Detecting Congestive Heart Failure by Extracting Multimodal Features and Employing Machine Learning Techniques

Table 6

Algorithm comparison of previous studies.

Author	Title of article	Method	Performance
Li et al. [87]	Combining convolutional neural network and distance distribution matrix for identification of congestive heart failure	CNN	TA = 81.9%
Isler and Kuntalp [81]	Combining classical HRV indices with wavelet entropy measures improves to performance in diagnosing congestive heart failure	KNN	ACC = 81.92%; Sens = 82.74%; Spec = 96.27%
Narin et al. [79]	Investigating the performance improvement of HRV indices in CHF using feature selection methods based on backward elimination and statistical significance	SVM	Sens = 79.33%; Spec = 94.47%
Isler and Kuntalp [80]	Heart rate normalization in the analysis of heart rate variability in congestive heart failure	KNN	Sens = 82.72%; Spec = 100.0%
Pecchia et al. [88]	Discrimination power of short-term heart rate variability measures for CHF assessment	CART	Sens = 89.75%; Spec = 100.0%
Elfadil and Ibrahim [89]	Self-organising neural network approach for identification of patients with congestive heart failure	Spectral; NN	ACC = 83.65%
Yang et al. [90]	A heart failure diagnosis model based on SVM	SVM; NB; CA	TA = 74.42%
Chang et al. [91]	Decision making model for early diagnosis of CHF using rough set and decision tree approaches	RS; DT	SEN = 97.53%
Our method	Extraction of multimodal features to predict congestive heart failure (CHF)	DT	Sens = 82%; Spec = 82%; TA = 81.9%
		SVM linear	Sens = 96%; Spec = 89%; TA = 93.1%
		EnsembleSubspace discriminant	Sens = 93%; Spec = 89%; TA = 91.4%