WO2013053290A1 - 一种提高胎心率数据减速识别准确性的装置和方法 - Google Patents

一种提高胎心率数据减速识别准确性的装置和方法 Download PDF

Info

Publication number
WO2013053290A1
WO2013053290A1 PCT/CN2012/082019 CN2012082019W WO2013053290A1 WO 2013053290 A1 WO2013053290 A1 WO 2013053290A1 CN 2012082019 W CN2012082019 W CN 2012082019W WO 2013053290 A1 WO2013053290 A1 WO 2013053290A1
Authority
WO
WIPO (PCT)
Prior art keywords
deceleration
heart rate
fetal heart
rate data
sequence
Prior art date
Application number
PCT/CN2012/082019
Other languages
English (en)
French (fr)
Inventor
饶箭
陈吴笋
曾永华
陈德伟
Original Assignee
深圳市理邦精密仪器股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 深圳市理邦精密仪器股份有限公司 filed Critical 深圳市理邦精密仪器股份有限公司
Priority to EP12839654.6A priority Critical patent/EP2767228B1/en
Priority to US14/349,326 priority patent/US9364197B2/en
Publication of WO2013053290A1 publication Critical patent/WO2013053290A1/zh

Links

Classifications

    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/52Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/5207Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves involving processing of raw data to produce diagnostic data, e.g. for generating an image
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/02Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
    • A61B5/024Detecting, measuring or recording pulse rate or heart rate
    • A61B5/02411Detecting, measuring or recording pulse rate or heart rate of foetuses
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/24Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316Modalities, i.e. specific diagnostic methods
    • A61B5/318Heart-related electrical modalities, e.g. electrocardiography [ECG]
    • A61B5/344Foetal cardiography
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/02Measuring pulse or heart rate
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/08Detecting organic movements or changes, e.g. tumours, cysts, swellings
    • A61B8/0866Detecting organic movements or changes, e.g. tumours, cysts, swellings involving foetal diagnosis; pre-natal or peri-natal diagnosis of the baby
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/48Diagnostic techniques
    • A61B8/488Diagnostic techniques involving Doppler signals
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/52Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/5215Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves involving processing of medical diagnostic data
    • A61B8/5223Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves involving processing of medical diagnostic data for extracting a diagnostic or physiological parameter from medical diagnostic data
    • GPHYSICS
    • G16INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16HHEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/30ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment

