This chapter describes the effectivity of the developed pull outing algorithm and observing algorithm described in Chapters III and IV. For this intent, existent recorded informations from the maternal abdominal tegument of healthy pregnant adult females ( 36, 37 and 38 hebdomads of gestation ) , most of which are corrupted with different degrees of noises have been used to prove these two new algorithms. The success of the proposed algorithm is important to both pull out the FECG signal and observe the FHR every bit good as the MHR in existent clip when utilizing the g-tec bio-amplifier straight from voluntary pregnant adult females.These algorithms are foremost tested utilizing Matlab 7.4 ( Matlab book ) and so implemented under Matlab and simulink ( Mathworks Inc. ) .
After implementing the algorithms utilizing Matlab 7.4, FECG extraction has been examined with 30 existent recorded informations and MHR sensing has been examined with 20 existent recorded informations. For the possibility of real-time sensing, the algorithms are implemented under MatlabA®/SimulinkA® and simulated for FECG extraction utilizing 20 existent recorded informations each of 5 seconds continuance.
FHR sensing has been examined as good with 30 existent recorded informations each of 1 infinitesimal continuance ( described in Section 4.3 ) . Ten of these estimated FECG signals are used to cipher the FHR in order to compare the consequences with estimated FHR obtained via ultrasound M manner and ICA ( Najafabadi 2008 ) . Finally, the FHR sensing system is tested in real-time. All the clinical recorded informations are explained in Section 3.7.1.
All these trials and consequences are explained in item in the undermentioned subdivisions.
Off-line Processing USING MATLAB 7.4
The consequences presented in this subdivision are obtained by off-line treating the algorithms implemented utilizing Matlab 7.4. All informations used are existent recorded informations which were acquired utilizing three leads from healthy pregnant voluntaries in the Medical Center at University Kebangsaan Malaysia ( PPUKM ) . The informations are preprocessed, and illustrations of the three lead recorded informations before and after the preprocessing phase are illustrated in Figures 5.1 and 5.
2 severally.Figure 5. 1 Sample of the 3-lead recorded informations before preprocessingAn illustration of the acquired signals ( X1, X2 and X3 ) from the pregnant venters is illustrated in Figure 5.1, as the input of the system, which consists of 3 signals. The first signal ( X1 ) in Figures 5.1 ( a ) is acquired from the upper portion from the venters ( mother signal merely ) utilizing lead P1. The other signals ( X2 and X3 ) in Figures 5.1 ( B ) and ( degree Celsius ) are acquired from the venters of the pregnant adult female ( AECG ) utilizing lead P3 and P4 ( refer to Figure 3.
3 ) , these signals are consist of the MECG assorted with the FECG. Preprocessing phase consists of the remotion of the DC signal, baseline wander and the power line intervention fading should be executed in order to feed these signals to the following phases. Hence these signals are preprocessed and the resulted signals ( Y1, Y2 and Y3 ) are shown in Figures 5.2. Preprocessed signal Y1 is used as a mention for the ANC to pull out the FECG, Y2 and Y3 are used as coveted signal for the ANC ( refer to Figure 4.9 ) .Figure 5.
2 The recorded signals after preprocessingComparison of ANC with ICAIn order to observe the foetal extremums from the AECG signal the FECG signal should be separated, by the extraction algorithm. This algorithm utilizing ANC ( described in Chapter IV ) is validated utilizing 30 recorded informations each of five seconds and compared with the algorithm implementing independent constituent analysis ( ICA ) .In this comparing the same preprocessing phase and the same postprocessing phase are applied for the ANC and ICA, where the postprocessing phase consists of the window signal created by MQRSW for taking the maternal residuary extremums and 1Hz IIR notch filter. The public presentations of the algorithms are so evaluated based on their sensitivenesss and positive predictivities ( ANSI/AAMI EC57: 1998 ) of extremum sensing in the extracted signal, which will be shown in Table 5.1. Figures 5.
3 and 5.4 shows illustrations of the extracted signals from the same part of an abdominal signal utilizing ICA and ANC severally.Figure 5.3 ( a ) shows that foetal extracted signal by ICA, is still corrupted with noise. Hence the window signal created by MQRSW in Figure 5.3 ( B ) is applied for taking the maternal residuary extremums followed by 1Hz IIR notch filter.
