Table 1.
Characteristics of study population.
Table 2.
Mean and standard deviation of the laboratory parameters of all 53 COVID-19 patients.
Fig 1.
The YACTA software performed a fully automatic segmentation of airways, vessels and lung parenchyma of an 84y/o male with extensive lung infiltrates who received high-flow oxygen therapy and invasive ventilation.
In the course of the disease he developed renal insufficiency, septic shock and died 21 days after symptom onset and 8 days after hospital admission. Consolidations are shown in red, ground-glass opacities in orange, healthy lung in blue, vessel voxels in grey are excluded. The 3D rendering shows the lung segmentation with their segmented lobes in different colors, in the statistical evaluation the lung parenchyma was considered as a whole.
Table 3.
QCT parameters of 53 COVID-19 patients in the soft kernel images IMR1.
Fig 2.
Heat map of Pearson’s correlation between QCT lung segmentation (y-axis) and laboratory parameters and clinical outcome (x-axis).
Significant codes: *** p < 0.001 / ** p < 0.01 / * p < 0.05.
Fig 3.
Correlations between LDH and QCT parameters with highest Pearson‘s r for 75th lung percentile in comparison to HLI-700 and FIBI-700.
Fig 4.
Heat map of coefficient of determination r2 for the correlation of outcome parameters with laboratory and QCT parameters.
For binary outcome parameters (NIV_Ventilation and Invasive_Ventilation) the Mc Fadden pseudo r2 was calculated, there is no p-value to specify for this value. Signif. codes: *** p < 0.001 / ** p < 0.01 / * p < 0.05 / ’ no p-value available.
Fig 5.
ROC curves of invasive ventilation QCT parameters alone with FIBI-700 (a) vs. 75th percentile lung density (b) vs. LDH (c) vs. Procalcitonin (d).
Fig 6.
ROC curves for invasive ventilation consisting of the combination of QCT and laboratory work: FIBI-700, LDH and Procalcitonin; (b) 75th percentile lung density, LDH and Procalcitonin.