CN111887824A - Arteriosclerosis detection device based on millimeter waves and neural network - Google Patents
Arteriosclerosis detection device based on millimeter waves and neural network Download PDFInfo
- Publication number
- CN111887824A CN111887824A CN202010751557.4A CN202010751557A CN111887824A CN 111887824 A CN111887824 A CN 111887824A CN 202010751557 A CN202010751557 A CN 202010751557A CN 111887824 A CN111887824 A CN 111887824A
- Authority
- CN
- China
- Prior art keywords
- wave
- neural network
- module
- millimeter wave
- feature
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 206010003210 Arteriosclerosis Diseases 0.000 title claims abstract description 28
- 208000011775 arteriosclerosis disease Diseases 0.000 title claims abstract description 28
- 238000001514 detection method Methods 0.000 title claims abstract description 24
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 21
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 21
- 238000000605 extraction Methods 0.000 claims abstract description 21
- 238000007781 pre-processing Methods 0.000 claims abstract description 19
- 238000003062 neural network model Methods 0.000 claims abstract description 18
- 238000004364 calculation method Methods 0.000 claims abstract description 5
- 238000012549 training Methods 0.000 claims abstract description 5
- 238000001914 filtration Methods 0.000 claims description 4
- 150000001875 compounds Chemical class 0.000 claims description 3
- 230000035487 diastolic blood pressure Effects 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 230000035488 systolic blood pressure Effects 0.000 claims description 3
- 238000000034 method Methods 0.000 description 11
- 238000005259 measurement Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 5
- 201000010099 disease Diseases 0.000 description 5
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 5
- 210000001367 artery Anatomy 0.000 description 4
- 239000002131 composite material Substances 0.000 description 4
- 238000010009 beating Methods 0.000 description 2
- 210000004556 brain Anatomy 0.000 description 2
- 206010008118 cerebral infarction Diseases 0.000 description 2
- 208000026106 cerebrovascular disease Diseases 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 210000005259 peripheral blood Anatomy 0.000 description 2
- 239000011886 peripheral blood Substances 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 206010002383 Angina Pectoris Diseases 0.000 description 1
- 206010002482 Angiosclerosis Diseases 0.000 description 1
- 208000024172 Cardiovascular disease Diseases 0.000 description 1
- 206010017711 Gangrene Diseases 0.000 description 1
- 206010022562 Intermittent claudication Diseases 0.000 description 1
- 230000003321 amplification Effects 0.000 description 1
- 238000002583 angiography Methods 0.000 description 1
- 210000000709 aorta Anatomy 0.000 description 1
- 210000001765 aortic valve Anatomy 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000003143 atherosclerotic effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000004204 blood vessel Anatomy 0.000 description 1
- 230000002490 cerebral effect Effects 0.000 description 1
- 238000002591 computed tomography Methods 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000002526 effect on cardiovascular system Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000002757 inflammatory effect Effects 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 208000021156 intermittent vascular claudication Diseases 0.000 description 1
- 238000002608 intravascular ultrasound Methods 0.000 description 1
- 210000003734 kidney Anatomy 0.000 description 1
- 210000005240 left ventricle Anatomy 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 238000002595 magnetic resonance imaging Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 208000010125 myocardial infarction Diseases 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000008085 renal dysfunction Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009885 systemic effect Effects 0.000 description 1
- 201000010875 transient cerebral ischemia Diseases 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/02007—Evaluating blood vessel condition, e.g. elasticity, compliance
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/05—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves
- A61B5/0507—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves using microwaves or terahertz waves
-
- 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
- A61B5/7203—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal
-
- 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
-
- 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
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
- G06F2218/04—Denoising
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- Veterinary Medicine (AREA)
- Heart & Thoracic Surgery (AREA)
- Artificial Intelligence (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Animal Behavior & Ethology (AREA)
- Pathology (AREA)
- Surgery (AREA)
- Physiology (AREA)
- Signal Processing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Psychiatry (AREA)
- Evolutionary Computation (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- Cardiology (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Vascular Medicine (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Fuzzy Systems (AREA)
- Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)
Abstract
The invention discloses an arteriosclerosis detecting device based on millimeter waves and a neural network, which comprises: the device comprises a millimeter wave detection device, a preprocessing module, a feature extraction module and a CNN neural network model, wherein the preprocessing module is respectively connected with the millimeter wave detection device and the feature extraction module, and the feature extraction module is connected with the CNN neural network model; the millimeter wave detection device is used for collecting pulse waves of a user; the preprocessing module is used for preprocessing the pulse wave; the feature extraction module comprises a feature point extraction module and a feature parameter calculation module and is used for extracting features of the preprocessed pulse waves; and the CNN neural network model is used for training and classifying according to the extracted characteristic parameters.
