R-peak detection is crucial in electrocardiogram (ECG) signal analysis. This study proposed an adaptive and time-efficient R-peak detection algorithm for ECG processing. First, wavelet multiresolution analysis was applied to enhance the ECG signal representation. Then, ECG was mirrored to convert large negative R-peaks to positive ones. After that, local maximums were calculated by the first-order forward differential approach and were truncated by the amplitude and time interval thresholds to locate the R-peaks. The algorithm performances, including detection accuracy and time consumption, were tested on the MIT-BIH arrhythmia database and the QT database. Experimental results showed that the proposed algorithm achieved mean sensitivity of 99.39%, positive predictivity of 99.49%, and accuracy of 98.89% on the MIT-BIH arrhythmia database and 99.83%, 99.90%, and 99.73%, respectively, on the QT database. By processing one ECG record, the mean time consumptions were 0.872 s and 0.763 s for the MIT-BIH arrhythmia database and QT database, respectively, yielding 30.6% and 32.9% of time reduction compared to the traditional Pan-Tompkins method.
Electrocardiogram (ECG) can describe the electrical activity of the heart and is an essential tool for the diagnosis of cardiovascular diseases (CAD). With the rapid development of wearable and wireless ECG techniques, real-time and routine ECG monitoring is attracting more and more attention due to the increasing popularization of medical health, especially for the elderly people [
ECG features are essential characteristics for CAD diagnosis. R-peak detection is the datum since all other features are extracted after the R-peak location [
Over the last decades, numerous techniques have been proposed for R-peak detection. In [
Time cost is important due to the fast-responding requirement in CAD early warning applications [
In this study, an adaptive and time-efficient ECG R-peak detection algorithm is proposed. The method takes advantage of wavelet-based multiresolution analysis (WMRA) and adaptive thresholding. WMRA is applied to strengthen ECG signal representation by extracting ECG frequency interval of interest from wide-range frequencies, which contain interference such as baseline drift and motion artifacts. All the noises produce considerable influence on the following thresholding operation. The adaptive thresholding is designed to exclude false R-peaks in the reconstructed signal by WMRA. The proposed algorithm was tested by the MIT-BIH arrhythmia database (MITDB) and the QT database (QTDB) [
The remainder of the paper is organized as follows. Section
The R-peak detection system is described in Figure
Block diagram of the proposed R-peak detection algorithm.
WMRA enhances signals using wavelet transform to extract both time and frequency domain information. This method is very suitable for ECG processing since ECG is essentially nonstationary with small amplitude (0.01~5 mV) and low frequency (0.05~100 Hz) [
Figure
Decomposition process of the eight-level WMRA. The sampling frequency is decomposed into two subbands: high frequency of detail coefficient (
For some ECG patterns, such as premature ventricular contraction (PVC) beat, R-peaks are presented with amplitude below the baseline but other features are above the baseline. To avoid the potential missing detection, signal mirroring is designed. The mirroring procedure for a PVC segment is described in Figure
Signal mirroring example with comparison range of 0.556 s and amplitude multiple of 1.5.
In some literatures [
Local maximums are located by implementing first-order forward differential in the mirrored signal. The procedure is illustrated as follows.
First-order forward difference is implemented on ECG signal with For all the elements in First-order forward difference is implemented on the updated
The following threshold procedure depends significantly on the amplitude threshold
Determination of
However, the threshold selection is strongly dependent on the noise;
Actually, most of the local maximums are not true R-peaks, such as burst points caused by high-frequency interference. The difficulty of R-peak detection lies in the recognition of false R-peaks with amplitudes approximate to or even larger than true R-peaks. To this end,
Tref = localmax1; Cref = localmax2; Rref=localmax3; WTref = width (Tref); WCref = width (Cref); ATref = amplitude (Tref); ACref = amplitude (Cref); If If WTref < WCref localmax1 is a true R-peak; Else if WTref > WCref localmax2 is a true R-peak; Else if ATref > ACref localmax1 is a true R-peak; Else localmax2 is a true R-peak; End if End if Else localmax1 is a true R-peak; End if End if If Rref is not NULL False R-peak is replaced by the Rref, and repeat the algorithm with new Tref and Cref; End if
Small amplitudes filtration by
The MITDB comprises 48 ECG records, and each contains 30-minute ECG signal [
The QTDB contains a total of 105 15-minute ECGs. ECGs in this database were chosen to represent a wide variety of QRS and ST-T morphologies with real-world variability to challenge the detection algorithms [
It should be noted that both databases provide two channels of ECG signals. In this study, only the first channel was used for algorithm development and test.
