Related Experiment Video
Updated: Mar 6, 2026
Author Spotlight: Investigating Early Events and Long-Term Effects of ACL Injuries for Osteoarthritis Progression
Published on: September 29, 2023
Visual assessment of AI-reconstructed knee MRI: A pilot study.
S S Ghotra1, L Buttex2, L Gallus3
1School of Health Sciences (HESAV), University of Applied Sciences and Arts Western Switzerland (HES-SO), Lausanne 1011, Switzerland; Department of Diagnostic & Interventional Radiology, Lausanne University Hospital (CHUV) and University of Lausanne (UNIL), Lausanne 1011, Switzerland; Radiography and Diagnostic Imaging, School of Medicine, University College Dublin, Ireland.
Artificial intelligence (AI) significantly improves knee MRI scans, enhancing image quality and reducing scan time by 36.9%. This AI reconstruction shows clinical acceptability and superior performance in evaluations.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Magnetic Resonance Imaging
Background:
- Artificial intelligence (AI) is transforming medical imaging by improving image quality and reducing scan times.
- AI reconstruction methods are being explored for knee MRI investigations.
Purpose of the Study:
- To evaluate the efficacy of AI reconstruction methods in knee MRI.
- To compare AI-enhanced protocols with standard 3T knee MRI protocols.
Main Methods:
- An exploratory comparison of AI-enhanced and standard 3T knee MRI protocols.
- Sequences included sagittal T1w FSE, PDw FSE FS, T2w FSE, and additional PDw FSE FS views.
- Image quality (IQ) and scan duration were optimized; AI protocol tested on 10 volunteers.
- Three MRI experts assessed images using ViewDEX, VGA, Kappa, and VGC for evaluation.
Main Results:
- AI reconstruction yielded equal or superior scores for anatomical and IQ criteria.
- AI reduced scan time by 36.9% (8:22 min vs 13:15 min).
- AI-enhanced sequences were clinically acceptable with moderate-to-good inter-observer agreement (Kappa).
- Visual grading characteristics (VGC) confirmed statistically higher performance for AI images (AUC 0.76-0.81, p ≤ 0.05).
Conclusions:
- AI integration into knee MRI protocols enhances image quality and reduces acquisition time by 36.9%.
- AI improves workflow and patient throughput by reducing scan time without compromising image quality.
- Further clinical validation is necessary to reinforce these findings.
