Related Experiment Video
Updated: Jan 9, 2026
06:59
Improved Registration of 3D CT Angiography with X-ray Fluoroscopy for Image Fusion During Transcatheter Aortic Valve Implantation
Published on: June 3, 2018
11.0K
AI-Driven CT-MRI Image Fusion and Segmentation for Automatic Preoperative Planning of ACL Reconstruction: Development
Haomiao Yu1,2, Jixiang Dong1,3, Long Wang1
1Senior Department of Orthopaedics, Chinese PLA General Hospital, Beijing, People's Republic of China.
The Journal of Bone and Joint Surgery. American Volume
|December 9, 2025
Summary
An AI system accurately plans anterior cruciate ligament (ACL) reconstruction using CT-MRI fusion, optimizing tunnel positioning for better surgical outcomes. This technology promises improved function and fewer complications in ACL repair.
Area of Science:
- Orthopedic Surgery
- Medical Imaging
- Artificial Intelligence
Background:
- Developing an AI-driven system for automated preoperative planning in anterior cruciate ligament (ACL) reconstruction.
- Integrating deep learning with computed tomography (CT)-magnetic resonance imaging (MRI) image fusion and segmentation.
Purpose of the Study:
- To develop and evaluate the accuracy of an AI-driven automated preoperative planning system for ACL reconstruction.
- To optimize ACL tunnel positioning using deep learning on fused CT-MRI images.
Main Methods:
- Manual annotation of CT and MRI scans from 200 knee joints.
- Dual-UNet registration architecture for CT-MRI image fusion and 3D reconstruction.
- Deep learning framework for optimizing femoral and tibial tunnel positioning.
- Validation in bone models and clinical testing with 3D-printed guides in 36 ACL reconstructions.
Main Results:
- High segmentation accuracy (Dice coefficient = 0.864) achieved with CT-MRI fusion.
- AI planning time averaged 192 ± 90.2 seconds per case.
- Mean deviation <1 mm between planned and executed tunnel/graft lengths in bone models.
- AI-guided surgery showed significantly smaller deviations in tunnel positioning compared to conventional methods in clinical testing.
Conclusions:
- AI-driven segmentation and planning system accurately and reproducibly generates optimal ACL tunnel positions.
- The technology offers potential for reduced complications and improved postoperative function in ACL reconstruction.
- Individualized, anatomical, and isometric characteristics are achieved through automated preoperative planning.