Experimental results are evaluated in terms of sensitivity
Time complexity is also tested, which quantifies the amount of time taken by an algorithm to run as a function of the length of string representing the input. It reflects the increment of time consumption when the input data increase. Time complexity of an algorithm is commonly expressed using
First, for both databases,
The testing results on MITDB are summarized in Table
Detection results of ECG signals from MITDB.
Record | Total beats |
|
|
|
|
+P (%) |
|
---|---|---|---|---|---|---|---|
100 | 2273 | 2273 | 0 | 0 | 100.00 | 100.00 | 100.00 |
101 | 1865 | 1864 | 1 | 1 | 99.95 | 99.95 | 99.89 |
102 | 2187 | 2187 | 0 | 0 | 100.00 | 100.00 | 100.00 |
103 | 2084 | 2084 | 0 | 0 | 100.00 | 100.00 | 100.00 |
104 | 2229 | 2222 | 7 | 10 | 99.69 | 99.55 | 99.24 |
105 | 2572 | 2528 | 44 | 50 | 98.29 | 98.06 | 96.41 |
106 | 2027 | 2004 | 23 | 22 | 98.87 | 98.91 | 97.80 |
107 | 2137 | 2121 | 16 | 6 | 99.25 | 99.72 | 98.97 |
108 | 1763 | 1739 | 24 | 18 | 98.64 | 98.98 | 97.64 |
109 | 2532 | 2532 | 0 | 0 | 100.00 | 100.00 | 100.00 |
111 | 2124 | 2117 | 7 | 4 | 99.67 | 99.81 | 99.48 |
112 | 2539 | 2539 | 0 | 0 | 100.00 | 100.00 | 100.00 |
113 | 1795 | 1795 | 0 | 0 | 100.00 | 100.00 | 100.00 |
114 | 1879 | 1872 | 7 | 10 | 99.63 | 99.47 | 99.10 |
115 | 1953 | 1953 | 0 | 0 | 100.00 | 100.00 | 100.00 |
116 | 2412 | 2393 | 19 | 5 | 99.21 | 99.79 | 99.01 |
117 | 1535 | 1534 | 1 | 1 | 99.93 | 99.93 | 99.87 |
118 | 2278 | 2277 | 1 | 0 | 99.96 | 100.00 | 99.96 |
119 | 1987 | 1987 | 0 | 0 | 100.00 | 100.00 | 100.00 |
121 | 1863 | 1860 | 3 | 3 | 99.84 | 99.84 | 99.68 |
122 | 2476 | 2476 | 0 | 0 | 100.00 | 100.00 | 100.00 |
123 | 1518 | 1518 | 0 | 0 | 100.00 | 100.00 | 100.00 |
124 | 1619 | 1617 | 2 | 2 | 99.88 | 99.88 | 99.75 |
200 | 2601 | 2593 | 8 | 3 | 99.69 | 99.88 | 99.58 |
201 | 1963 | 1962 | 1 | 1 | 99.95 | 99.95 | 99.90 |
202 | 2136 | 2123 | 13 | 6 | 99.39 | 99.72 | 99.11 |
203 | 2980 | 2953 | 27 | 21 | 99.09 | 99.29 | 98.40 |
205 | 2656 | 2640 | 16 | 2 | 99.40 | 99.92 | 99.32 |
207 | 2332 | 2018 | 314 | 328 | 86.54 | 86.02 | 75.86 |
208 | 2955 | 2932 | 23 | 3 | 99.22 | 99.90 | 99.12 |
209 | 3005 | 3005 | 0 | 1 | 100.00 | 99.97 | 99.97 |
210 | 2650 | 2629 | 21 | 13 | 99.21 | 99.51 | 98.72 |
212 | 2748 | 2748 | 0 | 0 | 100.00 | 100.00 | 100.00 |
213 | 3251 | 3245 | 6 | 2 | 99.82 | 99.94 | 99.75 |
214 | 2262 | 2253 | 9 | 10 | 99.60 | 99.56 | 99.16 |
215 | 3363 | 3360 | 3 | 4 | 99.91 | 99.88 | 99.79 |
217 | 2208 | 2193 | 15 | 10 | 99.32 | 99.55 | 98.87 |
219 | 2154 | 2154 | 0 | 0 | 100.00 | 100.00 | 100.00 |
220 | 2048 | 2048 | 0 | 0 | 100.00 | 100.00 | 100.00 |
221 | 2427 | 2417 | 10 | 5 | 99.59 | 99.79 | 99.38 |
222 | 2483 | 2480 | 3 | 3 | 99.88 | 99.88 | 99.76 |
223 | 2605 | 2585 | 20 | 0 | 99.23 | 100.00 | 99.23 |
228 | 2053 | 2032 | 21 | 14 | 98.98 | 99.32 | 98.31 |
230 | 2256 | 2256 | 0 | 0 | 100.00 | 100.00 | 100.00 |
231 | 1571 | 1571 | 0 | 0 | 100.00 | 100.00 | 100.00 |
