、低成本、实时监测等优势，能够广泛运用于儿童、青少年、老人等不同群体的糖尿病前期检测。
通过以下参照附图对本发明的示例性实施例的详细描述，本发明的其它特征及其优点将会变得清楚。
附图说明
被结合在说明书中并构成说明书的一部分的附图示出了本发明的实施例，并且连同其说明一起用于解释本发明的原理。
图1是根据本发明一个实施例的基于心电和脑电信息相结合的糖尿病前期检测系统的示意图；
图2是根据本发明一个实施例的从心电信号中获取其不同片段的特征信息的示意图；
图3是根据本发明一个实施例的所选取的心电特征和脑电特征的示意图。
具体实施方式
现在将参照附图来详细描述本发明的各种示例性实施例。应注意到：除非另外具体说明，否则在这些实施例中阐述的部件和步骤的相对布置、数字表达式和数值不限制本发明的范围。
以下对至少一个示例性实施例的描述实际上仅仅是说明性的，决不作为对本发明及其应用或使用的任何限制。
对于相关领域普通技术人员已知的技术、方法和设备可能不作详细讨论，但在适当情况下，所述技术、方法和设备应当被视为说明书的一部分。
在这里示出和讨论的所有例子中，任何具体值应被解释为仅仅是示例性的，而不是作为限制。因此，示例性实施例的其它例子可以具有不同的值。
应注意到：相似的标号和字母在下面的附图中表示类似项，因此，一 旦某一项在一个附图中被定义，则在随后的附图中不需要对其进行进一步讨论。
参见图1所示，本发明实施例提供的基于心电与脑电信息结合的糖尿病前期检测系统包括信号获取模块110、特征提取模块120、多模融合模块130和结果输出模块140。
信号获取模块110用于采集用户(或称被测者)的心电信号和脑电信号。例如，利用可穿戴设备实现同步获取用户在开展口服葡萄糖耐量试验时的心电和脑电信号。在一个实施例中，同步获取心电和脑电信号的具体过程是：首先，将六个电极片分别放置在用户胸部，将电极位置(或电极标识)标记为V1-V6，分别是胸骨右缘第四肋间V1；胸骨左缘第四肋间V2；V3的位置是V2和V4之间连线的中点；V4位于左锁骨中线和第五肋间交叉处；V5水平于腋前线；V6水平于腋中线。并且，用户在头部佩戴一个一次性脑电电极帽，电极帽中分布有6个电极用于监测脑电信息，6个电极分别位于大脑左半球及右半球的额叶、枕叶、顶叶。接下来，在用户佩戴好心电电极及脑电电极且检查无误后，静坐10分钟，待气息平稳后，开展口服葡萄糖耐量试验，例如服用75克葡萄糖。与此同时，开启心电采集设备以及脑电采集设备，实现同步获取用户的心电和脑电信号。2小时后，停止采集心电和脑电信号，并将数据导出保存。
特征提取模块120用于实现心电特征和脑电特征的降维，并选择出与血糖浓度变化最为相关的一些心电特征及脑电特征用于后续分析，以提高计算速度。
特征提取模块120可通过多种方式实现降维，例如利用主成分分析、独立成分分析、套索回归分析等算法，进而通过对多种方式降维后的特征进行相关性分析，选出与血糖浓度变化最为相关的一些心电特征及脑电特征。
在一个实施例中，特征提取模块120的具体实现过程包括：
步骤S210，从心电信号中提取心电特征。
例如，对心电信号进行分析，从心电信号中提取其不同片段的特征信息，包括RRI，R-H，P-H，QRS，PRQ，QT，QTC，ST，HR等9个不同片段的 特征信息，部分片段信息的标注如图2所示，详细含义如下表1。
表1：心电信号所提取的不同片段的特征信息
心电信号的不同片段特征信息名称 含义
RRI 心电信号中相邻两个R点的时间长度
R-H 心电信号中R波的高度
P-H 心电信号中P波的高度
QRS QRS波群的周期长度
PRQ 心电信号中R点到P点的时间长度
QT 从Q波开始到T波结束的时间长度
QTC 经过校准后的QT值
ST S点到T点中的直线段的时间距离
HR 一分钟内心跳的次数
步骤S220，从脑电信号中提取脑电特征。
具体地，对脑电信号进行分析，分别提取大脑左半球和右半球的额叶、枕叶、顶叶的不同频段的信号特征，例如，共提取30个脑电特征信息，如下表2。