Then the extracted foetal signal after postprocessing phase is shown in Figure 5.3 ( degree Celsius ) .Figure 5. 3 ( a ) Extracted FECG utilizing ICA ; ( B ) window signal, ( degree Celsius ) FECG signal after using remotion window signal and 1Hz notch filterFigure 5.4 ( a ) shows the fetal extracted signal by ANC, where the maternal residuary extremums are still among the foetal extremums, therefore the same postprocessing phase applied to the ICA is applied besides to the ANC. Finally, the extracted foetal signal is shown in Figure 5.4 ( degree Celsius ) .
In this comparing, the same signal is fed at the same time to the two algorithms for pull outing the FECG signal. These extracted signals are corrupted with different degrees of noise as shown in Figures 5.3 ( a ) and 5.4 ( a ) . The extracted signal by ICA has higher degree of noise than that extracted by ANC.
In signals with high degree of noise the ANC is more robust in pull outing the signal instead than ICA. This comparing besides shows the consequence of the postprocessing phase as an sweetening technique for rarefying the unwanted constituents in the FECG signal, as shown in Figures 5.3 ( degree Celsius ) and 5.4 ( degree Celsius ) .Figure 5. 4 ( a ) Extracted FECG utilizing ANC, ( B ) window signal and ( degree Celsius ) FECG signal after using remotion window signal and 1Hz notch filterThe ANC and ICA methods are evaluated in term of peak sensing, utilizing sensitiveness and positive predictivity of the foetal extremum sensing. Table 5.
1 shows the public presentation utilizing ICA and ANC based methods up to the foetal signal extraction phase.Table 5. 1 Performance of ICA and ANC based methodsWeeksNo. of SignalsANC methodICA method
( % )
( % )
( % )
( % )
35279.0 ( % )54.5 ( % )66.6 ( % )48.8 ( % )361388.
4 ( % )77.8 ( % )74.8 ( % )72.
2 ( % )37691.1 ( % )68.9 ( % )79.9 ( % )70.1 ( % )38984.8 ( % )69.
6 ( % )76.4 ( % )65.4 ( % )Overall norm87.23 ( % )72.09 ( % )75.75 ( % )68.
18 ( % )The consequences for FECG extraction from 30 signals are illustrated in Table 5.1, which were acquired from pregnant adult females ( scope between 36 to 38 hebdomads of gestation ) . The mean sensitiveness of the ANC based method is 87.23 % as compared to 75.75 % of the ICA based method. The mean positive predictivity of the ANC based method is 72.09 % as compared with that of the ICA based method which is 68.18 % .
It shows that the ANC based method was more successful in observing the FHR than ICA.Extraction with grading windowFigure 5.3 ( degree Celsius ) and Figure 5.4 ( degree Celsius ) shows the fetal extracted signal by ICA and ANC, where the window signal created by MQRSW is used for taking the maternal residuary extremums followed by 1Hz IIR notch filter.
In order to show the other ability of utilizing the window signal for scaling down the maternal residuary extremums, other tow illustrations of foetal signal each of 10 2nd continuance are shown in Figure 5.5 ( B ) and Figure 5.6 ( B ) , which are extracted by ICA and ANC severally. The window signal is applied for scaling down the maternal residuary extremums by multiplying them by 0.1 this is followed by 1Hz IIR notch filter. This ability can be used to avoid cutting the overlapped extremum in order to observe the foetal extremum subsequently. This is one of the advantages of this sweetening technique, where the public presentation of the algorithm will be better.Figure 5.
5 ( a ) Extracted FECG utilizing ICA ; ( B ) FECG signal after using scaling down window signal of MQRSWFigure 5. 6 ( a ) Extracted FECG utilizing ANC ; ( B ) FECG signal after usingscaling down window signal of MQRSWMaternal Peak Detection TestingThe truth of the algorithm to observe the maternal extremum was tested by using it to 20 recorded signals each of one minute long. Examples of the peak sensing are illustrated in Figures 5.
7 and 5.8, where the algorithm is applied to the signals without utilizing amplitude threshold. The signals show that all the local upper limit ( circles ) and minima ( stars ) in the signals are detected. The upper limit are the maternal extremums.Figure 5.
7 Detected extremums in AECG signals-1Figure 5. 8 Detected extremums in AECG signals-2Table 5.2 summarizes the public presentation of the sensing strategy on the 20 recorded AECG signals ( Appendix H ) . The mean Sensitivity ( ) of the algorithm is 99.05 % and its Positive Predictivity ( ) is 99.
79 % . There is a batch of noise particularly in two of the signals which lead to troubles in observing some maternal R extremums.Table 5. 2 Algorithm public presentation for maternal R extremum sensingWeeks of gestationNo. of SignalsANC method
( % )
( % )
36499.25 ( % )100 ( % )37599.60 ( % )100 ( % )381198.