Definitions

  • the invention relates to the field of biomedical signal processing, and in particular to a device for improving the accuracy of deceleration identification of fetal heart rate data and a method for realizing the same.
  • Fetal heart rate deceleration is the simplest and most effective method for providing reference and analysis for determining fetal safety during childbirth. During the whole process of development, about 50% ⁇ 70% of the cases will have fetal heart rate deceleration, which represents The situation is early deceleration, delayed deceleration and variable deceleration. Early deceleration is generally used as a reference for the phenomenon of fetal head compression, and has little to do with fetal hypoxia. However, if it occurs frequently in the early stage of labor, the possibility of hypoxia in the umbilical cord compressed fetus should also be considered.
  • references caused by hypoxia-induced vagal hyperthyroidism and/or inhibition of the myocardium mainly occur in the uterus-placental blood flow reduction, and fetal hypoxia caused by hypoxia.
  • the change in deceleration is mainly caused by umbilical cord compression.
  • the method of automatically identifying fetal heart rate deceleration data in clinical practice mainly based on the empirical parameters such as the duration and decreasing range of various decelerations.
  • the deceleration standard is calculated, that is, the fetal heart rate data is first collected by the fetal monitoring module and the fetal heart rate baseline is identified, and then the number of decelerations under the baseline and each are calculated according to the empirical parameter setting criteria of the clinical deceleration. Deceleration duration, amplitude, type, etc.
  • this method of identifying deceleration according to empirical parameters has the following disadvantages: First, the types of deceleration in the actual clinical environment are various, and the empirical parameters are neither accurate nor cover all situations, which may lead to deceleration identification data. The error is large; the second is that this method can not identify the continuous deceleration and eliminate the baseline variation, resulting in the identification of the number of deceleration data is less than the actual and the baseline variation is included in the deceleration; the third is likely to be in the labor process Multiple decelerations Explain the mix of book types. This method can only identify one type of each deceleration data collected, and cannot recognize the mixed deceleration data.
  • the empirical parameter can not meet the requirement of automatic identification and deceleration.
  • the purpose of the invention is to overcome the large deceleration error recognized by the existing automatic fetal heart rate deceleration identification method, and it is impossible to identify the continuous deceleration and reject the baseline variation, and cannot identify the mixed deceleration and the like.
  • dynamic threshold area method continuous peak detection and hybrid deceleration detection algorithms, the baseline variation can be effectively eliminated, and each deceleration and its type can be accurately identified.
  • an object of the present invention is to provide an apparatus and method for improving the accuracy of fetal heart rate data deceleration recognition.
  • fetal heart rate data collection module fetal heart rate data baseline identification module
  • fetal heart rate data preprocessing module fetal heart rate data deceleration identification module and output module
  • the fetal heart rate data acquisition module is configured to collect fetal heart rate data of a preset duration, and obtain a fetal heart rate data sequence H (n);
  • the fetal heart rate data baseline identification module is coupled to the fetal heart rate data acquisition module for receiving the fetal heart rate data sequence H(n) sent by the fetal heart rate data acquisition module, and identifying the fetal heart rate Rate baseline sequence B (n);
  • the fetal heart rate data pre-processing module is connected to the fetal heart rate data acquisition module, and configured to receive the fetal heart rate data sequence H(n) sent by the fetal heart rate data acquisition module, and The sequence H (n) is pretreated to obtain a fetal heart rate data sequence V (n);
  • the fetal heart rate data deceleration identification module is coupled to the fetal heart rate data baseline identification module and the fetal heart rate data preprocessing module for receiving a fetal heart rate baseline sequence sent by the fetal heart rate data baseline identification module B (n) and the pre-processed sequence V (n) sent by the fetal heart rate data pre-processing module, according to a preset deceleration judgment criterion and the fetal heart rate data baseline sequence B (n)
  • the pre-processed fetal heart rate data sequence V(n) is decelerated to obtain a deceleration data segment;
  • the output module described in the specification is connected to the fetal heart rate data deceleration identification module, and is configured to receive the deceleration data segment sent by the fetal heart rate data deceleration recognition module, and output the same.
  • the device for improving the accuracy of the fetal heart rate data deceleration identification further comprises: a fetal heart rate signal acquisition and processing conversion module, a fetal heart rate data deceleration verification module, a fetal heart rate data deceleration determination standard setting module, and a fetal heart rate data. Deceleration type judgment module, acquisition duration setting and judgment module, and fetal heart rate data deceleration attribute calculation module,
  • the fetal heart rate signal acquisition and processing conversion module is connected to the fetal heart rate data acquisition module, configured to collect a fetal heart rate signal, convert the fetal heart rate data, and send fetal heart rate data to the fetal heart rate Rate data acquisition module;
  • the fetal heart rate data deceleration verification module is connected to the fetal heart rate data deceleration identification module and the fetal heart rate data deceleration attribute calculation module, and is configured to receive the deceleration data sent by the fetal heart rate data deceleration identification module. Segment, according to the signal loss situation of each deceleration, verify whether each deceleration sequence segment is a real deceleration, and send the verification result to the fetal heart rate data deceleration attribute calculation module;
  • the fetal heart rate data deceleration determination standard setting module is connected with the fetal heart rate data deceleration recognition module, and is used for presetting the fetal heart rate data deceleration determination standard and transmitting to the fetal heart rate data deceleration.
  • Identification module is connected with the fetal heart rate data deceleration recognition module, and is used for presetting the fetal heart rate data deceleration determination standard and transmitting to the fetal heart rate data deceleration.
  • the fetal heart rate data deceleration type judging module is connected with the fetal heart rate data deceleration attribute calculation module, and is configured to perform a deceleration type judgment in the case of collecting the contraction data at the same time, and send the judgment result to the tire
  • the heart rate data deceleration attribute calculation module; the collection duration setting and determination module is connected with the fetal heart rate data acquisition module, configured to set the duration of the fetal heart rate data acquisition and determine whether the data acquisition time is If the set time is exceeded, if the set time is exceeded, a timeout signal is sent to the fetal heart rate data acquisition module.
  • the fetal heart rate data deceleration attribute calculation module and the fetal heart rate data are decelerated
  • the specification identification module and the output module are connected, configured to receive the deceleration data segment sent by the fetal heart rate data deceleration identification module, calculate a deceleration attribute value such as duration and amplitude of each deceleration data segment, and send the calculation result to The output module;
  • the fetal heart rate data preprocessing module further includes
  • the error data processing unit performs error data processing on the fetal heart rate data sequence H (n) to obtain an effective fetal heart rate data sequence V (n);
  • the interpolation processing unit fits the invalid data portion of the valid fetal heart rate data sequence V (n) by a linear interpolation method to obtain a fetal heart rate data sequence C (n).
  • the fetal heart rate data deceleration identification module further includes
  • a standard comparison unit configured to determine whether there is a sequence segment in the fetal heart rate data sequence C (n) that satisfies a preset deceleration determination criterion
  • the deceleration number unit is analyzed for analyzing the number of consecutive decelerations of the deceleration sequence segment in the sequence C (n) satisfying the criterion.
  • the output module further includes
  • a display unit configured to display the identified deceleration data segment and its attribute value
  • a printing unit configured to print out the identified deceleration data segment and its attribute value
  • the invention provides a method for improving the accuracy of deceleration identification of fetal heart rate data, comprising: Step 1: collecting fetal heart rate data within a predetermined time period to obtain a fetal heart rate data sequence Description H(n);
  • Step 2 performing baseline identification on the fetal heart rate data sequence H(n) to obtain a fetal heart rate baseline data sequence B(n);
  • Step 3 preprocessing the fetal heart rate data sequence H(n) to obtain a pre-processed fetal heart rate data sequence C(n);
  • Step 4 Decelerating and identifying the pre-processed fetal heart rate data sequence C(n) according to a preset deceleration judgment criterion and the fetal heart rate baseline data sequence B(n), and obtaining a deceleration data segment;
  • step 5 the calculation result of each of the deceleration data segments and the deceleration attribute values thereof is output.
  • the method further comprises: collecting and processing fetal heart rate signal conversion to obtain fetal heart rate data.
  • the step 3 further includes: Step 31: performing error data processing on the fetal heart rate data sequence H(n) to obtain a sequence V(n);
  • Step 32 performing interpolation on the sequence V(n) to obtain a pre-processed fetal heart rate data sequence C(n);
  • the step 4 further includes: Step 41: input the sequence C(n), B(n) into a preset deceleration judgment standard, and obtain a sequence C(n) a set of each sequence segment that satisfies the deceleration criterion ⁇ C, and its corresponding baseline sequence segment set, if there is no sequence segment that satisfies the deceleration judgment criterion, then return to step 1 described above, and re-calculate the fetal heart rate data. Acquisition
  • Step 42 Perform the difference between the sum and the obtained segment to obtain a sequence segment, and find that the threshold R is not exceeded in the sequence segment ⁇ A. a continuous sequence segment, if there is no sequence segment that satisfies this condition, it is judged that the sequence segment is a deceleration sequence segment, and if there is a sequence segment satisfying this condition in ⁇ , then the sequence segment satisfying the corresponding position of the condition in ⁇ Determined as the baseline variation, the baseline variation is the baseline variation
  • the book divides c ⁇ ' into several segments, denoted as ⁇ ; to determine whether the deceleration criterion is met, if it meets the deceleration criteria, it is the deceleration sequence segment, otherwise it is the baseline variation sequence segment, where R. For pre-set parameters;
  • Step 43 Find a continuous sequence segment in the deceleration sequence segment that deviates from a corresponding baseline value that does not exceed the threshold R1. If there is no sequence segment that satisfies such a condition, it is a deceleration. If there is a sequence segment that satisfies such a condition, Then it is recorded as ⁇ ⁇ '; ⁇ " 2 , and then it is analyzed whether each segment can meet the deceleration standard alone. If it can be sufficient, it will become deceleration independently. If it is not satisfied, it will need to be merged into the segment to make it The adjacent segments are continuously and together analyzed whether the deceleration criterion is met.
  • the continuous data segment is a deceleration, and the subsequent segment data is further subjected to the deceleration standard analysis. Otherwise, the baseline merges the segments until the segments are all merged, where R1 For pre-set parameters;
  • the method further includes: determining, according to the deceleration data segment, a signal loss condition of each deceleration data segment, and verifying whether each deceleration sequence segment is a real deceleration ;
  • the step of determining the signal loss of each deceleration data segment according to the deceleration data segment, and the step of verifying whether each deceleration sequence segment is a true deceleration further comprises:
  • c Calculate the degree of signal loss during the adjusted deceleration. If the threshold ⁇ is exceeded, the eligibility of the deceleration is revoked, otherwise the deceleration is retained, where f is the pre-set parameter.
  • the method further includes: calculating a deceleration attribute value of each of the deceleration data segments;
  • the calculating the deceleration attribute value of each of the deceleration data segments further includes:
  • Step 51 Calculate the duration according to the starting point and end point of each deceleration, calculate the maximum amplitude ⁇ from the baseline during deceleration, and record the position of the point (peak point)
  • step 52 it is determined whether the contraction data is collected at the same time. If the contraction data is not collected, the type of deceleration is not determined. If the contraction data is collected, the deceleration data segment is determined according to the preset deceleration type judgment condition. Early deceleration and / or delayed deceleration and / or change deceleration and / or extended deceleration.
  • the step 6 further includes: displaying and/or printing and/or storing and/or identifying the deceleration data segment and the attribute value calculation result.
  • the preset deceleration judgment standard described in step 4 further includes:
  • the fetal heart rate curve must be below the fetal heart rate baseline.
  • the duration below the fetal heart rate baseline must be greater than or equal to the threshold ⁇ .
  • the maximum deviation from the baseline must be greater than or equal to the threshold ⁇ .
  • the fetal heart rate data segment to be analyzed is a deceleration data segment, wherein ⁇ 0, ⁇ 0, "is a preset parameter.
  • the invention can effectively eliminate the baseline variation part, accurately identify each deceleration and its type, and avoid the existing method easily causing deceleration recognition error.
  • Continuous deceleration and rejection of baseline variability are not recognized, and deficiencies such as hybrid deceleration cannot be identified.
  • the program is suitable It is stated that the book deceleration occurs independently, and in the case of typical deceleration, it is suitable for continuous deceleration and hybrid deceleration.
  • FIG. 1 is a flow chart of an embodiment of a method for improving the accuracy of deceleration identification of fetal heart rate data according to the present invention
  • FIG. 2 is a flow chart showing another embodiment of a method for improving the accuracy of deceleration identification of fetal heart rate data according to the present invention
  • FIG. 3 is a block diagram of an embodiment of an apparatus for improving the accuracy of deceleration identification of fetal heart rate data according to the present invention
  • FIG. 4 is a structural diagram of another embodiment of an apparatus for improving the accuracy of deceleration identification of fetal heart rate data according to the present invention.
  • Figure 5 is a schematic diagram of the collected fetal heart rate data and the identified fetal heart rate baseline;
  • Figure 6 is a rendering of the fetal heart rate data sequence after pretreatment;
  • Figure 7 is a deceleration recognition effect diagram of the fetal heart rate data sequence
  • Figure 8 is a deceleration check effect diagram of the fetal heart rate data sequence
  • Fig. 9 is a diagram showing the effect of deceleration type recognition when there is contraction data of the fetal heart rate data sequence. detailed description
  • the technical solution for automatic identification of fetal heart rate deceleration mainly includes the upper computer and the lower computer.
  • the lower computer mainly obtains the fetal heart signal through the ultrasonic probe, and then calculates the fetal heart rate through hardware filtering processing and autocorrelation algorithm; Mainly to accept data from the lower computer, then display, store fetal heart rate data, depict fetal heart rate curve, analysis of fetal heart rate data (including identification of fetal heart rate data deceleration, etc.), display, print,
