CN110236594A - A kind of Fetal Movement Signal identification labeling method, system, medium, equipment - Google Patents
A kind of Fetal Movement Signal identification labeling method, system, medium, equipment Download PDFInfo
- Publication number
- CN110236594A CN110236594A CN201910460273.7A CN201910460273A CN110236594A CN 110236594 A CN110236594 A CN 110236594A CN 201910460273 A CN201910460273 A CN 201910460273A CN 110236594 A CN110236594 A CN 110236594A
- Authority
- CN
- China
- Prior art keywords
- movement
- foetus
- heart rate
- fetal heart
- value
- 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.)
- Granted
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/02—Measuring pulse or heart rate
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/08—Detecting organic movements or changes, e.g. tumours, cysts, swellings
- A61B8/0866—Detecting organic movements or changes, e.g. tumours, cysts, swellings involving foetal diagnosis; pre-natal or peri-natal diagnosis of the baby
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/48—Diagnostic techniques
- A61B8/488—Diagnostic techniques involving Doppler signals
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Veterinary Medicine (AREA)
- Biophysics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Pathology (AREA)
- Radiology & Medical Imaging (AREA)
- Engineering & Computer Science (AREA)
- Public Health (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Physics & Mathematics (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Gynecology & Obstetrics (AREA)
- Pregnancy & Childbirth (AREA)
- Cardiology (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
- Ultra Sonic Daignosis Equipment (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention discloses a kind of Fetal Movement Signals to identify labeling method, system, medium, equipment, the method includes the steps: obtain Fetal Heart Rate acceleration curve, uterine contraction curve and F/A figure;By carrying out baseline fetal heart rate calculating and Fetal Heart Rate acceleration identification to Fetal Heart Rate acceleration curve, obtains Fetal Heart Rate and accelerate information;It is smoothed by the signal to uterine contraction curve and seeks difference signal and search for movement of the foetus information, obtain the first F/A information;By extracting movement of the foetus information to the calculating of F/A figure curve baseline and dynamic threshold, the second fetus action message is obtained;Fetal Heart Rate is calculated to accelerate the total score of index score, uterine contraction movement of the foetus information index score value and F/A figure movement of the foetus index score and mark movement of the foetus event.The present invention integrates the weight size comprehensive descision of Fetal Heart Rate acceleration, uterine contraction movement of the foetus information and F/A figure movement of the foetus three kinds of signal sources of information and realizes that Fetal Movement Signal marks, so that the judgement source of Fetal Movement Signal is more comprehensive, increases the accuracy of movement of the foetus label.
Description
Technical field
The present invention relates to information technology field, in particular to a kind of movement of the foetus that can be used in electronic fetal monitoring software product
Signal identification labeling method, system, medium, equipment.
Background technique
Currently based on the movement of the foetus recognizer of Doppler ultrasound signal F/A figure (Actogram) it is more commonly used be
Threshold decision (including static threshold and dynamic threshold) method, and usually only with reference to the parameter that several are calculated, according to
Treaty rule carries out identification judgement.For example, Lu in 2014 et al. is in publication Bio-Medical Materials and
" the Fetal heart rate baseline estimation with analysis of delivered on Engineering
Fetal movement signal ", describe it is a kind of using Da-Jin algorithm detection F/A method, but this method use
It is static threshold detection method, the dynamic change of movement of the foetus active signal cannot be well adapted for, there is certain limitation;2014
What year Janusz WROBEL et al. delivered in publication Journal of Medical Informatics&Technologies
“Automated detection of fetal movements in Doppler ultrasound signals versus
Maternal perception " describes a kind of tire activity automatic detection algorithm, algorithm estimation point when each record starts
Class threshold value, and adaptively modified in recording process, although being based only on list the process employs dynamic threshold algorithm
One F/A figure signal carries out identification judgement, and the actual demand of hospital for obstetrics is not accounted in terms of real-time is shown.
In fact, movement of the foetus identification labelling technique needs to comprehensively consider together there are also some important parameters, such as fetal rhythm uterine contraction
The problems such as relevance between some specific forms of curve, the details and Fetal Heart Rate and movement of the foetus of F/A figure;Simultaneously in reality
When property shows and should also be marked with reference to above-mentioned many kinds of parameters.There are no the products for meeting the demand to go out currently on the market
Existing, this is exactly that the art of this patent will solve the problems, such as.
Summary of the invention
The purpose of the present invention is to overcome the shortcomings of the existing technology and deficiency, provides a kind of Fetal Movement Signal identification label side
Method, system, medium, equipment, the method and system, which can be realized, comprehensively considers association between all kinds of Fetal Movement Signals, a variety of ginsengs of tradeoff
The influence of several pairs of label results, improves the reliability of label result.
The purpose of the present invention is realized by the following technical solution:
A kind of Fetal Movement Signal identification labeling method, comprising steps of
Obtain Fetal Heart Rate acceleration curve, uterine contraction curve and F/A figure;
By carrying out baseline fetal heart rate calculating and Fetal Heart Rate acceleration identification to Fetal Heart Rate acceleration curve, obtains Fetal Heart Rate and accelerate
Information;
It is smoothed by the signal to uterine contraction curve and seeks difference signal and search for movement of the foetus information, obtain the first fetus
Action message;
Dynamic threshold movement of the foetus information is calculated and extracted by carrying out F/A figure curve baseline, obtains the second F/A
Information;
More criterions identify movement of the foetus, i.e., accelerate baseline various amplitude to assign different Fetal Heart Rates Fetal Heart Rate and accelerate index point
Value;Set F/A information peak value baseline value threshold range, according to where the peak value baseline value of F/A information not
Same threshold range assigns corresponding uterine contraction movement of the foetus information index score value, F/A figure movement of the foetus index score;Fetal Heart Rate is calculated to add
The total score of fast index score, uterine contraction movement of the foetus information index score value and F/A figure movement of the foetus index score;
When total score is more than or equal to threshold value T, movement of the foetus event of label.
Preferably, the reproduction replica for obtaining baseline fetal heart rate FHRCut and fetal heart rate curve is calculated by baseline fetal heart rate
After FHRTemp, in anodinia load test, the amplitude threshold higher than baseline amplitude bpm of definition is accelerated to define by meeting
For accthreshold, time duration threshold definitions are acckeeptime, and extract Fetal Heart Rate in accordance with the following steps and accelerate
Information:
Step S201: searching for backward from FHRCut starting point, writes down mark and is set to index value i, works as FHRCuti>
FHRTempi+ accthreshold searches for maximum value within the scope of 60 Fetal Heart Rates after the i of position, records the numerical value of maximum value
FHRPeak and position FHRPeakTime then executes step S202;Otherwise, continue to search for backward;
Step S202: it from the position FHRPeakTime 55 Fetal Heart Rate point search forward, writes down mark and is set to index value j;
If there is FHRCutj< FHRBaselinej+ 2, and the Fetal Heart Rate value in [j, FHRPeakTime] range is equal
No more than FHRPeak, position j is labeled as accstart;
Then it from the position FHRPeakTime 55 Fetal Heart Rate point search backward, writes down mark and is set to index value k;
If there is FHRCutk< FHRBaselinek+ 2, and the Fetal Heart Rate value in [FHRPeakTime, k] range is equal
No more than FHRPeak, position k is labeled as accend;
If accstart and accend are labeled, and the time interval between accstart and accend is more than
The range of fetal heart rate curve accstart to accend is judged as that Fetal Heart Rate of generation accelerates by acckeeptime;Then from
S201 is re-execute the steps after accelerating one minute of position FHRPeakTime.
Further, accelerate Fetal Heart Rate higher than the standard of baseline amplitude and duration requirement different pregnancy phases pregnant woman
It distinguishes:
If pregnant woman expires 32 pregnant weeks, accthreshold=15, acckeeptime=18;
If less than 32 pregnant week, accthreshold=10, acckeeptime=12.
