CN108420425A - Wearable dynamic maternal fetus electrocardioscanner - Google Patents
Wearable dynamic maternal fetus electrocardioscanner Download PDFInfo
- Publication number
- CN108420425A CN108420425A CN201710723391.3A CN201710723391A CN108420425A CN 108420425 A CN108420425 A CN 108420425A CN 201710723391 A CN201710723391 A CN 201710723391A CN 108420425 A CN108420425 A CN 108420425A
- Authority
- CN
- China
- Prior art keywords
- electrocardiosignal
- signal
- parent
- seconds
- pregnant woman
- 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
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]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0002—Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network
- A61B5/0004—Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network characterised by the type of physiological signal transmitted
- A61B5/0006—ECG or EEG signals
-
- 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
-
- 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]
- A61B5/344—Foetal cardiography
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/43—Detecting, measuring or recording for evaluating the reproductive systems
- A61B5/4306—Detecting, measuring or recording for evaluating the reproductive systems for evaluating the female reproductive systems, e.g. gynaecological evaluations
- A61B5/4343—Pregnancy and labour monitoring, e.g. for labour onset detection
- A61B5/4362—Assessing foetal parameters
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
- A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
- A61B5/6802—Sensor mounted on worn items
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
- A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
- A61B5/683—Means for maintaining contact with the body
- A61B5/6831—Straps, bands or harnesses
-
- 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
-
- 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
- A61B5/7203—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal
- A61B5/7207—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal of noise induced by motion artifacts
-
- 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
- A61B5/7225—Details of analog processing, e.g. isolation amplifier, gain or sensitivity adjustment, filtering, baseline or drift compensation
-
- 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
- A61B5/725—Details of waveform analysis using specific filters therefor, e.g. Kalman or adaptive filters
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B2503/00—Evaluating a particular growth phase or type of persons or animals
- A61B2503/02—Foetus
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Public Health (AREA)
- Molecular Biology (AREA)
- Pathology (AREA)
- Physics & Mathematics (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Biophysics (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Veterinary Medicine (AREA)
- Signal Processing (AREA)
- Physiology (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Psychiatry (AREA)
- Cardiology (AREA)
- Pediatric Medicine (AREA)
- Pregnancy & Childbirth (AREA)
- Gynecology & Obstetrics (AREA)
- Reproductive Health (AREA)
- Power Engineering (AREA)
- Computer Networks & Wireless Communication (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
Wearable dynamic maternal fetus electrocardioscanner, include the electrocardiosignal for acquiring pregnant woman and the ECG signal sampling module being connect with Analog-digital Converter module, microprocessor, data storage card for storing pregnant woman's electrocardiosignal and the maternal fetus ecg information stick generated according to the storage of preset storing frequencies and the wireless communication module that the maternal fetus ecg information stick of generation is wirelessly transmitted to hospital or monitor center, Analog-digital Converter module, data storage card and wireless communication module are connect with microprocessor, microprocessor is used for the real-time analysis of electrocardiosignal, data are stored to be controlled with communication, complete electrocardiosignal quality evaluation, parent electrocardio signal processing, Fetal ECG signal processing and generation maternal fetus ecg information stick.The device can realize the dynamic to parent and Fetal ECG signal, long-term monitoring and real-time analyzing processing, can be to parent and Fetal ECG long distance monitoring.
Description
Technical field
Noninvasive dynamic long route cardiogram monitoring is carried out to parent and fetus the present invention relates to a kind of, assesses electrocardiosignal matter in real time
Amount and extraction parent and fetal heart frequency information, while the device tool for the maternal fetus ecg information stick that can be wirelessly transferred is generated,
Body is that a kind of wearable maternal fetus electrocardio cellphone belongs to maternal fetus ECG detecting technical field.
Background technology
Since ecg wave form is recorded from body surface for the first time in Holland physiologist W.Einthoven in 1885, by more than 100 years
Development, ECG detecting has become clinical the most commonly used one of detection means.Ecg wave form reflects the electricity of cardiac excitation
Active procedure, in reflection heart basic function and its pathological change, analysis and the various arrhythmia cordis of discriminating assess myocardial damage
Degree and evolution instruct openheart surgery to carry out and indicate there is extremely important clinical reference valence in necessary drug-treated
Value.
Traditional maternal fetus ECG detecting carries out pregnant woman and lies in rest a few minutes on detection bed first in the following manner
Wait for that cardiovascular system activity is stablized, then by being placed in the noninvasive acquisition pregnant woman's electrocardiosignal of the electrocardioelectrode of abdomen, Yun Fuxin
Electric signal includes that (usual parent electrocardio signal strength is more than or is much larger than the fetus heart for the electrocardiosignal of parent and fetus simultaneously
Electric signal) then computer software Fetal ECG signal is extracted from pregnant woman's electrocardiosignal by automatically analyzing, finally provide
About Fetal ECG action message.The increase traditional detection method paid close attention to parent and foetus health with present society cannot
It is enough to meet the requirement that comprehensive dynamic long-time ECG detecting is carried out to parent and fetus, specifically its there are following four aspects
Deficiency：Lack to parent and fetus first, traditional maternal fetus ECG detecting is mainly ECG signal sampling under quiescent condition
The dynamic monitoring of electrocardio-activity, and some researches show that the electrocardiosignals under active state more likely to show potential disease
Information, it is therefore desirable to which dynamic monitoring is carried out to parent and Fetal ECG activity；Second is that traditional maternal fetus ECG detecting is mostly short
Journey detects, and detection time generally continues a few minutes shortage and tracked to parent and the movable long-range of Fetal ECG, and existing research table
The appearance that bright many carries the electrocardiosignal segment of specified disease information not necessarily receives the detection phase with sporadic in pregnant woman
Between present, it is therefore desirable to long-term monitoring is carried out to parent and Fetal ECG activity；Third, traditional more sides of maternal fetus ECG detecting
It overweights and medical explanation and judgement is made to Fetal ECG signal, have ignored the medical explanation to parent electrocardiosignal itself and sentence
It is disconnected, and the variation of parent electrocardio-activity itself equally contains very important diagnostic value, it is therefore desirable to parent and tire
Youngster's electrocardiosignal carries out the clinical of all-dimensional multi-angle and understands；Fourth, the computer automatic analysis of traditional maternal fetus ECG detecting
Method is suitable for the ecg analysis under plateau；And it is not suitable for the ECG Signal Analysis that long-range acquires under dynamic environment,
A large amount of not expected noise appearance can seriously affect judgement of the conventional method to parent and Fetal ECG feature under dynamic environment,
In addition, existing Computer aided analysis payes attention to not enough electrocardiosignal quality evaluation link at present, and electrocardiosignal quality is commented
It is the indispensable step for carrying out the premise and prerequisite of electrocardiosignal medicine deciphering, while being also the analysis of dynamic long route cardiogram signal to estimate
Suddenly, therefore, optimize conventional method, establish the new method gesture for being suitable for dynamic long-range parent and Fetal ECG signal processing must
Row.
Invention content
The present invention for the deficiency of existing clinical maternal fetus ECG detecting technology, provide one kind can realize to parent and
The noninvasive of Fetal ECG signal, dynamic, long-term monitoring can detection and analysis pregnant woman electrocardiosignal can be to parent and tire in real time
The wearable dynamic maternal fetus ECG detection device of youngster's electrocardio long distance monitoring.
