AI-assisted six-dimensional CT imaging for joint instability assessment

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Biomedical Engineering

Background:

• Assessing joint instability, a precursor to osteoarthritis, is challenging with current static imaging.
• Arthritis leads to significant functional disability and morbidity.

Purpose of the Study:

• To develop and validate an AI-assisted six-dimensional CT (6DCT) imaging technique for improved joint instability assessment.
• To integrate dynamic data including spatial, temporal, spectral information, and joint kinematics.

Main Methods:

• Developed a physics-informed motion correction network (ATOM) for dynamic CT images.
• Developed a physics-informed prior-assisted Bayesian network (PMBD) for multi-material decomposition and soft tissue differentiation.
• Derived kinematic metrics from motion-corrected images for joint motion pattern characterization.

Main Results:

• ATOM significantly reduced motion artifacts in dynamic wrist CT scans (p<0.05).
• PMBD improved material quantification accuracy and soft tissue differentiation, highlighting pathological features.
• Validated the 6DCT technique using patient, cadaver, and phantom scans.

Conclusions:

• AI-assisted 6DCT offers accurate visualization and quantitative assessment of dynamic joint pathology.
• This technology has the potential to revolutionize the early detection and management of joint instability and osteoarthritis.