表2：脑电信号所提取的不同频段下的特征信息
Figure PCTCN2020128559-appb-000001
Figure PCTCN2020128559-appb-000002
步骤S230，采用多种方式对心电特征和脑电特征集进行降维。
具体地，考虑到所提取的心电特征和脑电特征维度过长，为了实现快速计算，可采用主成分分析、独立成分分析、套索回归分析三种算法分别对心电特征和脑电特征数据进行降维处理。
在一个实施例中，基于主成分分析的降维过程如下：
以获取了共39个特征(包括9个心电特征和30个脑电特征)为例，假设每个特征有n个特征点，则所有的心电、脑电特征集(或称组合特征集)可表示为：
X＝{x (1),x (2),x (3),...,x (39)}      (1)
其中第i个特征可表示为：
x (i)＝(x 1 (i),x 2 (i),x 3 (i),...,x n (i)) T     (2)
则特征集的协方差矩阵为：
Figure PCTCN2020128559-appb-000003
其中，协方差矩阵为n×n大小的方阵，具有n个特征点。
计算协方差矩阵的特征向量及对应的特征值：
Uu＝λu          (4)
进一步地，将特征向量按特征值的大小按列排放，组成矩阵u＝[u 1,u 2,u 3,...,u n]，对应的特征值由大到小分别为：λ 1,λ 2,λ 3,...,λ n，则特征向量u 1为主特征向量(对应的特征值最大)，u 2为次特征向量，以此类推。通过截取矩阵u中例如前10％的值作为每个特征的新特征点，从而实现数据的降维。
在一个实施例中，基于独立成分分析的降维过程如下：
同样假设所有的心电、脑电特征集可表示为
X＝{x (1),x (2),x (3),...,x (39)}       (5)
其中第i个特征可表示为：
x (i)＝(x 1 (i),x 2 (i),x 3 (i),...,x n (i)) T     (6)
假设经过独立成分分析降维后的特征集可表示为：
Q＝{q (1),q (2),q (3),...,q (39)}      (7)
根据独立成分分析，则有：
X＝AQ       (8)
令W＝A -1，则q (i)＝A -1x (i)＝Wx (i)。
其中
Figure PCTCN2020128559-appb-000004
(即将W表示成另一种形式)，α，g为比例系数，其值大小采用极大似然法确定。当确定W矩阵的值后，则可计算出Q矩阵的值，该值即为经过独立成分分析降维后的特征集。
在一个实施例中，基于套索回归分析的降维过程如下：
同样假设所有的心电、脑电特征集可表示为
X＝{x (1),x (2),x (3),...,x (39)}       (9)
其中第i个特征可表示为：
x (i)＝(x 1 (i),x 2 (i),x 3 (i),...,x n (i)) T      (10)
假设经过套索回归分析降维后的特征集可表示为：
Y＝{y (1),y (2),y (3),...,y (39)}         (11)
套索回归分析过程可以视为凸优化的求解过程，即：
Figure PCTCN2020128559-appb-000005
其中J为代价函数，K为待求解的相关矩阵，通过利用含有范数的拉格朗日函数求解方法，解析出K矩阵的大小，则可以得到J函数的表达式。
步骤S240，对采用多种方式进行降维后的组合特征集进行相关性分析，进而选取满足设定相关性标准的心电特征和脑电特征构成优化的组合特征集。