The wearable dynamic maternal fetus ECG detection device of the present invention, including with lower part：
(1）ECG signal sampling module, including sequentially connected electrocardioelectrode, low-pass filter, high-pass filter, power frequency
Trapper and signal amplification circuit are connect with Analog-digital Converter module, the electrocardiosignal for acquiring pregnant woman, and complete signal
Filtering, denoising and amplification；
Signal is after electrocardioelectrode acquires by low-pass filter, high-pass filter, power frequency notch filter and signal amplification circuit
Low-pass filtering, high-pass filtering, notch filter and signal amplification are carried out successively.
(2) Analog-digital Converter module：It is connect with microprocessor, for realizing the conversion of analog signal to digital signal；
(3) data storage card：It is connect with microprocessor, for storing pregnant woman's electrocardiosignal, while according to preset
The maternal fetus ecg information stick that storing frequencies storage generates；7
(4) wireless communication module：It is connect with microprocessor, the maternal fetus ecg information stick of generation is wirelessly transmitted to doctor
Institute or monitor center (with picture of mobile telephone format)；
(5) microprocessor：It is connect with Analog-digital Converter module, data storage card and wireless communication module, is used for electrocardio
Real-time analysis, the data storage of signal are controlled with communication, complete electrocardiosignal quality evaluation, the parent electrocardio signal processing fetus heart
Electric signal processing and generation maternal fetus ecg information stick；
Pregnant woman's electrocardiosignal initially enters ECG signal sampling module, subsequently into Analog-digital Converter module, through simulation
Enter the signal storehouse of T seconds set time length of window after number conversion, update was primary per T/2 seconds for the storehouse, freshly harvested T/2
Second signal stacking, earliest collected T/2 seconds signal are popped, and the T second electrocardiosignal segments handled twice before and after ensureing exist
50% overlapping, to ensure can obtain continuous ecg characteristics information, avoid T seconds electrocardiosignal segment head and the tail characteristic informations without
Information is omitted caused by method detects；Electrocardiosignal quality is carried out first to the T seconds electrocardiosignal segment come out from signal storehouse to comment
Estimate, be then sequentially completed parent electrocardio signal processing Fetal ECG signal processing and generate maternal fetus ecg information stick, specifically
Process is as follows；
It is commented 1. electrocardiosignal quality evaluation carries out quality respectively to each lead signals of 3 lead pregnant woman's electrocardiosignal of T seconds
Estimate, it is good that each lead electrocardiosignal quality is divided into a kind of electrode delamination in following 4 class, motion artifacts, very noisy and quality
It is good；
2. parent electrocardio signal processing T seconds pregnant woman's electrocardiosignal good to signal quality is analyzed, 1. according to step
The pregnant woman's electrocardiosignal quality information provided automatically selects Principal Component Analysis Method or the peak detection based on parabola weight
Determine that parent R crest values point position according to parent R crest value points position, obtains parent RR interval series, finally according to parent RR
Interval series obtain maternal heart rate information, while constructing parent electrocardio template and reconstructing parent electrocardio waveform；
3. Fetal ECG signal processing T seconds pregnant woman's electrocardiosignal good to signal quality is analyzed, first from pregnant woman
Parent electrocardio signal is rejected in electrocardiosignal, obtains Fetal ECG signal, the pregnant woman's electrocardiosignal then 1. provided according to step
Quality information automatically selects Principal Component Analysis Method or determines fetus R crest value points based on the peak detection of parabola weight
It sets according to fetus R crest value points position, obtains fetus RR interval series, finally obtain fetal heart frequency according to fetus RR interval series
Information；
4. generating the maternal fetus ecg information that maternal fetus ecg information stick generates above-mentioned T seconds set time length of window
Stick.
Information stick is generated using picture format, is from top to bottom pregnant woman's essential information, packet comprising 4 partial information part 1s
It is the 3 lead pregnant woman's ecg wave forms removed after baseline drift, electrocardio letter to include name, age, pregnant week, height and weight part 2
Number quality information is shown in the 3 lead parent electrocardio waveforms that third portion above waveform be reconstruct and parent R crest value points position is marked
Note, if non-" quality the is good " type of a certain lead electrocardiosignal, which shows straight line, if 3 lead
It is not " quality is good " type to join electrocardiosignal, then 3 lead parent electrocardio waveforms show that straight line, maternal heart rate numerical value are shown
Above waveform, if 3 lead electrocardiosignals are not " quality is good " types, maternal heart rate numerical value vacancy；4th part be from
3 lead Fetal ECG waveforms after parent electrocardio signal and fetus R crest value points position mark are rejected in pregnant woman's electrocardiosignal, such as
Non- " quality the is good " type of a certain lead electrocardiosignal of fruit, then the lead Fetal ECG waveform show straight line, if 3 lead electrocardios
Signal is not " quality is good " type, then 3 lead Fetal ECG waveforms show that straight line fetal heart frequency numerical value is shown in waveform
Top, if 3 lead electrocardiosignals are not " quality is good " types, fetal heart frequency numerical value vacancy.
Above-mentioned steps 1. in, the method classified to electrocardiosignal quality is as follows：
First detect whether that current T seconds electrocardiosignal is divided into 5 sections for one section by " electrode delamination " by the T/5 seconds, if wherein 2 sections
There is constant in the above signal, then is determined as " electrode delamination ", if not " electrode delamination ", then judge whether that " motion artifacts " detect
The maximum value of current T second electrocardiosignals is used as threshold value using the 50% of the maximum value, will currently T seconds electrocardiosignals by the T/5 seconds one
Section is divided into 5 sections, if the time point more than the threshold value is not present in wherein 2 sections or more signals, be determined as " motion artifacts " if
It is not that " motion artifacts " judge that " very noisy " or " quality is good " is judged using Sample Entropy algorithm if the T seconds heart again
The sample entropy of electric signal is more than 1.5, then is determined as that " very noisy " is otherwise determined as " quality is good ".
The following T of computational methods of T seconds electrocardiosignal sample entropy：Second electrocardiosignal is resampled is fixed for 500 points first
Length, if the signal after resampling is data={ x1, x2 ..., xK }, K is that data point number carries out time series phase to data
Space Reconstruction takes m continuity point composition vector Y successivelym(i)=[xi, xi+ 1 ..., xi+m- 1], i=1~K-m+1 defines vector
Ym(i) and YmThe distance between (j) d (Ym(i), Ym(j))=max (| xi+ k-xj+k |), given threshold r=0.2 × σ X, wherein σ
X is the standard deviation of time series data, does not include element Y for K-m+1 m n dimensional vector ns statisticsm(i) K-m vector including
In with element Ym(i) distance is less than the element number of threshold value r, is denoted as template matches number Km(i), and K is calculatedm(i) with apart from total
NumberTo allMean value, be denoted as
Then space dimensionality is increased into m+1, B is recalculated according to above-mentioned stepsm+ 1 (r), then the Sample Entropy of signal data be
SiampEn (m, r, K)=ln [Bm(r)/Bm+ 1 (r)], parameter mr and K are respectively set to, and 2,0.2 and 500.
Above-mentioned steps are 2. middle to determine that the method for parent R crest value points position is as follows：If in 3 lead pregnant woman's electrocardiosignals at least
There are 2 lead electrocardiosignal quality good, then selects Principal Component Analysis Method；If only having 1 lead electrocardio in 3 lead pregnant woman's electrocardiosignals
Signal quality is good, then selects the peak detection based on parabola weight.