  • the main processing method flow in the manual is shown in Figure 1:
  • the host computer collects n-minute fetal heart rate data to obtain a fetal heart rate data sequence H(n), wherein, in this embodiment, it is required to collect at least ten minutes of data, according to "Fetal Electronic Monitoring", in general, It is necessary to determine the fetal heart rate baseline at least ten minutes of fetal heart rate data.
  • the fetal heart rate baseline data sequence B (n) is identified by low-pass filtering.
  • the baseline is a low-frequency signal relative to the fetal heart rate data, and the low-pass filter has the ability to pass low-frequency signals and high frequencies. The signal is blocked and can be used to extract low frequency signals such as fetal heart rate baseline.
  • the sequence H (n) can also be processed first, such as error data processing, mean filtering, linear interpolation, etc., and the processed data is then used to identify the fetal heart rate baseline.
  • Figure 5 shows the fetal heart rate collected. Rate data and identified fetal heart rate baseline.
  • the sequence H (n) is processed by error data processing and linear interpolation in sequence, and corresponding fetal heart rate data sequences corresponding to different preprocessing processes are obtained.
  • sequence segment corresponding to the condition is determined by a preset criterion, and according to the sequence segment and its corresponding baseline sequence segment, the sequence segments of the deceleration and baseline variation in the sequence segment are detected and distinguished, and the continuous deceleration of the deceleration sequence segment is analyzed. Number.
  • the duration is calculated from the start and end points of each deceleration, and the maximum amplitude value deviating from the baseline during deceleration is calculated.
  • This embodiment preferably displays and displays the recognized deceleration data segment and the calculation result.
  • the fetal heart signal received by the ultrasonic probe is uploaded to the upper computer after the lower-level machine calculates the fetal heart rate through the hardware filtering process and the autocorrelation algorithm.
  • the processing of the upper computer includes displaying, storing and depicting the fetal heart rate data into the fetal heart rate. Curves and related data processing and more.
  • the autocorrelation algorithm has the characteristics of enhancing the periodic signal and weakening the random noise, and is a common technique for calculating the fetal heart rate data.
  • the host computer collects n-minute fetal heart rate data to obtain a fetal heart rate data sequence H(n), wherein, in this embodiment, it is required to collect at least ten minutes of data, according to "Fetal Electronic Monitoring", in general, It is necessary to determine the fetal heart rate baseline at least ten minutes of fetal heart rate data.
  • the fetal heart rate baseline B (n) is identified by low-pass filtering.
  • the baseline belongs to the low-frequency signal relative to the fetal heart rate data.
  • the low-pass filter has the ability to pass the low-frequency signal and the high-frequency signal is The blocking feature can be used to extract low frequency signals such as fetal heart rate baseline.
  • the sequence H (n) can also be processed first, such as error data processing, mean filtering, linear interpolation, etc., and the processed data is then used to identify the fetal heart rate baseline.
  • Figure 5 shows the fetal heart rate collected. Rate data and identified fetal heart rate baseline.
  • there are many ways to identify the baseline of fetal heart rate data such as the average method commonly used in the industry.
  • the error data processing of the sequence H (n) is performed to obtain the effective fetal heart rate data sequence V (n). This step can effectively filter out the invalidity of the fetal heart rate due to poor fetal heart signal quality.
  • the manual incorrect data, the rest is valid data.
  • the linear interpolation method is used to fit the invalid data part of the sequence V (n) to obtain the sequence C (n).
  • the linear interpolation is a simple interpolation method that uses the two-point line principle to solve other points on the line.
  • Fig. 6 is an effect diagram of the fetal heart rate data sequence after preprocessing, and the fetal heart rate curve portion in the box is the curve segment obtained by interpolating the erroneous data.
  • the steps 1031 and 1032 described above are for pre-processing the collected fetal heart rate data.
  • the pre-processing steps of the fetal heart rate data are not limited to the above steps, and it is also possible to use the industry-known pair to determine how many minutes per minute.
  • Jump (bpm, beats per min ) is a method of removing fetal heart rate values in units of erroneous data and/or sliding average and/or interpolation fits, and the like.
  • the preset deceleration judgment criterion is:
  • the fetal heart rate curve must be below the fetal heart rate baseline.
  • the duration below the fetal heart rate baseline must be greater than or equal to the threshold r.
  • the maximum amplitude from the baseline ⁇ must be greater than or equal to the threshold ⁇ .
  • the area s of the fetal heart rate curve is obtained.
  • the process is as follows: Let the fetal heart rate curve and its corresponding baseline segment be ⁇ , [ ⁇ , will be Poor, get the sequence segment ⁇ , for the complex trapezoidal formula with the step size A, the area ⁇ S must be greater than or equal to the minimum value of the dynamic threshold " ⁇ ⁇ ⁇ ⁇ and the static threshold.
  • the judgment condition enriches the criterion of judgment, making the judgment more accurate, wherein the parameter
  • Hey. , oh. , ⁇ , « are preset empirical parameters, and different changes will be set as the parameters of the gestational age are input.
  • sequence segment ⁇ For each sequence segment, the difference between ⁇ and , gives the sequence segment ⁇ , and the value in the sequence segment ⁇ A does not exceed the threshold R.
  • Continuous sequence segment if there is no sequence segment satisfying this condition, the sequence segment is composed of one or continuous deceleration, and the number of decelerations will be analyzed by the continuous peak detection algorithm in the next step; if there is a sequence segment satisfying this condition, The sequence segment corresponding to the position is the baseline variation, and the baseline variation is ⁇ t, where ⁇ ⁇ , then the baseline variation will divide ⁇ into several 1 segments, denoted as ⁇ 2 , where ⁇ 3 ⁇ 4, for judging whether the deceleration standard is met, if it meets the deceleration standard, it is one or consists of continuous deceleration, and the number of decelerations will be analyzed by the continuous peak detection algorithm in the next ;; if it does not meet the deceleration standard, ij ⁇
  • this section can be used to distinguish between de
  • the process of analysis using the continuous peak detection algorithm is: looking for deviations from the corresponding baseline in the segment a continuous sequence segment exceeding the threshold Ri, if there is no sequence segment satisfying such a condition (all exceeding the threshold RI), the Bay is a deceleration; if there is a sequence segment that satisfies such a condition, it is recorded as
  • FIG. 7 is a deceleration recognition effect diagram.
  • the fetal heart rate curve part of the box is the baseline variation part, and the curve section indicated by the arrow is the identified fetal heart rate deceleration curve section.
  • the steps as described above: 1041, 1042, and 1043 are used to decelerate and recognize the collected fetal heart rate data, and the present invention can complete the identification of the deceleration data segment by the step, and can directly enter the following 105 steps. And 106 steps to perform attribute value calculation and result output or directly enter step 106 to output the result. In order to make the result more accurate, it is also preferable in the embodiment to employ the following deceleration checking step before the calculation and result output steps.
  • the deceleration in the above steps is obtained from the interpolated sequence, that is, without considering the signal loss.
  • the deceleration is verified, and the signal loss in each pre-deceleration is analyzed to determine whether a pre-deceleration is retained, split, and finally decelerated.
  • the step also includes the following processing:
  • Step 1 Compare the sequences V (n) and C (n) and mark the position with the interpolation to obtain the marker sequence M (n).
  • Step 2 For each deceleration, adjust the starting point and end point of the deceleration according to the sequence M (n) so that the starting point and end point are not the interpolation points and are closest to the interpolation point.
  • Step 3 Calculate the degree of signal loss in the adjusted deceleration, the signal loss is the number of invalid values in the fetal heart rate deceleration curve data and (the number of interpolation points) as a whole, if the threshold f is exceeded , the qualification for the deceleration is revoked; otherwise the eligibility for deceleration is retained.
  • Figure 8 is a deceleration check effect diagram, in which the arrow icon indicates deceleration.
  • the type of deceleration can be judged. If the instruction does not collect the contraction data, the type of deceleration is not judged, and the final deceleration display effect diagram is consistent with Fig. 8; if the contraction data is collected, the deceleration detection algorithm is used for each deceleration to determine the type of deceleration.
  • Judgment condition 1 The value obtained by subtracting ⁇ D is greater than or equal to the threshold value, minus the absolute value of the contraction starting point ⁇ is less than or equal to the threshold value, ⁇ minus the absolute value of the position of the contraction peak point ⁇ is less than or equal to the threshold value; 2: 1 ⁇ Subtracted value 1 ⁇ is greater than or equal to the threshold, minus 1 ⁇ obtained value s is greater than or equal to the threshold, D W salt to go to the peak position ⁇ is greater than or equal to the threshold 7 ⁇ and less than or equal to the threshold 7 ⁇ greater than the palace
  • the determined deceleration data satisfies the condition one, it is judged as early deceleration, and if condition 2 is satisfied, it is judged as delayed deceleration, and when condition 3 is satisfied, it is judged as change deceleration, and when condition 4 is satisfied, it is judged as deceleration.
  • Each deceleration data is input into the four determination conditions.
  • the deceleration is judged as a typical deceleration, that is, a single type deceleration; if a plurality of conditions are satisfied, the deceleration data is judged to be an atypical deceleration, That is, the type of the deceleration is mixed, and at the same time, the combination of which deceleration types the deceleration is determined according to the conditions satisfied by the deceleration.
  • Figure 9 shows the effect of the deceleration type recognition when there is contraction data. The figure shows the early deceleration, indicating the change deceleration, ! 1 means early deceleration and variable deceleration, 01 means delayed deceleration and variable deceleration.
  • the identified deceleration data segments are sent to the display, print, and storage modules, respectively, and each deceleration data segment and its attribute values can be identified and displayed on the fetal chart, and its attribute values can be stored and printed.
  • FIG. 1 An architectural diagram of an embodiment of an apparatus for improving the accuracy of deceleration identification of fetal heart rate data is shown in FIG.
  • Fetal heart rate data acquisition module 301 The instructions are used to collect the fetal heart rate data of the preset duration to obtain the fetal heart rate data sequence H(n); the fetal heart rate data baseline identification module 302:
  • fetal heart rate data acquisition module 301 Connected to the fetal heart rate data acquisition module 301, configured to receive the fetal heart rate data sequence H(n) sent by the fetal heart rate data acquisition module 301, and identify a fetal heart rate baseline sequence B(n);
  • Fetal heart rate data preprocessing module 303
  • fetal heart rate data acquisition module 301 Connected to the fetal heart rate data acquisition module 301, for receiving the fetal heart rate data sequence H(n) sent by the fetal heart rate data acquisition module 301, and preprocessing the sequence H(n) to obtain a fetal heart rate Rate data sequence V (n);
  • Fetal heart rate data deceleration identification module 304
  • the fetal heart rate data baseline identification module 302 and the fetal heart rate data pre-processing module 303 for receiving the fetal heart rate baseline baseline sequence B(n) sent by the fetal heart rate data baseline identification module 302 and the The pre-processed sequence V(n) sent by the fetal heart rate data pre-processing module 303, according to the preset deceleration judgment criterion and the fetal heart rate data baseline sequence B(n)
  • the heart rate data sequence V (n) is decelerated to obtain a deceleration data segment;
  • Fetal heart rate data deceleration attribute calculation module 305
  • fetal heart rate data deceleration identification module 304 configured to receive the deceleration data segment sent by the fetal heart rate data deceleration identification module 304, and calculate a deceleration attribute such as duration and amplitude of each deceleration data segment;
  • the fetal heart rate data deceleration attribute calculation module 305 is connected to receive the deceleration data segment and the deceleration attribute calculation result sent by the fetal heart rate data deceleration attribute calculation module 305, and output.
  • FIG. 1 An architectural diagram of another embodiment of an apparatus for improving the accuracy of deceleration identification of fetal heart rate data is shown in FIG.
  • a device for improving the accuracy of deceleration identification of fetal heart rate data further comprising: Instruction manual fetal heart signal acquisition and processing conversion module 300:
  • the fetal heart rate data acquisition module 301 Connected to the fetal heart rate data acquisition module 301, is an ultrasonic probe device for acquiring, processing and converting signals, which is used for collecting fetal heart rate signals, converting into fetal heart rate data, and transmitting fetal heart rate data to the clinic.
  • the fetal heart rate data deceleration identification module 304 and the fetal heart rate data deceleration attribute calculation module 305 are connected to receive the deceleration data segment sent by the fetal heart rate data deceleration identification module 304, according to each deceleration signal. Loss condition, check whether each deceleration sequence segment is a real deceleration, and send the verification result to the fetal heart rate data deceleration attribute calculation module 305;
  • Fetal heart rate data deceleration judgment standard setting module 313
  • the fetal heart rate data deceleration identification module 304 is connected to the fetal heart rate data deceleration determination module, and is sent to the fetal heart rate data deceleration recognition module 304;
  • Fetal heart rate data deceleration type judgment module 314
  • the fetal heart rate data deceleration attribute calculation module 305 is connected to the fetal heart rate data deceleration attribute calculation module 305, and the determination result is sent to the fetal heart rate data deceleration attribute calculation module 305;
  • the fetal heart rate data acquisition module 301 is connected to set the duration of the fetal heart rate data acquisition and determine whether the data collection time exceeds the set duration. If the set duration is exceeded, the fetal heart rate is The data collection module 301 sends a timeout signal;
  • the fetal heart rate data pre-processing module 303 further includes: an error data processing unit 307, performing error data processing on the fetal heart rate data sequence H(n) to obtain an effective fetal heart rate data sequence V. (n);
  • the fetal heart rate data deceleration identification module 304 described in the specification further includes: a standard comparison unit 309, configured to determine whether there is a sequence segment in the fetal heart rate data sequence C(n) that satisfies a preset deceleration determination criterion;
  • the deceleration and mutation unit 310 is configured to detect and distinguish the deceleration and baseline variation in the sequence C (n) that satisfies the criterion according to the fetal heart rate data sequence C (n) and the corresponding baseline sequence B (n) satisfying the criterion Sequence segment
  • the deceleration number unit 311 is configured to analyze the continuous deceleration number of the deceleration sequence segment in the sequence C (n) that satisfies the criterion;
  • the output module 306 further includes:
  • a display unit 315 configured to display the identified deceleration data segment and its attribute value
  • a printing unit 316 configured to print out the identified deceleration data segment and its attribute value
  • a storage unit 317 configured to store the identified deceleration data segment and its attribute value
  • An identifying unit 318 configured to identify the identified deceleration data segment and its attribute value

