CN111772628B - Electrocardiosignal atrial fibrillation automatic detection system based on deep learning - Google Patents
Electrocardiosignal atrial fibrillation automatic detection system based on deep learning Download PDFInfo
- Publication number
- CN111772628B CN111772628B CN202010688677.4A CN202010688677A CN111772628B CN 111772628 B CN111772628 B CN 111772628B CN 202010688677 A CN202010688677 A CN 202010688677A CN 111772628 B CN111772628 B CN 111772628B
- Authority
- CN
- China
- Prior art keywords
- atrial fibrillation
- interval
- electrocardiosignal
- electrocardiosignals
- network
- 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.)
- Active
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
- A61B5/72—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
-
- 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
Abstract
The invention discloses an automatic detection system for atrial fibrillation of electrocardiosignals based on deep learning, which comprises an electrocardiosignal processing module, a training module and an atrial fibrillation detection module; the electrocardiosignal processing module is used for carrying out R wave detection on the electrocardiosignals to obtain an RR interphase sequence; preprocessing the RR interval sequence, and sampling the preprocessed RR interval sequence from the initial time of the corresponding electrocardiosignal based on a sliding window to obtain an RR interval sample; wherein, the RR interval values in each RR interval sample have time correlation; the atrial fibrillation detection module is used for identifying whether each extracted RR interval sample is an atrial fibrillation signal or not by adopting an atrial fibrillation signal identification model based on a CLDNN (CLDNN network), so that the atrial fibrillation automatic detection of the electrocardiosignals is completed, not only can the inherent characteristics of the electrocardiosignals be extracted, but also the time correlation characteristics of the electrocardiosignals are considered, and the detection precision is higher; in addition, the system is an end-to-end system, and is more convenient and faster.
Description
Technical Field
The invention belongs to the technical field of electrocardiogram analysis, and particularly relates to an automatic electrocardiosignal atrial fibrillation detection system based on deep learning.
Background
Atrial fibrillation is the most common symptom of tachyarrhythmia in the clinic and is manifested by irregular activation of the atria. About 3350 thousands of people suffer from atrial fibrillation globally, wherein the number of patients suffering from atrial fibrillation in China exceeds 1000 thousands, and the patient suffering from atrial fibrillation is one of the most countries in the world. Atrial fibrillation, in addition to its manifestation of arrhythmia, often leads to other complications that can form left atrial mural thrombi and cause arterial emboli. The resulting embolism is 90% cerebral artery embolism (ischemic stroke), 10% peripheral artery embolism or mesenteric artery embolism, and the like, so atrial fibrillation is one of the important causes of cerebral stroke. However, atrial fibrillation generally develops to a later stage to cause other symptoms, so that it is important to detect and find atrial fibrillation at an early stage to prevent subsequent complications.
Currently, electrocardiographs are commonly used clinically to record cardiac Electrical Signals (ECGs) to detect and monitor cardiac conditions. Electrocardiographs, particularly portable electrocardiographs, are capable of non-invasively and conveniently recording electrocardiographic information for a long time, but also produce a large amount of electrocardiographic signal data. The interpretation and diagnosis of a large amount of electrocardiosignal data cause heavy burden for clinicians, and the accurate automatic identification and interpretation of the electrocardiosignals are carried out by utilizing a computer-aided diagnosis system, so that the diagnosis and treatment efficiency of the clinicians is greatly improved. With the development of deep learning technology, its powerful learning ability has received a great deal of attention and has been successfully applied in many fields, including computer-aided diagnosis. Therefore, the model constructed based on the deep learning method is applied to atrial fibrillation signal identification, accurate identification and monitoring of atrial fibrillation are hopefully realized, the burden of doctors is effectively reduced, and the working efficiency is improved.
