CN108577833B - Atrial fibrillation detection device and method - Google Patents
Atrial fibrillation detection device and method Download PDFInfo
- Publication number
- CN108577833B CN108577833B CN201810147914.9A CN201810147914A CN108577833B CN 108577833 B CN108577833 B CN 108577833B CN 201810147914 A CN201810147914 A CN 201810147914A CN 108577833 B CN108577833 B CN 108577833B
- Authority
- CN
- China
- Prior art keywords
- capacitor
- resistance
- data
- atrial fibrillation
- electrically connected
- 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
Classifications
-
- 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
- A61B5/361—Detecting fibrillation
-
- 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
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Cardiology (AREA)
- Physics & Mathematics (AREA)
- Biophysics (AREA)
- Surgery (AREA)
- Artificial Intelligence (AREA)
- Veterinary Medicine (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Molecular Biology (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Fuzzy Systems (AREA)
- Signal Processing (AREA)
- Psychiatry (AREA)
- Physiology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention discloses an atrial fibrillation detection device and method, which comprises a PVDF film, a single chip microcomputer, a server, a detection circuit and a WIFI module; the detection circuit comprises a first amplifying circuit, a first trap circuit, a second amplifying circuit and a second trap circuit which are electrically connected in sequence; the positive electrode and the negative electrode of the PVDF film are electrically connected with the first amplifying circuit; the invention utilizes the artificial intelligence algorithm and modeling of big data to analyze the data during sleeping, has no obvious dependence on the environment, has small environmental interference because the data acquisition process is carried out in the sleeping process, does not need the tested personnel to have professional knowledge in the monitoring process, and has good user experience.
Description
Technical field
The present invention relates to technical field, a kind of atrial fibrillation high more particularly, to detection efficiency, that accuracy is high, at low cost is detected
Device and method.
Background technique
Current atrial fibrillation monitoring technology depends on the acquisition of ECG signal, and monitoring device must carry for a long time, monitoring
Process need to carry out in the environment of hospital.Equipment using when need profession personnel carry out instrument and equipment connection, finally obtain
The data taken are also relied on the evaluation and test of medical practitioner.During monitoring, user experience is relatively poor；Entire monitoring process simultaneously
Need to expend higher cost；And discovery when be to have obvious atrial fibrillation sign.
Summary of the invention
Goal of the invention of the invention is needed to overcome atrial fibrillation detection method in the prior art to need to rely on medical practitioner
The deficiency for expending higher cost provides detection efficiency height, accuracy height, atrial fibrillation detection device and method at low cost.
To achieve the goals above, the invention adopts the following technical scheme:
A kind of atrial fibrillation detection device, including pvdf membrane, single-chip microcontroller, server, detection circuit and WIFI module；Detection circuit
Including the first amplifying circuit, the first trap circuit, the second amplifying circuit and the second trap circuit being sequentially connected electrically；Pvdf membrane
Anode and cathode are electrically connected with the first amplifying circuit, and single-chip microcontroller is electrically connected with WIFI module and the second trap circuit respectively,
WIFI module is electrically connected with server.
Intelligent algorithm and modeling of the present invention using big data, data when analysis is slept.It is unobvious to environment
Dependence, while data acquisition process be to be carried out during sleep, environmental disturbances are small, and tested personnel is not needed in monitoring
There is the knowledge of profession and there is good user experience.
Preferably, first amplifying circuit includes interface J4, amplifier U5-1, capacitor C14, capacitor C12, capacitor
C13, capacitor C17, capacitor C18, capacitor C43, capacitor C44 and capacitor C15, resistance R23, resistance R17, resistance R18, resistance R42
With resistance R27, the both ends capacitor C14 are electrically connected with interface J4, and the one end capacitor C12 is electric with resistance R23 and the one end capacitor C14 respectively
Connection, the capacitor C12 other end is electrically connected with the resistance R23 other end and the one end resistance R28 respectively, the resistance R28 other end respectively with
The one end capacitor C18, the one end resistance R18 and the electrical connection of amplifier U5-1 inverting input terminal, capacitor C18 and the resistance R18 other end are equal
It is electrically connected with amplifier U5-1 output end and the one end resistance R27, capacitor C43 and the one end capacitor C44 are grounded, capacitor C43 and electricity
Hold the C44 other end and meet power supply VCC-OP, capacitor C13, resistance R42, the one end capacitor C17 are grounded, capacitor C13 other end difference
Meet power vd D33 and the one end resistance R17, the resistance R17 other end respectively with the resistance R42 other end, the capacitor C17 other end and amplification
The electrical connection of device U5-1 non-inverting input terminal.
