Virtual reconstruction of pelvic tumor defects based on a gender-specific statistical shape model

Area of Science:

• Orthopedic surgery
• Biomedical engineering
• Medical imaging

Background:

• Pelvic tumors often necessitate bone removal, leading to complex anatomical defects.
• Accurate reconstruction is crucial for restoring function and stability.
• Current methods like contralateral mirroring have limitations in complex cases.

Purpose of the Study:

• To evaluate a statistical shape model (SSM)-based approach for virtual anatomical reconstruction of pelvic defects.
• To compare the accuracy of SSM-based reconstruction with the traditional mirroring method.
• To assess the utility of SSM for patient-specific implant design in pelvic tumor cases.

Main Methods:

• Utilized gender-specific pelvic statistical shape models (SSMs) (n=50 per gender).
• Generated patient-specific implant geometries using the SSM approach.
• Reconstructed eight tumor-damaged pelvic bone defects virtually.
• Quantitatively compared SSM reconstruction accuracy against the contralateral mirroring method.

Main Results:

• Achieved a mean deviation distance of 0.89 mm and 1.26 mm for defect reconstruction.
• Demonstrated that SSM-based reconstruction offers clinically acceptable accuracy.
• Found SSM reconstruction accuracy to be comparable to the mirroring method.

Conclusions:

• The SSM-based method is a valuable tool for planning pelvic reconstructive surgery.
• This approach facilitates the design of patient-specific implants for tumor defects.
• SSM provides a reliable alternative for complex pelvic reconstructions.