The conventional automatic detection system for atrial fibrillation of the electrocardiosignals is usually based on traditional machine learning algorithms such as random forests, SVM and the like and a deep learning algorithm such as a convolutional neural network, and the automatic detection system for atrial fibrillation of the electrocardiosignals based on the machine learning algorithm is relatively complex in feature extraction operation and over-parameter adjustment of partial models, does not consider the time correlation of the signals when processing the electrocardiosignals, and is relatively low in detection precision and efficiency. The conventional automatic detection system for the atrial fibrillation of the electrocardiosignals based on the deep learning algorithm does not consider the time correlation of the signals when the electrocardiosignals are processed, and has low detection precision.
Disclosure of Invention
Aiming at the defects or the improvement requirements of the prior art, the invention provides an automatic detection system for atrial fibrillation of electrocardiosignals based on deep learning, and aims to solve the technical problem of low detection precision caused by the fact that the time correlation of the electrocardiosignals is not considered in the prior art.
In order to achieve the above object, the present invention provides an automatic detecting system for atrial fibrillation of electrocardiographic signals based on deep learning, comprising:
the electrocardiosignal processing module is used for carrying out R wave detection on the collected electrocardiosignals, extracting the R wave time of each heartbeat in the electrocardiosignals, and calculating the RR interval numerical value of the electrocardiosignals to obtain an RR interval sequence; preprocessing the RR interval sequence, and sampling the preprocessed RR interval sequence from the initial time of the corresponding electrocardiosignal based on a sliding window to obtain an RR interval sample;
the training module is used for acquiring a plurality of electrocardiosignals containing atrial fibrillation labels, and respectively inputting the electrocardiosignals into the electrocardiosignal processing module to obtain RR interphase samples; marking atrial fibrillation of the electrocardiosignal corresponding to each RR interval sample as an RR interval sample label; training an atrial fibrillation signal recognition model by taking each RR interphase sample as input and the corresponding RR interphase sample label as output to obtain a trained atrial fibrillation signal recognition model; the electrocardiosignals comprise atrial fibrillation electrocardiosignals and atrial fibrillation-free electrocardiosignals, and the atrial fibrillation signal identification model is based on a CLDNN (CLDNN network);
the atrial fibrillation detection module is used for inputting the electrocardiosignals to be detected into the electrocardiosignal processing module to obtain RR interphase samples, inputting the RR interphase samples into the trained atrial fibrillation signal recognition model, and recognizing whether the RR interphase samples are atrial fibrillation signals, so that whether the electrocardiosignals to be detected contain the atrial fibrillation signals and the time when the atrial fibrillation occurs is determined.
Further preferably, when the RR interval sequence is preprocessed by the electrocardiograph signal processing module, if the number of times of continuous occurrence of the abnormal value is less than or equal to a preset number of times when the RR interval value has the abnormal value, performing interpolation correction on each abnormal RR interval value in the RR interval sequence; and if the continuous occurrence frequency of the abnormal value is more than the preset frequency, discarding the abnormal time in the RR interval sequence and the RR interval value in a preset time period before the abnormal time.
Further preferably, when the sliding window samples the preprocessed RR interval sequence from the initial time corresponding to the electrocardiographic signal, the electrocardiographic signal processing module moves the sliding window backward by a window size for the next extraction after each extraction is completed, so as to implement non-overlapping sampling.
Further preferably, the atrial fibrillation signal recognition model includes: the system comprises a cascaded input layer, a convolution sub-network, a bidirectional LSTM sub-network, a global maximum pooling layer, a full-connection mapping sub-network and an output layer;
the convolution sub-network is used for reducing the frequency domain variation of the RR interval samples and extracting the characteristics of the RR interval samples;
the bidirectional LSTM sub-network is used for carrying out time sequence analysis on the characteristics of the RR interval samples;
and the fully-connected mapping sub-network is used for carrying out mapping classification on the RR interval samples based on the time sequence analysis result.
Further preferably, the convolution sub-network is formed by two convolution layer cascades of 5 × 1 × 64 and 5 × 1 × 32; the bidirectional LSTM sub-network is composed of two LSTMs with opposite directions; the number of neurons in the two fully-connected layers in the fully-connected mapping sub-network is 32 and 16, respectively.