Description
Technical Field
The invention relates to the field of cardiovascular disease detection and judgment, in particular to an arteriosclerosis detection device based on millimeter waves and a neural network.
Background
Arteriosclerosis is a non-inflammatory lesion of an artery, and can thicken and harden the wall of an artery, lose elasticity and narrow a lumen. Arteriosclerosis is classified into various types, such as cerebral infarction in the brain, transient cerebral ischemia, angina pectoris in the heart, renal dysfunction in the kidney, and intermittent claudication in the peripheral blood vessels. Cerebral infarction, myocardial infarction and gangrene can occur in severe cases. Arteriosclerosis is a systemic disease, all known as atherosclerotic disease, and its harm is to affect the whole body, including cerebral vessels, cardiovascular vessels, renal vessels, peripheral vessels of brain, and even vessels to the fundus. Arteriosclerosis has become a next very common disease, and appears with the increase of age, and the rule is that the arteriosclerosis usually occurs in teenagers and aggravates and attacks the diseases in the middle-aged and old. More men than women are in the group of the disease, and in recent years, arteriosclerosis is gradually increased in China and becomes one of the main causes of death of the old. Therefore, early detection, early prevention, becomes critical away from arteriosclerosis.
With the continuous progress of scientific technology, the measurement of arteriosclerosis is accumulated for many years, and various methods are verified and discovered. The method is divided into two modes of invasive and non-invasive, wherein the invasive method comprises an angiography method and an intravascular ultrasound technology. Non-invasive methods fall into two broad categories: direct measurement and indirect measurement. The direct measurement uses magnetic resonance imaging, CT scanning, color doppler, etc. and the indirect measurement determines the arteriosclerosis degree by measuring the conduction velocity of the pulse wave, which is mainly divided into two modes: cervical-femoral pulse wave, brachial-ankle pulse wave. The main measuring instrument of the pulse wave is VP series of Japan ohm Longkolin, and the market share is higher in the world at present. Another instrument for measuring the velocity of the cervical-femoral pulse wave is the Complior detection system, available from French corporation. However, most of the above detection means are very complicated and the equipment is expensive.
Disclosure of Invention
In order to solve the defects of the prior art, realize accurate measurement, reduce complicated detection means and reduce cost, the invention adopts the following technical scheme:
an arteriosclerosis detecting device based on millimeter wave and neural network, comprising: the device comprises a millimeter wave detection device, a preprocessing module, a feature extraction module and a CNN neural network model, wherein the preprocessing module is respectively connected with the millimeter wave detection device and the feature extraction module, and the feature extraction module is connected with the CNN neural network model;
the millimeter wave detection device is used for collecting pulse waves of a user;
the preprocessing module is used for preprocessing the pulse wave;
the characteristic extraction module comprises a characteristic point extraction module and a characteristic parameter calculation module and is used for extracting the characteristics of the preprocessed pulse wave, the characteristic points comprise an initial point, a main wave peak point, a wave valley point of a pre-counterpulsation wave, a wave peak point of a pre-counterpulsation wave, a wave valley point of a counterpulsation wave and a wave peak point of a counterpulsation wave, the characteristic parameters comprise the relative height and the K value of the wave peak of the counterpulsation wave, K is (Pm-Pd)/(Ps-Pd), Ps represents the systolic pressure, Pd represents the diastolic pressure, Pm represents the average pressure, the characteristic parameter calculation module calculates the characteristic parameters through the characteristic points extracted by the characteristic extraction module, the relative height of the dicrotic wave peak can reflect arterial compliance, and is an index of the degree of arteriosclerosis, the larger the relative height of the dicrotic wave peak is, the better the arterial compliance is, and the K value can well reflect indexes of elasticity of a blood vessel wall, resistance of peripheral blood vessels and the like of a human body;
and the CNN neural network model is used for training and classifying according to the extracted characteristic parameters.
The method comprises the steps of acquiring pulse wave data of a user by using a millimeter wave detection device, preprocessing the pulse wave data, extracting features of the preprocessed data, using the extracted features as input of a CNN neural network model, training a large amount of effective data through the CNN neural network model to obtain the trained CNN neural network model, classifying normal pulse and arteriosclerosis pulse through the trained CNN neural network model, and judging whether the user suffers from arteriosclerosis or not according to the pulse wave detection condition of the user.
