Fig. 3

Schematic diagram of the concept for this study (a) and frequency spectrum analysis using DFT (b). Temporal changes in CSA of the right MPA are derived from respiration and cardiac pulsation. DVCT measures the combined wave of respiratory- and cardiac pulsation-derived changes in MPA-CSA (a). DFT was used to extract the cardiac pulsation-derived MPA-CSA change from the measured waveforms, because DFT can separate the combined wave into individual frequency waves (b). First, DFT was performed on MPA-CSA waveforms measured using DVCT. Subsequently, the amplitude of the waveform was quantified from the DFT complex data by calculating their absolute values. Then, the cardiac pulsation-derived components were extracted based on the heart rate. The cardiac pulsation-derived MPA-CSA changes were defined as peak-to-bottom of the waveforms, i.e., twice the waveform amplitude. Finally, the change ratio of MPA-CSA was calculated by dividing the cardiac-pulsation-derived MPA-CSA changes by the direct-current component of the DFT components (S₀) to eliminate patient-specific anatomical MPA-CSA differences. The direct current component indicates the MPA-CSA without respiratory- and cardiac pulsation-derived waves, which corresponds to the mean value of MPA-CSA along all scan phases. These procedures were performed for the data determined during the inhalation and exhalation phases, respectively. CSA, Cross-sectional area; DFT, Discrete Fourier transformation; DVCT, Dynamic ventilation computed tomography; MPA, Main pulmonary artery