Skip to main content
. Author manuscript; available in PMC: 2023 Dec 1.
Published in final edited form as: AI (Basel). 2023 Oct 10;4(4):875–887. doi: 10.3390/ai4040044

Table 2.

Feature classification and diagnostic performance of quantitative ultrasound features by different users.

Thickness Thickness Variation PID Nonlinearity Tortuosity Echo Intensity Echo Heterogeneity Overall Performance (AUC)
User 1 COVID-19 5.20 0.23 2.68 0.25 1.54 186.86 18.45 0.96
Normal 1.80 0.06 0.70 0.81 1.04 195.06 14.69
p-value 0.04 0.04 0.01 0.01 0.01 0.47 0.12
User 2 COVID-19 4.62 1.97 2.45 0.20 1.46 184.79 20.96 0.98
Normal 1.41 0.36 0.49 0.90 1.01 197.41 18.60
p-value 0.00 0.00 0.00 0.00 0.00 0.30 0.43
User 3 COVID-19 6.03 0.25 2.45 0.23 1.33 129.20 34.46 0.92
Normal 2.51 0.06 0.59 0.74 1.08 137.74 32.38
p-value 0.03 0.00 0.00 0.00 0.07 0.20 0.53
User 4 COVID-19 4.69 0.67 1.62 0.23 1.51 202.21 13.91 0.84
Normal 1.10 0.85 1.84 0.79 1.16 212.31 11.50
p-value 0.00 0.69 0.84 0.04 0.02 0.51 0.06
User 5 COVID-19 5.84 1.40 3.03 0.13 1.10 167.21 34.62 0.99
Normal 2.16 0.44 0.93 0.78 1.04 182.48 27.56
p-value 0.00 0.01 0.01 0.00 0.04 0.22 0.04