A metal artifact induced HU systematic bias field modeling and correction method and system for needle intervention quantitative CT temperature measurement
By employing paired scanning and robust low-dimensional modeling in CT-guided thermal ablation, the systematic HU bias field caused by metal artifacts was corrected, solving the problem of CT temperature measurement error and achieving more accurate temperature estimation and ablation threshold determination.
Patent Information
- Application Number
- CN202610271264.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-03-06
- Publication Date
- 2026-06-16
AI Technical Summary
During CT-guided thermal ablation, the beam hardening, photon starvation, and scattering effects introduced by the metal ablation needle cause systematic HU bias field errors in temperature estimation and ablation threshold boundary shifts, which are difficult to effectively correct quantitatively using existing methods.
We employ a method of paired scanning bias field measurement, needle coordinate reparameterization, robust low-dimensional modeling, and noise threshold saliency discretization. Through voxel-by-voxel HU difference calculation, cylindrical coordinate reparameterization, multi-scale low-pass decomposition, and sparse compensation strategy, we explicitly couple the bias model to correct the CT temperature measurement link.
It reduces the interference of mid-to-low frequency systematic bias on temperature inversion, improves the quantitative consistency and interpretability of CT thermometry, and enhances the accuracy of temperature estimation and the reliability of ablation threshold determination.
Smart Images

Figure CN122223178A_ABST