232 | 1780 | 1778 | 2 | 2 | 99.89 | 99.89 | 99.78 |
233 | 3079 | 3078 | 1 | 1 | 99.97 | 99.97 | 99.94 |
234 | 2753 | 2753 | 0 | 0 | 100.00 | 100.00 | 100.00 |
Total |
|
|
|
|
|
|
|
The testing results on QTDB are shown in Table
Detection results of ECG signals from QTDB.
Record | Total beats |
|
|
|
|
+P (%) |
|
---|---|---|---|---|---|---|---|
sel100 | 1134 | 1134 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel102 | 1088 | 1088 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel103 | 1048 | 1048 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel104 | 1109 | 1109 | 0 | 1 | 100.00 | 99.91 | 99.91 |
sel114 | 862 | 858 | 4 | 8 | 99.54 | 99.08 | 98.62 |
sel116 | 1185 | 1184 | 1 | 1 | 99.92 | 99.92 | 99.83 |
sel117 | 766 | 766 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel123 | 756 | 756 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel213 | 1642 | 1636 | 6 | 1 | 99.63 | 99.94 | 99.57 |
sel221 | 1247 | 1240 | 7 | 3 | 99.44 | 99.76 | 99.20 |
sel223 | 1309 | 1307 | 2 | 2 | 99.85 | 99.85 | 99.69 |
sel230 | 1077 | 1077 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel231 | 732 | 732 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel232 | 865 | 864 | 1 | 0 | 99.88 | 100.00 | 99.88 |
sel233 | 1533 | 1507 | 26 | 1 | 98.30 | 99.93 | 98.24 |
sel301 | 1351 | 1346 | 5 | 1 | 99.63 | 99.93 | 99.56 |
sel302 | 1500 | 1498 | 2 | 1 | 99.87 | 99.93 | 99.80 |
sel306 | 1040 | 1040 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel307 | 853 | 853 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel308 | 1294 | 1285 | 9 | 5 | 99.30 | 99.61 | 98.92 |
sel310 | 2012 | 1997 | 15 | 2 | 99.25 | 99.90 | 99.16 |
sel803 | 1026 | 1026 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel808 | 903 | 902 | 1 | 1 | 99.89 | 99.89 | 99.78 |
sel811 | 704 | 704 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel820 | 1159 | 1158 | 1 | 0 | 99.91 | 100.00 | 99.91 |
sel821 | 1557 | 1556 | 1 | 1 | 99.94 | 99.94 | 99.87 |
sel840 | 1180 | 1179 | 1 | 0 | 99.92 | 100.00 | 99.92 |
sel847 | 801 | 801 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel853 | 1113 | 1113 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel871 | 917 | 917 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel872 | 990 | 990 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel873 | 859 | 858 | 1 | 1 | 99.88 | 99.88 | 99.77 |
sel883 | 892 | 891 | 1 | 2 | 99.89 | 99.78 | 99.66 |
sel891 | 1267 | 1266 | 1 | 0 | 99.92 | 100.00 | 99.92 |
sel14046 | 1260 | 1260 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel14157 | 1081 | 1081 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel14172 | 663 | 663 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel15814 | 1036 | 1035 | 1 | 0 | 99.90 | 100.00 | 99.90 |
sel16265 | 1031 | 1031 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel16272 | 851 | 851 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel16273 | 1112 | 1111 | 1 | 0 | 99.91 | 100.00 | 99.91 |