Further preferably, the loss function of the atrial fibrillation signal identification model is a cross entropy function; during training, parameters in the atrial fibrillation signal recognition model are updated by using an Adam optimizer.
Further preferably, the automatic detection system for atrial fibrillation based on deep learning provided by the invention is applied to the technical field of electrocardiogram analysis.
In general, compared with the prior art, the above technical solution contemplated by the present invention can achieve the following beneficial effects:
1. the invention provides an automatic electrocardiosignal atrial fibrillation detection system based on deep learning, which comprises an electrocardiosignal processing module, a training module and an atrial fibrillation detection module; the electrocardiosignal processing module is used for carrying out R-wave detection on electrocardiosignals to obtain corresponding RR interval sequences, and sampling the RR interval sequences by adopting a sliding window to obtain a plurality of sections of RR interval samples; wherein, the RR interval values in each RR interval sample have time correlation; the atrial fibrillation detection module is used for identifying whether each RR interval sample is atrial fibrillation by adopting an atrial fibrillation signal identification model based on a CLDNN (CLDNN network), and compared with the existing automatic atrial fibrillation detection system based on a machine learning algorithm and a deep learning algorithm, the atrial fibrillation signal identification model disclosed by the invention not only can extract the inherent characteristics of the electrocardiosignals, but also considers the time correlation characteristics of the electrocardiosignals, and is higher in detection precision.
2. The invention provides an automatic atrial fibrillation detection system for electrocardiosignals based on deep learning, which is an end-to-end system.
3. In the automatic detection system for the atrial fibrillation of the electrocardiosignals based on deep learning, an atrial fibrillation detection module firstly adopts an electrocardiosignal processing module to process the electrocardiosignals to be detected into RR interval samples with time relevance before identification, and then adopts an atrial fibrillation signal identification model to identify each RR interval sample, so that whether the electrocardiosignals to be detected contain atrial fibrillation signals or not can be determined, and the time when the atrial fibrillation occurs can be obtained.
Drawings
Fig. 1 is a schematic structural diagram of an automatic detection system for atrial fibrillation of an electrocardiograph signal based on deep learning according to an embodiment of the present invention;
fig. 2 is a schematic diagram of sampling a pre-processed RR interval sequence based on a sliding window according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of an atrial fibrillation signal recognition model based on a CLDNN network according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
The invention provides an automatic detection system for atrial fibrillation of electrocardiosignals based on deep learning, which comprises the following components in percentage by weight as shown in figure 1: the device comprises an electrocardiosignal processing module, a training module and an atrial fibrillation detection module;
the electrocardiosignal processing module is respectively connected with the training module and the atrial fibrillation detection module, and the training module is connected with the atrial fibrillation detection module;
the electrocardiosignal processing module is used for carrying out R wave detection on the collected electrocardiosignals, extracting the R wave time of each heartbeat in the electrocardiosignals, and calculating the RR interval numerical value of the electrocardiosignals to obtain an RR interval sequence; preprocessing the RR interval sequence, and sampling the preprocessed RR interval sequence from the initial time of the corresponding electrocardiosignal based on a sliding window to obtain an RR interval sample; the reason for this is that the R peak is the most easily recognized peak in the electrocardiogram due to its large amplitude, and the detected R peak has higher noise immunity; in addition, atrial fibrillation appears as irregular beats on the electrocardiogram, which beats can be characterized to some extent using RR intervals without further features. Processing into RR interval samples is more suitable for model processing.
When the RR interval sequence is preprocessed by the electrocardiosignal processing module, when an abnormal value exists in an RR interval value, if the continuous occurrence frequency of the abnormal value is less than or equal to a preset frequency, performing interpolation correction on each abnormal RR interval value in the RR interval sequence; and if the continuous occurrence frequency of the abnormal value is more than the preset frequency, discarding the abnormal time in the RR interval sequence and the RR interval value in a preset time period before the abnormal time. In this embodiment, the preset number of times is 5 times, and the preset time period is 1 minute. And setting all abnormal RR interval values in the RR interval sequence to be 0, and when the number of times of continuous occurrence of the abnormal RR interval values in the RR interval sequence is more than 5 times, the abnormal RR interval values in the section of continuous abnormal values which are firstly abnormal and all RR interval values in one minute before the abnormal RR interval values are not used any more and are discarded. And when the number of continuous abnormal RR interval values in the RR interval sequence is less than or equal to 5, correcting the current abnormal RR interval value by adopting the average value of two non-abnormal RR interval values before and after the abnormal RR interval value.