具体地，对于利用主成分分析得到降维后的心电、脑电特征集u，利用独立成分分析得到降维后的心电、脑电特征集Q，以及利用套索回归分析降维后的心电、脑电特征集Y，利用Pearson相关分析方法分别分析上述降维后的3个心电、脑电特征集与血糖浓度的相关性，即分别分析血糖浓度值与RRI，R-H，P-H，QRS，PRQ，QT，QTC，ST，HR等9个心电特征以及 F1 T，F1 D，F1 A，F1 B，F1 G，O1 T，O1 D，O1 A，O1 B，O1 G等30个脑电特征的相关性。例如，如果满足相关系数k＞0.2且P值(双尾)≤0.05(其中k用于表示相关程度，P是对相关系数进行假设检验的概率，P≤0.05表示有显著相关关系)，则认为血糖值与该特征具有较强的相关性，则该特征暂时保留。
相关分析完成后，通过选取在主成分分析、独立成分分析、套索回归分析等三种算法中均满足相关性(k＞0.2，P值(双尾)≤0.05)的心电、脑电特征，将这些特征作为可使用的特征，如图3所示，最终所选取的特征包括RRI，QT，QRS三个心电特征以及O1 A，O1 D，P2 A，P2 B，P2 T五个脑电特征。
在本发明实施例中，通过主成分分析选择特征点投影具有最大方差的方向，能够获得低维特征集，主成分分析适用于样本呈高斯分布的情况，而独立成分分析不要求样本呈高斯分布，套索回归分析能够有效进行数据降维并准确识别更重要的特征。通过采用多种降维方法并结合相关性分析进行特征筛选，在有效进行数据降维的同时，能够准确识别出与血糖浓度具有强相关性的特征，提高了后续处理速度并且不影响检测精度，并且对特征数据的分布不敏感，适用范围更广泛。
多模融合模块130用于基于选出的心电特征和脑电特征集，利用不同方式获得糖尿病前期的多个检测结果。
具体地，可利用支持向量机、随机森林、卷积神经网络、长短期记忆网络、递归神经网络等，并将所提取的RRI，QT，QRS三个心电特征以及O1 A，O1 D，P2 A，P2 B，P2 T五个脑电特征分别作为该五种网络模型的输入，选择合适的参数，经过多次迭代训练后，分别得出每种网络模型的输出结果，即是否属于糖尿病前期。进一步地，通过采用投票法获得融合结果，例如，如果五种网络模型有三种或三种以上的输出结果为糖尿病前期，则判定为该用户为糖尿病前期，否则判定该用户为非糖尿病前期。采用融合结果作为最终的检测结果，提高了检测准确性。
结果输出模块140，用于向用户显示最终的判断结果。例如，可通过语音、文本等显示检测结果。
综上所述，本发明利用可穿戴设备同步获取心电信号和脑电信号，通过对心电特征和脑电特征进行降维处理并选择优化的组合特征集，进而基 于多种算法融合的糖尿病前期检测方法，能够为用户提供无创、无疼痛、便捷舒适、低成本的糖尿病前期检测方法，可广泛运用于儿童、青少年、老人等不同群体的糖尿病前期检测。
本发明可以是系统、方法和/或计算机程序产品。计算机程序产品可以包括计算机可读存储介质，其上载有用于使处理器实现本发明的各个方面的计算机可读程序指令。
计算机可读存储介质可以是可以保持和存储由指令执行设备使用的指令的有形设备。计算机可读存储介质例如可以是――但不限于――电存储设备、磁存储设备、光存储设备、电磁存储设备、半导体存储设备或者上述的任意合适的组合。计算机可读存储介质的更具体的例子(非穷举的列表)包括：便携式计算机盘、硬盘、随机存取存储器(RAM)、只读存储器(ROM)、可擦式可编程只读存储器(EPROM或闪存)、静态随机存取存储器(SRAM)、便携式压缩盘只读存储器(CD-ROM)、数字多功能盘(DVD)、记忆棒、软盘、机械编码设备、例如其上存储有指令的打孔卡或凹槽内凸起结构、以及上述的任意合适的组合。这里所使用的计算机可读存储介质不被解释为瞬时信号本身，诸如无线电波或者其他自由传播的电磁波、通过波导或其他传输媒介传播的电磁波(例如，通过光纤电缆的光脉冲)、或者通过电线传输的电信号。
这里所描述的计算机可读程序指令可以从计算机可读存储介质下载到各个计算/处理设备，或者通过网络、例如因特网、局域网、广域网和/或无线网下载到外部计算机或外部存储设备。网络可以包括铜传输电缆、光纤传输、无线传输、路由器、防火墙、交换机、网关计算机和/或边缘服务器。每个计算/处理设备中的网络适配卡或者网络接口从网络接收计算机可读程序指令，并转发该计算机可读程序指令，以供存储在各个计算/处理设备中的计算机可读存储介质中。
用于执行本发明操作的计算机程序指令可以是汇编指令、指令集架构(ISA)指令、机器指令、机器相关指令、微代码、固件指令、状态设置数据、或者以一种或多种编程语言的任意组合编写的源代码或目标代码，所述编程语言包括面向对象的编程语言—诸如Smalltalk、C++等，以及常规 的过程式编程语言—诸如“C”语言或类似的编程语言。计算机可读程序指令可以完全地在用户计算机上执行、部分地在用户计算机上执行、作为一个独立的软件包执行、部分在用户计算机上部分在远程计算机上执行、或者完全在远程计算机或服务器上执行。在涉及远程计算机的情形中，远程计算机可以通过任意种类的网络—包括局域网(LAN)或广域网(WAN)—连接到用户计算机，或者，可以连接到外部计算机(例如利用因特网服务提供商来通过因特网连接)。在一些实施例中，通过利用计算机可读程序指令的状态信息来个性化定制电子电路，例如可编程逻辑电路、现场可编程门阵列(FPGA)或可编程逻辑阵列(PLA)，该电子电路可以执行计算机可读程序指令，从而实现本发明的各个方面。
这里参照根据本发明实施例的方法、装置(系统)和计算机程序产品的流程图和/或框图描述了本发明的各个方面。应当理解，流程图和/或框图的每个方框以及流程图和/或框图中各方框的组合，都可以由计算机可读程序指令实现。
这些计算机可读程序指令可以提供给通用计算机、专用计算机或其它可编程数据处理装置的处理器，从而生产出一种机器，使得这些指令在通过计算机或其它可编程数据处理装置的处理器执行时，产生了实现流程图和/或框图中的一个或多个方框中规定的功能/动作的装置。也可以把这些计算机可读程序指令存储在计算机可读存储介质中，这些指令使得计算机、可编程数据处理装置和/或其他设备以特定方式工作，从而，存储有指令的计算机可读介质则包括一个制造品，其包括实现流程图和/或框图中的一个或多个方框中规定的功能/动作的各个方面的指令。
也可以把计算机可读程序指令加载到计算机、其它可编程数据处理装置、或其它设备上，使得在计算机、其它可编程数据处理装置或其它设备上执行一系列操作步骤，以产生计算机实现的过程，从而使得在计算机、其它可编程数据处理装置、或其它设备上执行的指令实现流程图和/或框图中的一个或多个方框中规定的功能/动作。
附图中的流程图和框图显示了根据本发明的多个实施例的系统、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上，流程 图或框图中的每个方框可以代表一个模块、程序段或指令的一部分，所述模块、程序段或指令的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。在有些作为替换的实现中，方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如，两个连续的方框实际上可以基本并行地执行，它们有时也可以按相反的顺序执行，这依所涉及的功能而定。也要注意的是，框图和/或流程图中的每个方框、以及框图和/或流程图中的方框的组合，可以用执行规定的功能或动作的专用的基于硬件的系统来实现，或者可以用专用硬件与计算机指令的组合来实现。对于本领域技术人员来说公知的是，通过硬件方式实现、通过软件方式实现以及通过软件和硬件结合的方式实现都是等价的。
以上已经描述了本发明的各实施例，上述说明是示例性的，并非穷尽性的，并且也不限于所披露的各实施例。在不偏离所说明的各实施例的范围和精神的情况下，对于本技术领域的普通技术人员来说许多修改和变更都是显而易见的。本文中所用术语的选择，旨在最好地解释各实施例的原理、实际应用或对市场中的技术改进，或者使本技术领域的其它普通技术人员能理解本文披露的各实施例。本发明的范围由所附权利要求来限定。