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Public Health (AREA)
  • Medical Informatics (AREA)
  • Biomedical Technology (AREA)
  • General Health & Medical Sciences (AREA)
  • Pathology (AREA)
  • Animal Behavior & Ethology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Biophysics (AREA)
  • Physics & Mathematics (AREA)
  • Veterinary Medicine (AREA)
  • Cardiology (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physiology (AREA)
  • Gynecology & Obstetrics (AREA)
  • Pregnancy & Childbirth (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Epidemiology (AREA)
  • Primary Health Care (AREA)
  • Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)
  • Ultra Sonic Daignosis Equipment (AREA)
  • Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)

Abstract

一种提高胎心率数据减速识别准确性的装置及方法。该方法包括:进行胎心率数据采集(101);对采集的胎心率数据进行基线识别(102);对采集的胎心率数据进行预处理(103);根据预设的减速判断标准和胎心率数据基线对预处理后的胎心率数据进行减速识别,得到减速数据段(104);计算每个减速数据段的减速属性值(105),将减速数据段和计算结果输出(106)。本发明通过采用动态阈值面积法、连续波峰检测和混合减速检测等方法,可以有效地剔除基线变异部分,准确地识别出每个减速及其类型,避免了现有的方法容易造成减速识别误差大,不能识别连续减速和剔除基线变异,不能识别混合减速等不足。本发明中的方案既适合减速独立出现,为典型减速的情况,又适合出现连续减速、混合减速的情况。