sel16420 | 1063 | 1063 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel16483 | 1087 | 1087 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel16539 | 922 | 922 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel16773 | 1008 | 1007 | 1 | 0 | 99.90 | 100.00 | 99.90 |
sel16786 | 925 | 925 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel16795 | 761 | 761 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel17152 | 1628 | 1628 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sel17453 | 1047 | 1047 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sele0104 | 804 | 804 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sele0106 | 896 | 896 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sele0107 | 812 | 806 | 6 | 2 | 99.26 | 99.75 | 99.02 |
sele0110 | 872 | 870 | 2 | 9 | 99.77 | 98.98 | 98.75 |
sele0111 | 907 | 907 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sele0112 | 684 | 675 | 9 | 12 | 98.68 | 98.25 | 96.98 |
sele0114 | 699 | 698 | 1 | 1 | 99.86 | 99.86 | 99.71 |
sele0116 | 558 | 558 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sele0121 | 1436 | 1431 | 5 | 0 | 99.65 | 100.00 | 99.65 |
sele0122 | 1415 | 1415 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sele0124 | 1121 | 1121 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sele0126 | 945 | 945 | 0 | 1 | 100.00 | 99.89 | 99.89 |
sele0129 | 671 | 644 | 27 | 23 | 95.98 | 96.55 | 92.80 |
sele0133 | 840 | 840 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sele0136 | 809 | 809 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sele0166 | 813 | 813 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sele0170 | 897 | 897 | 0 | 2 | 100.00 | 99.78 | 99.78 |
sele0203 | 1246 | 1245 | 1 | 1 | 99.92 | 99.92 | 99.84 |
sele0210 | 1063 | 1062 | 1 | 0 | 99.91 | 100.00 | 99.91 |
sele0211 | 1575 | 1573 | 2 | 4 | 99.87 | 99.75 | 99.62 |
sele0303 | 1045 | 1044 | 1 | 0 | 99.90 | 100.00 | 99.90 |
sele0405 | 1216 | 1216 | 0 | 1 | 100.00 | 99.92 | 99.92 |
sele0406 | 959 | 959 | 0 | 1 | 100.00 | 99.90 | 99.90 |
sele0409 | 1737 | 1737 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sele0411 | 1202 | 1202 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sele0509 | 1028 | 1028 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sele0603 | 870 | 869 | 1 | 0 | 99.89 | 100.00 | 99.89 |
sele0604 | 1031 | 1031 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sele0606 | 1442 | 1442 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sele0607 | 1184 | 1184 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sele0609 | 1127 | 1125 | 2 | 2 | 99.82 | 99.82 | 99.65 |
sele0612 | 751 | 751 | 0 | 0 | 100.00 | 100.00 | 100.00 |
sele0704 | 1094 | 1093 | 1 | 1 | 99.91 | 99.91 | 99.82 |
Total |
|
|
|
|
|
|
|
Our algorithm is also compared with several existing methods, including the most widely used Pan-Tompkins method, as shown in Table
Comparison of R-peak detection with other algorithms.