Principal Component Analysis Method determines that the specific method is as follows for parent R crest value points position：If the second best in quality lead number is
M, wherein M are 2 or 3 to set pregnant woman's electrocardiosignal of the M leads as { xm(n) }, wherein m=1,2 ..., M, n=1,2 ...,
N；N is that T seconds electrocardiosignal length generation electrocardiosignal matrix X are：
Wherein x (n)=[x1 (n), x2 (n) ..., xM (n)]TIt is formed in same sampling instant sampled value for M leads electrocardio
Vectorial A, indicate coefficient matrix Y；It indicates the signal source matrix of electrocardiosignal matrix X and meets orthonormalization conditioned signal source
The first row y1 (n) of matrix Y=[y1 (1), y1 (2) ..., y1 (N)] is the main composition of pregnant woman's electrocardiosignal, because of parent
Electrocardiosignal intensity is more than Fetal ECG signal so the i.e. corresponding parent electrocardio signal matrix A of the main composition and Y shape formula are as follows：
Carrying out Eigenvalues Decomposition to the correlation matrix R of electrocardiosignal matrix X is：
Wherein ∑ is the eigenvalue matrix ∪ of matrix R, is the eigenvectors matrix of matrix R, because of X=AY, then above formula is write
For：
Then there are A=U ∑s 1/2, that is, hasItsFor the feature of matrix R
The main composition of feature vector pregnant woman's electrocardiosignal (parent electrocardio signal) for matrix R character pair value λ m is then expressed as：
T seconds parent electrocardio signal y1 (n) are divided into 5 sections for one section by the T/5 seconds, the maximum value in every section is chosen and calculates this
The mean value of a little maximum values determines that the region in main composition higher than this threshold value is parent electrocardio letter using the 60% of mean value as threshold value
Number R waves region finally in this region of former pregnant woman's electrocardiosignal maximizing/minimum value as parent R crest value points
It sets.
It is determined parent R crest value points position the specific method is as follows at this time using the peak detection based on parabola weight
Only have 1 lead signals quality good in 3 lead electrocardiosignals, if the lead signals are p (n), band logical is carried out to the signal first
Filtering, retains 0.05Hz-40Hz frequency band signals filtered signals and is denoted as q (n), then generates the linear order that length is at 20 points
K1=[1,0.95,0.9,0.85 ..., 0.05], it is the square root sequence of sequence k1 to enable sequence k2, and it is sequence k2 to enable sequence k3
Overturning sequence, then parabola weight vectors obtained by w=[k3,1, k2], be 41 point sequences, then by the parabola weight
Vectorial w does to filtered electrocardiosignal q (n) related point by point successively, obtains coherent signal r (n) finally by T seconds time spans
Coherent signal r (n) is divided into 5 sections for one section by the T/5 seconds, chooses the maximum value in every section and calculates the mean value of these maximum values with equal
The 80% of value determines that the region in r (n) higher than this threshold value is parent electrocardio signal R waves region as threshold value, finally former pregnant
Maximizing/minimum value is as parent R crest value points position in this region of woman's electrocardiosignal.[0033] above-mentioned steps 2. in
Construction parent electrocardio template and reconstruct parent electrocardio waveform method it is as follows：The heart between the front and back adjacent R crest value points of extraction
Electric signal segment is denoted as vi (n), and vi (n) is single cardiac cycle electrocardiosignal, and wherein i=1,2 ..., L, L are week aroused in interest
Issue stretches L vi (n) to identical signal length, then calculates its coherence average valueAs
The parent electrocardio template of T seconds electrocardiosignals of current lead, then reconstructs parent electrocardio signal, is obtained according to R crest value points position
Phase between RR, by electro-cardiologic template signalStretch the identical signal of phase length between above-mentioned each RR successively, then will sequentially stretch
Signal head and the tail connection after contracting forms the parent electrocardio signal after reconstruct, at this time first R in current T seconds electrocardiosignal segment
Parent electrocardio signal vacancy before crest value point position and after the last one R crest value point is mended according to corresponding time relationship
Fill the correspondence signal patch of the parent electrocardio signal of T seconds electrocardiosignal segments of previous T seconds electrocardiosignal segment and the latter reconstruct
Section obtains complete T seconds parent electrocardio signal.
Above-mentioned steps 3. in from pregnant woman's electrocardiosignal reject parent electrocardio signal method be：First to T seconds pregnant woman's hearts
Electric signal xx (n) and the parent electrocardio signal yy (n) of reconstruct are adopted 10 again using 10 sampling frequencies of former sample frequency
Sample is denoted as xx* (n) and yy* (n) respectively, when being then set to 200 milliseconds of center in extraction yy* (n) with R crest value points
Between window signal segment yyi* (n) and the signal segment xxi* (n) of time corresponding with xx* (n) carry out relevant operation, note
Correlation is recorded, then carry out relevant operation with xxi* (n) again by moving left and right yyi* (n) and records correlation and determines one
The sampling number that maximum value and record in serial correlation move left and right at this time, and according to this sampling number adjustment yyi* (n)
Position in resampling stepmother body-centered electric signal yy* (n) obtains the parent after R wave fine positions finally by down-sampled
Electrocardiosignal, the parent electrocardio signal is subtracted from pregnant woman's electrocardiosignal realizes the rejecting of parent electrocardio signal.
Above-mentioned steps 3. in, determine parent R crest value points in determining the method for fetus R crest value points position and step 2.
The method set is identical.
Wearable maternal fetus electrocardio cellphone tool provided by the invention has the advantage that：
1, it realizes to parent and the dynamic of Fetal ECG signal, long-range, synchronous, monitoring in real time, it can be by maternal fetus electrocardio
Information is uploaded to hospital or monitor center in real time, can store the ecg wave form under pregnant woman's various activities state, improves clinical pregnant
Woman's cardiac monitoring technology has stronger clinical value；
2, the reliability for automatically analyzing result is enhanced to the necessary and accurate quality evaluation of pregnant woman's electrocardiosignal progress；
3, exclusive maternal fetus ecg information stick technology not only facilitates analysis of the medical staff to pregnant woman's electrocardiosignal,
Also it substantially increases pregnant woman itself and pregnant woman's physical and mental health is relatively beneficial to the reading and understanding ability of electrocardiosignal.
Description of the drawings
Fig. 1 is the structure principle chart of the wearable dynamic maternal fetus ECG detection device of the present invention.
Fig. 2 is the tape structural schematic diagram for bondage device of the present invention.
Fig. 3 is ecg signal acquiring processing flow schematic diagram of the present invention.
Fig. 4 is the exemplary plot of the 3 lead pregnant woman electrocardiosignals maternal fetus ecg information stick of " quality is good ".
Fig. 5 is the exemplary plot for the maternal fetus ecg information stick that lead signals in 3 leads belong to " electrode delamination ".
Fig. 6 is the exemplary plot for the maternal fetus ecg information stick that lead signals in 3 leads belong to " motion artifacts ".
Fig. 7 is the exemplary plot for the maternal fetus ecg information stick that lead signals in 3 leads belong to " very noisy ".
Fig. 8 is that 3 lead pregnant woman's electrocardiosignals are not admitted to " quality is good " type so that the parent tire that analysis can not carry out
The exemplary plot of youngster's ecg information stick.
Wherein：1, ECG signal sampling module, 2, Analog-digital Converter module, 3, microprocessor, 4, data storage card, 5,
Wireless communication module, 6, tape, 7, unisexuality magic tape, 8, pocket, 9, upper right electrocardioelectrode, 10, upper left electrocardioelectrode, 11, left
Lower electrocardioelectrode, 12, bottom right electrocardioelectrode.