Description

说 明 书
一种提髙胎心率数据减速识别准确性的装置和方法 技术领域
本发明涉及生物医学信号处理领域, 具体的说是一种提高胎心 率数据减速识别准确性的装置及其实现的方法。
背景技术
胎心率减速是在分娩期中用来为判定胎儿安危提供参考和分析 的最简单有效的方法, 在整个产程的进展中, 大约有 50%〜70%的产 例会出现胎心率减速, 其代表情形是早发减速、 迟发减速及变化减 速。 早发减速一般作为胎头受压所致的现象的参考, 与胎儿缺氧关 系不大, 但如果频发于产程早期, 也应考虑脐带受压胎儿缺氧的可 能;迟发减速一般作为由于缺氧导致迷走神经亢进及 /或对心肌的抑 制所致的参考, 主要发生在子宫 -胎盘血流量减少, 及胎盘功能低下 等所造成的胎儿缺氧时; 变化减速主要作为由脐带受压引起, 当出 现重度变化减速或不典型变化减速时提示胎儿窘迫的参考。
目前临床医学上对胎心率减速及其类型的判断没有统一的量化 标准, 这导致临床上自动识别胎心率减速数据的方法主要是根据各 类减速的持续时间、下降幅度等的经验参数设定的减速标准来计算, 即首先通过胎监模块采集到胎心率数据并识别出胎心率基线, 然后 根据临床上减速的经验参数设定标准去计算基线之下减速的个数以 及每个减速的持续时间、 幅度、 类型等。 但在临床应用中这种按照 经验参数去识别减速的方法有以下缺点: 一是实际临床环境中减速 的类型多种多样, 经验参数既不一定准确也无法覆盖到所有情况, 会导致减速识别数据的误差大; 二是这种方法不能识别连续减速和 剔除基线变异部分, 造成识别出的减速数据个数比实际偏少以及将 基线变异部分算在减速中去; 三是在产程中有可能会出现多种减速 说 明 书 类型的混合, 这种方法对采集的每一个减速数据只能识别出一种类 型, 不能识别混合减速数据的情况。
经验参数不能满足自动识别减速的要求, 本发明的目的是为了 克服现有的胎心率减速自动识别方法识别出的减速误差大, 不能识 别连续减速和剔除基线变异, 不能识别混合减速等不足, 通过采用 动态阈值面积法、 连续波峰检测和混合减速检测等算法, 可以有效 地剔除基线变异部分, 准确地识别出每个减速及其类型。
发明内容
为克服上述缺陷, 本发明的目的即在于一种提高胎心率数据减 速识别准确性的装置和方法。
本发明的目的是通过以下技术方案来实现的:
包括: 胎心率数据釆集模块、 胎心率数据基线识别模块、 胎心 率数据预处理模块、 胎心率数据减速识别模块和输出模块,
所述的胎心率数据采集模块,用于采集预设时长的胎心率数据, 得到胎心率数据序列 H (n) ;
所述的胎心率数据基线识别模块与所述的胎心率数据采集模块 连接, 用于接收所述胎心率数据采集模块发送的胎心率数据序列 H (n), 并且识别出胎心率基线序列 B (n) ;
所述的胎心率数据预处理模块与所述的胎心率数据采集模块连 接,用于接收所述胎心率数据采集模块发送的胎心率数据序列 H (n), 并且对所述的序列 H (n)进行预处理得到胎心率数据序列 V (n) ;
所述的胎心率数据减速识别模块与所述的胎心率数据基线识别 模块和胎心率数据预处理模块连接, 用于接收所述胎心率数据基线 识别模块发送的胎心率基线序列 B (n)和所述胎心率数据预处理模块 发送的预处理后的序列 V (n), 根据预设的减速判断标准和所述的胎 心率数据基线序列 B (n)对所述的预处理后的胎心率数据序列 V (n)进 行减速识别, 得到减速数据段; 说 明 书 所述的输出模块与所述的胎心率数据减速识别模块连接, 用于 接收胎心率数据减速识别模块发送的所述的减速数据段, 并进行输 出。
作为本发明进一歩的技术方案,
所述的提高胎心率数据减速识别准确性的装置还包括, 胎心信 号采集与处理转换模块、 胎心率数据减速校验模块、 胎心率数据减 速判断标准设定模块、 胎心率数据减速类型判断模块、 采集时长设 定与判断模块和胎心率数据减速属性计算模块,
所述的胎心信号采集与处理转换模块与所述的胎心率数据采集 模块相连接, 用于采集胎心信号, 并转换成胎心率数据, 发送胎心 率数据到所述的胎心率数据采集模块;
所述的胎心率数据减速校验模块与所述的胎心率数据减速识别 模块和胎心率数据减速属性计算模块连接, 用于接收所述的胎心率 数据减速识别模块发送的减速数据段, 根据每个减速的信号损失情 况, 校验每个减速序列段是否为真正的减速, 并将校验结果发送到 所述的胎心率数据减速属性计算模块;
所述的胎心率数据减速判断标准设定模块与所述的胎心率数据 减速识别模块相连接, 用于预先对胎心率数据减速判断标准进行设 定, 并发送到胎心率数据减速识别模块;
所述的胎心率数据减速类型判断模块与所述的胎心率数据减速 属性计算模块连接, 用于在同时釆集宫縮数据的情况下进行减速类 型的判断, 并将判断结果发送到胎心率数据减速属性计算模块; 所述的采集时长设定与判断模块与所述的胎心率数据采集模块 相连接, 用于对胎心率数据采集的时长进行设定并判断采集数据时 间是否超过设定时长, 若超过设定时长, 则向胎心率数据采集模块 发送超时信号。
所述的胎心率数据减速属性计算模块与所述的胎心率数据减速 说 明 书 识别模块和输出模块连接, 用于接收所述的胎心率数据减速识别模 块发送的减速数据段, 并计算每个减速数据段的持续时间和幅度等 减速属性值, 并将计算结果发送到所述的输出模块;
作为本发明更进一步的技术方案,
所述的胎心率数据预处理模块, 还进一步包括,
错误数据处理单元, 对所述的胎心率数据序列 H (n)进行错误数 据处理, 得到有效胎心率数据序列 V (n) ;
插值处理单元, 对所述的有效胎心率数据序列 V (n)中无效数据 部分采用线性插值方法进行拟合, 得到胎心率数据序列 C (n)。
作为本发明更进一步的技术方案, 所述的胎心率数据减速识别 模块, 还进一步包括,
标准比较单元, 用于判断所述的胎心率数据序列 C (n)中是否有 满足预设减速判断标准的序列段;
区分减速与变异单元, 用于根据满足判断标准的胎心率数据序 列 C (n)及其对应的基线序列 B (η) ,检测并区分满足标准的序列 C (n) 中减速和基线变异的序列段;
分析减速个数单元, 用于分析所述满足标准的序列 C (n)中减速 序列段的连续减速个数。
作为本发明更进一步的技术方案, 所述的输出模块, 还进一步 包括,
显示单元, 用于将识别到的减速数据段及其属性值进行显示; 打印单元, 用于将识别到的减速数据段及其属性值进行打印输 出;
存储单元, 用于将识别到的减速数据段及其属性值进行存储; 标识单元, 用于将识别到的减速数据段及其属性值进行标识。 本发明一种提高胎心率数据减速识别准确性的方法, 包括: 步骤 1,预订时长内进行胎心率数据的采集,得到胎心率数据序 说 明 书 列 H(n);
歩骤 2, 对所述的胎心率数据序列 H(n)进行基线识别, 得到胎 心率基线数据序列 B(n);
歩骤 3, 对所述的胎心率数据序列 H(n)进行预处理, 得到预处 理后的胎心率数据序列 C(n);
步骤 4,根据预设的减速判断标准和所述的胎心率基线数据序列 B(n)对预处理后的胎心率数据序列 C(n)进行减速识别, 得到减速数 据段;
步骤 5,将每个所述的减速数据段及其减速属性值的计算结果进 行输出。
作为本发明进一步的技术方案, 在所述的步骤 1之前还包括, 采集并处理胎心信号转换得到胎心率数据。
作为本发明更进一步的技术方案, 所述的步骤 3进一歩包括: 步骤 31,对所述的胎心率数据序列 H(n)进行错误数据处理得到 序列 V(n);
步骤 32, 对所述的序列 V(n) 进行插值处理得到预处理后的胎 心率数据序列 C(n);
作为本发明更进一步的技术方案, 所述的步骤 4进一步包括: 步骤 41, 将所述的序列 C(n)、 B(n)输入到预设的减速判断标准 中,得出序列 C(n)中满足减速标准的各个序列段的集合 {{C, 以及 其对应的基线序列段集合 ,如果没有满足减速判断标准的序 列段, 则回到所述的步骤 1, 重新对胎心率数据的采集;
步骤 42, 将所述的 . 与 作差, 得到序列段 , 在序 列段 {A 中寻找不超过阈值 R。的连续的序列段,如果 中没有满 足此条件的序列段, 则判断序列段 为减速序列段, 如果 {ς^中 有满足此条件的序列段, 则 {ς 中满足此条件对应位置的序列段判 断为基线变异部分,记基线变异部分为 ,所述的基线变异部 说 明 书 分将 c^'分割成为若干个片段,记为^ 对 ^; 判断是否满足 减速标准, 如果其满足减速标准则 为减速序列段, 否则则为基 线变异序列段, 其中 R。为预先设定的参数;
歩骤 43,在所述减速序列段 中寻找偏离其对应基线数值不 超过阈值 R1的连续的序列段, 如果没有满足这样条件的序列段, 则 为一个减速, 如果有满足这样条件的序列段, 则记为 {{ } '; }"2, 然后分析每个分段 是否能单独满足减速标准, 如果能 ^足,则 独立成为减速,如果不满足,则需要将 合并到片段 中, 使其相邻片段连续并一起分析是否满足减速 标准, 如果满足减速标准则此连续数据段为一个减速, 并继续对后 续的片段数据重新进行减速标准分析, 否则基线合并片段直到片段 全部被合并为止, 其中 R1为预先设定的参数;
作为本发明更进一步的技术方案,在所述的步骤 4之后还包括, 根据所述的减速数据段, 判断每个减速数据段的信号损失情况, 校 验每个减速序列段是否为真正的减速;
作为本发明更进一步的技术方案, 所述的根据所述的减速数据 段, 判断每个减速数据段的信号损失情况, 校验每个减速序列段是 否为真正的减速的步骤进一步包括:
a. 比较所述的序列 V (n)和 C (n), 对有插值的位置做标记, 得 到标记序列 M (n) ;
b.对于每个所述的减速数据段,根据序列 M (n)调整减速的起点、 终点的位置, 使起点、 终点都不为插值点, 且离插值点最近;
c 计算调整后的减速中的信号损失程度 , 如果 超过阈值 ^, 则撤销该减速的资格, 否则保留减速的资格, 其中 f为预先设定的 参数。
作为本发明更进一歩的技术方案, 所述的歩骤 5之前还包括, 计算每个所述的减速数据段的减速属性值; 说 明 书 作为本发明更进一步的技术方案, 所述的计算每个所述的减速 数据段的减速属性值, 还进一歩包括:
步骤 51, 根据每个减速的起点 、 终点 计算持续时间 , 计算减速中偏离基线的最大幅度 ^, 并记录该点 (波峰点) 的位置
DP
步骤 52,判断是否同时采集宫缩数据,如果没有采集宫缩数据, 则不判断减速的类型, 如果釆集了宫縮数据, 则根据预设减速类型 判断条件判断所述的减速数据段是否为早发减速和 /或迟发减速和 / 或变化减速和 /或延长减速。
作为本发明更进一步的技术方案, 所述的步骤 6进一步包括: 将减速数据段和属性值计算结果进行显示和 /或打印和 /或存储和 / 或标识。
作为本发明更进一步的技术方案, 步骤 4中所述的预设的减速 判断标准, 进一步包括:
第一: 该段胎心率曲线必须都在胎心率基线之下。
第二: 在胎心率基线之下的持续时间 必须大于等于阈值 ^。 第三: 偏离基线的最大幅度 必须大于等于阈值 ^。
第四:设该段胎心率曲线及其对应的基线段分别为 {ς 、 {Β^ , 将 与^¾作差,得到序列段 对 ¾采用步长为/ ζ的等分 的复化梯形公式得到面积 S, S必须大于等于动态阈值 α χ ί χ Γ和静 态阈值 中的最小值。
若满足以上所述的全部条件则判断所述的待分析的胎心率数据 段为减速数据段, 其中 Α0、 Τ0、 "是预先设定的参数。
本发明通过采用动态阈值面积法、 连续波峰检测和混合减速检 测等方法, 可以有效地剔除基线变异部分, 准确地识别出每个减速 及其类型, 避免了现有的方法容易造成减速识别误差大, 不能识别 连续减速和剔除基线变异, 不能识别混合减速等不足。 该方案既适 说 明 书 合减速独立出现, 为典型减速的情况, 又适合出现连续减速、 混合 减速的情况。
附图说明
为了易于说明, 本发明由下述的较佳实施例及附图作以详细描 述。
图 1为本发明一种提高胎心率数据减速识别准确性的方法的一 种实施例流程图;
图 2为本发明一种提高胎心率数据减速识别准确性的方法的另 一种实施例流程图;
图 3为本发明一种提高胎心率数据减速识别准确性的装置的一 种实施例架构图;
图 4为本发明一种提高胎心率数据减速识别准确性的装置的另 一种实施例架构图;
图 5为采集的胎心率数据和识别出的胎心率基线示意图; 图 6为胎心率数据序列预处理后的效果图;
图 7为胎心率数据序列的减速识别效果图;
图 8为胎心率数据序列的减速校验效果图;
图 9为胎心率数据序列的有宫缩数据时减速类型识别的效果图。 具体实施方式
为了使本发明的目的、 技术方案及优点更加清楚明白, 以下结 合附图及实施例, 对本发明进行进一步详细说明。 应当理解, 此处 所描述的具体实施例仅仅用以解释本发明, 并不用于限定本发明。
用于胎心率减速自动识别的技术方案实现主要包括上位机和下 位机, 下位机主要是通过超声探头来获取胎心信号, 然后经过硬件 滤波处理和自相关算法计算出胎心率; 上位机主要是接受来自下位 机的数据, 然后显示、 存储胎心率数据、 描绘胎心率曲线、 对胎心 率数据的分析 (包括对胎心率数据减速的识别等等)、 显示、 打印, 说 明 书 其中主要处理方法流程如图 1所示:
101.进行胎心率数据采集;
上位机采集 n分钟的胎心率数据, 得到胎心率数据序列 H (n), 其中, 本实施例中设定需要至少采集十分钟的数据, 根据 《胎儿电 子监护学》 , 一般而言, 确定胎心率基线至少需要十分钟的胎心率 数据。
102.对采集的胎心率数据进行基线识别;
根据序列 H (n), 采用低通滤波方法识别出胎心率基线数据序列 B (n), 相对于胎心率数据, 基线属于低频信号, 低通滤波器具有能 使低频信号通过而高频信号被阻断的特性, 可用于提取胎心率基线 这种低频信号。 另外在此步中也可以先对序列 H (n)进行数据处理, 如错误数据处理、 均值滤波、 线性插值等方法得到处理后的数据再 进行胎心率基线识别, 图 5为采集的胎心率数据和识别出的胎心率 基线。
103.对采集的胎心率数据进行预处理;
对序列 H (n)依次进行错误数据处理和线性插值的方法进行处 理, 得到不同预处理过程所对应的相应的胎心率数据序列。
