Automatic Anterior Cruciate Ligament Insertion Center Positioning Method Framework for Isometric Reconstruction and...

Area of Science:

• Orthopedic Surgery
• Medical Imaging
• Computer-Aided Surgery

Background:

• Anterior cruciate ligament (ACL) reconstruction demands high precision, particularly in graft placement.
• Limited arthroscopic visibility poses challenges for surgeons in determining accurate insertion points and angles.
• Current methods may lack the precision needed for optimal patient outcomes.

Purpose of the Study:

• To develop a novel framework for precise anterior cruciate ligament (ACL) insertion center localization.
• To enhance anatomical reconstruction accuracy in ACL surgery.
• To improve the reliability and success rates of ACL reconstruction procedures.

Main Methods:

• Proposed a framework integrating an isometric reconstruction method using 3D geometric processing algorithms.
• Developed a novel U-shaped network (EP-UNet) for medical image segmentation, incorporating axial edge and coordinate features.
• Utilized feature lines and points from bone morphology for accurate alignment with diverse bone shapes.

Main Results:

• The EP-UNet demonstrated superior performance in ACL segmentation, achieving high accuracy and robustness.
• The proposed method improved the mean Intersection over Union (mIoU) by 8.16% compared to baseline networks.
• The framework enables efficient and automatic determination of the patient's insertion center position.

Conclusions:

• The novel framework provides a reliable solution for the clinical challenge of ACL insertion center localization.
• The integration of 3D geometric processing and advanced segmentation networks enhances surgical precision.
• This approach has the potential to significantly improve ACL reconstruction success rates and patient recovery.