Dataset | Beats |
|
|
|
|
+P (%) |
| |
---|---|---|---|---|---|---|---|---|
Zidelmal et al. [ |
MITDB | 109494 | 109101 | 393 | 193 | 99.64 | 99.82 | 99.47 |
Pan and Tompkins [ |
MITDB | 116137 | 115860 | 277 | 507 | 99.76 | 99.56 | 99.33 |
Jung and Lee [ |
MITDB | 109541 | 108960 | 581 | 579 | 99.47 | 99.47 | 98.94 |
Chiarugi et al. [ |
MITDB | 109494 | 109288 | 266 | 210 | 99.76 | 99.81 | 99.57 |
Arzeno et al. [ |
MITDB | 109517 | 109099 | 354 | 405 | 99.68 | 99.63 | 99.31 |
Elgendi [ |
MITDB | 109985 | 109738 | 247 | 124 | 99.78 | 99.87 | 99.66 |
Christov [ |
MITDB | 110050 | 109615 | 240 | 239 | 99.74 | 99.65 | 99.56 |
Chouakri et al. [ |
MITDB | 110934 | 109488 | 1446 | 3068 | 98.68 | 97.24 | 96.03 |
Rodríguez et al. [ |
MITDB | 44715 | 42518 | 879 | 142 | 96.28 | 99.71 | 97.65 |
Yeh and Wang [ |
MITDB | 116137 | 115971 | 166 | 58 | 99.86 | 99.95 | 99.81 |
The proposed |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
The time consumption for each record from MITDB is described in Figure
(a) Time consumption of processing the 48 records in MITDB and (b) time consumption ratio of the proposed method over the Pan-Tompkins method.
Time of processing one record (MITDB)
Time ratio of processing one record (MITDB)
The time consumption for each record from QTDB is described in Figure
(a) Time consumption of processing the first 41 records in QTDB and (b) time consumption ratio of the proposed method over the Pan-Tompkins method.
Time of processing one record (first 41 records from QTDB)
Time ratio of processing one record (first 41 records from QTDB)
(a) Time consumption of processing the last 41 records in QTDB and (b) time consumption ratio of the proposed method over the Pan-Tompkins method.
Time of processing one record (last 41 records from QTDB)
Time ratio of processing one record (last 41 records from QTDB)
The time consumption reveals an important characteristic of the two methods. The number of sampling points of each QTDB ECG is 225000, and the number is 650000 of each MITDB ECG. Although the number has increased about two times from QTDB to MITDB, the time consumed increases only 12.5% using our method and 9.2% using the Pan-Tomkins method. It indicates that when data multiplies, the time consumption increases slightly instead of multiplying correspondingly. Both our method and the Pan-Tomkins method are not so sensitive to data increase.
However, for records 107, 109, 113, and 116 in MITDB, our method consumes the same and even more time than the Pan-Tompkins method. The disadvantage of our method is plotted in Figure
Multiples of time consumption versus multiples of samples.
Amplitude threshold takes a significant role in truncating burst points along the baseline; time interval threshold is a critical measure to further distinguish false R-peaks. According to (
Comparison of different threshold coefficients.
|
|
|||||||||
---|---|---|---|---|---|---|---|---|---|---|
0.10 | 0.15 | 0.20 | 0.25 | 0.30 | 0.35 | 0.40 | 0.45 | 0.50 | ||
SEN (%) | 0.36 | 98.65 | 98.93 | 98.68 | 97.87 | 96.45 | 93.87 | 90.78 | 87.54 | 84.16 |
+P (%) | 80.94 | 87.29 | 90.23 | 92.38 | 93.69 | 94.55 | 96.56 | 97.89 | 98.67 | |
ACC (%) | 80.06 | 86.47 | 89.16 | 90.56 | 90.57 | 89.05 | 87.94 | 85.92 | 83.22 | |
SEN (%) | 0.39 | 96.42 | 97.59 | 98.05 | 97.61 | 96.31 | 93.79 | 90.72 | 87.51 | 84.14 |