Further, because Fetal Heart Rate accelerate identifying purpose be in order to assist in identifying movement of the foetus, will be with above-mentioned extraction tire
Heart rate accelerates the recognition methods of information to carry out the identification of the small acceleration of fetal rhythm, and difference will be added in the case where being unsatisfactory for trueing add speed
Speed judges parameter change for accthreshold=10, acckeeptime=10, meets Rule of judgment and is judged as small acceleration.
Preferably, it is described from be smoothed by the signal to uterine contraction curve ask difference signal, search movement of the foetus information,
The step of obtaining the first F/A information are as follows:
Step S501, original uterine contraction curve is smoothed using 5 moving average methods, obtains smooth signal
UCSignal′;
Step S502, original uterine contraction signal UCSignal and UCSignal ' obtained in step S501 is subtracted each other, is obtained
To new Fetal Movement Signal array UCDiff;Index value i is initialized as 0;
Step S503, to uterine contraction Fetal Movement Signal UCDiffiValue scan for, if UCDiffi> 5, and do not have at this time
The movement of the foetus point being marked executes step S504;Otherwise, if there is the movement of the foetus marked, step S505 is executed;
Step S504, uterine contraction movement of the foetus accumulated time number fmpeaknums=0 is defined, i is risen labeled as new uterine contraction movement of the foetus
This movement of the foetus end point fmend is initialized as i, fmpeaknums=fmpeaknums+1, i=i+1 by initial point fmstart;
If movement of the foetus end point fmend step S505, marked and position i time difference, will within 5 seconds
The value of fmend is updated to i, fmpeaknums=fmpeaknums+1, i=i+1;Otherwise, terminate the mark of the movement of the foetus marked
Note state takes maximum value of the UCDiff in range [fmstart, fmend] as movement of the foetus range value fmpeak;By tuple
Movement of the foetus is deposited into uterine contraction movement of the foetus array by (fmstart, fmend, fmpeak, fmpeaknums), executes step S506;
Step S506, step S503 is continued to execute, is terminated until searching signal;If also when arriving signal end point
There is the movement of the foetus point in flag state, terminates this movement of the foetus flag state and be deposited into uterine contraction movement of the foetus array, uterine contraction movement of the foetus array
As the first F/A information.
Preferably, it is described by F/A figure curve baseline calculate the step of specifically:
If F/A signal is ActoSignal, F/A signal base line is ActoBaseline;
The calculation of ActoBaseline is as follows:
Firstly, every one minute calculating ActoSignal, by signal in one minute from low to high preceding 15% average value make
For the initial reference value ActoBase of the period baseline value;ActoBaseline is calculated by following trimming formula:
To filtering and backward filtering processing before carrying out to F/A signal base line, simplified place has been carried out to formula here
Reason;Forward direction Filtering Formula are as follows:
ActoBaselinei=0.8 × ActoBaselinei-1+0.2×ActoBaselinei
Backward Filtering Formula are as follows:
ActoBaselinei=0.8 × ActoBaselinei+1+0.2×ActoBaselinei;
Accelerate the baseline fetal heart rate calculating of information different from Fetal Heart Rate is obtained, fetal movements baseline need to only reflect F/A
Trend of the signal compared with lower part.Therefore, front and back is only done to F/A signal base line here to be filtered twice;
After obtaining ActoBaseline baseline, sequence of differences ActoDiff is defined:
ActoDiffi=ActoSignali-ActoBaselinei
Therefore, new Fetal Movement Signal ActoDiffSignal is calculated as follows:
Further, the dynamic threshold extracts F/A figure movement of the foetus information, obtains the second fetus action message
Step are as follows:
After improving the recognition effect of movement of the foetus using F/A signal base line, according to F/A in clinical data
The actual margin situation of signal devises a new dynamic threshold discriminant function and is used to extract initial movement of the foetus action message,
For carrying out more criterion identifications and movement of the foetus label later;Dynamic threshold discriminant function is mainly according to the range value of baseline to threshold value
Carry out the adjustment of size;Specific dynamic threshold discriminant function formula is as follows:
The second fetus action message is extracted as follows after completing adjusting thresholds:
Step S701, i=0 is initialized;
Step S702, it is searched for backward since the i of the position ActoDiffSignal, if there is ActoDiffSignali>
Threshold (i), and it is not on the movement of the foetus point of flag state at this time, execute step 3;Otherwise, if there is in label shape
The movement of the foetus of state executes step S704;
Step S703, i is labeled as new movement of the foetus starting point fmstart, and movement of the foetus end point fmend is initialized as i;
I=i+1;Execute step S705;
If the position fmend of step S704, marked movement of the foetus was differed with i within 5 seconds, the value of fmend is updated to
I, i=i+1;Otherwise, the label for terminating existing movement of the foetus, take ActoDiffSignal in [fmstart, fmend] range most
It is worth the peak value fmpeak marked as movement of the foetus greatly, and is fmtime by the position mark of maximum value, the baseline value of the position is set to
This movement of the foetus is stored in movement of the foetus array by fmbase by tuple (fmtime, fmpeak, fmbase, fmstart, fmend);It holds
Row step S705;
Step S705, step S702 is continued to execute, until searching the end ActoDiffSignal;If reaching end also
There is the movement of the foetus marked that flag state is not finished, that is, terminates this movement of the foetus and mark and be stored in movement of the foetus array, movement of the foetus array
As the second fetus action message.
Preferably, described most according to identification movement of the foetus, it assigns Fetal Heart Rate and accelerates index score acc_fm_score, uterine contraction movement of the foetus
The step of information index score value uc_fm_score, F/A figure movement of the foetus index score fmp_fm_score specifically:
Fetal Heart Rate is accelerated, is accelerated if it is big, acc_fm_score=1;If it is small acceleration, acc_fm_score
=0.5;
For uterine contraction movement of the foetus, if meeting fmpeak > 7 in uterine contraction movement of the foetus and fmpeaknums >=3 and fmend-
Fmstart > 5, uc_fm_score=0.5;Meet in uterine contraction movement of the foetus: peaknums >=3, uc_fm_score=0.3;Uterine contraction
Meet in movement of the foetus: fmpeak > 8 and peaknums < 3, uc_fm_score=0.2;
For F/A figure movement of the foetus, if F/A figure movement of the foetus meets: fmpeak >=1.2fmbase and
Fmend-fmstart > 6 and fmend-fmstart < 150, fmp_fm_score=0.8;
If F/A figure movement of the foetus meets: fmbase < fmpeak < 1.2fmbase and fmend-fmstart > 3,
Fmp_fm_score=0.3;If F/A figure movement of the foetus meets: fmpeak >=1.2fmbase and fmend-fmstart≤3,
Fmp_fm_score=0.3;
Under the assignment of above-mentioned index score, the threshold value T of total score is 1.
Further, the big of Fetal Heart Rate acceleration accelerates the normal acceleration that is positive, and is greater than for the raising of baseline fetal heart rate amplitude
Equal to 15bpm, the duration is more than or equal to 15s;It is small to accelerate to be that baseline fetal heart rate amplitude increases less than 15bpm but is more than or equal to
10bpm, duration are less than 15S but are more than or equal to 10s.