Further, as shown in fig. 2, when the pre-processed RR interval sequence is sampled from the initial time corresponding to the electrocardiographic signal based on the sliding window, the electrocardiographic signal processing module moves the sliding window backward by one window size for the next extraction after each extraction is completed, so as to implement non-overlapping sampling, and when the remaining data length of the RR interval sequence is smaller than the size of the sliding window, the RR interval sequence is discarded. In this embodiment, the size of the sliding window is 90 RR intervals, wherein one RR interval represents one heartbeat.
The training module is used for acquiring a plurality of electrocardiosignals containing atrial fibrillation labels, and respectively inputting the electrocardiosignals into the electrocardiosignal processing module to obtain RR interphase samples; marking atrial fibrillation of the electrocardiosignal corresponding to each RR interval sample as an RR interval sample label; training an atrial fibrillation signal recognition model by taking each RR interphase sample as input and the corresponding RR interphase sample label as output to obtain a trained atrial fibrillation signal recognition model; the electrocardiosignals comprise atrial fibrillation electrocardiosignals and atrial fibrillation-free electrocardiosignals, and the atrial fibrillation signal identification model is based on a CLDNN.
Specifically, as shown in fig. 3, the CLDNN network-based atrial fibrillation signal recognition model includes: the system comprises a cascaded input layer, a convolution sub-network, a bidirectional LSTM sub-network, a global maximum pooling layer, a full-connection mapping sub-network and an output layer; the convolution sub-network is used for reducing frequency domain variation of the RR interval samples and extracting the characteristics of the RR interval samples; the bidirectional LSTM sub-network is used for carrying out time sequence analysis on the characteristics of the RR interval samples; and the fully-connected mapping sub-network is used for carrying out mapping classification on the RR interval samples based on the time sequence analysis result. In this embodiment, the convolution sub-network is formed by cascading two convolution layers, 5 × 1 × 64 and 5 × 1 × 32; the bidirectional LSTM sub-network is composed of two LSTMs with opposite directions; the number of neurons in the two fully-connected layers in the fully-connected mapping sub-network is 32 and 16, respectively. The loss function of the atrial fibrillation signal identification model is a cross entropy function, specifically a cross entropy function
Wherein, yiLabels, p, representing RR interval samples iiRepresents the prediction probability of the atrial fibrillation signal identification model to RR interval sample i, and N represents the total number of RR interval samples. During training, parameters in the atrial fibrillation signal recognition model are updated by using an Adam optimizer.
In the embodiment of the invention, 20000 RR interval samples are respectively extracted from electrocardiosignals of 163 persistent atrial fibrillation cases and 200 non-atrial fibrillation cases, 40000 RR interval samples are taken as training sets for training an atrial fibrillation signal identification model, wherein when atrial fibrillation labels of the electrocardiosignals corresponding to all RR interval values in the RR interval samples are atrial fibrillation signals, the RR interval samples are marked as atrial fibrillation and are represented by 1; when at least one RR interval value exists in the RR interval samples, and the corresponding atrial fibrillation of the electrocardiosignal is marked as a non-atrial fibrillation signal, the RR interval sample is marked as the non-atrial fibrillation and is represented by 0. In order to further verify the effectiveness of the atrial fibrillation signal identification model trained in the embodiment, 29206 RR interval samples are extracted from the electrocardiosignals of 40 paroxysmal atrial fibrillation cases to serve as test data, wherein the test data and the training data are not repeated. In the embodiment of the invention, the sensitivity and specificity of all 29206 test data are used for measuring the performance of the atrial fibrillation signal identification model provided by the invention. The test result shows that the sensitivity of the atrial fibrillation signal recognition model is 97.63%, and the specificity is 99.18%.
