An automatic reduction method of 3D bone fragments based on a novel section contour point descriptor

Area of Science:

• Orthopedic Surgery
• Biomedical Engineering
• Medical Imaging

Background:

• Comminuted fractures present significant surgical challenges in precise bone fragment reduction and personalized internal fixation.
• Current clinical treatments struggle with the complexity of accurately aligning and stabilizing multiple bone fragments.

Purpose of the Study:

• To develop an automated method framework for precise reduction and internal fixation of comminuted fractures.
• To address the challenges of accurate fragment alignment and personalized implant design in complex fractures.

Main Methods:

• Utilized Gaussian mixture model (GMM) for distinguishing bone fragment section points from noise.
• Employed ellipse fitting and a novel descriptor for section point matching and feature description.
• Applied Convolution Auto-Encoder (CAE) and genetic algorithms for feature vector extraction and bone model registration.
• Reconstructed internal fixation plates based on registered bone models.

Main Results:

• The proposed automated method demonstrated high accuracy in precise reduction and internal fixation of comminuted fractures.
• Verification experiments confirmed the method's good efficiency in handling complex fracture patterns.
• The framework successfully enabled the reconstruction of personalized internal fixation plates.

Conclusions:

• The developed automated method framework offers a viable solution for precise reduction and internal fixation of comminuted fractures.
• This approach has the potential to significantly support and advance minimally invasive treatment strategies in orthopedic trauma.
• The study highlights the efficacy of integrating GMM, CAE, and genetic algorithms for complex orthopedic applications.