104.根据预设的减速判断标准和胎心率数据基线对预处理后的 胎心率数据进行减速识别, 得到减速数据段;
通过预设的判断标准判断出符合条件的序列段, 再根据此序列 段及其对应的基线序列段, 检测并区分此序列段中减速和基线变异 的序列段, 并分析减速序列段的连续减速个数。
105.计算每个减速数据段的减速属性值;
根据每个减速的起点、 终点计算其持续时间, 计算减速中偏离 基线的最大幅度值。
106.将减速数据段和计算结果输出;
本实施例优选为将识别到的减速数据段和计算结果进行显示和 说 明 书
/或打印和 /或存储。
为了更好的理解本发明, 作为本发明的另一个实施例, 一种提 高胎心率数据减速识别准确性的方法的流程图如图 2所示:
100.采集并处理胎心信号转换得到胎心率数据;
超声探头接收到的胎心信号在下位机经过硬件滤波处理和自相 关算法计算出胎心率后上传到上位机, 上位机的处理包括对胎心率 数据进行显示、 存储和描绘成胎心率曲线以及相关数据处理等等。 其中硬件滤波为了去除采集到的信号受到的频率干扰而自相关算法 是具有能使周期信号得到加强而随机噪声被减弱的特性, 是计算胎 心率数据的常用技术。
101. 进行胎心率数据采集, 得到胎心率数据序列 H (n) ;
上位机采集 n分钟的胎心率数据, 得到胎心率数据序列 H (n), 其中, 本实施例中设定需要至少采集十分钟的数据, 根据 《胎儿电 子监护学》 , 一般而言, 确定胎心率基线至少需要十分钟的胎心率 数据。
102. 对序列 H (n)进行胎心率基线识别得到序列 B (n);
根据序列 H (n), 采用低通滤波方法识别出胎心率基线 B (n), 相 对于胎心率数据, 基线属于低频信号, 低通滤波器具有能使低频信 号通过而高频信号被阻断的特性, 可用于提取胎心率基线这种低频 信号。 另外在此步中也可以先对序列 H (n)进行数据处理, 如错误数 据处理、 均值滤波、 线性插值等方法得到处理后的数据再进行胎心 率基线识别, 图 5为采集的胎心率数据和识别出的胎心率基线。 另 外, 进行胎心率数据基线识别的方式还有很多, 比如业界通常采用 的平均值法等等。
1031. 对 H (n)进行错误数据处理得到序列 V (n);
对序列 H (n)进行错误数据处理,得到有效胎心率数据序列 V (n), 此步可以有效的滤除胎心序列中因胎心信号质量差而产生的无效、 说 明 书 错误数据, 剩下的是有效数据。
1032. 对 V (n)进行插值处理得到序列 C (n);
对序列 V (n)中无效数据部分采用线性插值方法进行拟合, 得到 序列 C (n), 线性插值是利用两点成线原理求解直线上其他点的简单 的插值方法。 图 6为胎心率数据序列预处理后的效果图, 图中方框 内的胎心率曲线部分, 也即为对错误数据进行插值处理后所得的曲 线段。
如上所述的 1031和 1032步骤为对采集的胎心率数据进行预处 理, 此外, 对胎心率数据预处理步骤, 并不局限于如上步骤, 还可 以采用业界悉知的对以每分钟多少跳 (bpm, beats per min ) 为单位 记的胎心率值去除错误数据和 /或滑动平均和 /或插值拟合的方法等 等。
1041. 序列 C (n)中是否有满足预设减速判断标准的序列段; 在本实施例中优选预设的减速判断标准为:
对于一段胎心率曲线,必须满足以下全部条件才能确定为减速: 第一: 该段胎心率曲线必须都在胎心率基线之下。
第二: 在胎心率基线之下的持续时间 必须大于等于阈值 r。。 第三: 偏离基线的最大幅度 ^必须大于等于阈值 Λ。
第四: 根据复化数值积分原理对该段胎心率曲线求面积 s, 其 过程为: 设该段胎心率曲线及其对应的基线段分别为 {ς 、 [Β^ , 将 与^ 作差,得到序列段 {Α ,对 采用步长为 A的等分 的复化梯形公式得到面积^ S必须大于等于动态阈值《χ ί χ Γ和静 态阈值 中的最小值。
该判断条件丰富了判断的标准, 使得判断更加准确, 其中参数
Α。、 Τ。、 β、 «是预设的经验参数, 且随着孕周参数的输入的不同而 会设置不同变化。
将序列 C (η)、 Β (η)输入到减速判断标准中, 得出 C (η)中满足减 说 明 书 速标准的各个序列段的集合 {{c^L以及其对应的基线序列段集合 如果没有满足减速标准的序列段, 则说明该胎心率曲线中 没有减速, 此时即可结束胎心率减速的识别, 重新回到对胎心率数 据的采集。
1042.根据满足标准的序列 C (n)及其对应的基线序列 B (η) ,检测 并区分满足标准的序列 C (n)中减速和基线变异的序列段;
对于每个序列段 , 将 {ς 与 作差, 得到序列段 ^ , 在序列段 {A 中寻找值不超过阈值 R。的连续的序列段,如果 中 没有满足此条件的序列段,则序列段 为一个或者连续减速组成, 将在下一步用连续波峰检测算法分析减速个数; 如果 中有满足 此条件的序列段, 则 中对应位置的序列段即为基线变异部分, 记基线变异部分为 {{ t, 其中 {ς^ ^ς , 此时基线变异部分 会将 {ς 分割成为若干个1片段, 记为 ^}^2, 其中 ^ ¾, 对 判断是否满足减速标准, 如果其满足减速标准则 为一 个或者由连续减速组成, 将在下一歩用连续波峰检测算法分析减速 的个数; 如果其不满足减速标准, 贝 ij ^ 为基线变异部分, 通过这 一歩可以区分减速和基线变异部分。 '
1043. 分析减速序列段的连续减速个数;
对于每个待分析的胎心率序列片段(此处可能为 { 、ς 或者是 ) sk' , 其 中 { {q. , 采用连续波峰检测算法分析的过程为: 在片段中寻找偏离对 应基线不超过阈值 Ri的连续的序列段, 如果没有满足这样条件的序列段 (全部 超过阈值 RI), 贝 为一个减速; 如果有满足这样条件的序列段, 则记为
{{c"fc}^:} 2,其中 "^,' ,然后分析每个分段 '^,':是否能单独满 足减速标准, 如果能满足, 则 独立成为减速, 如果不满足, 则需要将 合并到片段 "^— 中, 使其相邻片段连续并一起分析是否满足 减速标准, 如果满足减速标准则此连续数据段为一个减速, 并继续 对后续的片段数据重新进行减速标准分析, 否则基线合并片段直到 说 明 书 片段全部被合并为止。 通过这一步将连续减速识别出来。 图 7为减 速识别效果图,图中方框中胎心率曲线部分为基线变异部分,箭头 所指的曲线段部分为识别到的胎心率减速曲线段。
如上所述的步骤: 1041、 1042、 1043为对与处理后的采集的胎 心率数据进行减速识别, 到本步骤为止已经可以完成本发明对减速 数据段进行识别, 可直接进入如下的 105步骤和 106步骤进行属性 值计算和结果输出或者直接进入 106步骤进行结果的输出。 为了使 得结果更加准确本实施例中还可以优选在计算和结果输出步骤之前 采用如下的减速校验步骤。
107. 根据每个减速的信号损失情况,校验每个减速序列段是否 为真正的减速;
如上步骤中的减速是根据插值后的序列求得的, 也就是在不考 虑信号损失的情况下求得的。 本步中要对减速进行校验, 分析每个 预减速中的信号损失, 以决定一个预减速是否被保留、 拆分, 最终 得到真正的减速。 该歩骤还包括以下处理过程:
第一步: 比较序列 V (n)和 C (n), 对有插值的位置做标记, 得到 标记序列 M (n)。
第二步: 对于每个减速, 根据序列 M (n)调整减速的起点、 终点 的位置, 使起点、 终点都不为插值点, 且离插值点最近。
第三步: 计算调整后的减速中的信号损失程度 , 信号损失度 为胎心率减速曲线数据中无效值的个数及(等同于插值点的个数)占 整体的比例, 如果 超过阈值 f, 则撤销该减速的资格; 否则保留减 速的资格。 图 8为减速校验效果图, 图中箭头图标表示减速。
105、 计算各个减速的持续时间、 幅度和 /或类型;
根据每个减速的起点 、终点 计算其持续时间 ,计算减速 中偏离基线的最大幅度 A, 并记录该点 (波峰点) 的位置 。
另外, 如果在同时采集宫缩数据的时候, 可以判断减速的类型, 说 明 书 如果没有采集宫缩数据, 则不判断减速的类型, 最终的减速显示效 果图与图 8—致; 如果采集了宫缩数据, 则对每个减速采用混合减 速检测算法确定该减速的类型的过程为: 判断条件一: 减去 ^得 到的值 D™大于等于阈值 , 减去宫缩起点 ^的绝对值小于等于 阈值 , ^减去宫缩波峰点位置 ^的绝对值小于等于阈值 ; 判断 条件二: 1 ^减去 得到的值1^大于等于阈值 , 减去1 ^得到的 值 s大于等于阈值 , DW咸去宫缩波峰点位置 ^大于等于阈值7^ 且小于等于阈值7^ 大于宫缩的终点 判断条件三: ^减去 得到的值1^小于阈值7^ ^大于等于阈值 且小于阈值 3, ^大 于等于阈值 判断条件四: ^大于等于阈值 且小于阈值 2, DA 大于等于阈值 4。 如果所判断的减速数据满足条件一则判断为早发 减速, 满足条件二则判断为迟发减速, 满足条件三则判断为变化减 速, 满足条件四则判断为延长减速。 将每个减速数据输入这四个判 断条件中, 如果仅满足一个条件, 则该减速判断为典型减速, 也即 单一类型减速; 如果满足多个条件, 则该减速数据判断为不典型减 速, 也即混合减速类型, 同时可以根据该减速所满足的条件确定该 减速为哪些减速类型的混合。 图 9为有宫缩数据时减速类型识别效 果图, 图中 表示早发减速, 表示变化减速, !1表示早发减速与变 化减速混合, 01表示迟发减速与变化减速混合。
106. 将减速数据段和属性值计算结果进行显示和 /或打印和 / 或存储和 /或标识;
将识别到的减速数据段分别送入显示、 打印、 存储模块, 并且 还可以在胎监图上标识并显示出每个减速数据段及其属性值, 并可 存储和打印其属性值。
为了更好的解释本发明, 一种提高胎心率数据减速识别准确性 的装置的一种实施例架构图, 如图 3所示:
胎心率数据采集模块 301 : 说 明 书 用于采集预设时长的胎心率数据, 得到胎心率数据序列 H (n); 胎心率数据基线识别模块 302 :
与所述的胎心率数据采集模块 301连接, 用于接收所述胎心率 数据采集模块 301发送的胎心率数据序列 H (n), 并且识别出胎心率 基线序列 B (n) ;
胎心率数据预处理模块 303 :
与所述的胎心率数据采集模块 301连接, 用于接收所述胎心率 数据采集模块 301发送的胎心率数据序列 H (n),并且对序列 H (n)进 行预处理得到胎心率数据序列 V (n);
胎心率数据减速识别模块 304:
与所述的胎心率数据基线识别模块 302和胎心率数据预处理模 块 303连接, 用于接收所述胎心率数据基线识别模块 302发送的胎 心率基线序列 B (n)和所述胎心率数据预处理模块 303发送的预处理 后的序列 V (n), 根据预设的减速判断标准和所述的胎心率数据基线 序列 B (n)对所述的预处理后的胎心率数据序列 V (n)进行减速识别, 得到减速数据段;
胎心率数据减速属性计算模块 305:
与所述的胎心率数据减速识别模块 304连接, 用于接收所述的 胎心率数据减速识别模块 304发送的减速数据段, 并计算每个减速 数据段的持续时间和幅度等减速属性;
输出模块 306 :
与所述的胎心率数据减速属性计算模块 305连接, 用于接收胎 心率数据减速属性计算模块 305发送的所述的减速数据段及其减速 属性计算结果, 并进行输出。
为了更好的解释本发明, 一种提高胎心率数据减速识别准确性 的装置的另一种实施例架构图, 如图 4所示:
一种提高胎心率数据减速识别准确性的装置, 还包括: 说 明 书 胎心信号采集与处理转换模块 300:
与所述的胎心率数据采集模块 301相连接, 是一个实现信号采 集、 处理和转换的超声探头装置, 用于采集胎心信号, 并转换成胎 心率数据, 发送胎心率数据到所述的胎心率数据采集模块 301 ; 胎心率数据减速校验模块 312 :
与所述的胎心率数据减速识别模块 304和胎心率数据减速属性 计算模块 305连接, 用于接收所述的胎心率数据减速识别模块 304 发送的减速数据段, 根据每个减速的信号损失情况, 校验每个减速 序列段是否为真正的减速, 并将校验结果发送到所述的胎心率数据 减速属性计算模块 305;
胎心率数据减速判断标准设定模块 313 :
与所述的胎心率数据减速识别模块 304相连接, 用于预先对胎 心率数据减速判断标准进行设定, 并发送到胎心率数据减速识别模 块 304;
胎心率数据减速类型判断模块 314:
与所述的胎心率数据减速属性计算模块 305连接, 用于在同时 采集宫缩数据的情况下进行减速类型的判断, 并将判断结果发送到 胎心率数据减速属性计算模块 305;
采集时长设定与判断模块 319:
与所述的胎心率数据采集模块 301相连接, 用于对胎心率数据 采集的时长进行设定并判断釆集数据时间是否超过设定时长, 若超 过设定时长, 则向胎心率数据采集模块 301发送超时信号;
另外, 所述的胎心率数据预处理模块 303, 进一步包括了: 错误数据处理单元 307, 对所述的胎心率数据序列 H (n)进行错 误数据处理, 得到有效胎心率数据序列 V (n) ;
插值处理单元 308, 对所述的有效胎心率数据序列 V (n)中无效 数据部分采用线性插值方法进行拟合, 得到胎心率数据序列 C (n); 说 明 书 所述的胎心率数据减速识别模块 304, 进一步包括了: 标准比较单元 309, 用于判断所述的胎心率数据序列 C (n)中是 否有满足预设减速判断标准的序列段;
区分减速与变异单元 310,用于根据满足判断标准的胎心率数据 序列 C (n)及其对应的基线序列 B (n) ,检测并区分满足标准的序列 C (n)中减速和基线变异的序列段;
分析减速个数单元 311, 用于分析所述满足标准的序列 C (n)中 减速序列段的连续减速个数;
所述的输出模块 306, 进一步包括了:
显示单元 315, 用于将识别到的减速数据段及其属性值进行显 示;
打印单元 316,用于将识别到的减速数据段及其属性值进行打印 输出;
存储单元 317, 用于将识别到的减速数据段及其属性值进行存 储;
标识单元 318, 用于将识别到的减速数据段及其属性值进行标 识;
以上所述仅为本发明的较佳实施例而已,并不用以限制本发明, 凡在本发明的精神和原则之内所作的任何修改、等同替换和改进等, 均应包含在本发明的保护范围之内。