+P (%) | 85.94 | 91.24 | 93.64 | 94.99 | 95.58 | 96.37 | 98.08 | 99.04 | 99.39 | |
ACC (%) | 83.28 | 89.24 | 91.93 | 92.83 | 92.20 | 90.59 | 89.14 | 86.78 | 83.71 | |
SEN (%) | 0.42 | 95.70 | 97.20 | 97.80 | 97.43 | 96.18 | 93.68 | 90.64 | 87.44 | 84.09 |
+P (%) | 89.98 | 94.56 | 96.42 | 97.40 | 97.79 | 98.27 | 98.98 | 99.38 | 99.56 | |
ACC (%) | 86.48 | 92.05 | 94.38 | 94.96 | 94.13 | 92.16 | 89.80 | 86.97 | 83.78 | |
SEN (%) | 0.45 | 94.57 | 96.47 | 97.17 |
|
95.73 | 93.36 | 90.42 | 87.28 | 83.94 |
+P (%) | 92.51 | 96.73 | 98.28 |
|
99.09 | 99.26 | 99.42 | 99.54 | 99.63 | |
ACC (%) | 90.85 | 96.42 | 98.54 |
|
97.90 | 95.72 | 92.94 | 89.93 | 86.68 | |
SEN (%) | 0.48 | 93.84 | 95.87 | 96.63 | 96.38 | 95.26 | 92.90 | 89.97 | 86.86 | 83.57 |
+P (%) | 93.54 | 97.20 | 98.59 | 99.07 | 99.26 | 99.38 | 99.48 | 99.56 | 99.64 | |
ACC (%) | 88.13 | 93.29 | 95.32 | 95.52 | 94.58 | 92.37 | 89.55 | 86.53 | 83.31 | |
SEN (%) | 0.51 | 93.06 | 95.28 | 96.08 | 95.83 | 94.69 | 92.30 | 89.35 | 86.24 | 82.98 |
+P (%) | 94.04 | 97.45 | 98.68 | 99.10 | 99.28 | 99.39 | 99.49 | 99.56 | 99.64 | |
ACC (%) | 87.87 | 92.96 | 94.86 | 95.01 | 94.04 | 91.78 | 88.94 | 85.92 | 82.73 | |
SEN (%) | 0.54 | 92.10 | 94.59 | 95.41 | 95.15 | 94.02 | 91.66 | 88.73 | 85.66 | 82.43 |
+P (%) | 94.26 | 97.63 | 98.73 | 99.12 | 99.30 | 99.41 | 99.50 | 99.57 | 99.65 | |
ACC (%) | 87.21 | 92.47 | 94.25 | 94.36 | 93.40 | 91.16 | 88.34 | 85.35 | 82.19 | |
SEN (%) | 0.57 | 90.90 | 93.54 | 94.41 | 94.14 | 93.03 | 90.73 | 87.86 | 84.85 | 81.65 |
+P (%) | 94.31 | 97.68 | 98.76 | 99.15 | 99.31 | 99.42 | 99.51 | 99.57 | 99.65 | |
ACC (%) | 86.17 | 91.51 | 93.30 | 93.38 | 92.44 | 90.25 | 87.48 | 84.54 | 81.42 | |
SEN (%) | 0.6 | 89.10 | 91.87 | 92.80 | 92.60 | 91.58 | 89.46 | 86.85 | 83.99 | 80.90 |
+P (%) | 94.28 | 97.67 | 98.76 | 99.16 | 99.32 | 99.43 | 99.52 | 99.58 | 99.65 | |
ACC (%) | 84.53 | 89.90 | 91.73 | 91.88 | 91.01 | 89.00 | 86.48 | 83.69 | 80.68 |
It can be seen that the optimal
The proposed method has two advantages. One is from the time efficiency as indicated in Section
Another advantage is from that there is no time length limitation for thresholding. As described in Section
From Table
Apart from record 207, there are still some missing and false recognitions on some other records. The main methodological defect of the algorithm is that amplitude threshold may fail to detect small R-peaks mixed in large ones (records 105, 106, 108, and 228). The
Missing detection due to the small amplitude of R-peaks.
Record 223 with different interface
Missing detection of small R-peaks on record 223
In this study, an adaptive and time-efficient methodology has been developed for automatic ECG R-peak detection. It is an adaptive method integrating WMRA, signal mirroring, local maximum detection, and amplitude and time interval thresholding. The accuracy performances were tested by using ECG records from MITDB and QTDB. Experimental results indicate that the proposed algorithm achieves average
The authors declare that they have no conflicts of interest.
This study is supported by the National Natural Science Foundation of China (61571113). This study is also supported by the Natural Science Foundation of Jiangsu Province of China (BK20160697), the International S&T Cooperation Program of China (2015DFA10490), and the China Scholarship Council (CSC).