Thickness accuracy of virtually designed patient-specific implants for large neurocranial defects

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Neurosurgery

Background:

• Patient-specific implants for cranial reconstruction are crucial for restoring form and function after trauma or surgery.
• Current computer-aided design (CAD) methods for implant generation can be complex and may not accurately capture patient-specific bone thickness.
• Existing reconstruction protocols often lack the precision needed for complex, extensive cranial defects.

Purpose of the Study:

• To compare the accuracy of thin-plate spline (TPS)-based implants versus traditional CAD-based implants for cranial reconstruction.
• To evaluate the ability of each method to reproduce patient-specific cranial bone thickness.
• To assess the geometric accuracy and thickness deviation of implants generated by both approaches.

Main Methods:

• Utilized computed tomography (CT) data from three human crania with virtually created defects.
• Performed 132 virtual reconstructions using both CAD-based and TPS-based implant design approaches.
• Quantified accuracy using root-mean-square error (RMSE) for geometric accuracy and implant thickness deviations.

Main Results:

• TPS-based reconstructions demonstrated superior geometric accuracy with an RMSE of 0.08–0.18 mm, compared to 0.50–1.25 mm for CAD-based implants.
• TPS implants accurately reproduced patient-specific thickness distribution (RMSE 0.63–0.70 mm), outperforming CAD-based implants (RMSE 0.63–1.67 mm).
• Both methods yielded highly accurate implants, but TPS excelled in replicating intricate thickness variations.

Conclusions:

• Thin-plate spline (TPS) offers a more accurate and precise method for generating patient-specific cranial implants compared to traditional CAD approaches.
• The TPS method's ability to replicate patient-specific thickness distribution is a significant advantage for complex cranial defect reconstruction.
• TPS-based implant design represents a promising advancement for improving outcomes in neurosurgical and craniofacial reconstructions.