Indeed, photon-counting CT (PCCT) offers a massive leap in spatial resolution for detecting submillimeter metastatic nodules in children. But as Joseph Cao called in the ARRS Online Course PCCT: It Counts in Children, Too, it isn’t a free lunch either. Maximizing performance requires a careful balance between resolution, noise, and kernel selection—especially for long-term follow-up of osteosarcoma, when the ability to identify tiny metastatic lung nodules as they age is so very vital.
Trade-off: Whereas ultra-high-resolution (UHR) mode at 0.2 mm offers incredible detail, alas, it does introduce a whole lotta noise.
- Cost: This noise must be compensated for via increased radiation dose or higher levels of iterative reconstruction, which can still result in grainy images.
- Alternative: Experts suggest that 0.4 mm reconstructions—already superior to prior scanner generations—may provide the sweet spot for detecting submillimeter nodules without the noise penalty.
Edge Watch: Kernel selection significantly impacts image quality in the lung parenchyma.
- Flaw: Using standard BL kernels on current PCCT platforms can cause a distinct loss of signal along the pleural interface.
- Fix: Quantitative kernels preserve this signal, maintaining the integrity of the lung edge across various iterative reconstruction levels.
RadFYI: Don’t rely solely on out-of-the-box vendor packages. Rads can run 20 or more different reconstruction variations to determine which settings work best for their specific pediatric populations.
Leave a Reply