Research Article
Detecting Congestive Heart Failure by Extracting Multimodal Features and Employing Machine Learning Techniques
Table 5
Features-based significance level to distinguish the CHF and NSR subjects.
| | Feature | CHF | NSR | -value | | Mean ± std | Mean ± std |
| | SDANN | 0.010 ± 0.015 | 0.018 ± 0.008 | 7.68 × 10−36 | | SDNN | 0.066 ± 0.032 | 0.086 ± 0.026 | 2.85 × 10−53 | | SDSD | 0.056 ± 0.045 | 0.028 ± 0.018 | 4.08 × 10−23 | | RMSSD | 0.063 ± 0.050 | 0.035 ± 0.020 | 1.04 × 10−24 | | TP | 347099 ± 316751 | 858649 ± 563951 | 1.11 × 10−24 | | ULF | 80616 ± 80069 | 228361 ± 175578 | 1.5 × 10−20 | | VLF | 178871 ± 166993 | 501651 ± 350518 | 1.32 × 10−22 | | LF | 42353 ± 50725 | 68217 ± 53525 | 2.6 × 10−21 | | HF | 45257 ± 58350 | 60419 ± 55206 | 3.05 × 10−18 | | LFHF | 1.442 ± 0.872 | 1.304 ± 0.389 | 6.55 × 10−46 | | MSEKD | 1.370 ± 0.293 | 1.464 ± 0.179 | 2.37 × 10−93 | | MApEn | 0.004 ± 0.006 | 0.0009 ± 0.003 | 2.07 × 10−5 | | WEShannon | 6594 ± 962 | 6151 ± 1351 | 1.38 × 10−84 | | WELogEn | −17460 ± 5702 | −14830 ± 5722 | 4.62 × 10−55 | | WETh | 19999 ± 0.347 | 19999 ± 0.201 | 0 | | WESure | −11085 ± 2692 | −9771 ± 2829 | 4.32 × 10−68 | | WENorm | 12613 ± 2043 | 13594 ± 2070 | 1.96 × 10−94 | | RMS | 0.660 ± 0.096 | 0.708 ± 0.098 | 4 × 10−99 | | Var | 0.005 ± 0.005 | 0.008 ± 0.005 | 4.26 × 10−27 | | Smoothness | 0.999 1.08 × 10−5 | 0.999 ± 1.11 × 10−5 | 0 | | Kurtosis | 40.2 75.4 | 5.125 ± 7.386 | 0.000108 | | Skewness | 1.996 ± 2.672 | 0.264 ± 0.656 | 8.93 × 10−7 |
|
|