Specific implementation mode
The present invention is more specifically described in detail with reference to specific embodiment, but embodiments of the present invention are unlimited
Routine techniques progress can refer to for not specifically specified technological parameter in this.
The wearable dynamic maternal fetus ECG detection device of the present invention includes ECG signal sampling module as shown in Figure 1
1 Analog-digital Converter module 2, DSP processing chips（Microprocessor 3) data storage card 4 and wireless communication module 5.Electrocardiosignal
Detection module 1 is connect with Analog-digital Converter module 2, the electrocardiosignal for acquiring pregnant woman, and completes the filtering of signal, denoising
And amplification, including sequentially connected electrocardioelectrode (upper right electrocardioelectrode 9, upper left electrocardioelectrode 10, lower-left electrocardioelectrode 11 and the right side
Lower electrocardioelectrode 12), low-pass filter, high-pass filter, 50Hz power frequency notch filters and signal amplification circuit, signal is through electrocardio electricity
Successively by by low-pass filter, high-pass filter, 50Hz power frequency notch filters and signal amplification circuit completion are low after the acquisition of pole
Pass filter, high-pass filtering, 50Hz notch filters and signal amplification.2 data storage card 4 of Analog-digital Converter module and wireless telecommunications
Module 5 is connect with microprocessor 3.Analog-digital Converter module is set as using 12 or more A/D converter sample frequencys
500Hz.Data storage card 4 is for storing pregnant woman's electrocardiosignal, while the mother generated according to the storage of preset storing frequencies
Body Fetal ECG information stick, wireless communication module is according to preset communication frequency, by the maternal fetus ecg information of generation
It is every 20 point 11 that stick is wirelessly transmitted to the storing frequencies of hospital or monitor center acquiescence and communication frequency with picture of mobile telephone format
Clock is primary, and user can reset storing frequencies or communication frequency as needed.Microprocessor 3 uses DSP processing chips, interior
ECG Signal Analysis software is set, the storage of real-time analysis data and the communication for electrocardiosignal control.
Fig. 2 gives the riding position of tape structure and tape top electrode for bondage above-mentioned apparatus.Tape 6 is by will not
Flexible rigid cloth is made, and both sides connect elastic fastener band 7, and elastic fastener band 7 can adjust the elasticity of bondage, tape
Outside is provided with the pocket 8 for placing whole device.3 lead ecg signal acquirings place mode, 4 electrodes using 4 electrodes of standard
It is individually positioned in the quadrangle of rectangle, respectively；Upper right electrocardioelectrode 9 (red), upper left electrocardioelectrode 10 (yellow), lower-left electrocardio
Electrode 11 (green) and bottom right electrocardioelectrode 12 (black), wherein bottom right electrocardioelectrode 12 are as with reference to electrode.1 electrocardio of lead
Signal is by upper right electrocardioelectrode 9 and reference electrode bottom right electrocardioelectrode 12) 2 electrocardiosignal of lead is obtained by upper left electrocardioelectrode 10
3 electrocardiosignal of lead is obtained with reference electrode to be obtained with reference electrode by lower-left electrocardioelectrode 11.Tape size be divided into it is big or middle,
The small corresponding electrode gap of three types is different, and specific size see the table below：
Cardiac diagnosis lead interface is set in ECG signal sampling module 1, and each electrocardioelectrode is connected by the cardiac diagnosis lead interface
It is connected to ECG signal sampling module 1, cardiac diagnosis lead interface is Universal USB data-interface, and pregnant woman's electrocardiosignal passes through when measurement
This interface enters ECG signal sampling module 1, and data can be exported by this interface when non-measured.
The ECG Signal Analysis software being solidificated on microprocessor 3, be used for electrocardiosignal real-time analysis, data storage with
Communication control, including electrocardiosignal quality assessment modules, parent electrocardio signal processing module Fetal ECG signal processing module and
Maternal fetus ecg information stick generation module is respectively completed electrocardiosignal quality evaluation, parent electrocardio signal processing Fetal ECG
Signal processing and generation maternal fetus ecg information stick.The object of all software module processing is that T seconds set time windows are long
(time window can sets itself, default value be 10 seconds within the scope of 5 seconds to 20 seconds to 3 lead electrocardiosignal segments of degree.) Fig. 3
Give ecg signal acquiring processing flow schematic diagram.User need to first confirm that personal essential information essential information includes：Name,
Age, pregnant week, height and weight, then connect electrocardioelectrode and tie up tape and power on and start timely collection signal,
Details are as follows for detailed process.
Signal initially enters ECG signal sampling module 1, subsequently into Analog-digital Converter module 2, turns through simulation numeral
Enter the signal storehouse of T seconds set time length of window after changing, update was primary per T/2 seconds for the storehouse, freshly harvested T/2 seconds signal
Collected T/2 seconds signal is popped earliest for stacking, before and after ensureing the T seconds electrocardiosignal segment that handles twice there are 50% overlapping,
To ensure that continuous ecg characteristics information can be obtained, avoids T seconds electrocardiosignal segment head and the tail characteristic informations that from can not detecting and make
At information omit.
Electrocardiosignal quality assessment modules are entered through the regular signal for after set time length of signal storehouse, to the T seconds 3
Each lead signals of lead pregnant woman's electrocardiosignal carry out quality evaluation respectively, and each lead electrocardiosignal quality is divided into following
A kind of electrode delamination, motion artifacts, very noisy and quality in 4 classes is good.If 3 lead pregnant woman's electrocardiosignals are without 1 lead
Electrocardiosignal quality is good, then program no longer carries out, and prompt the T seconds pregnant woman's electrocardiosignal segment " poor signal quality without, method into
Row analysis ".The method classified to electrocardiosignal quality using quality evaluation algorithm is as follows：First detect whether that " electrode is de-
It falls "Current T seconds electrocardiosignal is divided into 5 sections for one section by the T/5 seconds, if wherein constant occur in 2 sections or more signals, is determined as
" electrode delamination ", if not " electrode delamination then judges whether that " motion artifacts " detect the maximum value of current T seconds electrocardiosignal, with
The 50% of the maximum value is used as threshold value, currently will be divided into 5 sections for one section by the T/5 seconds by T seconds electrocardiosignals, if wherein 2 sections or more
The time point more than the threshold value is not present in signal, then is determined as " motion artifacts ", if nor " motion artifacts ", then judge be
" very noisy " still " quality is good " is judged using Sample Entropy algorithm, if the sample entropy of the T seconds electrocardiosignal is more than
1.5, then it is determined as that " very noisy " is otherwise determined as " quality is good ".
Calculate the following T of method of T seconds electrocardiosignal sample entropy：Second electrocardiosignal is resampled is fixed for 500 points first
Length, if the signal after resampling is data={ x1, x2 ..., xK }, K is data point number, and time series phase is carried out to data
Space Reconstruction takes m continuity point composition vector Y successivelym(i)=[xi, xi+ 1 ..., xi+ m-1] i ,=1~K-m+1, definition arrow
Measure Ym(i) and YmThe distance between (j) d (Ym(i), Ym(j））=max (| xi+ k-xj+k |) given threshold r=0.2 × σ X, wherein
The standard deviation that σ X12 are time series data does not include element Y for K-m+1 m n dimensional vector ns statisticsm(i) K-m arrow including
In amount with element Ym(i) distance is less than the element number of threshold value r, is denoted as template matches number Km(i), and K is calculatedm(i) and distance
SumTo allMean value, be denoted as
Then space dimensionality is increased into m+1, B is recalculated according to above-mentioned stepsm+ 1 (r), then the Sample Entropy of signal data be
SiampEn (m, r, K)=ln [Bm(r)/Bm+ 1 (r)], parameter mr and K are respectively set to 2,0.2 and 500.