The atrial fibrillation detection module is used for inputting the electrocardiosignals to be detected into the electrocardiosignal processing module to obtain RR interphase samples, inputting the RR interphase samples into the trained atrial fibrillation signal recognition model, and recognizing whether the RR interphase samples are atrial fibrillation signals, so that whether the electrocardiosignals to be detected contain the atrial fibrillation signals and the time when the atrial fibrillation occurs is determined.
It will be understood by those skilled in the art that the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the invention, and that any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (6)
1. The utility model provides an electrocardiosignal atrial fibrillation automatic check out system based on degree of depth study which characterized in that includes:
the electrocardiosignal processing module is used for carrying out R wave detection on the electrocardiosignal, extracting the R wave time of each heartbeat in the electrocardiosignal, and calculating the RR interval numerical value of the electrocardiosignal to obtain an RR interval sequence; preprocessing the RR interval sequence, and sampling the preprocessed RR interval sequence from the initial time of the corresponding electrocardiosignal based on a sliding window to obtain an RR interval sample;
the training module is used for acquiring a plurality of electrocardiosignals containing atrial fibrillation labels, and respectively inputting the electrocardiosignals into the electrocardiosignal processing module to obtain RR interphase samples; marking atrial fibrillation of the electrocardiosignal corresponding to each RR interval sample as an RR interval sample label; training an atrial fibrillation signal recognition model by taking each RR interphase sample as input and the corresponding RR interphase sample label as output to obtain a trained atrial fibrillation signal recognition model; wherein, electrocardiosignal is including the electrocardiosignal of atrial fibrillation and the electrocardiosignal of no atrial fibrillation, and atrial fibrillation signal identification model is the atrial fibrillation signal identification model based on CLDNN network, includes: the system comprises a cascaded input layer, a convolution sub-network, a bidirectional LSTM sub-network, a global maximum pooling layer, a full-connection mapping sub-network and an output layer; the convolution sub-network is used for reducing frequency domain variation of the RR interval samples and extracting the characteristics of the RR interval samples; the bidirectional LSTM subnetwork is used for carrying out time sequence analysis on the characteristics of the RR interval samples; the fully-connected mapping sub-network is used for carrying out mapping classification on the RR interval samples based on a time sequence analysis result; the fully-connected mapping sub-network comprises two fully-connected layers;
the atrial fibrillation detection module is used for inputting the electrocardiosignals to be detected into the electrocardiosignal processing module to obtain RR interphase samples, inputting the RR interphase samples into the trained atrial fibrillation signal recognition model, and recognizing whether the RR interphase samples are atrial fibrillation signals, so that whether the electrocardiosignals to be detected contain the atrial fibrillation signals and the time when the atrial fibrillation occurs is determined.
2. The system according to claim 1, wherein when the RR interval sequence is preprocessed by the ecg signal processing module, if the RR interval value has an abnormal value and the number of consecutive occurrences of the abnormal value is less than or equal to a preset number, performing interpolation correction on each abnormal RR interval value in the RR interval sequence; and if the continuous occurrence frequency of the abnormal value is more than the preset frequency, discarding the abnormal time in the RR interval sequence and the RR interval value in a preset time period before the abnormal time.
3. The system according to claim 1, wherein the module for processing the cardiac signal samples the pre-processed RR interval sequence from the initial time of the corresponding cardiac signal based on the sliding window, and moves the sliding window backward by a window size for the next extraction after each extraction is completed, so as to realize non-overlapping sampling.
4. The system according to claim 1, wherein said convolution sub-network is formed by cascading two convolution layers, 5 x 1 x 64 and 5 x 1 x 32; the bidirectional LSTM sub-network is composed of two LSTMs with opposite directions; the number of neurons of two fully-connected layers in the fully-connected mapping sub-network is 32 and 16 respectively.