Claims

权 利 要 求 书
1.一种提高胎心率数据减速识别准确性的装置, 其特征在于, 包括: 胎心率数据采集模块、胎心率数据基线识别模块、胎心率数据预处理模块、 胎心率数据减速识别模块和输出模块,
所述的胎心率数据采集模块, 用于采集预设时长的胎心率数据, 得到 胎心率数据序列 H (n) ;
所述的胎心率数据基线识别模块与所述的胎心率数据采集模块连接, 用于接收所述胎心率数据采集模块发送的胎心率数据序列 H (n),并且识别 出胎心率基线序列 B (n) ;
所述的胎心率数据预处理模块与所述的胎心率数据采集模块连接, 用 于接收所述胎心率数据采集模块发送的胎心率数据序列 H (n),并且对所述 的序列 H (n)进行预处理得到胎心率数据序列 V (n);
所述的胎心率数据减速识别模块与所述的胎心率数据基线识别模块 和胎心率数据预处理模块连接, 用于接收所述胎心率数据基线识别模块发 送的胎心率基线序列 B (n)和所述胎心率数据预处理模块发送的预处理后 的序列 V (n),根据预设的减速判断标准和所述的胎心率数据基线序列 B (n) 对所述的预处理后的胎心率数据序列 V (n)进行减速识别, 得到减速数据 段;
所述的输出模块与所述的胎心率数据减速识别模块连接, 用于接收胎 心率数据减速识别模块发送的所述的减速数据段, 并进行输出。
2.根据权利要求 1 所述的一种提高胎心率数据减速识别准确性的装 置, 其特征在于, 所述的提高胎心率数据减速识别准确性的装置还包括, 胎心信号采集与处理转换模块、 胎心率数据减速校验模块、 胎心率数据减 速判断标准设定模块、 胎心率数据减速类型判断模块、 采集时长设定与判 断模块和胎心率数据减速属性计算模块,
所述的胎心信号采集与处理转换模块与所述的胎心率数据采集模块 相连接, 用于采集胎心信号, 并转换成胎心率数据, 发送胎心率数据到所 述的胎心率数据采集模块; 权 利 要 求 书 所述的胎心率数据减速校验模块与所述的胎心率数据减速识别模块 和胎心率数据减速属性计算模块连接, 用于接收所述的胎心率数据减速识 别模块发送的减速数据段, 根据每个减速的信号损失情况, 校验每个减速 序列段是否为真正的减速, 并将校验结果发送到所述的胎心率数据减速属 性计算模块;
所述的胎心率数据减速判断标准设定模块与所述的胎心率数据减速 识别模块相连接, 用于预先对胎心率数据减速判断标准进行设定, 并发送 到胎心率数据减速识别模块;
所述的胎心率数据减速类型判断模块与所述的胎心率数据减速属性 计算模块连接, 用于在同时采集宫缩数据的情况下进行减速类型的判断, 并将判断结果发送到胎心率数据减速属性计算模块;
所述的采集时长设定与判断模块与所述的胎心率数据采集模块相连 接, 用于对胎心率数据采集的时长进行设定并判断采集数据时间是否超过 设定时长, 若超过设定时长, 则向胎心率数据采集模块发送超时信号。
所述的胎心率数据减速属性计算模块与所述的胎心率数据减速识别 模块和输出模块连接, 用于接收所述的胎心率数据减速识别模块发送的减 速数据段, 并计算每个减速数据段的持续时间和幅度等减速属性值, 并将 计算结果发送到所述的输出模块;
3. 根据权利要求 1或 2所述的一种提高胎心率数据减速识别准确性的 装置, 其特征在于, 所述的胎心率数据预处理模块, 还进一步包括,
错误数据处理单元, 对所述的胎心率数据序列 H (n)进行错误数据处理, 得到有效胎心率数据序列 V (n); 插值处理单元,对所述的有效胎心率数据序列 V (n)中无效数据部分采 用线性插值方法进行拟合, 得到胎心率数据序列 C (n)。
4. 根据权利要求 1或 2所述的一种提高胎心率数据减速识别准确性 的装置, 其特征在于, 所述的胎心率数据减速识别模块, 还进一步包括, 权 利 要 求 书 标准比较单元, 用于判断所述的胎心率数据序列 C (n)中是否有满足预 设减速判断标准的序列段;
区分减速与变异单元, 用于根据满足判断标准的胎心率数据序列 C (n) 及其对应的基线序列 B (n) ,检测并区分满足标准的序列 C (n)中减速和基线 变异的序列段;
分析减速个数单元,用于分析所述满足标准的序列 C (n)中减速序列段 的连续减速个数。
5. 根据权利要求 1或 2所述的一种提高胎心率数据减速识别准确性 的装置, 其特征在于, 所述的输出模块, 还进一步包括,
显示单元, 用于将识别到的减速数据段及其属性值进行显示; 打印单元, 用于将识别到的减速数据段及其属性值进行打印输出; 存储单元, 用于将识别到的减速数据段及其属性值进行存储; 标识单元, 用于将识别到的减速数据段及其属性值进行标识。
6. 一种提高胎心率数据减速识别准确性的方法, 其特征在于, 包括: 步骤 1, 预订时长内进行胎心率数据的采集, 得到胎心率数据序列
H (n) ;
步骤 2, 对所述的胎心率数据序列 H (n)进行基线识别, 得到胎心率基 线数据序列 B (n);
步骤 3, 对所述的胎心率数据序列 H (n)进行预处理, 得到预处理后的 胎心率数据序列 C (n) ;
步骤 4, 根据预设的减速判断标准和所述的胎心率基线数据序列 B (n) 对预处理后的胎心率数据序列 C (n)进行减速识别, 得到减速数据段;
步骤 5, 将每个所述的减速数据段及其减速属性值的计算结果进行输 出。
7.根据权利要求 6所述的一种提高胎心率数据减速识别准确性的方 法, 其特征在于, 在所述的步骤 1之前还包括, 采集并处理胎心信号转换 权 利 要 求 书 得到胎心率数据。
8. 根据权利要求 6所述的一种提高胎心率数据减速识别准确性的方 法, 其特征在于, 所述的步骤 3进一步包括:
步骤 31, 对所述的胎心率数据序列 H (n)进行错误数据处理得到序列 V (n) ;
步骤 32, 对所述的序列 V (n) 进行插值处理得到预处理后的胎心率数 据序列 C (n);
9. 根据权利要求 6所述的一种提高胎心率数据减速识别准确性的方 法, 其特征在于, 所述的步骤 4进一步包括:
步骤 41, 将所述的序列 C (n)、 B (n)输入到预设的减速判断标准中, 得出序列 C (n)中满足减速标准的各个序列段的集合 {C^' 以及其对应的 基线序列段集合 { ^ , 如果没有满足减速判断标准的序列段, 则回到 所述的步骤 1, 重新对胎心率数据的采集;
步骤 42,将所述的 与 ^作差,得到序列段 {A ,在序列段 中寻找不超过阈值 R。的连续的序列段, 如果 { , ^中没有满足此条件的序 列段, 则判断序列段 为减速序列段, 如果 {ς 中有满足此条件的序
Figure imgf000023_0001
步骤 43, 在所述减速序列段 中寻找偏离其对应基线数值不超过 阈值 R1的连续的序列段, 如果没有满足这样条件的序列段, 则 为一 个减速, 如果有满足这样条件的序列段, 则记为 然后分析每
Figure imgf000023_0002
个分段 '^^是否能单独满足减速标准,如果能满足,则 独立成为 减速, 如果不满足, 则需要将 '^^合并到片段 中, 使其相邻片段 连续并一起分析是否满足减速标准, 如果满足减速标准则此连续数据段为 权 利 要 求 书 一个减速, 并继续对后续的片段数据重新进行减速标准分析, 否则基线合 并片段直到片段全部被合并为止, 其中 R1为预先设定的参数;
10.根据权利要求 6所述的一种提高胎心率数据减速识别准确性的方 法, 其特征在于, 在所述的歩骤 4之后还包括, 根据所述的减速数据段, 判断每个减速数据段的信号损失情况, 校验每个减速序列段是否为真正的 减速;
11. 根据权利要求 10所述的一种提高胎心率数据减速识别准确性的 方法, 其特征在于, 所述的根据所述的减速数据段, 判断每个减速数据段 的信号损失情况, 校验每个减速序列段是否为真正的减速的步骤进一步包 括:
a. 比较所述的序列 V (n)和 C (n), 对有插值的位置做标记, 得到标记 序列 M (n) ;
b. 对于每个所述的减速数据段, 根据序列 M (n)调整减速的起点、 终 点的位置, 使起点、 终点都不为插值点, 且离插值点最近;
c 计算调整后的减速中的信号损失程度 , 如果 超过阈值 f, 则撤 销该减速的资格, 否则保留减速的资格, 其中 f为预先设定的参数。
12. 根据权利要求 6或 10所述的一种提高胎心率数据减速识别准确 性的方法, 其特征在于, 所述的步骤 5之前还包括, 计算每个所述的减速 数据段的减速属性值;
13. 根据权利要求 12所述的一种提高胎心率数据减速识别准确性的 方法, 其特征在于, 所述的计算每个所述的减速数据段的减速属性值, 还 进一步包括:
步骤 51, 根据每个减速的起点 、 终点 A计算持续时间 , 计算减 速中偏离基线的最大幅度 , 并记录该点 (波峰点) 的位置 >。
步骤 52, 判断是否同时釆集宫缩数据, 如果没有采集宫缩数据, 则不 判断减速的类型, 如果采集了宫缩数据, 则根据预设减速类型判断条件判 断所述的减速数据段是否为早发减速和 /或迟发减速和 /或变化减速和 /或 权 利 要 求 书 延长减速。
14. 根据权利要求 6所述的一种提高胎心率数据减速识别准确性的方 法, 其特征在于, 所述的步骤 6进一步包括: 将减速数据段和属性值计算 结果进行显示和 /或打印和 /或存储和 /或标识。
15. 根据权利要求 6所述的一种提高胎心率数据减速识别准确性的方 法, 其特征在于, 步骤 4中所述的预设的减速判断标准, 进一步包括: 第一: 该段胎心率曲线必须都在胎心率基线之下。
第二: 在胎心率基线之下的持续时间 必须大于等于阈值 Γ。。
第三: 偏离基线的最大幅度 ^必须大于等于阈值 Λ。
第四: 设该段胎心率曲线及其对应的基线段分别为 ^、 {Β^ , 将 ¾与^¾作差, 得到序列段 {Α , 对 ¾采用步长为 Α的等分的复化 梯形公式得到面积 S, S必须大于等于动态阈值《χ
Figure imgf000025_0001
Γ和静态阈值 中的 最小值。
若满足以上所述的全部条件则判断所述的待分析的胎心率数据段为 减速数据段, 其中 Α。、 Τ。、 β、 «是预先设定的参数。
PCT/CN2012/082019 2011-10-13 2012-09-26 一种提高胎心率数据减速识别准确性的装置和方法 WO2013053290A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP12839654.6A EP2767228B1 (en) 2011-10-13 2012-09-26 Device and method for improving identification accuracy of fetal heart rate deceleration
US14/349,326 US9364197B2 (en) 2011-10-13 2012-09-26 Device and method for improving identification accuracy of fetal heart rate deceleration