Claims (9)

  1. 一种基于心电与脑电信息结合的糖尿病前期检测系统，包括：
    信号获取模块：利用可穿戴设备以无创方式同步获取用户的心电信号和脑电信号；
    特征提取模块：利用多种方式对由心电特征和脑电特征构成的组合特征集进行降维处理，获得多个降维的组合特征集，并通过分析所述多个降维的组合特征集与血糖浓度值的相关性筛选出满足设定相关性标准的心电特征和脑电特征，构成优化的组合特征集；
    多模融合模块：用于将所述优化的组合特征集分别输入至经训练的多种类型的神经网络模型，通过融合所述多种类型的神经网络的输出结果，获得用户糖尿病前期的检测结果。
  2. 根据权利要求1所述的系统，其中，所述同步获取用户的心电信号和脑电信号包括：
    将用于监测心电信号的六个电极片V1至V6分别放置在用户胸部，其中V1设置在胸骨右缘第四肋间，V2设置在胸骨左缘第四肋间，V3设置在V2与V4之间连线的中点，V4设置在左锁骨中线与第五肋间交叉处，V5水平于腋前线，V6水平于腋中线；
    用户在头部佩戴脑电电极帽，该脑电电极帽中分布设置用于监测脑电信号的六个电极，分别对应大脑左半球和右半球的额叶、枕叶、顶叶；
    开展葡萄糖耐量试验，并开启心电采集设备和脑电采集设备，以同步获取用户的心电信号和脑电信号。
  3. 根据权利要求1所述的系统，其中，所述特征提取模块执行以下过程：
    从所述心电信号中提取多个不同片段的特征信息，从所述脑电信号中分别提取对应大脑不同位置的不同频段的多个脑电特征信息，构成所述组合特征集；
    利用主成分分析对所述组合特征集进行降维处理，获得第一组合特征集；
    利用独立成分分析对所述组合特征集进行降维处理，获得第二组合特 征集；
    利用套索回归分析对所述组合特征集进行降维处理，获得第三组合特征集；
    分别分析所述第一组合特征集、所述第二组合特征集和所述第三组合特征集与血糖浓度的相关性，进而筛选出满足设定相关性标准的心电特征和脑电特征，构成所述优化的组合特征集。
  4. 根据权利要求3所述的系统，其中，利用Pearson相关分析法分析相关性，所述相关性标准设置为相关性k＞0.2且P≤0.05，P表示对相关性系数进行假设检验的概率。
  5. 根据权利要求3所述的系统，其中，所述利用主成分分析对所述组合特征集进行降维处理包括：
    计算所述组合特征集中各特征的特征点的协方差矩阵；
    计算所述协方差矩阵的特征向量和对应的特征值：
    将特征向量按特征值的大小排序，组成矩阵u＝[u 1,u 2,u 3,...,u n]，对应的特征值由大到小分别是λ 1,λ 2,λ 3,...,λ n，从矩阵u中截取排序靠前的设定比例的值作为每个特征的新特征点，以实现数据降维。
  6. 根据权利要求1所述的系统，其中，所述多种类型的神经网络模型包括支持向量机、随机森林、卷积神经网络、长短期记忆网络、递归神经网络中的至少两种类型。
  7. 根据权利要求1所述的系统，其中，所述多模融合模块采用投票法融合所述多种类型的神经网络的输出结果，获得用户是否属于糖尿病前期的检测结果。
  8. 一种基于心电与脑电信息结合的糖尿病前期检测方法，包括以下步骤：
    利用可穿戴设备以无创方式同步获取用户的心电信号和脑电信号；
    利用多种方式对由心电特征和脑电特征构成的组合特征集进行降维处理，获得多个降维的组合特征集，并通过分析所述多个降维的组合特征集与血糖浓度值的相关性筛选出满足设定相关性标准的心电特征和脑电特征，构成优化的组合特征集；
    将所述优化的组合特征集分别输入至经训练的多种类型的神经网络模型，通过融合所述多种类型的神经网络的输出结果，获得用户糖尿病前期的检测结果。
  9. 一种计算机可读存储介质，其上存储有计算机程序，其中，该程序被处理器执行时实现以下步骤：
    利用可穿戴设备以无创方式同步获取用户的心电信号和脑电信号；
    利用多种方式对由心电特征和脑电特征构成的组合特征集进行降维处理，获得多个降维的组合特征集，并通过分析所述多个降维的组合特征集与血糖浓度值的相关性筛选出满足设定相关性标准的心电特征和脑电特征，构成优化的组合特征集；
    将所述优化的组合特征集分别输入至经训练的多种类型的神经网络模型，通过融合所述多种类型的神经网络的输出结果，获得用户糖尿病前期的检测结果。
PCT/CN2020/128559 2020-05-29 2020-11-13 基于心电与脑电信息结合的糖尿病前期检测系统和方法 WO2021238092A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/312,946 US20220313172A1 (en) 2020-05-29 2020-11-13 Prediabetes detection system and method based on combination of electrocardiogram and electroencephalogram information