The millimeter wave detection device comprises a millimeter wave transmitting and receiving device, and can be used for carrying out non-contact signal acquisition and acquiring pulse beating analog signals.
The preprocessing module comprises a filtering processing module and a low-noise amplifying circuit, the filtering processing module is used for filtering out clutter, and the low-noise amplifying circuit is used for amplifying pulse wave signals.
The CNN neural network adopts an EfficientNeT neural network, and the model accuracy rate formula is as follows:
representing the entire convolutional network, w represents the size of the convolutional kernel, d represents the neural network depth, r represents the resolution, i represents the ith convolutional layer, 1 … s represents the signals of a plurality of convolutional layers with the same structure, FiDenotes the convolution operation on the i-th layer, LiIs represented by FiIn the ith plurality of convolution layers of the same structure, there is LiA convolution layer of uniform structure, XiRepresenting the input tensor, Hi、Wi、CiRespectively represent the dimensions of the input of the ith layer,representing a custom operator, achieves the best model accuracy by optimizing the parameters d, w, and r.
The EfficientNeT neural network uses a composite coefficientAnd (3) carrying out compound adjustment on the parameters of d, w and r:
s.t.α*β2*γ2≈2(α≥1,β≥1,γ≥1)
alpha, beta and gamma are constants, the optimal alpha, beta and gamma are determined through the adjustment of the basic network, and the reference network is expanded through the optimal alpha, beta and gamma, so that the expanded network also has higher accuracy and efficiency.
The invention has the advantages and beneficial effects that:
through the technique that millimeter wave and deep learning combined together for arteriosclerosis detection device improves the convenience of using when guaranteeing to measure the accuracy, reduces product cost, reduction measurement cost.
Drawings
Fig. 1 is a schematic structural view of the present invention.
Fig. 2a is a diagram of a baseline network in accordance with the present invention.
FIG. 2b is a diagram of an expanded network for increasing the receptive field in the present invention.
Fig. 2c is an extended network diagram of increasing network depth in the present invention.
Fig. 2d is a diagram of an extended network for increased resolution in the present invention.
Fig. 2e is a diagram of a hybrid parameter extended network in accordance with the present invention.
Detailed Description
The following detailed description of embodiments of the invention refers to the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating the present invention, are given by way of illustration and explanation only, not limitation.
As shown in fig. 1, an arteriosclerosis detecting device based on millimeter waves and a neural network includes a millimeter wave detecting device, a preprocessing module, a feature extraction module, a CNN neural network model, and an output module.
The millimeter wave detection device is used for collecting the pulse waves of a user in a time period, and non-contact signal collection is carried out through the millimeter wave transmitting and receiving device to obtain pulse beating analog signals.
The preprocessing module preprocesses the pulse wave, filters the acquired analog signal, filters out clutter, and amplifies the signal by using a low-noise amplifying circuit.
The feature extraction module performs feature extraction on the processed pulse waves, extracts six feature points of the pulse waves and selects feature parameters. The characteristic points of the pulse wave include: the initiation point, i.e. the point at which the aortic valve opens; a main wave peak point, namely a maximum amplitude point; the wave valley point of the pre-dicrotic wave, namely the point when the pressure of the aorta expansion begins to drop; the peak point of the wave before dicrotic beat, i.e. the beginning of the relaxation of the left ventricle; the wave valley point of the counterpulsation wave, namely the starting point of the counterpulsation wave; the peak point of the dicrotic wave, i.e. the point at which the dicrotic wave reaches maximum pressure. Calculating the relative height (f/H) and K value of the peak of the dicrotic wave as characteristic parameters through six characteristic points, wherein the relative height of the peak of the dicrotic wave reflects the compliance of the artery and is an index of the degree of angiosclerosis, and the larger the f/H is, the better the compliance of the artery is; k is (Pm-Pd)/(Ps-Pd), Ps and Pd are systolic and diastolic pressures, respectively, and Pm is the mean pressure. The K value can well reflect indexes of elasticity of the vessel wall, resistance of peripheral vessels and the like of a human body.