Preferably, first trap circuit includes capacitor C28, capacitor C27, capacitor C29, capacitor C30, capacitor C31,
Amplifier U5-2, resistance R32, resistance R34, resistance R38, resistance R29, resistance R35 and resistance R39；The one end capacitor C28 and resistance
The electrical connection of the one end R32, the capacitor C28 other end are electrically connected with the one end capacitor C27 and the one end resistance R38 respectively, the resistance R32 other end
It is electrically connected respectively with capacitor C29, capacitor C30 and resistance R34 one end, capacitor C29, capacitor C30 and the resistance R38 other end are distinguished
Be electrically connected with resistance R35 and the one end resistance R39, the resistance R39 other end ground connection, the resistance R35 other end respectively with resistance R29 mono-
The output end electrical connection at end, the inverting input terminal of amplifier U5-2 and amplifier U5-2, the resistance R34 other end respectively with capacitor
The electrical connection of the non-inverting input terminal of the C27 other end and amplifier U5-2, the resistance R29 other end are electrically connected with the one end capacitor C31, capacitor
C31 other end ground connection.
A kind of method of atrial fibrillation detection device, includes the following steps:
(4-1)
(4-1-1) tests each non-atrial fibrillation tester and atrial fibrillation tester, obtains non-atrial fibrillation tester and atrial fibrillation
The BCG data of tester；
(4-1-2) removes the body in BCG data and moves data, remaining BCG data is divided into several segments, every section is 30
Second；
(4-1-3) medical practitioner makes the judgement of non-atrial fibrillation data or atrial fibrillation data to every segment data；
(4-1-4) is for every segment data xiFeature extraction is carried out, i is the serial number of data segment；
Define sample moment
Wherein, N xiIn data amount check, xi[n] indicates xiNth strong point,For xiMean value；
6 temporal signatures, 6 frequency domain characters and 3 supplementary features are extracted, matrix is formed
M is xiNumber, 15 numbers being characterized,0 in Y indicates
Non- atrial fibrillation, 1 indicates atrial fibrillation, randomly selects 80% data in X and Y as training set, in addition 20% is used as test set, training
Collection is trained in training pattern, and the accuracy of model is finally verified on test set；
(4-2) detection obtains the BCG data of a subject, and the body removed in BCG data moves data, by remaining BCG
Data are divided into several segments, and every section is 30 seconds；
6 temporal signatures for extracting subject, 6 frequency domain characters and 3 supplementary features of repeating the above steps simultaneously normalize
Processing；
(4-3) utilizes 15 characteristic values after normalized to find most like data line, the data line in X
Corresponding Y value is 1 or 0, exports subject's atrial fibrillation or normal result.
Preferably, temporal signatures include poor standard deviation, the degree of bias, kurtosis, maximum amplitude, maximum amplitude difference ratio and maximum
Difference in magnitude standard deviation；
Utilize formulaExtraction standard difference std (xi)；
Utilize formulaExtract degree of bias skeweness (xi)；
Utilize formulaExtract kurtosis kurtosis (xi)；
Utilize formula PP (xi)=max (xi)-min(xi) extract maximum amplitude it is poor, wherein max (xi) it is xiIn maximum
Value, min (xi) it is xiIn minimum value；
By xiIt is divided into 10 equal portions data xi j, j=1 ..., 10 utilizes formula PP (x for any equal portionsi j)=max
(xi j)-min(xi j) calculate jth equal portions data maximum amplitude difference PP (xi j), calculate PP (xi i) ..., PP (xi 10) in maximum
Value max (PP (xi)), calculate PP (xi i) ..., PP (xi 10) average value mean (PP (xi)),
Utilize max (PP (xi))/mean(PP(xi)) maximum amplitude difference ratio is calculated,
Utilize formulaCalculate the standard deviation of maximum amplitude difference.