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010475003.6A CN111631704B (zh) 2020-05-29 2020-05-29 基于心电与脑电信息结合的糖尿病前期检测系统和方法
CN202010475003.6 2020-05-29

Publications (1)

Publication Number Publication Date
WO2021238092A1 true WO2021238092A1 (zh) 2021-12-02

Family

ID=72324267

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/128559 WO2021238092A1 (zh) 2020-05-29 2020-11-13 基于心电与脑电信息结合的糖尿病前期检测系统和方法

Country Status (3)

Country Link
US (1) US20220313172A1 (zh)
CN (1) CN111631704B (zh)
WO (1) WO2021238092A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113951879A (zh) * 2021-12-21 2022-01-21 苏州百孝医疗科技有限公司 血糖预测方法和装置、监测血糖水平的系统

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111631704B (zh) * 2020-05-29 2021-06-25 中国科学院深圳先进技术研究院 基于心电与脑电信息结合的糖尿病前期检测系统和方法
CN113205884B (zh) * 2021-05-12 2023-02-24 中国科学院深圳先进技术研究院 一种血糖预测方法、系统及其应用
CN116010783B (zh) * 2023-01-06 2024-06-07 上海大学 一种rsvp弱隐目标诱发脑电识别方法、装置及存储介质
CN117898687B (zh) * 2024-01-12 2024-07-23 杭州光特医疗科技有限公司 多维度生命体征监测方法、装置、系统及智能戒指

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2002069798A1 (en) * 2001-02-28 2002-09-12 University Of Technology, Sydney A non-invasive method and apparatus for determining onset of physiological conditions
US6572542B1 (en) * 2000-03-03 2003-06-03 Medtronic, Inc. System and method for monitoring and controlling the glycemic state of a patient
WO2016103191A1 (en) * 2014-12-22 2016-06-30 Medicus Engineering Aps Closed-loop control of insulin infusion
CN111631704A (zh) * 2020-05-29 2020-09-08 中国科学院深圳先进技术研究院 基于心电与脑电信息结合的糖尿病前期检测系统和方法

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007508076A (ja) * 2003-10-13 2007-04-05 ノボ・ノルデイスク・エー／エス 生理学的状態を診断する装置及び方法
US20180146879A9 (en) * 2004-08-30 2018-05-31 Kalford C. Fadem Biopotential Waveform Data Fusion Analysis and Classification Method
EP2096996A2 (en) * 2006-11-14 2009-09-09 Novo Nordisk A/S Adaptive hypoglycaemia alert system and method
CN109171756A (zh) * 2018-06-18 2019-01-11 广州普麦健康咨询有限公司 基于深度置信网络模型的糖尿病指标预测方法及其系统

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6572542B1 (en) * 2000-03-03 2003-06-03 Medtronic, Inc. System and method for monitoring and controlling the glycemic state of a patient
WO2002069798A1 (en) * 2001-02-28 2002-09-12 University Of Technology, Sydney A non-invasive method and apparatus for determining onset of physiological conditions
WO2016103191A1 (en) * 2014-12-22 2016-06-30 Medicus Engineering Aps Closed-loop control of insulin infusion
CN111631704A (zh) * 2020-05-29 2020-09-08 中国科学院深圳先进技术研究院 基于心电与脑电信息结合的糖尿病前期检测系统和方法

