Fifty-eighth annual meeting of the american association of physicists in medicine
MO-DE-207A-03: Characterization of Potential Gains in a Novel Exponential 4DCT Reconstruction Algorithm
To sample the breathing cycle and obtain accurate target volumes, 4DCT requires the entire field of view (FOV) to be illuminated during one breathing period. However, interplay effects between internal motion and couch pitch may adversely impact the temporal resolution, thereby blurring object edges. This work performs a characterization of exponential 4DCT reconstruction to highlight potential gains over standard cos2 weighted reconstruction algorithm.
A respiratory motion platform translated several objects in the superior-inferior direction at varied breathing rates (8–20 breaths/minute) and couch pitches (0.06–0.1 A.U.) to evaluate the interplay between parameters. Ten-phase 4DCTs were acquired (0.5s rotation) and data were reconstructed with cos2 and exponential weighting factors. To isolate a metric to quantify temporal resolution (i.e. remove couch interplay effects), a small object was translated in the anterior-posterior direction. Full-width half maximum (FWHM) intensity distributions were quantified between reconstruction algorithms and a static case. 4DCT sinogram data for fifteen lung and abdomen patients were retrospectively reconstructed using cos2 and exponential weighting factors. Image subtractions were generated to elucidate intensity and boundary differences.
After taking the static object size into account, the FWHM of exponential weighting was 1.5±1.2 mm (range: 0.1–4.1) as compared to the FWHM of cos2 3.4±2.4 mm (range: 0.3–8.0). This translated to estimated improvements in temporal resolution of 33.5±48.2 ms (range: 0.03–170.9). Slower breathing periods, faster couch pitches, and intermediate 4DCT phases where velocity is highest showed a tendency to have the largest improvements in temporal resolution with exponential reconstruction. For patient cases, coronal views showed less blurring at object boundaries and local intensity differences near moving features.
Exponential weighted 4DCT offers potential for improving temporal resolution in 4DCT, thus improving the image sharpness near the boundaries. Understanding the potential implications on delineation ability is an important next step of this work.
The submitting institution holds research agreements with Philips Healthcare.