Fourteen normal pulmonary CT scans of pigs which underwent diagnostic CT to exclude pneumonic infiltrates method before other experiments were studied. A recruitment maneuver was performed before these CTs to reinflate atelectasis potentially obscuring small infiltrates. Forty-one porcine CT scans showing diffuse pulmonary opacifications resulting from experimental lung injury (repeated lung lavage with normal saline [13,20]) were analyzed.Sheep experimentsOne hundred sixty-eight whole-lung CT scans of mechanically ventilated sheep (range of body weight 46 to 70 kg) were indentified in our database. Sixty-six CT scans of normal lungs of sheep which underwent diagnostic CT for the exclusion of pneumonic infiltrates were studied. A recruitment maneuver was performed before these CTs to reinflate atelectasis potentially obscuring small infiltrates.

We also studied 102 lung CT scans of sheep with bilateral focal opacifications (dependent atelectasis) which developed during mechanical ventilation with pure oxygen. For 58 of the ovine CTs, a corresponding CT performed before and after an increase of airway pressure was available. These 58 pairs of consecutive CT scans reflecting substantially different lung conditions were analyzed in order to explore whether the extrapolation method can also be used to assess intra-individual changes of quantitative CT parameters.Quantitative CT analysis and extrapolation methodImages used had been generated by two different models of multi-slice CT scanners, either a Somatom Volume Zoom (120-kV tube voltage, 165-mA tube current, 4 �� 2.

5-mm collimation; Siemens, Erlangen, Germany) or a Philips MX8000 IDT 16 (120-kV tube voltage, 170-mA tube current, 16 �� 1.5-mm collimation; Philips Medical Systems, Hamburg, Germany). All CT images were reconstructed with standard reconstruction filters (“B35 f” on the Siemens and “B” on the Philips scanner) and slice thicknesses between 5 and 10 mm [21,22]. All quantitative analyses of CT data were performed at the University Hospital Leipzig. Six members of the research group at Leipzig University Hospital performed the manual image segmentations, while three other authors monitored the segmentations and performed the extrapolations. The distribution of pulmonary opacifications Entinostat was classified by two independent observers considering all CT slices available [4]. The Osiris software (University Hospital Geneva, Switzerland) was used for manual segmentation of the lung parenchyma. Identical to our previous studies, major pulmonary vessels, trachea and main bronchi were excluded [18,22,23]. Total lung volume (Vtotal) and mass (Mtotal) were calculated voxel-by-voxel from all lung voxels within the -1,000 to +100 Hounsfield units (HU) range.