This project involves the development of reconstruction algorithms for multislice CT using helical data acquisition. With the recent introduction of multi-row detectors and the ability to acquire data over 360 degrees in less than 500 ms, x-ray CT is undergoing a new phase of rapid innovation. Compared to single-slice CT scanners, CT scanners with multi-row detectors can cover larger volumes, achieve higher axial resolution, avoid motion artifacts due to respiration, and improve the detectability of low-contrast details. These advantages have been identified with 4-row scanners and will become increasingly significant with greater numbers of detector rows. Eventually, a powerful imaging system with the capability of achieving early and reliable diagnosis of numerous diseases will be available for biomedical research. With more than 4 rows it is known that the cone-beam (CB) divergence of the beams cannot be neglected during reconstruction. However, from a mathematical point-of-view, designing an algorithm that accurately accounts for this divergence poses a challenge. The design of helical CB reconstruction algorithms is not a priority for CT manufacturers due to the numerous technological problems hindering the development of scanners with multi-row detectors. However, the future of multi-slice CT depends on progress that can be made in this field. Highly accurate reconstruction algorithms are needed to realize the full potential and development of multi-row scanners. This research project aims to satisfy that need.
The specific aims are (1) to implement, characterize, and compare existing helical CB reconstruction algorithms, using a collection of figures-of-merit, and to disseminate the coded algorithms; (2) to derive, implement, and characterize new helical CB reconstruction algorithms that provide 3D images with isotropic spatial resolution and high local temporal resolution (< 300 ms); and (3) to derive, implement and characterize new helical CB reconstruction algorithms that provide 3D images with isotropic spatial resolution and high detectability of low-contrast details (possibly at the expense of temporal resolution). Indirectly, this project will have significant impacts on all aspects of medical imaging - particularly in oncology, angiography, evaluation of infections and cardiac diseases, trauma, and radiotherapy.
| Dennerlein, Frank; Noo, Frederic; Hornegger, Joachim et al. (2007) Fan-beam filtered-backprojection reconstruction without backprojection weight. Phys Med Biol 52:3227-40 |
| Pack, Jed D; Noo, Frederic; Clackdoyle, Rolf (2005) Cone-beam reconstruction using the backprojection of locally filtered projections. IEEE Trans Med Imaging 24:70-85 |
| Pack, Jed D; Noo, Frederic; Kudo, H (2004) Investigation of saddle trajectories for cardiac CT imaging in cone-beam geometry. Phys Med Biol 49:2317-36 |
| Noo, Frederic; Clackdoyle, Rolf; Pack, Jed D (2004) A two-step Hilbert transform method for 2D image reconstruction. Phys Med Biol 49:3903-23 |
| Kudo, Hiroyuki; Rodet, Thomas; Noo, Frederic et al. (2004) Exact and approximate algorithms for helical cone-beam CT. Phys Med Biol 49:2913-31 |
| Noo, Frederic; Defrise, Michel; Kudo, Hiroyuki (2004) General reconstruction theory for multislice X-ray computed tomography with a gantry tilt. IEEE Trans Med Imaging 23:1109-16 |
| Noo, Frederic; Pack, Jed; Heuscher, Dominic (2003) Exact helical reconstruction using native cone-beam geometries. Phys Med Biol 48:3787-818 |
