. 2020 Apr 27;10(12):5613–5622. doi: 10.7150/thno.45985

Table 2.

Performance for predicting progression to severe illness with logistic regression analysis

	Day 0	P	Day 4	P	Changes from Day 0 to Day 4	P
APACHE-II	0.78(0.69~0.88)	0.554	0.77(0.66~0.89)	0.076	0.82(0.72~0.91)	0.046
NLR	0.78(0.67~0.88)	0.636	0.84(0.75~0.93)	0.156	0.78(0.67~0.88)	0.001
D-dimer	0.75(0.64~0.85)	0.410	0.78(0.67~0.88)	0.007	0.78(0.67~0.88)	0.001
PGV	0.76(0.65~0.86)	0.753	0.83(0.73~0.93)	0.015	0.84(0.74~0.93)	0.015
PSV	0.76(0.65~0.86)	0.644	0.87(0.77~0.97)	0.256	0.92(0.86~0.99)	0.464
PCV	0.76(0.66~0.86)	0.738	0.88(0.79~0.97)	0.572	0.91(0.85~0.98)	0.190
CT features	0.76(0.66~0.86)	Reference	0.89(0.80~0.97)	Reference	0.93(0.87~0.99)	Reference
NLR + CT features	0.78(0.67~0.88)	0.551	0.89(0.80~0.97)	0.432	0.93(0.87~0.99)	0.336

Note:

(a) Results are presented as the area under the receiver operating characteristic curve (AUC) along with 95% CI.

(b) PGV=Percentage of GGO volume; PSV=Percentage of semi-consolidation volume; PCV=Percentage of consolidation volume.