Title: High-Accuracy X-ray-to-CT Registration Enables Electromagnetic Navigation for Femoral Osteotomy

Abstract: Executing preoperative plans with precision during corrective femoral osteotomies is a persistent clinical challenge. Existing methodologies are often constrained by inconsistent accuracy, high invasiveness, and significant radiation exposure. Specifically, free-hand techniques and patient-specific instrumentation (PSI) typically necessitate more than 30 and 6 fluoroscopic images, respectively. To address these limitations, we have developed an integrated navigation system utilizing electromagnetic tracking (EMT) designed for femoral osteotomies, aiming to reduce both surgical dissection and the need for intraoperative fluoroscopy.

The proposed solution integrates CT-based preoperative planning with a single intraoperative C-arm calibration step. By acquiring just two fluoroscopic images during initialization, the system achieves precise X-ray-to-CT registration. This capability allows for real-time, fluoroscopy-free EMT navigation of bone fragments and the saw blade in relation to the preoperative plan, supporting both uniplanar and biplanar osteotomy procedures.

In a feasibility study involving 18 synthetic femora, EMT guidance demonstrated superior performance compared to free-hand execution in terms of total angular error. The EMT group recorded an error of $(3.05 \pm 0.75)^\circ$, significantly lower than the $(6.32 \pm 2.36)^\circ$ observed in the free-hand group ($p=0.031$). These results were achieved under conditions of identical minimal surgical exposure for both techniques. Notably, no trials guided by EMT exceeded the $>5^\circ$ clinical threshold, whereas the free-hand method resulted in 4 outlier errors out of 6 trials.

Furthermore, the system exhibited statistical equivalence to PSI regarding both total angular error ($p \le 0.02$) and total translational error ($p=0.048$), with thresholds set at $\pm 2^\circ$ and $\pm 2\text{mm}$. User questionnaire scores showed no significant differences between the methods. Because the system transfers preoperative plans using only two fluoroscopic images while maintaining accuracy comparable to PSI without requiring additional surgical exposure, it provides a strong foundation for future cadaveric and clinical validation studies.

Source: arXiv Generated at: 2026-06-03 00:00:00 UTC