Magnetic resonance imaging of articular cartilage in osteoarthritis: studies on robustness and acceleration of quantitative imaging

ABSTRACT

Osteoarthritis (OA) is a prevalent chronic condition that affects all the components of the knee joint, resulting in joint pain and functional impairment. It is characterized by degeneration of articular cartilage, which in turn is associated with the deformation of subchondral bone. Non-invasive methods that could detect early changes and eventually contribute towards preventing or diagnosing OA are lacking, obscuring the relative contributions of these processes. Methods that detect and assess compositional changes due to OA are needed.

MR imaging techniques enable non-invasive evaluation of changes in articular cartilage. Quantitative Magnetic Resonance Imaging (qMRI) parameters accurately depict the characteristics of articular cartilage and are more robust and objective than qualitative anatomical imaging alone. T1ρ relaxation is a qMRI parameter, which describes relaxation along a spin-locking field in the rotating frame of reference. Continuous wave (CW) T1ρ and its dispersion, i.e., dependence on spin-lock amplitude is highly sensitive to the collagen fiber network, proteoglycan content, and cartilage deterioration in general. Transverse relaxation (T2) is a parameter that is highly sensitive to the orientation and integrity of the collagen fiber network.

The native T1 relaxation is regarded as a biomarker for macromolecular content and interstitial water content. Despite its promise, qMRI necessitates the utilization of multiple signal weightings to facilitate the quantification of the respective parameters, resulting in prolonged acquisition times.

Furthermore, T1ρ methods are susceptible to field distortions. Therefore, it is essential to develop methods that are both insensitive to field imperfections and have as short scan times as possible. Developing, validating, and comparing methods for more robust T1ρ measurements and accelerating the acquisition speed of quantitative measurements to clinically acceptable timeframe (no more than 5–10 minutes per measurement) was one of the goals of the thesis. The thesis also examines the feasibility of using accelerated CWT1ρ, and T1 relaxation time mapping methods, for detecting subtle degeneration of articular cartilage in spontaneous OA of human articular cartilage and in posttraumatic OA (PTOA) of the equine model, respectively. All the experiments were conducted on a 9.4 T preclinical MRI scanner, utilizing both human and animal osteochondral samples, as well as phantom samples. The assessment of the cartilage was conducted by evaluating the T1ρ dispersion and T1 parameters in conjunction with faster acquisition techniques. These parameters were correlated with the reference parameters, including the biomechanical properties of the samples, digital densitometry of Safranin-O-stained histological sections for proteoglycan assessment, and OARSI grading.

The findings of this dissertation demonstrate that the artifacts resulting from the field imperfections in T1ρ-weighted imaging can be effectively suppressed by various refocused spin-locking pulse schemes. Double-refocused adiabatic BIR-4 preparation was found to be the most robust, but due to the excessive RF energy deposition using powerful adiabatic pulses, the use of the adiabatic method is likely restricted to the pre-clinical setting. The earlier reported single-refocus hard-pulse T1ρ method, and the triple-refocused hard pulse method proposed in the thesis, were the most robust of the least RF-intensive methods. Furthermore, the results indicate that T1ρ dispersion combined with the compressed sensing technique remains sensitive to degenerative changes in the articular cartilage despite significant undersampling. T1 as measured by variable flip angle multiband SWIFT (VFA-MB-SWIFT) was sensitive to modest tissue changes caused by PTOA but could not distinguish the mildest forms of damage in the regions adjacent to damaged tissue, which were also undetectable via biomechanical testing.