Preferably, the step of more criterion identification movement of the foetus specifically:
Step S1001, i=0 is initialized;
Step S1002, it searches for fetal monitoring signal backward since the i of position, accelerates in conjunction with Fetal Heart Rate, uterine contraction movement of the foetus information
Judged with F/A figure movement of the foetus information;Define movement of the foetus index fmscore be used to measure there is movement of the foetus in current location can
Letter degree, fmscore are initialized as 0;Accelerate for Fetal Heart Rate, uterine contraction movement of the foetus information and F/A figure movement of the foetus information define
Movement of the foetus index score acc_fm_score, uc_fm_score and fmp_fm_score,;
Step S1003, whether searching position i is near the range of any Fetal Heart Rate accelerated events (including small acceleration), i.e.,
In [accstart, accend] ± 6 range of the accelerated events, if it does, executing step S1004;If it does not, executing
Step S1005;
Step S1004, it enables:
Fmscore=fmscore+acc_fm_score
Then search range [accstart, accend] ± 6 whether there is uterine contraction movement of the foetus and F/A figure movement of the foetus event,
If it does, updating fmscore:
Fmscore=fmscore+uc_fm_score+fmp_fm_score
If fmscore > 1.0 is determined as movement of the foetus event, by movement of the foetus position mark fmpoint=accstart;I updates
Subsequent first position of maximum value between accelerated events, uterine contraction movement of the foetus and F/A figure movement of the foetus three's end point;Continue
Step 3 is executed, until searching fetal monitoring signal end;
Step S1005, whether searching position i is attached in any uterine contraction movement of the foetus event and F/A figure movement of the foetus event range
Closely, i.e., in [fmstart, the fmend] of movement of the foetus event ± 6 range, if it does, being updated by formula in step S1004
fmscore;If fmscore > 1.0 is determined as movement of the foetus event, the time is taken in uterine contraction movement of the foetus and F/A figure movement of the foetus earlier above
Fmstart, by movement of the foetus position mark fmpoint=fmstart;I is updated to both uterine contraction movement of the foetus and F/A figure movement of the foetus
The latter point of maximum value between end point;Step S1003 is continued to execute, until searching monitored signal end.
Preferably, the movement of the foetus recognized is marked by following strategy:
Step S1101, every five seconds carries out movement of the foetus identification calculating;
Step S1102, after recognizing movement of the foetus position, if current time is within the scope of movement of the foetus, first choice is current
Moment identifies movement of the foetus, otherwise reacquires Fetal Heart Rate acceleration curve, uterine contraction curve and F/A figure.
A kind of Fetal Movement Signal identification tagging system, comprising:
Monitoring modular, for obtaining Fetal Heart Rate acceleration curve, uterine contraction curve and F/A figure;
Fetal Heart Rate accelerates information calculating and identification module, by by carrying out based on baseline fetal heart rate to Fetal Heart Rate acceleration curve
It calculates and Fetal Heart Rate accelerates identification, obtain Fetal Heart Rate and accelerate information;
First F/A information calculates and search module, asks for being smoothed by the signal to uterine contraction curve
Difference signal simultaneously searches for movement of the foetus information, obtains the first F/A information;
Second fetus activity information meter is calculated and extraction module, is used to calculate by progress F/A figure curve baseline and dynamic
State threshold value extracts movement of the foetus information, obtains the second action message of fetus;
More criterion identification modules identify movement of the foetus for more criterions, i.e., accelerate baseline various amplitude to assign Fetal Heart Rate different
Fetal Heart Rate accelerate index score;The threshold range for setting F/A information peak value baseline value, according to F/A information peak
Peak value baseline value where different threshold ranges, assign corresponding uterine contraction movement of the foetus information index score value, F/A figure movement of the foetus
Index score;It calculates Fetal Heart Rate and accelerates index score, uterine contraction movement of the foetus information index score value and F/A figure movement of the foetus index score
Total score;
Movement of the foetus mark module is used for when total score is more than or equal to threshold value T, movement of the foetus event of label.
Preferably, the monitoring modular includes fetal rhythm detector, uterine contraction pressure detector, and the fetal rhythm detector is for obtaining
Take Fetal Heart Rate acceleration curve and F/A figure;The uterine contraction pressure detector is for detecting uterine contraction signal.
Further, the fetal rhythm detector is two wireless blue tooth ultrasound fetal rhythm probes, to detect twins.
A kind of storage medium is stored with program, when described program is executed by processor, in conjunction with the number of monitoring modular detection
According to realizing above-mentioned Fetal Movement Signal identification labeling method.
A kind of calculating equipment, including processor and for the memory of storage processor executable program, the processing
When device executes the program of memory storage, in conjunction with the data that monitoring modular detects, above-mentioned Fetal Movement Signal recognition methods is realized.
Compared with the prior art, the invention has the following advantages and beneficial effects:
1, the present invention combines Fetal Heart Rate acceleration, uterine contraction movement of the foetus information and F/A figure movement of the foetus information, believes in conjunction with three kinds
The weight size comprehensive descision in number source simultaneously realizes that Fetal Movement Signal marks, so that the judgement source of Fetal Movement Signal is more comprehensive, increases
The accuracy and real-time of movement of the foetus label.
2, present invention uses histogram analysis and the method for filtering to carry out baseline calculating to the Fetal Movement Signal that probe obtains,
And Fetal Movement Signal is extracted using dynamic threshold, can effective filter out interference signal, reduce signal fluctuation it is larger brought by accidentally
Sentence, detects more accurate Fetal Movement Signal.
Detailed description of the invention
Fig. 1 is that the present invention is based on the flow charts of the Fetal Movement Signal of more criterions identification labeling method.
Fig. 2 is the flow chart that the present invention obtains Accelerated heart rate information from fetal heart rate curve.
Fig. 3 is the flow chart that the present invention obtains the first F/A information from uterine contraction curve.
Fig. 4 is the flow chart that the present invention obtains the second fetus action message from F/A figure.
Fig. 5 is Fetal Movement Signal identification tagging system schematic diagram of the present invention.
Specific embodiment
For a better understanding of the technical solution of the present invention, the implementation that the present invention is described in detail provides with reference to the accompanying drawing
Example, embodiments of the present invention are not limited thereto.
Embodiment 1
As shown in Fig. 2, present embodiment discloses a kind of Fetal Movement Signals to identify labeling method, comprising steps of
S1, Fetal Heart Rate acceleration curve, uterine contraction curve and F/A figure are obtained;
It pops one's head in especially by ultrasonic fetal rhythm and obtains Fetal Heart Rate acceleration curve and F/A figure;It is obtained by uterine contraction pressure probe
Take uterine contraction curve.
S2, accelerate identification by carrying out baseline fetal heart rate calculating and Fetal Heart Rate to Fetal Heart Rate acceleration curve, obtain Fetal Heart Rate
Accelerate information;
S2.1, baseline fetal heart rate calculate;
Andersson S academic dissertation Acceleration and deceleration has been continued to use in baseline fetal heart rate calculating
Method in and baseline estimation.It is main that calculate that steps are as follows described:
S2.1.1: fetal heart rate curve is broken and partially (zero parts) is removed, the average value of fetal heart rate curve is then found out
FHRmean;
S2.1.2: the invalid Fetal Heart Rate value of removal fetal heart rate curve stem and tail portion.Since fetal heart rate curve starting point
Search, if Fetal Heart Rate value is than FHR obtained in step 2.1.1meanSmall 60bpm or more, continues searching, until searching first
A Fetal Heart Rate value ratio FHRmeanWithin small 60bpm, these parts (invalid Fetal Heart Rate part) of the front are then removed;Then
It is searched for forward from fetal heart rate curve distal point, if Fetal Heart Rate value is than FHR obtained in step S2.1.1meanSmall 60bpm or more,
It continues searching, until searching first Fetal Heart Rate value ratio FHRmeanWithin small 60bpm, the invalid tire of the back is then removed
Heart rate part.New fetal heart rate curve FHRCut is finally obtained after truncation;
S2.1.3: using the breakpoint in the method removal FHRCut of linear interpolation.From FHRCutiStarting point starts to search backward
Rope, as Abs (FHRCuti+1-FHRCuti) > 30 (Abs be take absolute value function) when by coordinate i labeled as istart, from istartIt opens
Begin to search for 400 sample points backward, writes down and be designated as j, as Abs (FHRCutj+1-FHRCutj) < 10 and Abs (FHRCutj+1-
FHRmean) < 50 when, write down coordinate j be iend.By FHRCut in istartTo iendBetween segment use linear interpolation method
It is replaced, from position i after replacementendFirst point afterwards starts to continue to search for by step S2.1.3 to carry out linear interpolation, until
All breakpoints of FHRCut are interpolated replacement;
S2.1.4: the FHRCut signal (as unit of bpm value) that step S2.1.3 is obtained is switched to as unit of millisecond
Gradually heartbeat interval value array (inter beat interval, IBI) FHRPS, transform mode are as follows:
After the completion of conversion, element of the numerical value except 300ms to 600ms range in FHRPS is removed.Then straight by frequency
Square figure finds out the frequency of each IBI in FHRPS, and taking the highest IBI value of frequency is PSmax, all IBI value frequencies are taken and are
PSSum.Then frequency array is added up according to the order of IBI value from low to high, when accumulated value is greater than 0.875 × PSSum, if
IBI value is PST at this timestart.From PSTstartStart, is searched for from high to low in frequency array by IBI value, until what is searched
IBI value FHRPSjMeet the following conditions:
1) at least 12.5% IBI value is greater than current PSmax;
2) this IBI value frequency ratio 5 continuous IBI value frequencies to be continued searching are all big;
3) this IBI value frequency is greater than 0.5% × PSSum or Abs (PSmax-PSTstart)≤30ms。
If there is such IBI value value, by PSmaxCorrect IBI value value thus.Then by PSmaxBeing converted to unit is bpm
Fetal Heart Rate
S2.1.5: the FHRCut for taking step S2.1.3 to obtain copies as FHRTemp.Data in FHRTemp are carried out 5 times
Filtering and 4 pruning modes, filtering and trimming are alternately.A basic Fetal Heart Rate value FHRBase is initialized before filtering,
FHRBase initial value is PSMaxFHR obtained in step S2.1.4.Calculating is re-started to FHRBase before each filtering.