72 ( % )99.63 ( % )Overall norm99.05 ( % )99.79 ( % )
Off-line Processing utilizing SIMULINK
In the old subdivision of this chapter, the algorithms were implemented utilizing Matlab 7.
4 in order to both determine the effectivity of these developed algorithm and trial or compare these algorithms utilizing existent recorded informations. The concluding executions of these algorithms were executed under MatlabA®/SimulinkA® and their consequence are given in this subdivision.FECG ExtractionThe effectivity of the FECG pull outing algorithm implemented utilizing simulink was examined which was examined with 20 existent recorded informations each of five seconds continuance. The normalized least average square ( NLMS ) algorithm ( described in Section 2.5.
3 ) , was utilized to pull out the FECG signal from the AECG signal and compared with the proposed RLS algorithm. Figure 5.9 shows the complete execution of both algorithms utilizing the same preprocessing phase and the postprocessing hart. The public presentation of both algorithms was compared in footings of FHR sensing as ( described in Section 4.2.5 ) . The public presentations of the algorithms were so evaluated based on their sensitivenesss and positive predictivities.
Figure 5. 9 FECG extraction utilizing Simulink blocks of RLS & A ; NLMSFigures 5.10 ( a ) , and 5.
11 ( a ) show tow illustrations of the input preprocessed AECG signals and tow end products from each of the tow complete algorithms. The end products in Figures 5.10 ( B ) and ( degree Celsius ) are about the same, but the end product from the algorithm RLS appear to heighten the foetal signal better than the NLMS algorithm as shown in Figures 5.11 ( B ) and ( degree Celsius ) .
Figure 5. 10 ( a ) AECG signal, ( B ) extracted signal by RLS ( degree Celsius ) by NLMS of towend products.Figure 5. 11 ( a ) AECG signal, ( B ) extracted signal by RLS ( degree Celsius ) by NLMS of towend products.The obtained consequences are summarized in Table 5.3.
Average values of sensitiveness ( ) and positive anticipation ( ) of the RLS based method are 88.59 % and 82.78 % , severally, compared to 80.44 % and 72.09 % of the NLMS based method.
Table 5. 3 Performance of RLS and NLMS based methodsWeeksNo. of SignalsRLS methodNLMS method
( % )
( % )
( % )
( % )
22 ( % )82.94 ( % )81.88 ( % )72.94 ( % )37592.4 ( % )78.
8 ( % )74.9 ( % )66.7 ( % )38689 ( % )85.86 ( % )82.9 ( % )75.33 ( % )Overall norm88.
59 ( % )82.78 ( % )80.44 ( % )72.09 ( % )Fetal Peak DetectionThe complete RLS algorithm as shown in Figure 5.
9 for foetal peak sensing is farther tested on more informations of longer continuance. Thirty existent recorded informations each of 60 2nd continuance are used. The algorithm public presentation is farther improved by using a peak place rectification technique. The preciseness in the designation of QRS complex extremums is of great importance for ciphering FHR. For this intent, two conditions were used to measure the FHR extremums after FHR sensing. These are:The fetal detected extremums were compared with those foetal extremums in the preprocessed AECG signal. If the difference between the locations of the two extremums is two samples or less, the extremum was assumed as true, otherwise false.If it is hard to compare with those foetal extremums in the preprocessed AECG signal, the fetal detected extremums could be compared with the foetal extremums in the extracted signal.
Figure 5.12 to 4.14 show the AECG signal and the detected extremums from the extracted foetal signal. The labels on the AECG or the foetal extracted signal wave form indicate the true or false extremum locations as follows:T – True detected foetal R extremum.FF – false detected foetal R extremum.OL – overlapped ( foetal R extremums detected within the MQRS ) .
MS – missed foetal R extremums.The fetal detected extremums ( B ) of Figure 5.12 were compared with the ascertained foetal extremums in the preprocessed AECG signal ( a ) of Figure 5.12 as suggested in status ( I ) .
Most of these extremums are true detected ( labeled with T ) . The false, lost and overlapped extremums are labeled with FF, MS and OL severally. In Figure 5.12, FF is located at sample figure 7, OL at 490 and MS at 1271.