5. The automatic detection system for atrial fibrillation of electrocardiosignals according to claim 1, wherein the loss function of the atrial fibrillation signal identification model is a cross entropy function; during training, parameters in the atrial fibrillation signal recognition model are updated by using an Adam optimizer.
6. The automatic atrial fibrillation detecting system for cardiac signals according to any one of claims 1 to 5, wherein the system is used in the technical field of electrocardiogram analysis.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010688677.4A CN111772628B (en) | 2020-07-16 | 2020-07-16 | Electrocardiosignal atrial fibrillation automatic detection system based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010688677.4A CN111772628B (en) | 2020-07-16 | 2020-07-16 | Electrocardiosignal atrial fibrillation automatic detection system based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111772628A CN111772628A (en) | 2020-10-16 |
| CN111772628B true CN111772628B (en) | 2021-10-08 |
Family
ID=72763177
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010688677.4A Active CN111772628B (en) | 2020-07-16 | 2020-07-16 | Electrocardiosignal atrial fibrillation automatic detection system based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111772628B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113876331B (en) * | 2021-10-26 | 2022-08-02 | 华中科技大学 | Electrocardiosignal-oriented semi-supervised atrial fibrillation automatic detection system |
| CN113940682A (en) * | 2021-11-10 | 2022-01-18 | 浙江好络维医疗技术有限公司 | Atrial fibrillation identification method based on statistical characteristics |
| CN113995419B (en) * | 2021-12-10 | 2023-04-11 | 华中科技大学 | Atrial fibrillation risk prediction system based on heartbeat rhythm signal and application thereof |
| CN114305442B (en) * | 2022-01-20 | 2024-03-29 | 复旦大学 | Detection method of atrial fibrillation occurrence start and stop points based on sliding window coding |
| CN114504324A (en) * | 2022-01-20 | 2022-05-17 | 武汉中旗生物医疗电子有限公司 | Method and device for detecting dynamic atrial fibrillation starting and stopping points and electronic equipment |
| CN115607167A (en) * | 2022-11-18 | 2023-01-17 | 中山大学 | Lightweight model training method, atrial fibrillation detection method, device and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104392137A (en) * | 2014-12-02 | 2015-03-04 | 北京海思敏医疗技术有限公司 | Signal abnormity detecting method and device based on entropy |
| CN108968941A (en) * | 2018-05-25 | 2018-12-11 | 深圳市太空科技南方研究院 | A kind of arrhythmia detection method, apparatus and terminal |
| CN109864736A (en) * | 2019-03-22 | 2019-06-11 | 深圳市理邦精密仪器股份有限公司 | Processing method, device, terminal device and the medium of electrocardiosignal |
| CN110766099A (en) * | 2019-11-08 | 2020-02-07 | 哈尔滨理工大学 | Electrocardio classification method combining discriminant deep belief network and active learning |
| CN110811608A (en) * | 2019-11-19 | 2020-02-21 | 中电健康云科技有限公司 | Atrial fibrillation monitoring method based on ECG (ECG) signals |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1566199B1 (en) * | 2004-02-23 | 2014-03-19 | Biotronik CRM Patent AG | Adaptive ventricular rate smoothing during atrial fibrillation |
| US20190328243A1 (en) * | 2016-12-21 | 2019-10-31 | Emory University | Methods and Systems for Determining Abnormal Cardiac Activity |
| CN106901723A (en) * | 2017-04-20 | 2017-06-30 | 济南浪潮高新科技投资发展有限公司 | A kind of electrocardiographic abnormality automatic diagnosis method |
| CN108334830B (en) * | 2018-01-25 | 2022-10-04 | 南京邮电大学 | Scene recognition method based on target semantic and depth appearance characteristic fusion |
| US10602940B1 (en) * | 2018-11-20 | 2020-03-31 | Genetesis, Inc. | Systems, devices, software, and methods for diagnosis of cardiac ischemia and coronary artery disease |
| CN110226921B (en) * | 2019-06-27 | 2022-07-29 | 广州视源电子科技股份有限公司 | Electrocardiosignal detection and classification method and device, electronic equipment and storage medium |