Preferably, frequency domain character include PSD standard deviation, the PSD degree of bias, PSD kurtosis, peak away from standard deviation, peak away from the degree of bias and
Peak is away from kurtosis；
X is calculated using autoregressive methodiEnergy spectral density PSD, the time window of calculating is 5s, movement one in each second
The PSD being calculated is referred to as S [f, t] by secondary window, ifIt is averaged to obtain on t axis for S [f, t]；
Utilize formulaCalculate the PSD standard deviation of S [f, t]；
Utilize formulaCalculate the PSD degree of bias of S [f, t]；
Utilize formulaCalculate the PSD kurtosis of S [f, t]；
If Δ fpesk[k] is the peak peak distance that S [f, t] is obtained on f axis,
Utilize formulaPeak is calculated away from standard deviation std (Δ fpesk
[k])；
Utilize formulaPeak is calculated away from degree of bias skeweness
(Δfpeak[k])；
Utilize formulaPeak is calculated away from kurtosis kurtosis (Δ
fpeak[k])。
Preferably, training pattern is support vector machines, random forest Random Forest or linear discriminant analysis
LDA。
Preferably, the calculation method of supplementary features is as follows:
Supplementary features include zero-crossing rate, xdiStandard deviation and signal area；
Calculate xiZero passage number/N is defined as zero-crossing rate by the number of zero passage；
Calculate xiSlope between consecutive points, each slope form set xdi, calculate xdiStandard deviation std (xdi)；
Calculate xiSignal area
Therefore, the invention has the following beneficial effects: the intelligent algorithm for utilizing big data and modelings, when analysis is slept
Data；Environment is not relied on significantly, while the process of data acquisition is carried out during sleep, environmental disturbances
Small, tested personnel is not needed in monitoring to be had the knowledge of profession and has good user experience.
Detailed description of the invention
Fig. 1 is a kind of functional block diagram of the invention；
Fig. 2 is a kind of circuit diagram of the first amplifying circuit of the invention；
Fig. 3 is a kind of circuit diagram of the first trap circuit of the invention；
Fig. 4 is a kind of flow chart of the invention.
In figure: pvdf membrane 1, single-chip microcontroller 2, server 3, WIFI module 4, the first amplifying circuit 5, the first trap circuit 6,
Two amplifying circuits 7, the second trap circuit 8.
Specific embodiment
The present invention will be further described with reference to the accompanying drawings and detailed description.
Embodiment as shown in Figure 1 is a kind of atrial fibrillation detection device, including pvdf membrane 1, single-chip microcontroller 2, server 3, detection
Circuit and WIFI module 4；Detection circuit includes the first amplifying circuit 5 being sequentially connected electrically, the amplification of the first trap circuit 6, second
Circuit 7 and the second trap circuit 8；Pvdf membrane anode and cathode be electrically connected with the first amplifying circuit, single-chip microcontroller respectively with
WIFI module and the electrical connection of the second trap circuit, WIFI module are electrically connected with server.
As shown in Fig. 2, first amplifying circuit includes interface J4, amplifier U5-1, capacitor C14, capacitor C12, capacitor
C13, capacitor C17, capacitor C18, capacitor C43, capacitor C44 and capacitor C15, resistance R23, resistance R17, resistance R18, resistance R42
With resistance R27, the both ends capacitor C14 are electrically connected with interface J4, and the one end capacitor C12 is electric with resistance R23 and the one end capacitor C14 respectively
Connection, the capacitor C12 other end is electrically connected with the resistance R23 other end and the one end resistance R28 respectively, the resistance R28 other end respectively with
The one end capacitor C18, the one end resistance R18 and the electrical connection of amplifier U5-1 inverting input terminal, capacitor C18 and the resistance R18 other end are equal
It is electrically connected with amplifier U5-1 output end and the one end resistance R27, capacitor C43 and the one end capacitor C44 are grounded, capacitor C43 and electricity
Hold the C44 other end and meet power supply VCC-OP, capacitor C13, resistance R42, the one end capacitor C17 are grounded, capacitor C13 other end difference
Meet power vd D33 and the one end resistance R17, the resistance R17 other end respectively with the resistance R42 other end, the capacitor C17 other end and amplification
The electrical connection of device U5-1 non-inverting input terminal.