Then signal enters parent electrocardio signal processing module, module T seconds pregnant woman's electrocardio letter good to signal quality
It number is analyzed, the pregnant woman's electrocardiosignal quality information provided according to electrocardiosignal quality assessment modules automatically selects main composition point
Analysis method determines parent R crest values point position according to parent R crest value points position based on the peak detection of parabola weight,
Parent RR interval series are obtained, finally calculate maternal heart rate information according to parent RR interval series, while utilizing coherence average skill
Art constructs parent electrocardio template and reconstructs parent electrocardio waveform.Difference according to electrocardiosignal quality automatically selects parent electrocardio R
The method of crest value point recognizer is as follows：If at least 2 lead electrocardiosignal quality are good in 3 lead pregnant woman's electrocardiosignals,
Then select Principal Component Analysis Method；If only having in 3 lead pregnant woman's electrocardiosignals, 1 lead electrocardiosignal quality is good, and selection is based on
The peak detection of parabola weight.
Wherein determine that the specific method is as follows for parent R crest value points position using Principal Component Analysis Method；If the second best in quality lead
Connection number is M, and wherein M is 2 or 3, if pregnant woman's electrocardiosignal of the M leads is { xm(n) }, wherein m=1,2 ..., M, n=1,
(N is T seconds electrocardiosignal length to 2 ..., N.) generate electrocardiosignal matrix X be：
Wherein x (n)=[x1 (n), x2 (n) ..., xM (n)T] it is that M leads electrocardio is formed in same sampling instant sampled value
Vectorial A, indicate coefficient matrix Y；It indicates the signal source matrix of electrocardiosignal matrix X and meets orthonormalization conditioned signal source
The first row y1 (n) of matrix Y=[y1 (1), y1 (2) ..., y1 (N)] is the main composition of pregnant woman's electrocardiosignal, because of parent
This electrocardiosignal.Matrix
A and
Carrying out Eigenvalues Decomposition to the correlation matrix R of electrocardiosignal matrix X is：
Wherein ∑ is the eigenvalue matrix ∪ of matrix R, is the eigenvectors matrix of matrix R, because of X=AY, then above formula can
It is written as：
Then there are A=U ∑s 1/2 to haveItsFor the feature of matrix RFor matrix
The feature vector of R character pair value λ m.The main composition of pregnant woman's electrocardiosignal (parent electrocardio signal) can be expressed as：
T seconds parent electrocardio signal y1 (n) are divided into 5 sections for one section by the T/5 seconds, the maximum value in every section is chosen and calculates this
The mean value of a little maximum values determines that the region in main composition higher than this threshold value is parent electrocardio letter using the 60% of mean value as threshold value
Number R waves region, finally in this region of former pregnant woman's electrocardiosignal maximizing/minimum value as parent R crest value points
It sets.
The peak detection based on parabola weight is wherein used to determine that the specific method is as follows for parent R crest value points position：
Only have 1 lead signals quality good in 3 lead electrocardiosignals at this time, if the lead signals are p (n), the signal is carried out first
Bandpass filtering retains 0.05Hz-40Hz frequency band signals, and filtered signal is denoted as q (n), then generates the line that length is at 20 points
Property sequence k1=[1,0.95,0.9,0.85 ..., 0.05], enable sequence k2 be sequence k1 square root sequence, enable sequence k3 be sequence
The overturning sequence of k2 is arranged, then parabola weight vectors are obtained by w=[k3,1, k2], are 41 point sequences, then by the parabola
Weight vectors w does to filtered electrocardiosignal q (n) related point by point successively, coherent signal r (n) is obtained, finally by T seconds time
The coherent signal r (n) of length is divided into 5 sections for one section by the T/5 seconds, chooses the maximum value in every section and calculates the equal of these maximum values
Value determines that the region in r (n) higher than this threshold value is parent electrocardio signal R waves region, finally using the 80% of mean value as threshold value
Maximizing/minimum value is as parent R crest value points position in this region of former pregnant woman's electrocardiosignal.
Wherein utilize coherence average technical construction parent electrocardio template and reconstruct parent electrocardio signal method it is as follows, extraction
Electrocardiosignal segment between front and back adjacent R crest value points is denoted as vi (n) v, and i (n) is single cardiac cycle electrocardiosignal,
Wherein i=1,2 ..., L, L are cardiac cycle number, and L vi (n) is stretched to identical signal length, and it is relevant flat then to calculate it
Mean value The parent electrocardio template of as current lead T seconds electrocardiosignals, then reconstructs parent electrocardio
Signal obtains the phase between RR, by electro-cardiologic template signal according to R crest value points positionPhase between above-mentioned each RR of stretching successively is long
Spend identical signal, then sequentially will it is flexible after the connection of signal head and the tail form the parent electrocardio signal after reconstruct, at this time when
Parent electrocardio letter in preceding T seconds electrocardiosignal segment before first R crest value points position and after the last one R crest value point
Number vacancy is supplemented T seconds electrocardiosignal segments of previous T second electrocardiosignal segment and the latter and is reconstructed according to corresponding time relationship
Parent electrocardio signal correspondence signal segment i.e. obtain complete T seconds parent electrocardio signal.
Then signal enters Fetal ECG signal processing module module T seconds pregnant woman's electrocardiosignal good to signal quality
It is analyzed, rejects parent electrocardio signal from pregnant woman's electrocardiosignal first with R wave fine position technologies, obtain Fetal ECG
Then signal automatically selects principal component analysis according to pregnant woman's electrocardiosignal quality information that electrocardiosignal quality assessment modules provide
Method determines that fetus R crest values point position according to fetus R crest value points position, obtains based on the peak detection of parabola weight
To fetus RR interval series, finally fetal heart frequency information is calculated according to fetus RR interval series.Using R wave fine position technologies from
The method that parent electrocardio signal is rejected in pregnant woman's electrocardiosignal is as follows, the mother first to T seconds pregnant woman electrocardiosignal xx (n) and reconstruct
Body-centered electric signal yy (n) carries out resampling using 10 sampling frequencies of former sample frequency, is denoted as xx* (n) and yy* respectively
(n), the signal segment yyi* (n) of 200 milliseconds of time windows at center is then set in extraction yy* (n) with R crest value points
And the signal segment xxi* (n) of time corresponding with xx* (n) carries out relevant operation, correlation is recorded, then by moving left and right
Yyi* (n) carries out relevant operation with xxi* (n) again and records correlation determining maximum value and note in a series of correlations
The sampling number moved left and right at this time is recorded, and according to this sampling number adjustment yyi* (n) in resampling stepmother body-centered electric signal
Position in yy* (n) obtains the parent electrocardio signal after R wave fine positions finally by down-sampled, believes from pregnant woman's electrocardio
The parent electrocardio signal is subtracted in number realizes the rejecting of parent electrocardio signal.The automatic choosing of difference according to electrocardiosignal quality
The method of Fetal ECG R crest value point recognizers is selected with Principal Component Analysis Method and base involved by parent electrocardio signal processing module
In parabola weight peak detection also with step parent electrocardio signal processing module.