The CNN neural network is trained by the extracted features, and the EfficientNeT neural network is run by using a TensorFlow framework. The EfficientNet utilizes the compound coefficient to uniformly scale all dimensions of the model, and consists of Stem +16 Blocks + Con2D + GlobavalagePooling 2D + Dense in total, so that the effect of highest precision and highest efficiency is achieved. The method comprises three coefficients of w, d and r, wherein w represents the size of a convolution kernel and determines the size of a receptive field; d represents the neural network depth; r represents the resolution size. As shown in fig. 2, fig. 2a is a base network, which may also be understood as a small network; FIG. 2b is a manner of expanding the network to increase the receptive field; FIG. 2c is a manner of extending the network in a manner that increases the depth of the network; FIG. 2d is a manner of increasing resolution to expand the network; FIG. 2e is a hybrid parameter expansion approach; the efficiency of model scaling relies heavily on the baseline network, which is developed using a network structure search and then scaled to obtain a series of models called EfficientNets.
The mathematical formula of the neural network is as follows:
wherein: xiIs the input tensor so that,representing the entire convolutional network, i represents the ith convolutional layer, 1 … s represents stage (multiple convolutional layers of the same structure)Signal, FiDenotes the convolution operation on the i-th layer, LiIs represented by FiIn the ith stage, there is LiA convolution layer of uniform structure Hi、Wi、CiRespectively represent the dimension of the i-th layer input, where Hi、WiRepresents the convolution kernel size, C, of the ith convolution layeriIndicates the number of channels in the ith convolutional layer,is a custom operator that can be defined as | _ indicating a multiply-by-multiply operation. The formula shows how the parameters d, w and r are optimized to achieve the best model accuracy.
The normalized composite parameter adjusting method of EfficientNet uses a composite coefficientTo make composite adjustments to the d, w and r parameters.
s.t.α*β2*γ2≈2(α≥1,β≥1,γ≥1)
Wherein alpha, beta and gamma are constants which can be obtained by grid search, andoptimization by manual adjustment is required. The optimization method comprises the following steps: in the first step, the optimal alpha, beta and gamma can be adjusted and determined through a base network, and then the parameter is used for expanding or amplifying the base network into a large network, so that the large network also has higher accuracy and efficiency. Incorporating the selected features intoThe number is used as input to train the model.
The classification and identification of normal pulse waves and pulse waves with arteriosclerosis are completed through a trained model, the pulse waves of a detected user are subjected to preprocessing such as noise reduction and amplification, characteristic parameters are extracted, the characteristic parameters are used as network input, and whether the user has the arteriosclerosis or not is identified according to the model classification.
The method comprises the steps of acquiring pulse wave data of a user by using a millimeter wave detection device, preprocessing the pulse wave data, extracting features of the preprocessed data, using the extracted features as input of a CNN neural network model, training a large amount of effective data through the CNN neural network model to obtain the trained CNN neural network model, classifying normal pulse and arteriosclerosis pulse through the trained CNN neural network model, and judging whether the user suffers from arteriosclerosis or not according to the pulse wave detection condition of the user.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (5)
1. An arteriosclerosis detecting device based on millimeter wave and neural network, characterized by comprising: the device comprises a millimeter wave detection device, a preprocessing module, a feature extraction module and a CNN neural network model, wherein the preprocessing module is respectively connected with the millimeter wave detection device and the feature extraction module, and the feature extraction module is connected with the CNN neural network model;
the millimeter wave detection device is used for collecting pulse waves of a user;
the preprocessing module is used for preprocessing the pulse wave;
the feature extraction module comprises a feature point extraction module and a feature parameter calculation module, and is used for performing feature extraction on the preprocessed pulse wave, wherein the feature point comprises an initial point, a main peak point, a wave valley point of a pre-counterpulsation wave, a wave peak point of a pre-counterpulsation wave, a wave valley point of a counterpulsation wave and a wave peak point of a counterpulsation wave, the feature parameter comprises the relative height and the K value of the wave peak of the counterpulsation wave, K is (Pm-Pd)/(Ps-Pd), Ps represents the systolic pressure, Pd represents the diastolic pressure, Pm represents the average pressure, and the feature parameter calculation module calculates the feature parameter through the feature point extracted by the feature extraction module;
and the CNN neural network model is used for training and classifying according to the extracted characteristic parameters.
2. The arteriosclerosis detecting device based on millimeter wave and neural network as claimed in claim 1, wherein the millimeter wave detecting device comprises millimeter wave transmitting and receiving device.
3. The arteriosclerosis detecting device according to claim 1, wherein the preprocessing module comprises a filtering processing module and a low noise amplifying circuit.
