High-fidelity prefiltration using a double bowtie design for quantitative low-dose pediatric spectral CT imaging
In parallel with efforts to minimize radiation exposure in pediatric patients, computed tomography (CT) remains the primary imaging modality for children, owing to its rapid scan times and excellent spatial resolution. Spectral CT, which enables the generation of virtual monoenergetic images (VMIs), can improve image contrast and reduce artifacts without increasing radiation dose. However, most CT systems are optimized for adult-sized patients, posing challenges in pediatric imaging and leading to missed opportunities to for example improve spectral separability—a crucial factor for enhanced image quality and accurate material decomposition. This study introduces a novel “double bowtie” filtration design that integrates traditional Teflon bowtie filters with K-edge material filtration to enhance spectral signal-to-noise ratio (SNR) in VMIs for pediatric patients. The double bowtie design involves optimizing the shape of both the Teflon and K-edge materials, as well as adjustments to system parameters, to maximize spectral performance while maintaining radiation doses similar to conventional pediatric CT protocols. Through simulations using pediatric-sized phantoms, significant improvements in SNR were observed with the double bowtie design, particularly when gadolinium was used as the K-edge material. This approach not only improves image quality but also opens the possibility of reducing radiation doses, thus making spectral CT more suitable for pediatric diagnostics. The proposed double bowtie filter can be integrated into existing CT systems without requiring specialized pediatric equipment, potentially improving diagnostic capabilities in spectral CT for pediatric patients. Although further clinical evaluations are necessary, these findings suggest a promising avenue for enhancing pediatric CT imaging while maintaining strict radiation safety standards.
Authors
- Olivia F. Sandvold (ORCID: https://orcid.org/0000-0002-2247-3445)
- Ying Ge (ORCID: https://orcid.org/0000-0003-1114-0204)
- Roland Proksa
- Amy E. Perkins (ORCID: https://orcid.org/0000-0002-1150-8972)
- Peter B. Noël (ORCID: https://orcid.org/0009-0001-0859-4240)
Institutions
- California University of Pennsylvania (US)
- Philips (United States) (US)
Publication Details
- Published
- 2025-04-08
- DOI
- https://doi.org/10.1117/12.3046287
- Citations
- 1
- Primary Topic
- Advanced X-ray and CT Imaging
- Type
- article
- Field-Weighted Citation Impact
- 0.67