Last signal enters maternal fetus ecg information stick generation module, which generates above-mentioned T seconds set time window
The maternal fetus ecg information stick of length, information stick are generated using picture format, include 4 partial information part 1s from top to bottom
For pregnant woman's essential information, including name, age, pregnant week, height and weight part 2 are that 3 leads after removing baseline drift are pregnant
Woman's ecg wave form, electrocardiosignal quality information are shown in the parent electrocardio waveform and parent R waves that third portion above waveform is reconstruct
Peak point position marks, if non-" quality the is good " type of a certain lead electrocardiosignal, the lead parent electrocardio waveform are shown
If 3 lead electrocardiosignal of straight line is not " quality is good " type, 3 lead parent electrocardio waveforms show the straight line parent heart
Rate score is shown in above waveform, if 3 lead electrocardiosignals are not " quality is good " types, maternal heart rate numerical value vacancy the
4 parts are the Fetal ECG waveform and fetus R crest value points position mark after rejecting parent electrocardio signal in pregnant woman's electrocardiosignal
Note, if non-" quality the is good " type of a certain lead electrocardiosignal, which shows straight line, if 3 lead
It is not " quality is good " type to join electrocardiosignal, then 3 lead Fetal ECG waveforms show that straight line, fetal heart frequency numerical value are shown
Above waveform, if 3 lead electrocardiosignals are not " quality is good " types, fetal heart frequency numerical value vacancy.The parent of generation
Fetal ECG information stick real-time display on a display screen, while sending data storage card to store root according to preset storing frequencies
It is wirelessly transmitted to hospital or monitor center according to preset communication frequency, the storing frequencies and communication frequency of acquiescence are every 20
Minute, user can reset storing frequencies or communication frequency as needed.
Fig. 4,5,6,7,8 give the maternal fetus ecg information stick example of the present invention, and wherein Fig. 4 is 3 lead pregnant woman's hearts
Electric signal " quality is good " example, Fig. 5 is that a wherein lead signals belong to " electrode delamination " example, and Fig. 6 is a wherein lead
Signal belongs to " motion artifacts " example, and Fig. 7 is that a wherein lead signals belong to " very noisy " example, and Fig. 8 is 3 lead pregnant woman's electrocardios
Signal is not admitted to " quality is good " type so that analysis can not carry out example, and the set time window in example takes 10 seconds.
Each maternal fetus ecg information stick shows the pregnant woman of 10 second set time length of window, parent and Fetal ECG with picture format
It is pregnant woman's essential information, including name, age, pregnant week, height and weight that information, which includes 4 partial information part 1s from top to bottom,
Part 2 is that 3 lead pregnant woman's ecg wave form lead, 1,2,3 signal after removing baseline drift is shown from top to bottom, electrocardiosignal
Quality information is shown in the 3 lead parent electrocardio waveforms and parent R crest value points position mark that third portion above waveform is reconstruct
1,2,3 signal of (symbol " ■ ") lead is shown from top to bottom, should if non-" quality the is good " type of a certain lead electrocardiosignal
If lead parent electrocardio waveform shows that 3 lead electrocardiosignal of straight line is not " quality is good " type, 3 lead parent electrocardios
Waveform shows that straight line maternal heart rate numerical value is shown in above waveform, if 3 lead electrocardiosignals are not " quality is good " types,
Then the 4th part of maternal heart rate numerical value vacancy is the 3 lead Fetal ECGs after rejecting parent electrocardio signal in pregnant woman's electrocardiosignal
Waveform and fetus R crest value points position mark 1,2,3 signal of (symbol "●") lead are shown from top to bottom, if a certain lead heart
Non- " quality the is good " type of electric signal, then the lead Fetal ECG waveform show straight line, if 3 lead electrocardiosignals are not
" quality is good " type, then 3 lead Fetal ECG waveforms show that straight line, fetal heart frequency numerical value are shown in wave
Above shape, if 3 lead electrocardiosignals are not " quality is good " types, fetal heart frequency numerical value vacancy.
Claims (6)
1. a kind of wearable dynamic maternal fetus ECG detection device, characterized in that including with lower part：
(1) ECG signal sampling module, including sequentially connected electrocardioelectrode, low-pass filter, high-pass filter, notch filter
Device and signal amplification circuit are connect with Analog-digital Converter module, the electrocardiosignal for acquiring pregnant woman, and complete the filter of signal
Wave, denoising and amplification；
(2) Analog-digital Converter module：It is connect with microprocessor, for realizing the conversion of analog signal to digital signal；
(3) data storage card：It is connect with microprocessor, for storing pregnant woman's electrocardiosignal, while according to preset storage
The maternal fetus ecg information stick that frequency storage generates；
(4) wireless communication module：Connect with microprocessor, by the maternal fetus ecg information stick of generation be wirelessly transmitted to hospital or
Person's monitor center；
(5) microprocessor：It is connect with Analog-digital Converter module, data storage card and wireless communication module, is used for electrocardiosignal
Real-time analysis, data storage controlled with communication, complete electrocardiosignal quality evaluation, parent electrocardio signal processing, Fetal ECG
Signal processing and generation maternal fetus ecg information stick；
Pregnant woman's electrocardiosignal initially enters ECG signal sampling module, subsequently into Analog-digital Converter module, through simulation numeral
Enter the signal storehouse of T seconds set time length of window after conversion, update was primary per T/2 seconds for the storehouse, freshly harvested T/2 seconds letter
Number stacking, earliest collected T/2 second signal are popped, and there are 50% weights for the T seconds electrocardiosignal segment handled twice before and after guarantee
It is folded, to ensure that continuous ecg characteristics information can be obtained, avoid T seconds electrocardiosignal segment head and the tail characteristic informations that from can not detecting
Caused by information omit, to from signal storehouse come out T seconds electrocardiosignal segment carry out electrocardiosignal quality evaluation first, then
It is sequentially completed parent electrocardio signal processing, Fetal ECG signal processing and generates maternal fetus ecg information stick, detailed process is such as
Under；
1. electrocardiosignal quality evaluation：To the T seconds, each lead signals of 3 lead pregnant woman's electrocardiosignal carry out quality evaluation respectively,
Each lead electrocardiosignal quality is divided into one kind in following 4 class：Electrode delamination, motion artifacts, very noisy and quality are good；
2. parent electrocardio signal processing：T seconds pregnant woman's electrocardiosignal good to signal quality is analyzed, and is 1. provided according to step
Pregnant woman's electrocardiosignal quality information automatically select Principal Component Analysis Method or based on parabola weight peak detection determine
Parent R crest value points position obtains parent RR interval series, finally according to the phase between parent RR according to parent R crest value points position
Sequence obtains maternal heart rate information, while constructing parent electrocardio template and reconstructing parent electrocardio waveform；
3. Fetal ECG signal processing：T seconds pregnant woman's electrocardiosignal good to signal quality is analyzed, first from pregnant woman's electrocardio
Parent electrocardio signal is rejected in signal, obtains Fetal ECG signal, the pregnant woman's electrocardiosignal quality then 1. provided according to step
Information automatically selects Principal Component Analysis Method or determines fetus R crest value points position root based on the peak detection of parabola weight
According to fetus R crest value points position, fetus RR interval series are obtained, finally show that fetal heart frequency is believed according to fetus RR interval series
Breath；
4. generating maternal fetus ecg information stick：Generate the maternal fetus ecg information stick of above-mentioned T seconds set time length of window.
2. wearable dynamic maternal fetus ECG detection device according to claim 1, it is characterized in that the step is 1.