| CN110464334A (en) * | 2019-07-23 | 2019-11-19 | 苏州国科视清医疗科技有限公司 | A kind of electrocardiographic abnormality detection method based on composite depth learning network |
| CN111053549A (en) * | 2019-12-23 | 2020-04-24 | 威海北洋电气集团股份有限公司 | Intelligent biological signal abnormality detection method and system |
-
2020
- 2020-07-16 CN CN202010688677.4A patent/CN111772628B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104392137A (en) * | 2014-12-02 | 2015-03-04 | 北京海思敏医疗技术有限公司 | Signal abnormity detecting method and device based on entropy |
| CN108968941A (en) * | 2018-05-25 | 2018-12-11 | 深圳市太空科技南方研究院 | A kind of arrhythmia detection method, apparatus and terminal |
| CN109864736A (en) * | 2019-03-22 | 2019-06-11 | 深圳市理邦精密仪器股份有限公司 | Processing method, device, terminal device and the medium of electrocardiosignal |
| CN110766099A (en) * | 2019-11-08 | 2020-02-07 | 哈尔滨理工大学 | Electrocardio classification method combining discriminant deep belief network and active learning |
| CN110811608A (en) * | 2019-11-19 | 2020-02-21 | 中电健康云科技有限公司 | Atrial fibrillation monitoring method based on ECG (ECG) signals |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111772628A (en) | 2020-10-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111772628B (en) | Electrocardiosignal atrial fibrillation automatic detection system based on deep learning | |
| CN109171712B (en) | Atrial fibrillation identification method, atrial fibrillation identification device, atrial fibrillation identification equipment and computer readable storage medium | |
| CN106725428B (en) | Electrocardiosignal classification method and device | |
| CN103110417B (en) | Automatic electrocardiogram recognition system | |
| CN110638430B (en) | Method for building cascade neural network ECG signal arrhythmia classification model | |
| CN113499079B (en) | Atrial fibrillation detection method in electrocardiogram | |
| CN112971795B (en) | Electrocardiosignal quality evaluation method | |
| CN112617849A (en) | Atrial fibrillation detection and classification method based on CNN + LSTM | |
| CN105877739A (en) | Clinical examination method of electrocardio intelligent analyzing system | |
| CN113995419A (en) | Atrial fibrillation risk prediction system based on heartbeat rhythm signal and application thereof | |
| Behbahani et al. | Detection of QRS complexes in the ECG signal using multiresolution wavelet and thresholding method | |
| CN113180688B (en) | Coronary heart disease electrocardiogram screening system and method based on residual error neural network | |
| Kharshid et al. | Classification of short-time single-lead ECG recordings using deep residual CNN | |
| CN109199361B (en) | Method, storage medium and device for extracting R peak time in electrocardiosignal data | |
| Upasani et al. | Automated ECG Diagnosis | |
| CN111528833A (en) | Method and system for quickly identifying and processing electrocardiosignals | |
| CN110960207A (en) | Tree model-based atrial fibrillation detection method, device, equipment and storage medium | |
| Liu et al. | Automatic Arrhythmia Detection Based on Convolutional Neural Networks. | |
| CN114098754A (en) | Atrial fibrillation signal preprocessing method, detection system, equipment and storage medium | |
| CN108836312B (en) | Clutter rejection method and system based on artificial intelligence | |
| Kammath et al. | Detection of Bundle Branch Blocks using Machine Learning Techniques | |
| Jiang et al. | An SPC-based forward-backward algorithm for arrhythmic beat detection and classification | |
| Liu et al. | Adaptive atrial fibrillation detection focused on atrial activity analysis | |
| Aravind et al. | ECG Classification and Arrhythmia Detection Using Wavelet Transform and Convolutional Neural Network | |
| Sayem et al. | A novel 1D generative adversarial network-based framework for atrial fibrillation detection using restored wrist photoplethysmography signals |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |