Abstract
C-arm X-ray imaging systems are widely applied in surgeries. Overlaying X-ray with optical images during the surgery has been shown to be an efficient approach. Moreover, overlaying needed data from different modalities in an augmented reality (AR) manner can improve the accuracy of surgical procedures, decrease the variability of surgical outcomes, reduce trauma to the critical structures, increase the reproducibility of surgeons’ performance, and reduce radiation exposure. C-Arm geometric calibration and recovering the C-arm pose are essential for surgical navigation and AR applications in operating rooms. Therefore, in this paper, existing researches for calibration and pose estimation of C-arm devices in surgical AR applications are evaluated from photogrammetric point of view. Then, a proposed marker-based method for C-arm pose estimation is introduced. For this purpose, a marker is designed to facilitate tracking and pose estimation in mixed reality based on golden section principle, and perspective invariants such as cross-ratios, collinearity, and intersection. Moreover, a procedure is also proposed for fast determination of these fiducial markers. The experiments show benefits of such a structure which has a limited occlusion with consistency to different conditions such as narrow field of view and at the same time, even in images with high projectivity. It also results that the distortion correction step is important and the effect of distortion of X-ray images can cause inconsistency in the perspective invariants.
Introduction
In modern surgery, mobile C-arm X-ray systems are indispensable for guiding orthopedic and neurosurgical procedures. However, they impose a significant "mental mapping" burden on surgeons, who must look away from the patient to a separate monitor, interpret 2D shadows of 3D anatomy, and mentally align that information with the physical surgical site.
Augmented Reality (AR) addresses this by overlaying X-ray data directly onto the surgeon's view. The primary technical challenge is pose estimation: the system must precisely track the C-arm's position relative to the patient. As shown in the taxonomy of tracking methods, researchers have explored various avenues, including external tracking systems and additional sensors attached directly to the C-arm.
Existing solutions like optical trackers often suffer from line-of-sight obstructions in crowded operating rooms, while magnetic trackers are prone to interference from metallic surgical tools. This research proposes a robust marker-based approach designed to be fast and minimally obstructive.
Methodology
The authors introduce a direct marker-based method utilizing a unique pentagram (star-shaped) fiducial marker.
Marker Design and Invariance
The marker consists of radiopaque ball bearings arranged in a pattern rooted in the Golden Section principle. Because the ratios of distances in a pentagram are tied to the golden ratio ($\phi \approx 1.618$), the "cross-ratio"—a projective invariant—remains constant even under perspective tilt or distortion. This allows the system to reliably identify individual points in a 2D X-ray image regardless of the C-arm's orientation.
Processing Pipeline
- Distortion Correction: X-ray images are pre-processed using Bernstein polynomials to correct for "pincushion" and "S-shape" hardware distortions.
- Point Identification: A blob detector locates the markers, and a "Minimum Volume Enclosing Ellipsoid" (MVEE) algorithm defines the marker's boundary.
- Pose Estimation: The Infinitesimal Plane-based Pose Estimation (IPPE) algorithm calculates the C-arm's 3D position and orientation.
Results and Comparison
The proposed system was evaluated using 190 simulated X-ray images and real-world data from a Siemens Siremobil Iso-C C-arm. The system demonstrated high efficiency and clinical relevance compared to other sensor-based methods.
Key performance metrics include:
- Accuracy: The system achieved a translation error of 5.86 mm and a rotation error of 0.91°, which is considered acceptable for many AR-guided surgical tasks.
- Speed: The total latency for identification, pose estimation, and AR augmentation was approximately 73 ms, enabling real-time interaction.
- Robustness: The marker remained identifiable even at steep angles, though accuracy began to decline when points overlapped at rotations exceeding 60°.
Discussion and Conclusion
By prioritizing geometric math over visual complexity, the authors developed a "sparse" marker that provides high-speed tracking without obscuring critical anatomy. The use of cross-ratios ensures robustness against perspective changes, though the study emphasizes that rigorous distortion correction is mandatory to maintain these mathematical invariants. Future work will focus on mitigating self-occlusion at extreme angles, potentially through the use of smaller markers or multi-marker arrays.
Summary updated