In, the method classified to electrocardiosignal quality is as follows：
First detecting whether " current T seconds electrocardiosignal is divided into 5 sections for one section by electrode delamination by the T/5 seconds, if wherein 2 sections or more
There is constant in signal, then is determined as " electrode delamination ", if not " electrode delamination ", then judge whether " motion artifacts ", detection is worked as
The maximum value of preceding T seconds electrocardiosignal, using the 50% of the maximum value as threshold value, will current T second electrocardiosignals by one section of T/5 seconds
5 sections are divided into, if wherein the time point more than the threshold value is not present in 2 sections or more signals, is determined as " motion artifacts ", if also not
It is " motion artifacts ", then judges " very noisy " or " quality is good ", is judged using Sample Entropy algorithm, if the T seconds
The sample entropy of electrocardiosignal is more than 1.5, then is determined as " very noisy ", is otherwise determined as " quality is good ".
3. wearable dynamic maternal fetus ECG detection device according to claim 2, it is characterized in that the T seconds electrocardio
The following T of computational methods of sample of signal entropy：Second electrocardiosignal is resampled first as 500 regular lengths, if after resampling
Signal be data={ x1, x2 ..., xK }, K be data point number, to data carry out time series phase space reconfiguration, take successively
M continuity point forms vector Ym(i)=[xi, xi+1 ..., xi+m-1], i=1~K-m+1 define vector Ym(i) and Ym(j)
The distance between d (Ym(i), Ym(j))=max (| xi+k-xj+k |), given threshold r=0.2 × σ X, wherein σ X are time series
The standard deviation of data does not include element Y for K-m+1 m n dimensional vector ns statisticsm(i) in K-m vector including with element Ym(i)
Distance be less than threshold value r element number, be denoted as template matches number Km(i), and K is calculatedm(i) and apart from total K-mTo allMean value,
It is denoted asThen space dimensionality is increased into m+1, is counted again according to above-mentioned steps
Calculate Bm+ 1 (r), then the Sample Entropy of signal data be：SampEn (m, r, K)=ln ［ Bm(r)/Bm+ 1 (r)], mr and K points of parameter
It is not set as, 20.2 and 500.
4. wearable dynamic maternal fetus ECG detection device according to claim 1, it is characterized in that the step is 2.
The method of middle determining parent R crest values point position is as follows：If at least 2 lead electrocardiosignal matter in 3 lead pregnant woman's electrocardiosignals
Amount is good, then selects Principal Component Analysis Method；If it is good to only have 1 lead electrocardiosignal quality in 3 lead pregnant woman's electrocardiosignals, select
Select the peak detection based on parabola weight；
Principal Component Analysis Method determines that the specific method is as follows for parent R crest value points position：If the second best in quality lead number is M,
Middle M is 2 or 3, if pregnant woman's electrocardiosignal of the M leads is { xm(n) }, wherein m=1,2 ..., M, n=1,2 ...,
N；N is T seconds electrocardiosignal length, generates electrocardiosignal matrix XTFor：
Wherein x (n)=[x1 (n), x2 (n) ..., xM (n)] be M leads electrocardio same sampling instant sampled value form to
Amount；A indicates coefficient matrix Y；It indicates the signal source matrix of electrocardiosignal matrix X and meets orthonormalization condition, signal source square
The first row y1 (n)=[y1 (1), y1 (2) ..., y1 (N)] of battle array Y is the main composition of pregnant woman's electrocardiosignal, because of the parent heart
Electrical signal intensity is more than Fetal ECG signal so the i.e. corresponding parent electrocardio signal matrix A of the main composition and Y shape formula are as follows：
Carrying out Eigenvalues Decomposition to the correlation matrix R of electrocardiosignal matrix X is：
Wherein ∑ is the eigenvalue matrix of matrix R, and U is the eigenvectors matrix of matrix R, because of X=AY, then above formula is written as：
Then there are A=U ∑s1/2, that is, haveWherein λ is the characteristic value of matrix R, and u=[u, u ..., u] is R pairs of matrix
Answer eigenvalue λmFeature vector, the main composition of pregnant woman's electrocardiosignal is then expressed as：m m1 m2
T seconds parent electrocardio signal y1 (n) are divided into 5 sections for one section by the T/5 seconds, the maximum value in every section is chosen and calculates these most
The mean value being worth greatly determines that the region in main composition higher than this threshold value is parent electrocardio signal R using the 60% of mean value as threshold value
Wave region, finally in this region of former pregnant woman's electrocardiosignal maximizing/minimum value as parent R crest value points position；
It determines parent R crest value points position using the peak detection based on parabola weight the specific method is as follows and 3 lead at this time
It is good to join only 1 lead signals quality in electrocardiosignal, if the lead signals are p (n), band logical filter is carried out to the signal first
Wave retains 0.05Hz-40Hz frequency band signals, and filtered signal is denoted as q (n), then generates the linear order that length is at 20 points
K1=[1,0.95,0.9,0.85 ..., 0.05], it is the square root sequence of sequence k1 to enable sequence k2, and it is sequence k2 to enable sequence k3
Overturning sequence, then parabola weight vectors obtained by w=[k3,1, k2], be 41 point sequences, then by the parabola weight
Vectorial w does to filtered electrocardiosignal q (n) related point by point successively, coherent signal r (n) is obtained, finally by T seconds time spans
Coherent signal r (n) be divided into 5 sections by one section of T/5 seconds, choose the maximum value in every section and calculate the mean value of these maximum values,
It determines that the region in r (n) higher than this threshold value is parent electrocardio signal R waves region using the 80% of mean value as threshold value, finally exists
Maximizing/minimum value is as parent R crest value points position in this region of former pregnant woman's electrocardiosignal.
5. wearable dynamic maternal fetus ECG detection device according to claim 1, it is characterized in that the step is 2.
In, construct parent electrocardio template and reconstruct parent electrocardio waveform method it is as follows：
Electrocardiosignal segment between the front and back adjacent R crest value points of extraction is denoted as vi (n), and vi (n) is single cardiac cycle
Electrocardiosignal, wherein i=1,2 ..., L, L are cardiac cycle number, and L vi (n) is stretched to identical signal length, is then calculated
Its coherence average value The parent electrocardio template of as current lead T seconds electrocardiosignals, then reconstructs
Parent electrocardio signal obtains the phase between RR, by electro-cardiologic template signal according to R crest value points positionStretch successively for it is above-mentioned each
The identical signal of phase length between RR, the signal head and the tail connection after then sequentially stretching form the parent electrocardio after reconstructing and believe
Number, at this time in current T seconds electrocardiosignal segment before first R crest value points position and after the last one R crest value point
Parent electrocardio signal vacancy supplements T seconds electrocardio letters of previous T seconds electrocardiosignal segment and the latter according to corresponding time relationship
The correspondence signal segment of the parent electrocardio signal of number segment reconstruct obtains complete T seconds parent electrocardio signal.
6. wearable dynamic maternal fetus ECG detection device according to claim 1, characterized in that the step is 3.