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
LARSEN ANINE, HØJLUND KURT, POULSEN MIKAEL KJAER, MADSEN RASMUS ELSBORG, JUHL CLAUS B: "Hypoglycemia-Associated Electroencephalogram and Electrocardiogram Changes Appear Simultaneously Author Affiliations", JOURNAL OF DIABETES SCIENCE AND TECHNOLOGY, vol. 7, no. 1, 31 January 2013 (2013-01-31), pages 93 - 99, XP055871335, DOI: 10.1177/193229681300700111 *
TOBORE IGBE; KANDWAL ABHISHEK; LI JINGZHEN; YAN YAN; OMISORE OLATUNJI MUMINI; ENITAN EFETOBORE; SINAN LI; YUHANG LIU; WANG LEI; NI: "Towards adequate prediction of prediabetes using spatiotemporal ECG and EEG feature analysis and weight-based multi-model approach", KNOWLEDGE-BASED SYSTEMS, ELSEVIER, AMSTERDAM, NL, vol. 209, 21 September 2020 (2020-09-21), AMSTERDAM, NL , XP086303293, ISSN: 0950-7051, DOI: 10.1016/j.knosys.2020.106464 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113951879A (zh) * 2021-12-21 2022-01-21 苏州百孝医疗科技有限公司 血糖预测方法和装置、监测血糖水平的系统

Also Published As

Publication number Publication date
CN111631704B (zh) 2021-06-25
CN111631704A (zh) 2020-09-08
US20220313172A1 (en) 2022-10-06

Similar Documents

Publication Publication Date Title
WO2021238092A1 (zh) 基于心电与脑电信息结合的糖尿病前期检测系统和方法
Çınar et al. Classification of normal sinus rhythm, abnormal arrhythmia and congestive heart failure ECG signals using LSTM and hybrid CNN-SVM deep neural networks
US11517212B2 (en) Electrocardiogram information dynamic monitoring method and dynamic monitoring system
EP3698709A1 (en) Electrocardiogram information processing method and electrocardiogram workstation system
CN110664395B (zh) 图像处理方法、图像处理装置以及存储介质
US20210204858A1 (en) Automatic sensing of features within an electrocardiogram
Chen et al. Machine learning method for continuous noninvasive blood pressure detection based on random forest
CN105574322B (zh) 生理参数指标运算系统及方法
US20160242664A1 (en) An apparatus and method for evaluating multichannel ecg signals
CN113317797B (zh) 一种结合医疗领域知识的可解释心律失常分类方法
Yadav et al. Arrhythmia Detection on ECG Signal Using Neural Network Approach
Dang et al. A novel deep convolutional neural network for arrhythmia classification
CN113901893A (zh) 基于多重级联深度神经网络的心电信号的识别与分类方法
CN117651523A (zh) 心电图解析辅助装置、程序、心电图解析辅助方法、心电图解析辅助系统、峰值推测模型生成方法、以及区间推测模型生成方法
CN115336973A (zh) 基于自注意力机制和单导联心电信号的睡眠分期系统构建方法、及睡眠分期系统
Yang et al. A novel method for automated congestive heart failure and coronary artery disease recognition using THC-Net
CN112869753A (zh) 一种针对心电图qrst波形的分析方法、设备、介质和心电图仪
Zaman et al. Estimating reliability of signal quality of physiological data from data statistics itself for real-time wearables
CN117425431A (zh) 心电图解析辅助装置、程序、心电图解析辅助方法、以及心电图解析辅助系统
CN118197592A (zh) 一种基于数据扩充的多模态心律失常分类辅助诊断系统
Li et al. A real-time affective computing platform integrated with ai system-on-chip design and multimodal signal processing system
CN112989971A (zh) 一种不同数据源的心电数据融合方法及装置
Jenkins et al. Improving diffusion models for ecg imputation with an augmented template prior
Shankar et al. Wavelet based Machine Learning Approaches towards Precision Medicine in Diabetes Mellitus.
Wang et al. Multi-feature Fusion Network of ECG and VCG for coronary artery disease detection

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20937678

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20937678

Country of ref document: EP

Kind code of ref document: A1