4. The arteriosclerosis detecting device based on millimeter wave and neural network as claimed in claim 1, wherein the CNN neural network adopts EfficientNeT neural network, and the model accuracy formula is as follows:
representing the entire convolutional network, w represents the size of the convolutional kernel, d represents the neural network depth, r represents the resolution, i represents the ith convolutional layer, 1 … s represents the signals of a plurality of convolutional layers with the same structure, FiDenotes the convolution operation on the i-th layer, LiIs represented by FiIn the ith plurality of convolution layers of the same structure, there is LiA convolution layer of uniform structure, XiRepresenting the input tensor, Hi、Wi、CiRespectively represent the dimensions of the input of the ith layer,representing a custom operator.
5. The arteriosclerosis detecting device as claimed in claim 4, wherein the EfficientNeT neural network uses complex coefficientAnd (3) carrying out compound adjustment on the parameters of d, w and r:
s.t.α*β2*γ2≈2(α≥1,β≥1,γ≥1)
alpha, beta and gamma are constants, optimal alpha, beta and gamma are determined through base line network adjustment, and the reference network is expanded through the optimal alpha, beta and gamma.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010751557.4A CN111887824A (en) | 2020-07-30 | 2020-07-30 | Arteriosclerosis detection device based on millimeter waves and neural network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010751557.4A CN111887824A (en) | 2020-07-30 | 2020-07-30 | Arteriosclerosis detection device based on millimeter waves and neural network |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111887824A true CN111887824A (en) | 2020-11-06 |
Family
ID=73183439
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010751557.4A Pending CN111887824A (en) | 2020-07-30 | 2020-07-30 | Arteriosclerosis detection device based on millimeter waves and neural network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111887824A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112735480A (en) * | 2021-01-25 | 2021-04-30 | 杭州电子科技大学 | Vocal cord pathological change detection device based on neural network |
CN113729649A (en) * | 2021-09-29 | 2021-12-03 | 九次元(重庆)智能技术有限公司 | Cardiovascular and cerebrovascular disease identification system based on echo state network |
AU2021107074B4 (en) * | 2021-08-25 | 2022-06-16 | Rudder Technology Pty Ltd | Remote detection for blood pressure with radar sensor |
CN114642409A (en) * | 2022-01-19 | 2022-06-21 | 北京邮电大学 | Human body pulse wave sensing, heart rate monitoring and blood pressure monitoring method and related device |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102397064A (en) * | 2011-12-14 | 2012-04-04 | 中国航天员科研训练中心 | Continuous blood pressure measuring device |
CN102429649A (en) * | 2011-12-14 | 2012-05-02 | 中国航天员科研训练中心 | Continuous blood pressure measuring device |
CN102488503A (en) * | 2011-12-14 | 2012-06-13 | 中国航天员科研训练中心 | Continuous blood pressure measurer |
CN102824166A (en) * | 2012-07-05 | 2012-12-19 | 华东师范大学 | Device for sorting treatment of pulse wave |
CN107427236A (en) * | 2015-05-22 | 2017-12-01 | 谷歌公司 | Painstaking effort tube sensor for cardiovascular monitoring is synchronous |
CN108478203A (en) * | 2018-02-08 | 2018-09-04 | 南京理工大学 | A kind of blood pressure measuring method monitoring radar based on single vital sign |
CN111310149A (en) * | 2018-12-11 | 2020-06-19 | 阿里巴巴集团控股有限公司 | Method and system for authentication based on correlation of multiple pulse signals |
CN111449637A (en) * | 2020-04-07 | 2020-07-28 | 上海市第十人民医院 | Evaluation system and method for arteriovenous internal fistula blood vessel |
CN111462093A (en) * | 2020-04-02 | 2020-07-28 | 北京小白世纪网络科技有限公司 | Method for classifying diseases based on fundus images |
-
2020
- 2020-07-30 CN CN202010751557.4A patent/CN111887824A/en active Pending
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102397064A (en) * | 2011-12-14 | 2012-04-04 | 中国航天员科研训练中心 | Continuous blood pressure measuring device |
CN102429649A (en) * | 2011-12-14 | 2012-05-02 | 中国航天员科研训练中心 | Continuous blood pressure measuring device |
CN102488503A (en) * | 2011-12-14 | 2012-06-13 | 中国航天员科研训练中心 | Continuous blood pressure measurer |
CN102824166A (en) * | 2012-07-05 | 2012-12-19 | 华东师范大学 | Device for sorting treatment of pulse wave |
