New CBCT framework swaps data completeness for clinical sufficiency

A new review paper from Yi Jia, Rongjun Ge, Yang Chen, Yan Xi, and Wenjun Xia challenges the long-standing assumption that intraoperative cone-beam CT (CBCT) should strive for 'data completeness' akin to fan-beam CT. The authors argue that under a single circular trajectory—standard for mobile C-arms—complete data is mathematically unattainable in three-dimensional cone-beam geometry. Moreover, blindly increasing sampling to approach completeness only worsens the inherent trade-offs between image quality (Q), imaging time (T), and radiation dose (D). Instead, the paper reframes the evaluation criterion around 'data sufficiency,' defined as meeting the minimum image-quality threshold required for specific clinical tasks such as screw placement or tumor resection. This shift acknowledges that approximation errors are acceptable when they do not compromise clinical decision-making, thereby enabling a better Q-T-D balance.

The framework synthesizes evidence from multiple surgical scenarios to show that task-specific minimum quality thresholds can be established, allowing operators to reduce scan time and dose while still obtaining actionable images. For example, in orthopaedic procedures where high-contrast bone visualization is needed, lower dose protocols may suffice compared to soft-tissue imaging. By moving from a one-size-fits-all completeness standard to a task-driven sufficiency model, the approach promises to make intraoperative CBCT faster, safer, and more practical for real-time guidance. The paper is available on arXiv and has implications for medical physics, computer vision, and image processing—potentially influencing future C-arm system design and clinical protocols.