As shown in figure 3, first trap circuit includes capacitor C28, capacitor C27, capacitor C29, capacitor C30, capacitor
C31, amplifier U5-2, resistance R32, resistance R34, resistance R38, resistance R29, resistance R35 and resistance R39；The one end capacitor C28 with
The electrical connection of the one end resistance R32, the capacitor C28 other end are electrically connected with the one end capacitor C27 and the one end resistance R38 respectively, and resistance R32 is another
One end is electrically connected with capacitor C29, capacitor C30 and resistance R34 one end respectively, and capacitor C29, capacitor C30 and the resistance R38 other end are equal
Be electrically connected respectively with resistance R35 and the one end resistance R39, the resistance R39 other end ground connection, the resistance R35 other end respectively with resistance R29
The output end of one end, the inverting input terminal of amplifier U5-2 and amplifier U5-2 is electrically connected, the resistance R34 other end respectively with capacitor
The electrical connection of the non-inverting input terminal of the C27 other end and amplifier U5-2, the resistance R29 other end are electrically connected with the one end capacitor C31, capacitor
C31 other end ground connection.
As shown in figure 4, a kind of method of atrial fibrillation detection device, includes the following steps:
Step 101, each non-atrial fibrillation tester and atrial fibrillation tester are tested, obtains non-atrial fibrillation tester and atrial fibrillation
The BCG data of tester；The full name of BCG is Ballistoeardiogram, and Chinese is meant that ballistocardiography.
Step 102, the body removed in BCG data moves data, remaining BCG data is divided into multistage, every section is 30 seconds；
Step 103, medical practitioner makes the judgement of non-atrial fibrillation data or atrial fibrillation data to every segment data；
Step 104, for every segment data xiFeature extraction is carried out, i is the serial number of data segment；
Define sample moment
Wherein, N xiIn data amount check, xi[n] indicates xiNth strong point,For xiMean value；
Temporal signatures include poor standard deviation, the degree of bias, kurtosis, maximum amplitude, maximum amplitude difference ratio and maximum amplitude difference mark
It is quasi- poor；
Utilize formulaExtraction standard difference std (xi)；
Utilize formulaExtract degree of bias skeweness (xi)；
Utilize formulaExtract kurtosis kurtosis (xi)；
Utilize formula PP (xi)=max (xi)-min(xi) extract maximum amplitude it is poor, wherein max (xi) it is xiIn maximum
Value, min (xi) it is xiIn minimum value；
By xiIt is divided into 10 equal portions data xi j, j=1 ..., 10 utilizes formula PP (x for any equal portionsi j)=max
(xi j)-min(xi j) calculate jth equal portions data maximum amplitude difference PP (xi j), calculate PP (xi i) ..., PP (xi 10) in maximum
Value max (PP (xi)), calculate PP (xi i) ..., PP (xi 10) average value mean (PP (xi)),
Utilize max (PP (xi))/mean(PP(xi)) maximum amplitude difference ratio is calculated,
Utilize formulaCalculate the standard deviation of maximum amplitude difference.
Frequency domain character include standard deviation, the degree of bias, kurtosis, peak away from standard deviation, peak away from the degree of bias and peak away from kurtosis；
X is calculated using autoregressive methodiEnergy spectral density PSD, the time window of calculating is 5s, movement one in each second
The PSD being calculated is referred to as S [f, t] by secondary window, ifIt is averaged to obtain on t axis for S [f, t]；
The PSD standard deviation of S [f, t] is calculated using formula
The PSD degree of bias of S [f, t] is calculated using formula
The PSD kurtosis of S [f, t] is calculated using formula
If Δ fpesk[k] is the peak peak distance that S [f, t] is obtained on f axis,
Utilize formulaPeak is calculated away from standard deviation std (Δ fpesk
[k])；
Utilize formulaPeak is calculated away from degree of bias skeweness
(Δfpeak[k])；
Utilize formulaPeak is calculated away from kurtosis kurtosis (Δ
fpeak[k])。
The calculation method of supplementary features is as follows:
Supplementary features include zero-crossing rate, xdiStandard deviation and signal area；
Calculate xiZero passage number/N is defined as zero-crossing rate by the number of zero passage；
Calculate xiSlope between consecutive points, each slope form set xdi, calculate xdiStandard deviation std (xdi)；
Calculate xiSignal area
6 temporal signatures, 6 frequency domain characters and 3 supplementary features are extracted, matrix is formed
M is xiNumber, 15 numbers being characterized,0 in Y indicates
Non- atrial fibrillation, 1 indicates atrial fibrillation, randomly selects 80% data in X and Y as training set, in addition 20% is used as test set, training
Collection is trained in training pattern, and the accuracy of model is finally verified on test set；M xiIt is normal for 1000 to 2000
The data of people and patients with atrial fibrillation.
Step 200, detection obtains the BCG data of a subject, and the body removed in BCG data moves data, will be remaining
BCG data are divided into 100 sections, and every section is 30 seconds；
6 temporal signatures for extracting subject, 6 frequency domain characters and 3 supplementary features of repeating the above steps simultaneously normalize
It handles, every segment data can obtain 15 characteristic values after normalized, and 100 segment datas will obtain 100 groups of characteristic values；
Step 300, most like data line, the number are found in X using 100 groups of characteristic values after normalized
It is 1 or 0 according to the corresponding Y value of row, exports subject's atrial fibrillation or normal result.
It should be understood that this embodiment is only used to illustrate the invention but not to limit the scope of the invention.In addition, it should also be understood that,
After having read the content of the invention lectured, those skilled in the art can make various modifications or changes to the present invention, these etc.
Valence form is also fallen within the scope of the appended claims of the present application.
Claims (7)
1. a kind of atrial fibrillation detection device, characterized in that including pvdf membrane (1), single-chip microcontroller (2), server (3), detection circuit and
WIFI module (4)；Detection circuit includes the first amplifying circuit (5) being sequentially connected electrically, the first trap circuit (6), the second amplification
Circuit (7) and the second trap circuit (8)；The anode and cathode of pvdf membrane are electrically connected with the first amplifying circuit, single-chip microcontroller difference
It is electrically connected with WIFI module and the second trap circuit, WIFI module is electrically connected with server；Based on the atrial fibrillation detection device
Detection method include the following steps:
(4-1)
(4-1-1) tests each non-atrial fibrillation tester and atrial fibrillation tester, obtains non-atrial fibrillation tester and atrial fibrillation test
The BCG data of person；
(4-1-1) removes the body in BCG data and moves data, remaining BCG data is divided into several segments, every section is 30 seconds；
(4-1-2) medical practitioner makes the judgement of non-atrial fibrillation data or atrial fibrillation data to every segment data；
(4-1-3) is for every segment data xiFeature extraction is carried out, i is the serial number of data segment；
Define sample moment
Wherein, N xiIn data amount check, xi[n] indicates xiNth strong point,For xiMean value；
6 temporal signatures, 6 frequency domain characters and 3 supplementary features are extracted,
Form matrix
M is xiNumber, 15 numbers being characterized,0 in Y indicates non-room
It quivers, 1 indicates atrial fibrillation, randomly selects 80% data in X and Y as training set, in addition 20% is used as test set, training set exists
It is trained in training pattern, the accuracy of model is finally verified on test set；3 supplementary features are respectively zero-crossing rate, xdi
Standard deviation and signal area；Calculate xiSlope between consecutive points, xdiFor the set of each slope composition；
(4-2) detection obtains the BCG data of a subject, and the body removed in BCG data moves data, by remaining BCG data
Several segments are divided into, every section is 30 seconds；
Repeat the above steps 6 temporal signatures for extracting subject, 6 frequency domain characters and 3 supplementary features and normalized；
(4-3) utilizes 15 characteristic values after normalized to find most like data line in X, corresponding to the row data
Y value be 1 or 0, Y value is expressed as atrial fibrillation when being 1, and Y value is expressed as normal when being 0.
2. atrial fibrillation detection device according to claim 1, characterized in that first amplifying circuit includes interface J4, is put
Big device U5-1, capacitor C14, capacitor C12, capacitor C13, capacitor C17, capacitor C18, capacitor C43, capacitor C44 and capacitor C15, electricity
Resistance R23, resistance R17, resistance R18, resistance R42 and the both ends resistance R27, capacitor C14 are electrically connected with interface J4, capacitor C12 mono-
End be electrically connected respectively with resistance R23 and the one end capacitor C14, the capacitor C12 other end respectively with the resistance R23 other end and resistance R28
One end electrical connection, the resistance R28 other end are electric with the one end capacitor C18, the one end resistance R18 and amplifier U5-1 inverting input terminal respectively
Connection, capacitor C18 and the resistance R18 other end be electrically connected with amplifier U5-1 output end and the one end resistance R27, capacitor C43 with
The one end capacitor C44 is grounded, and capacitor C43 and the capacitor C44 other end meet power supply VCC-OP, capacitor C13, resistance R42, capacitor
The one end C17 is grounded, and the capacitor C13 other end meets power vd D33 and the one end resistance R17 respectively, the resistance R17 other end respectively with electricity
Hinder the R42 other end, the capacitor C17 other end and the electrical connection of amplifier U5-1 non-inverting input terminal.
3. atrial fibrillation detection device according to claim 1, characterized in that first trap circuit includes capacitor C28, electricity
Hold C27, capacitor C29, capacitor C30, capacitor C31, amplifier U5-2, resistance R32, resistance R34, resistance R38, resistance R29, resistance
R35 and resistance R39；The one end capacitor C28 is electrically connected with the one end resistance R32, the capacitor C28 other end respectively with the one end capacitor C27 and
The electrical connection of the one end resistance R38, the resistance R32 other end are electrically connected with capacitor C29, capacitor C30 and resistance R34 one end respectively, capacitor
C29, capacitor C30 and the resistance R38 other end are electrically connected with resistance R35 and the one end resistance R39 respectively, another termination of resistance R39
Ground, the resistance R35 other end output end with the one end resistance R29, the inverting input terminal of amplifier U5-2 and amplifier U5-2 respectively
Electrical connection, the resistance R34 other end are electrically connected with the non-inverting input terminal of the capacitor C27 other end and amplifier U5-2 respectively, resistance R29
The other end is electrically connected with the one end capacitor C31, capacitor C31 other end ground connection.
4. atrial fibrillation detection device according to claim 1, characterized in that temporal signatures include standard deviation, the degree of bias, kurtosis,
Maximum amplitude is poor, maximum amplitude difference ratio and maximum amplitude difference standard deviation；
Utilize formulaExtraction standard difference std (xi)；
Utilize formulaExtract degree of bias skeweness (xi)；
Utilize formulaExtract kurtosis kurtosis (xi)；
Utilize formula PP (xi)=max (xi)-min(xi) extract maximum amplitude it is poor, wherein max (xi) it is xiIn maximum value,
min(xi) it is xiIn minimum value；
By xiIt is divided into 10 equal portions data xi j, j=1 ..., 10 utilizes formula PP (x for any equal portionsi j)=max (xi j)-min
(xi j) calculate jth equal portions data maximum amplitude difference PP (xi j), calculate PP (xi i) ..., PP (xi 10) in maximum value max (PP
(xi)), calculate PP (xi i) ..., PP (xi 10) average value mean (PP (xi)),
Utilize max (PP (xi))/mean(PP(xi)) maximum amplitude difference ratio is calculated,
Utilize formulaCalculate the standard deviation of maximum amplitude difference.
5. atrial fibrillation detection device according to claim 1, characterized in that frequency domain character include PSD standard deviation, the PSD degree of bias,
PSD kurtosis, peak are away from standard deviation, peak away from the degree of bias and peak away from kurtosis；
X is calculated using autoregressive methodiEnergy spectral density PSD, the time window of calculating is 5s, the window of movement in each second
Mouthful, the PSD being calculated is referred to as S [f, t], ifIt is averaged to obtain on t axis for S [f, t]；
Utilize formulaCalculate the PSD standard deviation of S [f, t]
Utilize formulaCalculate the PSD degree of bias of S [f, t]
Utilize formulaCalculate the PSD kurtosis of S [f, t]
If Δ fpesk[k] is the peak peak distance that S [f, t] is obtained on f axis,
Utilize formulaPeak is calculated away from standard deviation std (Δ fpesk[k])；
Utilize formulaPeak is calculated away from degree of bias skeweness (Δ
fpeak[k])；
Utilize formulaPeak is calculated away from kurtosis kurtosis (Δ fpeak
[k])。
6. atrial fibrillation detection device according to claim 1, characterized in that training pattern is support vector machines, random gloomy
Woods Random Forest or linear discriminant analysis LDA.
7. atrial fibrillation detection device according to claim 1 or 2 or 3 or 4 or 5 or 6, characterized in that the calculating of supplementary features
Method is as follows:
Calculate xiZero passage number/N is defined as zero-crossing rate by the number of zero passage；
Calculate xdiStandard deviation std (xdi)；
Calculate xiSignal area
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810147914.9A CN108577833B (en) | 2018-02-13 | 2018-02-13 | Atrial fibrillation detection device and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810147914.9A CN108577833B (en) | 2018-02-13 | 2018-02-13 | Atrial fibrillation detection device and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108577833A CN108577833A (en) | 2018-09-28 |
| CN108577833B true CN108577833B (en) | 2019-09-13 |
Family
ID=63608849
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810147914.9A Active CN108577833B (en) | 2018-02-13 | 2018-02-13 | Atrial fibrillation detection device and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108577833B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111449622A (en) * | 2020-03-20 | 2020-07-28 | 复旦大学 | Atrial fibrillation recognition system based on BCG detection |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104382591A (en) * | 2014-12-11 | 2015-03-04 | 福州康为网络技术有限公司 | SRA (Sequence Read Archive) atrial fibrillation detector |
| CN106659407A (en) * | 2014-08-14 | 2017-05-10 | 心脏起搏器股份公司 | Atrial fibrillation detection using ventricular rate variability |
| CN107307860A (en) * | 2017-05-16 | 2017-11-03 | 哈尔滨医科大学 | A kind of atrial fibrillation detection means |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9314181B2 (en) * | 2009-11-03 | 2016-04-19 | Vivaquant Llc | Method and apparatus for detection of heartbeat characteristics |
| KR101910982B1 (en) * | 2012-02-13 | 2019-01-04 | 삼성전자주식회사 | Method and apparatus for eliminating motion artifact of biosignal using personalized biosignal pattern |
-
2018
- 2018-02-13 CN CN201810147914.9A patent/CN108577833B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106659407A (en) * | 2014-08-14 | 2017-05-10 | 心脏起搏器股份公司 | Atrial fibrillation detection using ventricular rate variability |
| CN104382591A (en) * | 2014-12-11 | 2015-03-04 | 福州康为网络技术有限公司 | SRA (Sequence Read Archive) atrial fibrillation detector |
| CN107307860A (en) * | 2017-05-16 | 2017-11-03 | 哈尔滨医科大学 | A kind of atrial fibrillation detection means |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108577833A (en) | 2018-09-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103190905B (en) | Multi-channel surface electromyography signal collection system based on wireless fidelity (Wi-Fi) and processing method thereof | |
| CN106214123A (en) | A kind of electrocardiogram compressive classification method based on degree of depth learning algorithm | |
| CN107358196A (en) | A kind of sorting technique of heart beat type, device and electrocardiogram equipment | |
| CN109480783B (en) | Apnea detection method and device and computing equipment | |
| US20180132744A1 (en) | Systems and methods for physiological sign analysis | |
| CN105550659A (en) | Real-time electrocardiogramclassification method based on random projection | |
| CN104490387A (en) | Wireless, portable and wearable electrocardiogram detector capable of restraining motion interference | |
| CN106066933A (en) | The method obtaining atrial fibrillation identification artificial neural network weighted value matrix | |
| CN107296604A (en) | A kind of atrial fibrillation determination methods | |
| CN110801212B (en) | BCG signal heart rate extraction method based on neural network | |
| CN108577833B (en) | Atrial fibrillation detection device and method | |
| CN109288515A (en) | Periodical monitoring method and device based on premature beat signal in wearable ECG signal | |
| CN108720839B (en) | Human motion evaluation method and device based on motion energy collection | |
| CN111449622A (en) | Atrial fibrillation recognition system based on BCG detection | |
| Srinivasan et al. | A new phase space analysis algorithm for cardiac arrhythmia detection | |
| CN103315767B (en) | Determining method and system for heart sound signals | |
| US10925546B2 (en) | System and method for evaluating effects of antiarrhythmic agent | |
| CN109009005A (en) | A kind of wearable Chinese medicine pulse acquisition and analysis system | |
| Pranata et al. | Detrended fluctuation analysis on ECG device for home environment | |
| TWI590802B (en) | System and method for arrhythmia judgment data base and ecg arrhythmia judgment | |
| CN115607167A (en) | Lightweight model training method, atrial fibrillation detection method, device and system | |
| CN105232041A (en) | Measuring device for electrical characteristic distribution of human body surface, data terminal and method | |
| Prawiro et al. | A heart rate detection method for low power exercise intensity monitoring device | |
| CN116595401A (en) | Load identification method based on POCS clustering algorithm | |
| CN108836312A (en) | A kind of method and system of the novel progress clutter rejecting based on artificial intelligence |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220428 Address after: 310000 713, building 2, Guigu building, No. 2000, Shixin North Road, ningwei street, Xiaoshan District, Hangzhou, Zhejiang Province Patentee after: Hangzhou Hebo Technology Co.,Ltd. Address before: 310003 room 407, No. 26, Zijinguan lane, Shangcheng District, Hangzhou City, Zhejiang Province Patentee before: HANGZHOU BOBO TECHNOLOGY Co.,Ltd. |
|
| TR01 | Transfer of patent right |