Diagnostic radiograph based 3D bone reconstruction framework: application to the femur

Area of Science:

• Medical Imaging
• Orthopedic Surgery
• Biomedical Engineering

Background:

• Three-dimensional (3D) anatomical visualization is crucial for image-guided orthopedic surgery and minimally invasive procedures.
• Direct 3D imaging like Computed Tomography (CT) is limited to complex cases, necessitating reliance on 2D radiographic images for planning.
• Surgeons often mentally reconstruct 3D anatomy from 2D images, which can be challenging.

Purpose of the Study:

• To apply and validate a bi-planar 3D reconstruction methodology using orthogonal radiographs.
• To benchmark the accuracy of the proposed reconstruction against 3D CT scan data.
• To demonstrate the general applicability of the framework beyond long bones.

Main Methods:

• A bi-planar 3D reconstruction technique was developed utilizing prominent bony anatomy edges and contours from orthogonal radiographs.
• The methodology incorporates both outer contours and key interior edges of the bony anatomy.
• The human femur was used as the primary anatomical model for validation.

Main Results:

• The proposed methodology successfully reconstructed 3D anatomical models from 2D radiographic images.
• Reconstruction accuracy was validated by comparison with 3D CT scan data.
• The framework demonstrated applicability to various bony anatomies, not just long bones.

Conclusions:

• The bi-planar 3D reconstruction method offers a viable alternative to direct 3D imaging for orthopedic applications.
• This technique enhances anatomical visualization from conventional 2D radiographs, supporting surgical planning.
• The method's ability to incorporate interior bony edges improves reconstruction fidelity and broadens its applicability.