TABLE 5. Comparison of Pulmonary Disease and Asthma Cough Diagnosis.

Ref	Disease	Method	Dataset	Size of the data	Specificity	Sensitivity	F1-score	Accuracy
[4]	Asthma	Signal processing techniques	Cough sound	12 asthmatic and 12 healthy male and female	–	–	–	100%
[52]	Asthma	Prediction model (LASSO-penalized logistic regression)	Cough sound, survey and questionnaire	1226 symptomatic children (345 (28%) had asthma)	71%	72%	–	–
[28]	Asthma	FeNO measurements and an airway responsiveness test	Questionnaire	Cold air and talking sounds from 163 patients	81%.	44%	–	–
[64]	Asthma	CNN	Cough sound	6737 cough samples and 8854 control sounds by 5 different recording devices from 43 subjects	–			90.9%
[83]	Pulmonary disease or asthma	LR	Cough sound, questionnaire and vital signs	54 patients (22 healthy)	81%	81%	–	81%
[84]	Pulmonary disease	RF	Cough sound, survey	100 coughs	82%	80%	–	80.67%
[7]	Pulmonary disease or asthma	LR and Bayesian Network (BNN)	Cough sound and questionnaire	325 patients	84%	84%	–	90%
[59]	Pulmonary disease or asthma	–	Cough sound	228 COPD patients	76%	90%	–	–
[85]	Asthma	Gaussian Mixture Model–Universal Background Model (GMM-UBM)	Cough sound, vital signs and questionnaire	1192 patient cough sounds, and 1140 healthy cough sounds	84.76%	82.81%	–	80%
[33]	Pulmonary disease	DNN and confusion matrix	Cough sound, vital signs and questionnaire	108 Subjects		18%	67%	41%
[68]	Pulmonary disease	HMM	Cough sound	–	–	–	–	92%