The “X-Ray On” synchronization signal for three scans. The transition points from HIGH to LOW indicated the start time of a CT image acquisition. The “X-Ray On” time is about and the “X-Ray Off” time is about .
Results of lung segmentation and fractional air content conversion. (a) Original CT slice. (b) Fractional air content.
Internal motion measured by the 3D template matching algorithm. Template locations (▵) at expiration (a) coronal view, and (b) sagittal view. The larger ▵ indicates the location of the template used in (d)–(f). (c) Breathing curve for air content. (d) Measured motion in 3D, CC, AP, and LR directions for a template in a couch position that intercepts the diaphragm. (e) Trajectory in sagittal plane. (f) Fitting 3D motion as a function of internal air content.
The internal air content and spirometer-measured tidal volume for the 15 scans in a single couch position (12 slices, total thickness). (a) Before correction. (b) After systematic time offset, phase difference, and spirometry drift corrections. Note that the amplitude of the air content curves is normalized for displaying purpose.
Improvement in the cross correlations between the air content and tidal volume. Couch positions 6 and 17 correspond to the apex and inferior extents of the lungs, respectively. The shown curves are (1) before correction, (2) spirometry drift correction only, (3) systematic time offset correction only, (4) systematic time offset plus phase difference corrections, and (5) all three corrections, respectively. Note that curves (3) and (4) almost overlap for this patient.
Fitting lines between air content and tidal volume for three couch positions within the lung.
(a) Correlation between internal air content and tidal volume (solid), and ratios of air content to tidal volume change (dashed), as a function of couch position. (b) The air content at normal tidal exhalation (solid) and phase difference (dashed) as a function of couch position.
Results of abutting the air content surfaces between two neighboring couch acquisitions. (a) The surfaces mismatch along the tidal volume axis due to spirometry drift. (b) Result of the two-dimensional, second-order polynomial fits to the first four slices (1–4) for the higher surface and the last four slices (9–12) for the lower surface in (a), respectively. ◻ and 엯: original data, × and +: fitted data. The solid and dotted lines are the extrapolation of the two fitted surfaces at the abutment. (c) The same as in (b) but after tidal volume drift correction. The two lines now overlap with each other. (d) After tidal volume drift correction, the surfaces match.
Slice air content as a function of the tidal volume and CT slice position. This is a piecewise joined surface that has not been smoothed. The apex and diaphragm are located near 100 and slice position, respectively.
Comparison between the air content integrated over all CT slices and the spirometer-measured tidal volume . This individual-slice analysis agrees well with the analysis conducted using the couch-based air content.
Volume analysis results for 12 four-dimensional computed tomography (CT) patients. Entries with dashes indicate patients that did not have CT synchronization signals available.
Article metrics loading...
Full text loading...