Figure 5. 12 ( a ) Preprocessed AECG signal ( B ) detected foetal extremums from the extracted FECG signalFigure 5. 13 ( a ) Preprocessed AECG signal ( B ) detected foetal extremums from the extracted FECG signalFigure 5.12 replicate Figure 5.13 with some of the tried peak locations labeled in the preprocessed AECG. For illustration, the foetal extremum at location 394 is considered true because it is within 2 samples of the detected extremum in Figure 5.
13 ( B ) as required by status ( two ) .However, in some signals it is hard to compare the foetal detected extremums, due to the absence of the foetal extremums in the pre-processed AECG signal. One instance is illustrated in Figure 5.14 ( degree Celsius ) , as the fetal detected extremums are compared to the extracted signal by the ANC before using postprocessing phase Figure 5.14 ( B ) .
Figure 5. 14 ( a ) AECG signal ; ( B ) observed foetal extremums in the Extracted signal by ANC, before postprocessing phase ; ( degree Celsius ) FECG signal after using postprocessing phaseFigure 5.14 reflects the effectivity of these developed algorithms, where the foetal signal can non be observed in the AECG signal Figure 5.14 ( a ) . The extraction algorithm was able to pull out the FECG signal, utilizing ANC as shown in Figure 5.
14 ( B ) , where the foetal extremums are marked. In Figure 5.14 ( degree Celsius ) the signal is postprocessed and the foetal extremums are detected. These detected extremums are compared with the pronounced extremums in figure 5.
14 ( B ) harmonizing the 5.3.2 ( two ) . Most of the detected extremums are true ( T ) , one extremum is false ( FF ) and one extremum is overlapped ( OL ) .All the 30 signals used for FHR sensing were processed harmonizing to the above- mentioned conditions as shown in Figures 5.12 to 5.14.
The consequences of these sensings are illustrated in Table 5. 4 after using the peak place rectification technique to rectify the overlapped extremum place described in Section 4.2.7. The overall norm of the sensitiveness consequence is 79.76 % and the positive predectivity consequence is 77.
49 % ( Appendix G ) .Table 5. 4 Algorithm public presentation for foetal R extremum sensingWeeks of gestationNo. of SignalsANC method
( % )
( % )
28 ( % )80.99 ( % )37884.30 ( % )82.43 ( % )381071.
90 ( % )69.43 ( % )Overall norm79.76 ( % )77.49 ( % )
comparision of FHR with ultrasound measurment
In this subdivision, the consequences obtained were compared with the old work done in Najafabadi ( 2008 ) which compared the FHR sensing utilizing ultrasound with the ICA technique. The ultrasound M manner, which is used in that work to mensurate the FHR, was done one time before each acquired signal so as to utilize it as a mention. Furthermore, the ICA used short parts of the signals so as to observe the FHR and to compare it with the measured values by ultrasound denoted by usm.In order to compare the consequences of the present work on the same signals, the proposed algorithm to observe FHR is applied to the signals of 60 seconds continuance. The norm of the FHR of each signal was used in this comparing.
In add-on per centum of mistake was besides calculated. The consequence is illustrated in Table 5.5.Table 5.
5 Performance of peak sensing algorithm with US and ICASignalNo.Lead system protocolsMeasured FHR by UltrasoundusmEstimated FHR by ICA algorithm% MistakeEstimated FHR by the proposed algorithm% Mistake36_136_236_3F156130138141156133
73.8The columns in Table 5.5 are described below:Column one: capable info including gestation age in hebdomads.Column two: lead system protocol, as desired in Najafabadi ( 2008 ) .Column three: the mensural mean FHR by ultrasound M manner before enteringColumn four: the norm of ultrasound measured values ( usm ) in Column 3.
Column five: the estimated FHR by the ICA method for successful sensing instances,Column six: the per centum mistake between estimated FHR by ICA method and ultrasound.Column seven: the estimated FHR by the proposed algorithm for 60 seconds.Column eight: the per centum mistake between estimated FHR by the proposed algorithm and usm.( 5.
1 )As can be seen in the Table 5.5, the consequence of the estimated FHR by the proposed algorithm from 10 signals each of 60 seconds are compared to the ultrasound estimated FHR and to the ICA estimated FHR from little part of each signal, the mistake per centum of the proposed algorithm is ( 0.7 to 4.9 ) compared to ( 0 to 2.
3 ) of ICA. Our consequences can be compared with the mistake per centum of old research, which used the FHR measured by ultrasound foetal monitoring ( BIOSIS Co. , LTD ) as mention for the estimated FHR by the algorithm proposed by Ibrahimy, where the mistake per centum is ( 3.31 to 6.
