Table 4

Test set precision, recall and F1 scores for each input variable combination when using previously unseen data obtained from the next 271 patients

Combination of input variable	Lung pattern	Precision for test set (%)	Recall for test set (%)	F1 score for test set (%)
Biometrics with spirometry and plethysmography	Normal	99.20±1.69	90.40±8.98	94.37±5.32
	Obstructive	96.60±7.17	88.20±9.39	91.94±7.11
	Restrictive	94.60±1.43	100.00	97.22±0.76
	Obstructive and restrictive	90.40±6.13	88.00±6.53	88.94±4.02
Biometrics with spirometry	Normal	99.50±1.58	92.10±8.31	95.51±5.29
	Obstructive	96.20±8.07	92.20±5.98	94.15±6.99
	Restrictive	94.60±0.97	100.00	97.22±0.52
	Obstructive and restrictive	91.80±5.43	94.60±31.45	91.53±11.83
Spirometry and plethysmography	Normal	62.90±21.53	44.00±11.73	48.99±10.32
	Obstructive	28.30±21.18	13.80±13.48	Undefined
	Restrictive	79.50±6.57	79.40±14.09	78.80±8.22
	Obstructive and restrictive	48.40±8.57	94.70±5.08	63.63±7.34
Spirometry	Normal	44.00±11.12	67.10±13.53	51.76±8.17
	Obstructive	42.10±14.43	53.40±34.77	43.90±20.78
	Restrictive	75.20±8.20	56.50±20.12	62.87±17.40
	Obstructive and restrictive	39.40±27.83	42.70±36.57	Undefined

Mean±SD for the test sets of 10 experimental runs are shown. Data from the initial 477 patients were used for training and data from the next 271 patients were randomly assigned to the validation (136 patients) and test (135 patients) sets. For each experimental run, the model randomly selected one test per patient. Biometric variables are sex, age, height. Spirometry metrics included are FVC, FEV₁, FEF_25–75 and plethysmography metrics are RV and TLC. Undefined F1 score indicates that either precision, recall or both for that particular class was equal to zero for at least 1 of the 10 experimental runs.
FEF, forced mid-expiratory flow; FEV₁, forced expiratory volume in 1 s; FVC, forced vital capacity; RV, residual volume; TLC, total lung capacity.