FHRBase is once updated first, update method are as follows:
It is searched for forward since the end FHRTemp, is designated as i under mark position, as Abs (FHRTempi- FHRBase) < 50
When, it enables:
FHRBase=0.975 × FHRBase+0.025 × FHRTempi
Then it is filtered operation.Filtering operation is filtered to filtering with backward before being divided into.The specific behaviour of forward direction filtering
Work is: searching for backward from FHRTemp starting point, is designated as m under mark position, if Abs (FHRTempm- FHRBase) < 50, is pressed
Following formula updates:
Otherwise, it is updated as follows:
The concrete operations filtered backward are: traversing from FHRTemp from tip forward, n is designated as under mark position, using as follows
Formula updates:
FHRTempn=0.975 × FHRTempn+1+0.025×FHRBase
Cut operation is divided into upper limit trimming and trims with lower limit.Concrete operations are as follows:
1) new sequence NewFHRTemp is defined, trimming upper limit value v is re-definedhWith trimming lower limit value vl.Since it is desired that carrying out
4 trimmings, each vhValue be followed successively by 20,15,10 and 5, vlValue be followed successively by 20,15,10 and 10;
2) traversal compares FHRTemp and FHRCut from the beginning, writes down and is designated as i.If there is FHRCuti> FHRTempi+vh
And FHRCuti+1< FHRTempi+1, i is labeled as pend.It is searched for forward since the position pend, writes down and be designated as k.If there is
FHRCutk< FHRTempk, k is labeled as pstart.Then NewFHRTemp is updated as follows:
Otherwise, if there is FHRCuti< FHRTempi-vlWith and FHRCuti+1> FHRTempi+1, i is labeled as pend.
It is searched for forward since the position pend, writes down and be designated as k.If there is FHRCutk> FHRTempk, k is labeled as pstart.Then
NewFHRTemp is updated by above-mentioned formula;
3) continue to traverse FHRTemp backward by step 2) and FHRCut is trimmed, until updating terminates.
By filtering and trimming, finally obtained NewFHRTemp is replaced FHRTemp.
S2.1.6: passing through above step, and fetal heart rate curve originally here will because break-in operation length is changed
New fetal heart rate curve is aligned with former fetal heart rate curve.FHRCut is copied as into FHRCutCopy, FHRTemp is copied as
FHRTempCopy.If the truncation starting point in step S2.1.2 of FHRCut is cutstart, truncation end point is
cutend.It enables:
Finally, FHRTemp is considered as baseline fetal heart rate, and it is defined as FHRBaseline.
S2.2, Fetal Heart Rate accelerate identification;
Fetal Heart Rate acceleration search is mainly continued to use Cao Meng firm on the basis of accelerating and defining and is believed based on the fetal rhythm uterine contraction of cloud service
Design and realization [D] the Ji'nan University of analysis system, 2016. searching method are ceased, while increasing and small accelerated events is searched
Rope.Here it will meet and the amplitude threshold higher than baseline amplitude bpm of definition is accelerated to be defined as in anodinia load test
Accthreshold, time duration threshold definitions are acckeeptime.Because the judgement accelerated relies on baseline fetal heart rate
It calculates.After obtaining baseline fetal heart rate by baseline fetal heart rate calculating, Fetal Heart Rate is extracted in accordance with the following steps and accelerates information.
S2.2.1: searching for backward from FHRCut starting point, writes down mark and is set to i, works as FHRCuti> FHRTempi+
Accthreshold searches for maximum value within the scope of 60 Fetal Heart Rates after the i of position, record maximum value numerical value FHRPeak and
Position FHRPeakTime then executes step S2.2.2.Otherwise, continue to search for backward;
S2.2.2: it from the position FHRPeakTime 55 Fetal Heart Rate point search forward, writes down mark and is set to j.If there is
FHRCutj< FHRBaselinej+ 2, and the Fetal Heart Rate value in [j, FHRPeakTime] range is no more than FHRPeak, by j
Labeled as accstart.Then it from the position FHRPeakTime 55 Fetal Heart Rate point search backward, writes down mark and is set to k.If deposited
In FHRCutk< FHRBaselinek+ 2, and the Fetal Heart Rate value in [FHRPeakTime, k] range is no more than FHRPeak,
K is labeled as accend.If accstart and accend are labeled, and the time interval between accstart and accend
More than acckeeptime, the range of fetal heart rate curve accstart to accend is judged as that Fetal Heart Rate of generation accelerates.It connects
Re-execute the steps S2.2.1 after accelerating one minute of position FHRPeakTime.
Here according to the definition that Fetal Heart Rate accelerates, the acceleration Fetal Heart Rate of different pregnancy phases pregnant woman be higher than baseline amplitude and it is lasting when
Between required standard it is different.If pregnant woman expires 32 pregnant weeks, accthreshold=15, acckeeptime=18;If not
Full 32 pregnant weeks, accthreshold=10, acckeeptime=12.In addition, Fetal Heart Rate here accelerate identifying purpose be in order to
Assist in identifying movement of the foetus, therefore, the recognition methods of information will be accelerated to carry out the small acceleration of fetal rhythm with above-mentioned extraction Fetal Heart Rate here
Identification, difference be in the case where being unsatisfactory for trueing add speed, will accelerate judge parameter change for accthreshold=10,
Acckeeptime=10 meets Rule of judgment and is judged as small acceleration.
S3, it is smoothed by the signal to uterine contraction curve and asks difference signal, search movement of the foetus information, obtain the first tire
Youngster's action message;
S3.1, original uterine contraction curve is smoothed using 5 moving average methods, finally obtains smooth signal
UCSignal′;5 moving average methods for N number of Non-stationary Data { Yi }, regard as the minizone of every m adjacent data
It is inside close to smoothly, i.e. its mean value is close to constant.Then it can use the average value of every m adjacent data, to indicate this m number
According to any of value, and regard it as inhibiting the measurement result of random error or eliminate the signal of noise.It is usually multi-purpose
The mean value indicates the measurement result or signal of its middle data or end-point data.It is 5 sliding averages when m takes 5.Pass through
After sliding average, frequent random fluctuation in data can be filtered, shows smooth variation tendency, while can also obtain random error
Change procedure, so as to estimate its statistical characteristic value.Referring in particular to Pei Yixuan, the original substantially of Guo Min moving average method
Reason and application [J] Canon launching and control journal, 2001 (1): 21-23.
S3.2, UCSignal ' obtained in original uterine contraction signal UCSignal and S3.1 is subtracted each other, obtains new tire
Dynamic signal array UCDiff.I is initialized as 0;
S3.3, to uterine contraction Fetal Movement Signal UCDiffiValue scan for, if UCDiffi> 5, and at this time without
The movement of the foetus point being marked executes step S3.4;Otherwise, if there is the movement of the foetus marked, step S3.5 is executed;
S3.4, uterine contraction movement of the foetus accumulated time number fmpeaknums=0 is defined.I is labeled as new uterine contraction movement of the foetus starting point
This movement of the foetus end point fmend is initialized as i, fmpeaknums=fmpeaknums+1, i=i+1 by fmstart;
If S3.5, the movement of the foetus end point fmend marked and position i time difference within 5 seconds, by fmend's
Value is updated to i, fmpeaknums=fmpeaknums+1, i=i+1.Otherwise, terminate the flag state of the movement of the foetus marked,
Take maximum value of the UCDiff in range [fmstart, fmend] as movement of the foetus range value fmpeak.By tuple (fmstart,
Fmend, fmpeak, fmpeaknums) movement of the foetus is deposited into uterine contraction movement of the foetus array.Execute step S3.6;
S3.6, step S3.3 is continued to execute, terminated until searching signal.If there are also locate when arriving signal end point
In the movement of the foetus point of flag state, terminates this movement of the foetus flag state and be deposited into uterine contraction movement of the foetus array.
S4, by calculating F/A figure curve baseline and dynamic threshold extracts movement of the foetus information, it is living to obtain the second fetus
Dynamic information;
S4.1, F/A figure curve baseline calculate;
If F/A signal is ActoSignal, F/A signal base line is ActoBaseline.
The calculation of ActoBaseline is as follows:
Firstly, every one minute calculate ActoSignal mono- minute in signal from low to high preceding 15% average value be used as this
The initial reference value ActoBase of period baseline value.ActoBaseline is calculated by following trimming formula:
Then according to the thought for calculating baseline fetal heart rate, to filtering and backward filtering before being carried out to F/A signal base line
Processing, has carried out simplified processing to formula here.Forward direction Filtering Formula are as follows:
ActoBaselinei=0.8 × ActoBaselinei-1+0.2×ActoBaselinei
Backward Filtering Formula are as follows:
ActoBaselinei=0.8 × ActoBaselinei+1+0.2×ActoBaselinei
Accelerate the baseline fetal heart rate calculating of information different from Fetal Heart Rate is obtained, fetal movements baseline need to only reflect F/A
Trend of the signal compared with lower part.Therefore, front and back is only done to F/A signal base line here to be filtered twice.
After obtaining ActoBaseline baseline, sequence of differences ActoDiff is defined:
ActoDiffi=ActoSignali-ActoBaselinei
Therefore, new Fetal Movement Signal ActoDiffSignal is calculated as follows.
S4.2, dynamic threshold extract movement of the foetus information;
After improving the recognition effect of movement of the foetus using F/A signal base line, according to F/A in clinical data
The actual margin situation of signal devises a new dynamic threshold discriminant function and is used to extract initial movement of the foetus information, for
Next section carries out more criterion identifications and movement of the foetus label.Dynamic threshold discriminant function be mainly according to the range value of baseline to threshold value into
The adjustment of row size.Specific dynamic threshold discriminant function formula is as follows:
Then movement of the foetus information is tentatively extracted as follows:
S4.2.1 initializes i=0;
S4.2.2 is searched for backward since the i of the position ActoDiffSignal, if there is ActoDiffSignali>
Threshold (i), and it is not on the movement of the foetus point of flag state at this time, execute step S4.2.3;Otherwise, if there is being in
The movement of the foetus of flag state executes step S4.2.4;
I is labeled as new movement of the foetus starting point fmstart by S4.2.3, and movement of the foetus end point fmend is initialized as i.I=
i+1.Execute step S4.2.5;
If the position fmend of the marked movement of the foetus of S4.2.4 was differed with i within 5 seconds, the value of fmend is updated to i, i
=i+1;Otherwise, the label for terminating existing movement of the foetus, takes maximum value of the ActoDiffSignal in [fmstart, fmend] range
It is fmtime as the peak value fmpeak of movement of the foetus label, and by the position mark of maximum value, the baseline value of the position is set to
This movement of the foetus is stored in movement of the foetus array by fmbase by tuple (fmtime, fmpeak, fmbase, fmstart, fmend).It holds
Row step S4.2.5;
S4.2.5 continues to execute step S4.2.2, until searching the end ActoDiffSignal.If reaching end also
There is the movement of the foetus marked that flag state is not finished, that is, terminates this movement of the foetus and mark and be stored in movement of the foetus array.
S5, more criterions identify movement of the foetus, i.e., accelerate baseline various amplitude to assign different Fetal Heart Rates Fetal Heart Rate and accelerate index
Score value;The threshold range for setting F/A information peak value baseline value, according to where the peak value baseline value of F/A information
Different threshold ranges assign corresponding uterine contraction movement of the foetus information index score value, F/A figure movement of the foetus index score;Calculate Fetal Heart Rate
Accelerate the total score of index score, uterine contraction movement of the foetus information index score value and F/A figure movement of the foetus index score;
S5.1, initialization i=0;
S5.2, fetal monitoring signal is searched for backward since the i of position, in conjunction with Fetal Heart Rate acceleration, uterine contraction movement of the foetus information and tire
Youngster's activity diagram movement of the foetus information judges.It defines movement of the foetus index fmscore and the credible journey of movement of the foetus occurs for measuring current location
Degree, fmscore are initialized as 0.Accelerate for Fetal Heart Rate, uterine contraction movement of the foetus information and F/A figure movement of the foetus information define movement of the foetus
Index score acc_fm_score, uc_fm_score and fmp_fm_score, specific value are chosen according to the following rules:
1) for accelerating, accelerate if it is big, acc_fm_score=1;If it is small acceleration, acc_fm_score=
0.5;
2) for uterine contraction movement of the foetus, if meeting fmpeak > 7 in uterine contraction movement of the foetus and fmpeaknums >=3 and fmend-
Fmstart > 5, uc_fm_score=0.5;Meet in uterine contraction movement of the foetus:
Fmpeaknums >=3, uc_fm_score=0.3;Meet in uterine contraction movement of the foetus: fmpeak > 8 and peaknums < 3,
Uc_fm_score=0.2.
3) for F/A figure movement of the foetus, F/A figure movement of the foetus meets: fmpeak >=1.2fmbase and fmend-
Fmstart > 6 and fmend-fmstart < 150, fmp_fm_score=0.8;
F/A figure movement of the foetus meets: fmbase < fmpeak < 1.2fmbase and fmend-fmstart > 3, fmp_
Fm_score=0.3;F/A figure movement of the foetus meets: fmpeak >=1.2fmbase and fmend-fmstart≤3, fmp_fm_
Score=0.3.
Whether S5.3, searching position i add near the range of any Fetal Heart Rate accelerated events (including small acceleration), i.e., at this
[accstart, accend] ± 6 range of fast event, if it does, executing step S5.4.If it does not, executing step
S5.5;
S5.4, order:
Fmscore=fmscore+acc_fm_score
Then search range [accstart, accend] ± 6 whether there is uterine contraction movement of the foetus and F/A figure movement of the foetus event,
If it does, updating fmscore:
Fmscore=fmscore+uc_fm_score+fmp_fm_score
If fmscore > 1.0 is determined as movement of the foetus event, by movement of the foetus position mark fmpoint=accstart.I updates
Subsequent first position of maximum value between accelerated events, uterine contraction movement of the foetus and F/A figure movement of the foetus three's end point.Continue
Step S5.3 is executed, until searching fetal monitoring signal end;
Whether S5.5, searching position i exist near any uterine contraction movement of the foetus event and F/A figure movement of the foetus event range
[fmstart, fmend] ± 6 range of movement of the foetus event, if it does, updating fmscore by step S5.4, middle formula.If
Fmscore > 1.0 is determined as movement of the foetus event, takes the fmstart of time earlier above in uterine contraction movement of the foetus and F/A figure movement of the foetus, will
Movement of the foetus position mark fmpoint=fmstart.I is updated to maximum between both uterine contraction movement of the foetus and F/A figure movement of the foetus end point
The latter point of value.Step S5.3 is continued to execute, until searching monitored signal end.
S6, when total score be more than or equal to 1 when, movement of the foetus event of label, otherwise reacquire Fetal Heart Rate acceleration curve, palace
Contracting curve and F/A figure.
The movement of the foetus recognized is marked by following strategy:
Every five seconds carries out movement of the foetus identification and calculates;
It is preferred to be marked at current time if current time is within the scope of movement of the foetus after algorithm recognizes movement of the foetus position
Know movement of the foetus.
For the best identified effect of keeping method, the F/A figure signal obtained from probe, hardware aspect needs to adjust
The signal strength of probe output, signal amplitude is greater than 10 unit values as far as possible when making to occur movement of the foetus, if hardware view can not be adjusted
It is whole, Fetal Movement Signal value when movement of the foetus occurs can be made in software algorithm level by the processing of F/A figure amplifying of signal
Greater than 10 unit values.
Embodiment 2
Present embodiments provide a kind of Fetal Movement Signal identification tagging system, comprising: monitoring modular adds for obtaining Fetal Heart Rate
Fast curve, uterine contraction curve and F/A figure;
Fetal Heart Rate accelerates information calculating and identification module, by by carrying out based on baseline fetal heart rate to Fetal Heart Rate acceleration curve
It calculates and Fetal Heart Rate accelerates identification, obtain Fetal Heart Rate and accelerate information;
First F/A information calculates and search module, asks for being smoothed by the signal to uterine contraction curve
Difference signal simultaneously searches for movement of the foetus information, obtains the first F/A information;
Second fetus activity information meter is calculated and extraction module, is used to calculate by progress F/A figure curve baseline and dynamic
State threshold value extracts movement of the foetus information, obtains the second action message of fetus;
More criterion identification modules identify movement of the foetus for more criterions, i.e., accelerate baseline various amplitude to assign Fetal Heart Rate different
Fetal Heart Rate accelerate index score;Set the peak value base of movement of the foetus range value, movement of the foetus accumulated time number, movement of the foetus duration, movement of the foetus label
The threshold range of line value, the peak value baseline value institute marked according to movement of the foetus range value, movement of the foetus accumulated time number, movement of the foetus duration, movement of the foetus
Different threshold ranges, assign corresponding uterine contraction movement of the foetus information index score value, F/A figure movement of the foetus index score;Calculate tire
Heart rate accelerates the total score of index score, uterine contraction movement of the foetus information index score value and F/A figure movement of the foetus index score;
Movement of the foetus mark module is used for when total score is more than or equal to 1, movement of the foetus event of label.
The monitoring modular includes fetal rhythm detector, uterine contraction pressure detector, and the fetal rhythm detector is super for wireless blue tooth
Sound fetal rhythm is popped one's head in (FHR_PB), for obtaining Fetal Heart Rate acceleration curve and F/A figure;The uterine contraction pressure detector is nothing
Line bluetooth uterine contraction pressure probe (TOCO_PB), for detecting uterine contraction signal.The fetal rhythm detector configures two probes, so as to
Detect twins.Three probe in MCU module communicated by various communication interfaces and peripheral hardware, realize data acquisition,
Control and storage.Ultrasound works frequency are as follows: nominal frequency 1.0MHz, error are not more than ± 10%.
Embodiment 3
A kind of storage medium is present embodiments provided, which is stored with one or more programs, described program quilt
When processor executes, according to the data that 1 monitoring module monitors of embodiment arrive, realize that the Fetal Movement Signal of above-described embodiment 1 identifies mark
Note method is as follows:
Obtain Fetal Heart Rate acceleration curve, uterine contraction curve and F/A figure;
By carrying out baseline fetal heart rate calculating and Fetal Heart Rate acceleration identification to Fetal Heart Rate acceleration curve, obtains Fetal Heart Rate and accelerate
Information;
It is smoothed by the signal to uterine contraction curve and seeks difference signal and search for movement of the foetus information, obtain the first fetus
Action message;
It is calculated and dynamic threshold extraction movement of the foetus information, the second F/A of acquisition by carrying out F/A figure curve baseline
Information;
More criterions identify movement of the foetus, i.e., accelerate baseline various amplitude to assign different Fetal Heart Rates Fetal Heart Rate and accelerate index point
Value;Set F/A information peak value baseline value threshold range, according to where the peak value baseline value of F/A information not
Same threshold range assigns corresponding uterine contraction movement of the foetus information index score value, F/A figure movement of the foetus index score;Fetal Heart Rate is calculated to add
The total score of fast index score, uterine contraction movement of the foetus information index score value and F/A figure movement of the foetus index score;
When total score is more than or equal to 1, movement of the foetus event of label.
Storage medium described in the present embodiment can be the media such as ROM, RAM, disk, CD.
Embodiment 4
A kind of calculating equipment is present embodiments provided, which includes processor and memory, and memory is stored with
One or more programs when processor executes the program of memory storage, realize that the Fetal Movement Signal of above-described embodiment 1 identifies label
Method is as follows:
Obtain Fetal Heart Rate acceleration curve, uterine contraction curve and F/A figure;
By carrying out baseline fetal heart rate calculating and Fetal Heart Rate acceleration identification to Fetal Heart Rate acceleration curve, obtains Fetal Heart Rate and accelerate
Information;
It is smoothed by the signal to uterine contraction curve and seeks difference signal and search for movement of the foetus information, obtain the first fetus
Action message;
It is calculated and dynamic threshold extraction movement of the foetus information, the second F/A of acquisition by carrying out F/A figure curve baseline
Information;
More criterions identify movement of the foetus, i.e., accelerate baseline various amplitude to assign different Fetal Heart Rates Fetal Heart Rate and accelerate index point
Value;Set F/A information peak value baseline value threshold range, according to where the peak value baseline value of F/A information not
Same threshold range assigns corresponding uterine contraction movement of the foetus information index score value, F/A figure movement of the foetus index score;Fetal Heart Rate is calculated to add
The total score of fast index score, uterine contraction movement of the foetus information index score value and F/A figure movement of the foetus index score;
When total score is more than or equal to 1, movement of the foetus event of label.
It is hand-held eventually that calculating equipment described in the present embodiment can be desktop computer, laptop, smart phone, PDA
End, tablet computer or other terminal devices having a display function.
The above embodiment is a preferred embodiment of the present invention, but embodiments of the present invention are not by above-described embodiment
Limitation, other any changes, modifications, substitutions, combinations, simplifications made without departing from the spirit and principles of the present invention,
It should be equivalent substitute mode, be included within the scope of the present invention.
Claims (10)
1. a kind of Fetal Movement Signal identifies labeling method, which is characterized in that comprising steps of
Obtain Fetal Heart Rate acceleration curve, uterine contraction curve and F/A figure;
By carrying out baseline fetal heart rate calculating and Fetal Heart Rate acceleration identification to Fetal Heart Rate acceleration curve, obtains Fetal Heart Rate and accelerate letter
Breath;
It is smoothed by the signal to uterine contraction curve and seeks difference signal and search for movement of the foetus information, obtain the first F/A
Information;
It is calculated and dynamic threshold extraction movement of the foetus information, acquisition the second F/A letter by carrying out F/A figure curve baseline
Breath;
More criterions identify movement of the foetus, i.e., accelerate baseline various amplitude to assign different Fetal Heart Rates Fetal Heart Rate and accelerate index score;If
The threshold range for determining F/A information peak value baseline value, according to the different threshold values where the peak value baseline value of F/A information
Range assigns corresponding uterine contraction movement of the foetus information index score value, F/A figure movement of the foetus index score;It calculates Fetal Heart Rate and accelerates index
The total score of score value, uterine contraction movement of the foetus information index score value and F/A figure movement of the foetus index score;
When total score is more than or equal to threshold value T, movement of the foetus event of label.
2. Fetal Movement Signal according to claim 1 identifies labeling method, which is characterized in that obtained by baseline fetal heart rate calculating
After the reproduction replica FHRTemp for obtaining baseline fetal heart rate FHRCut and fetal heart rate curve, in anodinia load test, it will expire
Foot accelerates the amplitude threshold higher than baseline amplitude bpm of definition to be defined as accthreshold, time duration threshold definitions
For acckeeptime, and Fetal Heart Rate is extracted in accordance with the following steps and accelerates information:
Step S201: searching for backward from FHRCut starting point, writes down mark and is set to index value i, works as FHRCuti>FHRTempi+
Accthreshold searches for maximum value within the scope of 60 Fetal Heart Rates after the i of position, record maximum value numerical value FHRPeak and
Position FHRPeakTime then executes step S202;Otherwise, continue to search for backward;
Step S202: it from the position FHRPeakTime 55 Fetal Heart Rate point search forward, writes down mark and is set to index value j;
If there is FHRCutj<FHRBaselinej+ 2, and the Fetal Heart Rate value in [j, FHRPeakTime] range no more than
Position j is labeled as accstart by FHRPeak;
Then it from the position FHRPeakTime 55 Fetal Heart Rate point search backward, writes down mark and is set to index value k;
If there is FHRCutk<FHRBaselinek+ 2, and the Fetal Heart Rate value in [FHRPeakTime, k] range no more than
Position k is labeled as accend by FHRPeak;
If accstart and accend are labeled, and the time interval between accstart and accend is more than
The range of fetal heart rate curve accstart to accend is judged as that Fetal Heart Rate of generation accelerates by acckeeptime;Then from
S201 is re-execute the steps after accelerating one minute of position FHRPeakTime.
3. Fetal Movement Signal according to claim 1 identifies labeling method, which is characterized in that described from by uterine contraction curve
Signal be smoothed ask difference signal, search movement of the foetus information, obtain the first F/A information the step of are as follows:
Step S501, original uterine contraction curve is smoothed, obtains smooth signal UCSignal ';
Step S502, original uterine contraction signal UCSignal and UCSignal ' obtained in step S501 is subtracted each other, is obtained new
Fetal Movement Signal array UCDiff;Index value i is initialized as 0;
Step S503, to uterine contraction Fetal Movement Signal UCDiffiValue scan for, if UCDiffi> 5, and at this time without
The movement of the foetus point being marked executes step S504;Otherwise, if there is the movement of the foetus marked, step S505 is executed;
Step S504, uterine contraction movement of the foetus accumulated time number fmpeaknums=0 is defined, i is labeled as new uterine contraction movement of the foetus starting point
This movement of the foetus end point fmend is initialized as i, fmpeaknums=fmpeaknums+1, i=i+1 by fmstart;
If movement of the foetus end point fmend step S505, marked and position i time difference within 5 seconds, by fmend's
Value is updated to i, fmpeaknums=fmpeaknums+1, i=i+1;Otherwise, terminate the flag state of the movement of the foetus marked,
Take maximum value of the UCDiff in range [fmstart, fmend] as movement of the foetus range value fmpeak;By tuple (fmstart,
Fmend, fmpeak, fmpeaknums) movement of the foetus is deposited into uterine contraction movement of the foetus array, execute step S506;
Step S506, step S503 is continued to execute, is terminated until searching signal;If there are also locate when arriving signal end point
In the movement of the foetus point of flag state, terminate this movement of the foetus flag state and be deposited into uterine contraction movement of the foetus array, uterine contraction movement of the foetus array is
First F/A information.
4. Fetal Movement Signal according to claim 1 identifies labeling method, which is characterized in that described by F/A figure
The step of curve baseline calculates specifically:
If F/A signal is ActoSignal, F/A signal base line is ActoBaseline;ActoBaseline's
Calculation is as follows:
Firstly, every one minute calculating ActoSignal, using signal in one minute from low to high preceding 15% average value as should
The initial reference value ActoBase of period baseline value;ActoBaseline is calculated by following trimming formula:
To filtering and backward filtering processing before carrying out to F/A signal base line, simplified processing has been carried out to formula here;Before
To Filtering Formula are as follows:
ActoBaselinei=0.8 × ActoBaselinei-1+0.2×ActoBaselinei
Backward Filtering Formula are as follows:
ActoBaselinei=0.8 × ActoBaselinei+1+0.2×ActoBaselinei;
Accelerate the baseline fetal heart rate calculating of information different from Fetal Heart Rate is obtained, front and back is only done to F/A signal base line and is filtered twice
Wave processing;
After obtaining ActoBaseline baseline, sequence of differences ActoDiff is defined:
ActoDiffi=ActoSignali-ActoBaselinei
Therefore, new Fetal Movement Signal ActoDiffSignal is calculated as follows:
5. Fetal Movement Signal according to claim 1 identifies labeling method, which is characterized in that it is described most according to identification movement of the foetus,
It assigns Fetal Heart Rate and accelerates index score acc_fm_score, uterine contraction movement of the foetus information index score value uc_fm_score, F/A figure
The step of movement of the foetus index score fmp_fm_score specifically:
Fetal Heart Rate is accelerated, is accelerated if it is big, acc_fm_score=1;If it is small acceleration, acc_fm_score=
0.5;
For uterine contraction movement of the foetus, if meeting fmpeak > 7 in uterine contraction movement of the foetus and fmpeaknums >=3 and fmend-fmstart > 5,
Uc_fm_score=0.5;Meet in uterine contraction movement of the foetus: peaknums >=3, uc_fm_score=0.3;Meet in uterine contraction movement of the foetus:
Fmpeak>8 and peaknums<3, uc_fm_score=0.2;
For F/A figure movement of the foetus, if F/A figure movement of the foetus meets: fmpeak >=1.2fmbase and fmend-
Fmstart>6 and fmend-fmstart<150, fmp_fm_score=0.8;
If F/A figure movement of the foetus meets: fmbase<fmpeak<1.2fmbase and fmend-fmstart>3, fmp_fm_
Score=0.3;If F/A figure movement of the foetus meets: fmpeak >=1.2fmbase and fmend-fmstart≤3, fmp_fm_
Score=0.3;
Under the assignment of above-mentioned index score, the threshold value T of total score is 1.
6. Fetal Movement Signal according to claim 1 identifies labeling method, which is characterized in that more criterion identification movement of the foetus
Step specifically:
Step S1001, i=0 is initialized;
Step S1002, fetal monitoring signal is searched for backward since the i of position, in conjunction with Fetal Heart Rate acceleration, uterine contraction movement of the foetus information and tire
Youngster's activity diagram movement of the foetus information judges;It defines movement of the foetus index fmscore and the credible journey of movement of the foetus occurs for measuring current location
Degree, fmscore are initialized as 0;Accelerate for Fetal Heart Rate, uterine contraction movement of the foetus information and F/A figure movement of the foetus information define movement of the foetus
Index score acc_fm_score, uc_fm_score and fmp_fm_score,;
Step S1003, whether searching position i is near the range of any Fetal Heart Rate accelerated events, i.e., in the accelerated events
[accstart, accend] ± 6 range, if it does, executing step S1004;If it does not, executing step S1005;
Step S1004, it enables:
Fmscore=fmscore+acc_fm_score
Then search range [accstart, accend] ± 6 whether there is uterine contraction movement of the foetus and F/A figure movement of the foetus event, if
In the presence of update fmscore:
Fmscore=fmscore+uc_fm_score+fmp_fm_score
If fmscore > 1.0, it is determined as movement of the foetus event, by movement of the foetus position mark fmpoint=accstart;I is updated to add
Subsequent first position of maximum value between fast event, uterine contraction movement of the foetus and F/A figure movement of the foetus three's end point;It continues to execute
Step 3, until searching fetal monitoring signal end;
Step S1005, whether searching position i is near any uterine contraction movement of the foetus event and F/A figure movement of the foetus event range, i.e.,
In [fmstart, fmend] ± 6 range of movement of the foetus event, if it does, updating fmscore by formula in step S1004;If
Fmscore > 1.0 are determined as movement of the foetus event, the fmstart of time earlier above in uterine contraction movement of the foetus and F/A figure movement of the foetus are taken, by tire
Dynamic position mark fmpoint=fmstart;I is updated to maximum value between both uterine contraction movement of the foetus and F/A figure movement of the foetus end point
The latter point;Step S1003 is continued to execute, until searching monitored signal end.
7. Fetal Movement Signal according to claim 1 identifies labeling method, which is characterized in that by following strategy to recognizing
Movement of the foetus is marked:
Step S1101, every five seconds carries out movement of the foetus identification calculating;
Step S1102, after recognizing movement of the foetus position, if current time is within the scope of movement of the foetus, first choice is at current time
Movement of the foetus is identified, Fetal Heart Rate acceleration curve, uterine contraction curve and F/A figure are otherwise reacquired.
8. a kind of Fetal Movement Signal identifies tagging system characterized by comprising
Monitoring modular, for obtaining Fetal Heart Rate acceleration curve, uterine contraction curve and F/A figure;
Fetal Heart Rate accelerate information calculate and identification module, for by Fetal Heart Rate acceleration curve carry out baseline fetal heart rate calculate and
Fetal Heart Rate accelerates identification, obtains Fetal Heart Rate and accelerates information;
First F/A information calculates and search module, seeks difference for being smoothed by the signal to uterine contraction curve
Signal simultaneously searches for movement of the foetus information, obtains the first F/A information;
Second fetus activity information meter is calculated and extraction module, for by carrying out the calculating of F/A figure curve baseline and dynamic threshold
Value extracts movement of the foetus information, obtains the second action message of fetus;
More criterion identification modules identify movement of the foetus for more criterions, i.e., accelerate baseline various amplitude to assign different tires Fetal Heart Rate
Heart rate accelerates index score;The threshold range for setting F/A information peak value baseline value, according to F/A information peak value base
Different threshold ranges where the peak value baseline value of line value assign corresponding uterine contraction movement of the foetus information index score value, F/A figure
Movement of the foetus index score;It calculates Fetal Heart Rate and accelerates index score, uterine contraction movement of the foetus information index score value and F/A figure movement of the foetus index
The total score of score value;
Movement of the foetus mark module is used for when total score is more than or equal to threshold value T, movement of the foetus event of label.
9. a kind of storage medium, is stored with program, which is characterized in that when described program is executed by processor, in conjunction with monitoring modular
The data of detection realize that the described in any item Fetal Movement Signal of claim 1-7 identifies labeling method.
10. a kind of calculating equipment, including processor and for the memory of storage processor executable program, feature exists
In when the processor executes the program of memory storage, in conjunction with the data that monitoring modular detects, realization claim 1-7 appoints
Fetal Movement Signal recognition methods described in one.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910460273.7A CN110236594B (en) | 2019-05-30 | 2019-05-30 | Fetal movement signal identification marking method, system, medium and equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910460273.7A CN110236594B (en) | 2019-05-30 | 2019-05-30 | Fetal movement signal identification marking method, system, medium and equipment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110236594A true CN110236594A (en) | 2019-09-17 |
CN110236594B CN110236594B (en) | 2022-06-07 |
Family
ID=67885465
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910460273.7A Active CN110236594B (en) | 2019-05-30 | 2019-05-30 | Fetal movement signal identification marking method, system, medium and equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110236594B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112971752A (en) * | 2019-12-13 | 2021-06-18 | 深圳市理邦精密仪器股份有限公司 | Fetal heart rate deceleration type correction method and device and fetal monitoring equipment |
CN113827217A (en) * | 2021-10-14 | 2021-12-24 | 深圳京柏医疗科技股份有限公司 | Fetal heart detection method, device, equipment and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103190913A (en) * | 2013-04-15 | 2013-07-10 | 深圳邦健生物医疗设备股份有限公司 | Fetal movement recognition method, fetal movement recognition equipment and fetal monitor |
CN106725511A (en) * | 2017-02-10 | 2017-05-31 | 苏州萌动医疗科技有限公司 | Movement of the foetus detection method based on multi-feature fusion |
CN109589137A (en) * | 2018-11-02 | 2019-04-09 | 深圳市理邦精密仪器股份有限公司 | Method, apparatus, terminal and the computer readable storage medium of movement of the foetus identification |
-
2019
- 2019-05-30 CN CN201910460273.7A patent/CN110236594B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103190913A (en) * | 2013-04-15 | 2013-07-10 | 深圳邦健生物医疗设备股份有限公司 | Fetal movement recognition method, fetal movement recognition equipment and fetal monitor |
CN106725511A (en) * | 2017-02-10 | 2017-05-31 | 苏州萌动医疗科技有限公司 | Movement of the foetus detection method based on multi-feature fusion |
CN109589137A (en) * | 2018-11-02 | 2019-04-09 | 深圳市理邦精密仪器股份有限公司 | Method, apparatus, terminal and the computer readable storage medium of movement of the foetus identification |
Non-Patent Citations (3)
Title |
---|
MEI-JIA HUANG、HUI-JIN WANG: "Automated Detection of Fetal Movements Based on Baseline Calculation and Fetal Heart Rate Accelerations", 《MODERN APPLIED SCIENCE》 * |
李晓: "胎心率电子监护检测与分析新方法的研究", 《中国优秀硕士学位论文全文数据库》 * |
蔡弘宇: "一种新型的医院智能胎心监护系统的设计与实现", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112971752A (en) * | 2019-12-13 | 2021-06-18 | 深圳市理邦精密仪器股份有限公司 | Fetal heart rate deceleration type correction method and device and fetal monitoring equipment |
CN113827217A (en) * | 2021-10-14 | 2021-12-24 | 深圳京柏医疗科技股份有限公司 | Fetal heart detection method, device, equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN110236594B (en) | 2022-06-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US9888862B2 (en) | Electroanatomical mapping | |
Jezewski et al. | A novel technique for fetal heart rate estimation from Doppler ultrasound signal | |
CA2137670C (en) | Medical signal analyser | |
CN101810476B (en) | Classification method of heart beat template of dynamic electrocardiogram | |
CN110123304B (en) | Dynamic electrocardio noise filtering method based on multi-template matching and correlation coefficient matrix | |
Smrdel et al. | Automated detection of transient ST-segment episodes in 24h electrocardiograms | |
WO2023226223A1 (en) | Ppg signal quality evaluation method and apparatus and ppg signal processing method and system | |
CN110236594A (en) | A kind of Fetal Movement Signal identification labeling method, system, medium, equipment | |
CN103957813B (en) | For identifying the system and method for high risk pregnancy | |
US20230309899A1 (en) | Determination method and determination apparatus for beginning of t-wave, storage medium and computer program product | |
CN1981698A (en) | Woundless blood-pessure testing method based on waveform characteristic identification | |
CN110491504B (en) | Method for acquiring medical index data of heart sound signal | |
CN109589137B (en) | Fetal movement identification method, fetal movement identification device, terminal and computer-readable storage medium | |
CN116350206A (en) | Method for deriving respiration waves from cardiac-related signals and for calculating respiration rate from the respiration waves | |
Kyle et al. | Computer identification of systolic time intervals | |
CN111084621B (en) | QRS wave group form identification method and device based on depth self-encoder | |
Bakardjian | Ventricular beat classifier using fractal number clustering | |
CN109567867A (en) | A kind of processing method of Fetal Heart Rate amplitude-variation | |
Wrobel et al. | Recognition of fetal movements–automated detection from Doppler ultrasound signals compared to maternal perception | |
CN111898445A (en) | Active segment detection method for multi-channel facial surface electromyographic signals | |
JPH0622915A (en) | Heart arrhythmia detecting device based on ecg signal | |
Angyal et al. | Algorithm of oscillometric blood pressure measurement applied on microcontroller | |
CN102438514A (en) | Electrocardiogram lead recognition method and its equipment | |
CN111772627B (en) | Online fetal electrocardiosignal extraction device and method | |
US20230210467A1 (en) | Adaptive Correlation Methods for Heartbeat Detection |
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 |