63 ) . This comparing shows that the consequences of estimated FHR by the proposed algorithm, ICA and the algorithm proposed by Ibrahimy are similar.
Real-time FHR Detection
With the execution of the algorithms utilizing Simulink blocks through the old phases, real-time trial on pregnant voluntaries, utilizing g-tec bio-amplifier connected to a personal computing machine is performed. For this intent, three signals ( channels ) from their venters are acquired utilizing three electrodes, a common and a land, as shown in Figure 3.4.The recordings were made at the place of the pregnant voluntaries while the pregnant adult females are in supine place. The entering continuance ranged between one to five proceedingss.
Some illustrations of the recorded signals and bosom rate hints with different degrees of noise are illustrated in Figure 5.15 to 5.17. The bosom rate hints of two proceedingss as in Figure 5.15 represent the MHR and FHR hints for a adult female of 36 hebdomads of gestation.
The MHR sensing were successful in most instances. The FHR hint is found to be of utile quality except for little parts. The good quality is due to the consistent presence of big foetal signal which resulted from good extraction. The overall norm of the sensitiveness consequence is 84.95 % and the positive predectivity consequence is 86.29 % for foetal peak sensing from the information of the pregnant adult female ( 36 hebdomads ) illustrated in Figure 5.15.
The sensitiveness consequence is 98,96 % and the positive predectivity consequence is 99.25 % for maternal extremum sensing.Figure 5.
15 ( a ) AECG signal, ( B ) MHR and ( degree Celsius ) FHR hints of good quality ( 36 hebdomads )Table 5.6 Algorithm public presentation for R extremum sensingR Peak sensingANC method
( % )
( % )
foetal Peak sensing84.95 ( % )86.
29 ( % )maternal Peak sensing98.96 ( % )99.25 ( % )The AECG signals with a little foetal R extremum compared to maternal R extremum or unwanted constituents would bring forth FHR hints of medium or hapless quality. Figure 5.
16 is an illustration with FHR hint, which can be considered of hapless quality. This signal is of tow proceedingss, which is acquired from a pregnant in the thirty-seventh hebdomad of gestation.Figure 5. 16 ( a ) MHR and ( B ) FHR hints of hapless quality ( 37 hebdomads )Another illustration is illustrated in Figure 5.17. The signal is acquired from a pregnant voluntary in the 38th hebdomad of gestation and is five proceedingss long.
This signal is corrupted with noise in some subdivisions of the FHR hint but some information can be still obtained from the other subdivisions. This hint can be considered of medium quality. In general, the per centum of successful MHR is high, while the per centum of successful FHR depends on the success of FECG extraction.Figure 5. 17 ( a ) MHR and ( B ) FHR hints of medium quality ( 38 hebdomads )
In this work, two new developed algorithms are implemented for FECG signal processing. The first algorithm is the FECG extraction and the 2nd algorithm is the peak sensing algorithm. The extraction algorithm is new developed due to the new sweetening technique used to heighten the extracted signal by utilizing the IIR notch filter ( alpha = 0.
85 ) , and by MQRS window and its window signal for scaling down the maternal residuary extremums and adjust them once more in order to be shorter than the foetal extremums for heightening the sensing of the foetal extremums. The sensing algorithm is besides new developed foremost, due to its ability to observe the extremum without amplitude threshold, and due to the new sweetening technique for sensing by finding the overlapped extremums for rectifying the foetal RR interval about every overlapped extremum, moreover it is the first sensing algorithm that is threshold amplitude free sensing without pre-determined extremum ‘s threshold comparison with Pan & A ; Tompkins ( 1985 ) and Karvounis et Al. ( 2006 ) .
It merely depends on the normal bosom rate ( maternal or foetal ) and the sampling frequence. The extraction algorithm is based on ANC, RLS technique. For proving the proposed extraction algorithms, it is compared with two other methods ICA and NLMS. The findings obtained from ICA, which are shown in Table 5.1 disclosed that the public presentation of the proposed algorithm was better than that of ICA. The highest value of sensitiveness of the proposed algorithm ( ) was 87.
23 % , while ICA value was 75.75 % . The positive predectivity value ( ) of the proposed algorithm was 72.09 % , whereas the value of ICA was 68.18 % . These findings reflect the hardiness of the proposed algorithm compared to ICA in pull outing more enlightening foetal signals for foetal QRS.On the other manus, the consequences obtained by comparing the proposed algorithm with NLMS algorithm show that the public presentation of these algorithms was better than those of NLMS. The highest value of the proposed algorithm sensitiveness ( ) was 88.