In, the method that parent electrocardio signal is rejected from pregnant woman's electrocardiosignal is：First to T seconds pregnant woman electrocardiosignal xx (n) and reconstruct
Parent electrocardio signal yy (n) carry out resampling using 10 sampling frequencies of former sample frequency, be denoted as xx respectively*(n) and
yy*(n), yy is then extracted*(n) the signal segment yyi of 200 milliseconds of time windows at center is set in R crest value points*
(n) and and xx*(n) the signal segment xxi of corresponding time*(n) relevant operation is carried out, record correlation is then by moving left and right
yyi*(n) again with xxi*(n) relevant operation is carried out and records correlation to determine maximum value and record in a series of correlations
The sampling number moved left and right at this time, and adjust yyi according to this sampling number*(n) in resampling stepmother body-centered electric signal yy*
(n) position in obtains carrying out relevant operation through R waves, correlation is recorded, then by moving left and right finally by down-sampled
yyi*(n) again with xxi*(n) relevant operation is carried out and records correlation to determine maximum value and record in a series of correlations
The sampling number moved left and right at this time, and adjust yyi according to this sampling number*(n) in resampling stepmother body-centered electric signal yy*
(n) position in is obtained finally by down-sampled through R waves.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710723391.3A CN108420425A (en) | 2017-08-22 | 2017-08-22 | Wearable dynamic maternal fetus electrocardioscanner |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710723391.3A CN108420425A (en) | 2017-08-22 | 2017-08-22 | Wearable dynamic maternal fetus electrocardioscanner |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108420425A true CN108420425A (en) | 2018-08-21 |
Family
ID=63155643
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710723391.3A Pending CN108420425A (en) | 2017-08-22 | 2017-08-22 | Wearable dynamic maternal fetus electrocardioscanner |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108420425A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109214318A (en) * | 2018-08-22 | 2019-01-15 | 北京天泽智云科技有限公司 | A method of finding the faint spike of unstable state time series |
| CN109662709A (en) * | 2018-12-27 | 2019-04-23 | 索思（苏州）医疗科技有限公司 | A kind of Fetal ECG monitoring system |
| CN109893121A (en) * | 2019-03-26 | 2019-06-18 | 深圳理邦智慧健康发展有限公司 | Acquisition method, device, terminal and the computer readable storage medium of electrocardiosignal |
| CN112120688A (en) * | 2019-06-25 | 2020-12-25 | 深圳市理邦精密仪器股份有限公司 | Electrocardiosignal processing method, electrocardiosignal processing equipment and computer-readable storage medium |
| CN112741632A (en) * | 2019-10-31 | 2021-05-04 | 深圳市理邦精密仪器股份有限公司 | Electrocardiosignal detection method, system, equipment and computer storage medium |
| CN113520356A (en) * | 2021-07-07 | 2021-10-22 | 浙江大学 | Heart disease early diagnosis system based on Korotkoff sounds |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103720469A (en) * | 2014-01-02 | 2014-04-16 | 山东大学 | Wearable type dynamitic maternal fetus electrocardio-detecting device |
| CN106667478A (en) * | 2016-12-07 | 2017-05-17 | 成都亿咖极科技有限公司 | Multi-lead joint detection intelligent fetal electrocardiogram detection method and system |
-
2017
- 2017-08-22 CN CN201710723391.3A patent/CN108420425A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103720469A (en) * | 2014-01-02 | 2014-04-16 | 山东大学 | Wearable type dynamitic maternal fetus electrocardio-detecting device |
| CN106667478A (en) * | 2016-12-07 | 2017-05-17 | 成都亿咖极科技有限公司 | Multi-lead joint detection intelligent fetal electrocardiogram detection method and system |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109214318A (en) * | 2018-08-22 | 2019-01-15 | 北京天泽智云科技有限公司 | A method of finding the faint spike of unstable state time series |
| CN109214318B (en) * | 2018-08-22 | 2021-10-22 | 北京天泽智云科技有限公司 | Method for searching weak peak of unsteady time sequence |
| CN109662709A (en) * | 2018-12-27 | 2019-04-23 | 索思（苏州）医疗科技有限公司 | A kind of Fetal ECG monitoring system |
| CN109893121A (en) * | 2019-03-26 | 2019-06-18 | 深圳理邦智慧健康发展有限公司 | Acquisition method, device, terminal and the computer readable storage medium of electrocardiosignal |
| CN109893121B (en) * | 2019-03-26 | 2021-11-05 | 深圳理邦智慧健康发展有限公司 | Electrocardiosignal acquisition method, electrocardiosignal acquisition device, electrocardiosignal acquisition terminal and computer-readable storage medium |
| CN112120688A (en) * | 2019-06-25 | 2020-12-25 | 深圳市理邦精密仪器股份有限公司 | Electrocardiosignal processing method, electrocardiosignal processing equipment and computer-readable storage medium |
| CN112741632A (en) * | 2019-10-31 | 2021-05-04 | 深圳市理邦精密仪器股份有限公司 | Electrocardiosignal detection method, system, equipment and computer storage medium |
| CN112741632B (en) * | 2019-10-31 | 2024-03-12 | 深圳市理邦精密仪器股份有限公司 | Electrocardiosignal detection method, system, equipment and computer storage medium |
| CN113520356A (en) * | 2021-07-07 | 2021-10-22 | 浙江大学 | Heart disease early diagnosis system based on Korotkoff sounds |
| CN113520356B (en) * | 2021-07-07 | 2024-04-30 | 浙江大学 | Early diagnosis system for heart diseases based on Korotkoff sounds |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103720469B (en) | Wearable type dynamitic maternal fetus electrocardio-detecting device | |
| CN108420425A (en) | Wearable dynamic maternal fetus electrocardioscanner | |
| CN109157202B (en) | Cardiovascular disease early warning system based on multi-physiological signal deep fusion | |
| US11344243B2 (en) | Artificial intelligence self-learning-based static electrocardiography analysis method and apparatus | |
| CN106214145B (en) | Electrocardiogram classification method based on deep learning algorithm | |
| WO2019161608A1 (en) | Multi-parameter monitoring data analysis method and multi-parameter monitoring system | |
| JPH1170089A (en) | Cardiac waveform characteristics displaying | |
| CN109199375A (en) | A kind of noninvasive Fetal ECG detection device and ecg signal data processing method | |
| JP2010538802A (en) | Method and apparatus for measuring susceptibility to arrhythmias using Wedensky modulated electrocardiography and alternans analysis | |
| WO2017059569A1 (en) | Pacing signal processing method, system and electrocardiogram monitor | |
| CN110897631B (en) | Real-time pregnancy monitoring device and method | |
| CN107411736A (en) | Fetal ECG signal detection system | |
| CN109288515B (en) | Periodicity monitoring method and device based on premature beat signal in wearable electrocardiosignal | |
| Fanelli et al. | Telefetalcare: A first prototype of a wearable fetal electrocardiograph | |
| JP7277970B2 (en) | Electrocardiogram measurement method and system using wearable device | |
| Marchon et al. | ECG electrode configuration to extract real time FECG signals | |
| Rodríguez-Jorge et al. | Internet of things-assisted architecture for QRS complex detection in real time | |
| Hsu et al. | Heart rate and respiratory rate monitoring using seismocardiography | |
| US20210298683A1 (en) | Bed-based ballistocardiogram apparatus and method | |
| WO2021184818A1 (en) | Particle filter-based device and method for measuring heart rate of human body beat by beat | |
| CN109431493A (en) | Wearable body surface potential acquisition device and method based on range segment separating weighting algorithm | |
| CN106108885A (en) | Assessment physical motion state and electrocardio disturb method and the instrument thereof of size | |
| CN105455800A (en) | Wearable device for monitoring adult heart rate and fetal heart rate simultaneously and method | |
| CN113288156B (en) | Method for generating electrocardiographic data of any lead visual angle | |
| Fanelli | TeleFetal Care: development of a wearable system for fetal monitoring during pregnancy |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180821 |
|
| WD01 | Invention patent application deemed withdrawn after publication |