CN107427236A (en) * | 2015-05-22 | 2017-12-01 | 谷歌公司 | Painstaking effort tube sensor for cardiovascular monitoring is synchronous |
CN108478203A (en) * | 2018-02-08 | 2018-09-04 | 南京理工大学 | A kind of blood pressure measuring method monitoring radar based on single vital sign |
CN111310149A (en) * | 2018-12-11 | 2020-06-19 | 阿里巴巴集团控股有限公司 | Method and system for authentication based on correlation of multiple pulse signals |
CN111462093A (en) * | 2020-04-02 | 2020-07-28 | 北京小白世纪网络科技有限公司 | Method for classifying diseases based on fundus images |
CN111449637A (en) * | 2020-04-07 | 2020-07-28 | 上海市第十人民医院 | Evaluation system and method for arteriovenous internal fistula blood vessel |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112735480A (en) * | 2021-01-25 | 2021-04-30 | 杭州电子科技大学 | Vocal cord pathological change detection device based on neural network |
AU2021107074B4 (en) * | 2021-08-25 | 2022-06-16 | Rudder Technology Pty Ltd | Remote detection for blood pressure with radar sensor |
CN113729649A (en) * | 2021-09-29 | 2021-12-03 | 九次元(重庆)智能技术有限公司 | Cardiovascular and cerebrovascular disease identification system based on echo state network |
CN113729649B (en) * | 2021-09-29 | 2024-05-14 | 重庆岱普科技有限责任公司 | Cardiovascular and cerebrovascular disease recognition system based on echo state network |
CN114642409A (en) * | 2022-01-19 | 2022-06-21 | 北京邮电大学 | Human body pulse wave sensing, heart rate monitoring and blood pressure monitoring method and related device |
CN114642409B (en) * | 2022-01-19 | 2022-10-18 | 北京邮电大学 | Human body pulse wave sensing method, heart rate monitoring method and blood pressure monitoring device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111887824A (en) | Arteriosclerosis detection device based on millimeter waves and neural network | |
US20220395212A1 (en) | Methods and systems for tracking physiological parameters of mother and fetus during pregnancy | |
WO2017024457A1 (en) | Blood-pressure continuous-measurement device, measurement model establishment method, and system | |
CN108577883A (en) | A kind of Screening for coronary artery disease device, screening system and signal characteristic extracting methods | |
CN112806977B (en) | Physiological parameter measuring method based on multi-scale fusion network | |
CN102509267B (en) | Retrospective off-line gating method for intravascular ultrasound image sequence | |
US20230301581A1 (en) | Fusion signal processing for maternal uterine activity detection | |
CN111839488B (en) | Non-invasive continuous blood pressure measuring device and method based on pulse wave | |
WO2023185873A1 (en) | Non-cuff blood pressure measuring apparatus based on multi-stage multi-modal learning and method | |
CN110120031A (en) | A kind of method and apparatus obtaining vascular flow deposit score | |
CN110840428B (en) | Noninvasive blood pressure estimation method based on one-dimensional U-Net network | |
CN101625756B (en) | Method for improving visual effect of ultrasonic image in coronary artery | |
Rooijakkers et al. | Low-complexity R-peak detection in ECG signals: A preliminary step towards ambulatory fetal monitoring | |
CN111783715A (en) | Identity recognition method based on pulse signal feature extraction | |
CN113243902B (en) | Feature extraction method based on photoplethysmography | |
CN111329508A (en) | Heart murmur intelligent analysis method for precordial disease screening | |
CN113729653A (en) | Human body pulse wave signal acquisition method | |
CN113539522A (en) | Continuous blood pressure monitoring method based on single-path cardiac shock signal | |
CN116649924A (en) | Pulse analysis method and device | |
US20100274134A1 (en) | Method for evaluation of renal vascular perfusion using power doppler ultrasonography | |
CN114145725B (en) | PPG sampling rate estimation method based on noninvasive continuous blood pressure measurement | |
CN110811673A (en) | Heart sound analysis system based on probabilistic neural network model | |
CN114947782A (en) | System and method for central arterial pressure waveform reconstruction based on PPG signal | |
CN117197096B (en) | Blood vessel function assessment method and system based on blood vessel image | |
Janney et al. | Recognition of dicrotic notch in arterial blood pressure pulses using signal processing techniques |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201106 |