Detection of Lungs Status Using Morphological Complexities of Respiratory Sounds

Table 2

Comparison of performances between the proposed and baseline methods [3].

Method	Feature vector	Type of; classifier	SEN (%)	SPE (%)	CA (%)	Training time (ms)	Testing time (ms)
Proposed	[𝜒,𝜉,𝜁,𝛼]	ELM	86.30	86.90	92.86	4.20	3.70
Method	[𝜒,𝜉,𝜁,𝛼]	SVM	86.30	85.80	91.50	23.60	8.10
Baseline	[𝑓25%,𝑓50%,𝑓75%]	ELM	80.47	79.71	87.66	2.62	1.10
Method [4]	[𝑓25%,𝑓50%,𝑓75%]	SVM	90.90	72.60	86.33	10.70	7.75

SEN: sensitivity; SPE: specificity; CA: classification accuracy; 𝜒: Kurtosis; 𝜉: Skewness; 𝜁: Lacunarity; 𝛼: sample entropy; 𝑓_25%,𝑓_50%,𝑓_75%: quartiles features.