59 % , while NLMS value was 80.448 % . The positive predectivity value ( ) of the proposed algorithm was 82.78 % , whereas the value of ICA was 72.
09 % .The proposed sensing algorithm is based on an amplitude threshold free technique. The algorithm is used to observe the maternal signal extremums every bit good as the foetal signal extremums. In add-on, the consequences from the sensing of extremums from FECG signal were compared with consequences of the detected extremums of the same signal by utilizing ultrasound.With respect to maternal extremum sensing, over 99 % ability of the maternal QRS sensing is shown in Table 5.2. The primary ground of the mistake in the maternal QRS sensing is the intervention due to the noise which leads to troubles in observing some maternal R extremum. This consequence is comparable to the consequence of the maternal extremum detected in existent clip, where the sensitiveness consequence is 98.
96 % and the positive predectivity consequence is 99.25 % . Beside that this consequence can besides be compared with the consequence of maternal extremum sensing by Ibrahimy Muhammad ( 2001 ) , which is around 99 % .For the foetal QRS sensing, most of the foetal QRS composites are in general detected with a per centum of 79.76 % , as illustrated in Table 5.4.
This consequence is comparable to the consequence of the foetal extremum detected in existent clip, where the sensitiveness consequence is 84.95 % and the positive predectivity consequence is 86.29 % when the FHR hints are of good quality, i.e.
the foetal QRS composites are systematically above the noise degree. Beside that this consequence can besides be compared with the consequence of maternal extremum sensing by Ibrahimy Muhammad ( 2001 ) , which is around 83 % . The consequences of the proposed algorithm for mensurating the mean FHR from 10 signals have besides been compared with measurings by an ultrasound manner M machine, in order to asses the dependability of the algorithm. The mistake rate of the proposed algorithm to the ultrasound measurings is in the scope of 0.7 to 4.9 % ) . All the consequences are illustrated in Table 5.
7.Table 5.7 Algorithm public presentation for R extremum sensingProposed algorithm for:
( % )
( % )
FECG extraction ( Matlab 7.4. )RLS87.
23 ( % )72.09 ( % )ICA75.75 ( % )68.18 ( % )FECG extraction ( under Matlab simulink. )RLS88.59 ( % )82.78 ( % )NLMS80.44 ( % )72.
09 ( % )Peak sensingMaternal ( off-line )99.05 ( % )99.79 ( % )fetal ( simulink )79.
76 ( % )77.49 ( % )Peak sensing ( R. Time. )maternal98,96 ( % )99.25 ( % )fetal84.95 ( % )86.29 ( % )Peak sensing ( R.
Time. ) ( by Ibrahim )maternalAround 99 ( % )fetalAround 83 ( % )Percentage of mistake( ensuing from Comparison with ultrasound )Developed algorithm0.7 ( % )4.9 ( % )ICA02.
3 ( % )Ibrahimy3.31 ( % )6.63 ( % )The restriction of the algorithm for observing the foetal signal extremum is due to the low degree of the SNR. The foetal signal extremum, has little values despite sweetening when compared to the chief arise beginnings, such as gesture artefact, electrode contact and the maternal signal. Detection trouble besides arises when the foetal signal extremum is overlapped with the maternal. Furthermore in the current design of the algorithm, 5 seconds buffer is used for each processing frame. This leads to a few mistakes when the extremums have to be detected at the terminal of the frame.In the attempt to emulate existent simulation, the chosen recorded trial signals include those corrupted with different degrees of noise.
The algorithm has incorporated a few stairss to better its public presentation on such signals. These include scaling utilizing the MQRSW, 1 Hz notch filter in the postprocessing phase, hunt interval without utilizing amplitude threshold and foetal signal extremum place rectification based on cognition of the maternal signal extremum locations. With these stairss, the appraisals described in this chapter show the ability of the algorithm to execute in existent clip comparable to the widely used ultrasound machine.
The bosom rate measurings antecedently acquired by off line processing of AECG signals and by real-time processing of straight acquired AECG signals by the proposed algorithm consequence in the FHR and MHR sensing of around 80 % and 99 % severally.
The algorithm was shown to be better than other similar techniques due to its threshold amplitude free sensing, which has to be adjusted antecedently for the other similar techniques, and comparable consequences were obtained from the ultrasound machine. Beside that the sweetening technique, introduces the possibility to find the overlapped QRS composites and to rescale their amplitude in order to avoid cutting these composites.