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN2011103096799A CN102319064B (zh) 2011-10-13 2011-10-13 一种提高胎心率数据减速识别准确性的装置和方法
CN201110309679.9 2011-10-13

Publications (1)

Publication Number Publication Date
WO2013053290A1 true WO2013053290A1 (zh) 2013-04-18

Family

ID=45447001

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2012/082019 WO2013053290A1 (zh) 2011-10-13 2012-09-26 一种提高胎心率数据减速识别准确性的装置和方法

Country Status (4)

Country Link
US (1) US9364197B2 (zh)
EP (1) EP2767228B1 (zh)
CN (1) CN102319064B (zh)
WO (1) WO2013053290A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104586379A (zh) * 2015-01-21 2015-05-06 深圳市理邦精密仪器股份有限公司 一种胎心率曲线的参数输出方法及装置

Families Citing this family (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102940485B (zh) * 2012-10-10 2015-04-08 广州三瑞医疗器械有限公司 一种胎心率基线自动识别算法
CN103565433B (zh) * 2013-10-30 2015-02-25 深圳市理邦精密仪器股份有限公司 一种提高胎儿监护效率的方法和装置
CN103784130B (zh) * 2014-01-24 2016-10-05 深圳市理邦精密仪器股份有限公司 一种提高胎心率曲线效率的方法及系统
CN104586378A (zh) * 2015-01-21 2015-05-06 深圳市理邦精密仪器股份有限公司 一种胎心率曲线的输出方法及装置
CN105193451B (zh) * 2015-10-21 2018-02-27 北京工业大学 一种胎儿心率曲线基线的识别方法
EP3189776A1 (en) * 2016-01-08 2017-07-12 Koninklijke Philips N.V. An apparatus and method for generating fetal heart rate data
CN105866011B (zh) * 2016-03-31 2018-10-26 艾康生物技术(杭州)有限公司 脉冲基线值计算方法及血球分析仪的粒子计数方法
CN107865673A (zh) * 2016-09-23 2018-04-03 深圳市理邦精密仪器股份有限公司 胎心率的获取装置及方法、胎心监测设备
US10589109B2 (en) * 2017-04-10 2020-03-17 West Affum Holdings Corp. Wearable cardioverter defibrillator (WCD) system computing patient heart rate by multiplying ECG signals from different channels
CN108261176B (zh) * 2017-12-28 2021-06-04 深圳京柏医疗科技股份有限公司 胎心监护数据处理方法、装置、系统、存储介质和计算机设备
CN108742599A (zh) * 2018-03-20 2018-11-06 深圳大学 一种胎儿健康预警系统及方法
CN108903930A (zh) * 2018-04-26 2018-11-30 李坚强 一种胎心率曲线分类系统、方法及装置
CN109567868A (zh) * 2018-10-15 2019-04-05 广东宝莱特医用科技股份有限公司 一种ctg胎心评分方法及系统
CN110432884B (zh) * 2019-07-08 2022-07-01 暨南大学 基于胎心率减速区面积的胎儿状况测评方法及系统
CN112971753A (zh) * 2019-12-13 2021-06-18 深圳市理邦精密仪器股份有限公司 胎心率减速类型的识别方法、装置及胎儿监护设备
CN112971752A (zh) * 2019-12-13 2021-06-18 深圳市理邦精密仪器股份有限公司 胎心率减速类型的修正方法、装置及胎儿监护设备
CN114159040A (zh) * 2021-12-30 2022-03-11 徐智策 一种具备预警功能的胎心监护仪及监护方法

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1323012A (zh) * 2000-05-09 2001-11-21 权基澈 胎儿护理遥测系统
US20040133115A1 (en) * 2002-11-01 2004-07-08 Hamilton Emily F. Method and apparatus for identifying heart rate feature events
US20060229518A1 (en) * 2005-03-31 2006-10-12 Physical Logic Ag Method for continuous, non-invasive, non-radiating detection of fetal heart arrhythmia
CN1942135A (zh) * 2004-02-24 2007-04-04 诺文塔医药股份公司 通过胎儿心率分析对胎儿反应进行评定
US20090054797A1 (en) * 2007-06-13 2009-02-26 Miller Lisa A Electronic fetal monitoring assessment system and method
GB2471667A (en) * 2009-07-06 2011-01-12 Monica Healthcare Ltd Monitoring uterine activity with motion artifact removal
CN102048534A (zh) * 2010-12-17 2011-05-11 深圳市理邦精密仪器股份有限公司 一种可回放胎心多普勒音频信号的监护方法和装置
CN102048558A (zh) * 2011-01-28 2011-05-11 深圳市理邦精密仪器股份有限公司 一种胎心率信号处理方法及其装置
CN102090883A (zh) * 2010-12-30 2011-06-15 深圳市理邦精密仪器股份有限公司 一种胎动自动识别方法及其装置
CN102210586A (zh) * 2011-06-03 2011-10-12 泰安市迈迪医疗电子有限公司 用于胎儿监护装置的自动分析方法

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102319063B (zh) * 2011-07-26 2013-10-16 深圳市理邦精密仪器股份有限公司 一种提高胎心率数据加速识别准确性的装置和方法

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1323012A (zh) * 2000-05-09 2001-11-21 权基澈 胎儿护理遥测系统
US20040133115A1 (en) * 2002-11-01 2004-07-08 Hamilton Emily F. Method and apparatus for identifying heart rate feature events
CN1942135A (zh) * 2004-02-24 2007-04-04 诺文塔医药股份公司 通过胎儿心率分析对胎儿反应进行评定
US20060229518A1 (en) * 2005-03-31 2006-10-12 Physical Logic Ag Method for continuous, non-invasive, non-radiating detection of fetal heart arrhythmia
US20090054797A1 (en) * 2007-06-13 2009-02-26 Miller Lisa A Electronic fetal monitoring assessment system and method
GB2471667A (en) * 2009-07-06 2011-01-12 Monica Healthcare Ltd Monitoring uterine activity with motion artifact removal
CN102048534A (zh) * 2010-12-17 2011-05-11 深圳市理邦精密仪器股份有限公司 一种可回放胎心多普勒音频信号的监护方法和装置
CN102090883A (zh) * 2010-12-30 2011-06-15 深圳市理邦精密仪器股份有限公司 一种胎动自动识别方法及其装置
CN102048558A (zh) * 2011-01-28 2011-05-11 深圳市理邦精密仪器股份有限公司 一种胎心率信号处理方法及其装置
CN102210586A (zh) * 2011-06-03 2011-10-12 泰安市迈迪医疗电子有限公司 用于胎儿监护装置的自动分析方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP2767228A4 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104586379A (zh) * 2015-01-21 2015-05-06 深圳市理邦精密仪器股份有限公司 一种胎心率曲线的参数输出方法及装置

Also Published As

Publication number Publication date
US20140243674A1 (en) 2014-08-28
EP2767228A4 (en) 2015-06-17
EP2767228A1 (en) 2014-08-20
CN102319064B (zh) 2013-08-14
US9364197B2 (en) 2016-06-14
EP2767228B1 (en) 2022-08-03
CN102319064A (zh) 2012-01-18

Similar Documents

Publication Publication Date Title
WO2013053290A1 (zh) 一种提高胎心率数据减速识别准确性的装置和方法
WO2013013646A2 (zh) 一种提高胎心率数据加速识别准确性的装置和方法
US9198634B2 (en) Medical decision support system
JP5815705B2 (ja) 胎児モニタリング中に超音波信号の不明確性を削減する方法
WO2018010107A1 (zh) 判断脉搏波是否有效的方法及装置
WO2017084546A1 (zh) 基于可穿戴设备的用户关注信息确定方法、装置和可穿戴设备
WO2023226223A1 (zh) Ppg信号质量评估方法及装置以及ppg信号处理方法及系统
US11457873B2 (en) Method and apparatus for identifying homology of physiological signals
GB2520628A (en) Abnormal noise detection system
CN111839494A (zh) 一种心率监测方法及系统
EP3189776A1 (en) An apparatus and method for generating fetal heart rate data
CN109567869B (zh) 一种处理胎心率曲线上加速活动的方法及系统
CN101866423A (zh) 一种动脉压信号逐拍分割方法
JP3974492B2 (ja) 異常音検出装置
TWI418338B (zh) 睡眠腦波監測裝置
Perera et al. Automated diagnosis of cardiac abnormalities using heart sounds
CN113598810B (zh) 一种基于分割网络的胎心率基线自动计算方法
CN109567866B (zh) 一种胎心率周期变异的处理方法
TWI672127B (zh) 基於小波分析之即時qrs波偵測演算法
RU2762818C9 (ru) Способ обнаружения движения плода из доплеровского ультразвукового сигнала
EP3672493B1 (en) A fetal movement detection method from doppler ultrasound signal
JP2002196783A (ja) 時系列信号の識別方法及び装置
EP3230751B1 (en) Signal processing method and apparatus
CN112580751A (zh) 一种基于zynq与深度学习的鼾声辨识装置
Roj et al. Improving the periodicity measurement in fetal heart activity signal

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 12839654

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2012839654

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 14349326

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE