(Monday, 7/15/2019) 1:15 PM - 1:45 PM
Room: Exhibit Hall | Forum 9
Purpose: To measure how the modulation transfer function (MTF) changes as the display field of view (DFOV) is changed using a range of reconstruction algorithms. These changes can significantly degrade the expected image sharpness throughout the image when reconstructions use larger DFOVs.
Methods: The MTF on a GE 750 HD scanner was measured using an edge method with the edge angled at 3 degrees. Measurements were taken with a slice thickness of 3.75mm, using the Plus reconstruction option. These measurements were performed at DFOVs of 5cm, 10cm, 20cm, 30cm, 40cm, and 50cm, with reconstruction algorithms of Standard, Detail, Bone, and Edge for each DFOV. The MTF across reconstruction algorithm were compared for each DFOV, and the variation in MTFs for varying DFOVs were compared for each algorithm.
Results: Detail and sharper reconstruction algorithms have degraded MTFs at the large DFOVs of 40cm and 50 cm; the Standard at 50 cm. The Detail algorithm is reported to have better sharpness than the Standard. This is true for the smaller DFOVs. However, we found that the MTF for the Detail kernel is actually less sharp than the Standard algorithm with the largest DFOVs.
Conclusion: The DFOV can significantly affect image sharpness, sometimes in unexpected ways. Clinicians, technicians, and physicists need to be aware of these limitations in image sharpness when using DFOVs larger than 30 cm. They also need to be aware of how particular reconstruction algorithms behave as the DFOV is varied. One might think that a chest or abdomen scan of a large adult would be sharper using the Detail algorithm compared to the Standard algorithm. Actually, changing from a Standard to Detail algorithm here would result in a less sharp image, even though with head scans and small adult scans the Detail algorithm would actually be sharper, as expected.
Funding Support, Disclosures, and Conflict of Interest: FNR supplies CT protocols to GE Healthcare under a licensing agreement.
Keywords
Not Applicable / None Entered.
Taxonomy
Not Applicable / None Entered.
Contact Email
