Premium
CE n Pi T : Helical cardiac CT reconstruction
Author(s) -
Bontus Claas,
Koken Peter,
Köhler Thomas,
Grass Michael
Publication year - 2006
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.2219776
Subject(s) - iterative reconstruction , projection (relational algebra) , radon transform , tomography , algorithm , detector , reconstruction algorithm , trajectory , computer science , radon , artificial intelligence , computer vision , physics , optics , quantum mechanics , astronomy
Computer tomography (CT) scanners with an increasing number of detector rows offer the potential of shorter scanning times. Nevertheless, the reconstruction problem becomes more challenging, since cone beam artifacts are likely to enter. Here, we consider helical cardiac CT. We analyze how a relationship can be established between exact reconstruction algorithms and the demand to perform a cardiac gating. Utilizing the redundancies requires the consideration of all kinds of Radon planes. For the reconstruction algorithm proposed here, we separate the data into two parts. The first part contains contributions of Radon planes, which are measured with a large number of redundancies. The second part contains the remaining contributions. As it turns out, the second part contributes rather to the low‐frequency contents of trans‐axial slices. Therefore, we propose to perform a gated back‐projection only for the first part, while the second part is back‐projected in an ungated way. Data from the complete source trajectory are employed in the reconstruction process in contrary to conventional helical cardiac reconstruction methods. Moreover, all different types of Radon planes are taken into account in the reconstruction, though an ECG‐dependent cardiac gating is applied. The reconstruction results, which we present for clinical and simulated data, demonstrate the high potential of CE n Pi T for helical cardiac CT with large cone angle systems.