Automated continuous quantitative measurement of proximal airways on dynamic ventilation CT: initial experience using an ex vivo porcine lung phantom

Figures & data

Figure 1 Flow chart of the algorithm.

Figure 2 An example of curved multi-planar reconstruction images with and without anatomical synchronization.

Notes: (A) The reference (inspiratory) phase. (B) The middle expiratory phase without registration. (C) The full expiratory phase without registration. (D) The middle expiratory phase with registration. (E) The full expiratory phase with registration. The location of the crosscut point (short red lines on each image) is indicated by the blue arrow. It is obvious that the crosscut points of the airway (enlarged in the small windows) slide peripherally on the unregistered images (B and C), while they keep locations on the registered images (D and E).

Figure 3 Illustration of the chest phantom. A porcine lung placed in the chest phantom (artiCHEST) inflates and deflates by a negative pressure cylinder.

Figure 4 A sample image for the quantitative tracheal measurement provided by the software.

Notes: A crosscut section (center square, surrounded by a yellow line) demonstrates the following: outer contour and cross-sectional area, inner contour and luminal area, long axis, short axis, wall area percent, respectively. Corresponding measurement points for all 24 frames are shown in the bottom of the image. The green plots in the upper right corner demonstrate continuous change in tracheal luminal area for this series of 24 frames.

Figure 5 Continuous changes in Ai and MLD. It is obvious that the two time curves change almost simultaneously.

Notes: A large, negative cross-correlation coefficient was obtained (R=−0.96), suggesting that inflation/deflation of the lung and dilation/constriction of the trachea occur simultaneously in the phantom. Blue dots indicate MLD and red dots indicate tracheal Ai.
Abbreviations: Ai, tracheal luminal size; MLD, mean lung density; HU, Hounsfield units.

Figure S1 (A–C) Original CT images. (D) An example of airway lumen segmentation. (E) An example of centerline tree construction. (F) The original crosscut image.(G) An example of wall contour estimation (green indicates the inner wall contour, yellow indicates the outer wall contour, blue dotted line indicates the maximum diameter, and pink dotted line indicates the minimum diameter).