Comparison of model‐free Lorentzian and spinlock model‐based fittings in quantitative CEST imaging of acute stroke

PurposeIn this work, the use of joint Total Generalized Variation (TGV) regularization to improve Multipool‐Lorentzian fitting of chemical exchange saturation transfer (CEST) Spectra in terms of stability and parameter signal‐to‐noise ratio (SNR) was investigated.Theory and MethodsThe joint TGV term was integrated into the nonlinear parameter fitting problem. To increase convergence and weight the gradients, preconditioning using a voxel‐wise singular value decomposition was applied to the problem, which was then solved using the iteratively regularized Gauss–Newton method combined with a Primal‐Dual splitting algorithm. The TGV method was evaluated on simulated numerical phantoms, 3T phantom data and 7T in vivo data with respect to systematic errors and robustness. Three reference methods were also implemented: The standard nonlinear fitting, a method using a nonlocal‐means filter for denoising and the pyramid scheme, which uses downsampled images to acquire accurate start values.ResultsThe proposed regularized fitting method showed significantly improved robustness (compared to the reference methods). In testing, over a range of SNR values the TGV fit outperformed the other methods and showed accurate results even for large amounts of added noise. Parameter values found were closer or comparable to the ground truth. For in vivo datasets, the added regularization increased the parameter map SNR and prevented instabilities.ConclusionThe proposed fitting method using TGV regularization leads to improved results over a range of different data‐sets and noise levels. Furthermore, it can be applied to all Z‐spectrum data, with different amounts of pools, where the improved SNR and stability can increase diagnostic confidence.

Section: Applications Of Tgv To Other Cest Modelsmentioning

confidence: 97%

Improved quantification in CEST‐MRI by joint spatial total generalized variation

Huemer,

Stilianu,

Maier

et al. 2024

“…Third, the nonlinear increase in CEST contrast observed between 5.26 × 10 3 and 5.26 × 10 8 vg/μL suggests that labeling efficiency will not correlate linearly with underlying capsid content. Although quantitative approaches toward CEST imaging have been more recently explored, 70,71 CEST contrast correlations with capsid density are semi-quantitative at best. Thus, a direct correlation of CEST contrast to encapsulated viral particle density in future studies may require additional verification to create a calibration curve.…”

Section: F I G U R Ementioning

confidence: 99%

Imaging of adeno‐associated viral capsids for purposes of gene editing using CEST NMR/MRI

Lam,

Velasquez,

Ogiyama

et al. 2024

PurposeGene therapy using adeno‐associated virus (AAV) vector‐mediated gene delivery has undergone substantial growth in recent years with promising results in both preclinical and clinical studies, as well as emerging regulatory approval. However, the inability to quantify the efficacy of gene therapy from cellular delivery of gene‐editing technology to specific functional outcomes is an obstacle for efficient development of gene therapy treatments. Building on prior works that used the CEST reporter gene lysine rich protein, we hypothesized that AAV viral capsids may generate endogenous CEST contrast from an abundance of surface lysine residues.MethodsNMR experiments were performed on isolated solutions of AAV serotypes 1–9 on a Bruker 800‐MHz vertical scanner. In vitro experiments were performed for testing of CEST‐NMR contrast of AAV2 capsids under varying pH, density, biological transduction stage, and across multiple serotypes and mixed biological media. Reverse transcriptase–polymerase chain reaction was used to quantify virus concentration. Subsequent experiments at 7 T optimized CEST saturation schemes for AAV contrast detection and detected AAV2 particles encapsulated in a biocompatible hydrogel administered in the hind limb of mice.ResultsCEST‐NMR experiments revealed CEST contrast up to 52% for AAV2 viral capsids between 0.6 and 0.8 ppm. CEST contrast generated by AAV2 demonstrated high levels of CEST contrast across a variety of chemical environments, concentrations, and saturation schemes. AAV2 CEST contrast displayed significant positive correlations with capsid density (R2 > 0.99, p < 0.001), pH (R2 = 0.97, p = 0.01), and viral titer per cell count (R2 = 0.92, p < 0.001). Transition to a preclinical field strength yielded up to 11.8% CEST contrast following optimization of saturation parameters. In vivo detection revealed statistically significant molecular contrast between viral and empty hydrogels using both mean values (4.67 ± 0.75% AAV2 vs. 3.47 ± 0.87% empty hydrogel, p = 0.02) and quantile analysis.ConclusionAAV2 viral capsids exhibit strong capacity as an endogenous CEST contrast agent and can potentially be used for monitoring and evaluation of AAV vector‐mediated gene therapy protocols.

Quasi‐steady‐state (QUASS) reconstruction enhances T₁ normalization in apparent exchange‐dependent relaxation (AREX) analysis: A reevaluation of T₁ correction in quantitative CEST MRI of rodent brain tumor models

Sun

2024

Self Cite

PurposeThe apparent exchange‐dependent relaxation (AREX) analysis has been proposed as an effective means to correct T1 contribution in CEST quantification. However, it has been recognized that AREX T1 correction is not straightforward if CEST scans are not performed under the equilibrium condition. Our study aimed to test if quasi‐steady‐state (QUASS) reconstruction could boost the accuracy of the AREX metric under common non‐equilibrium scan conditions.Theory and MethodsNumerical simulation and in vivo scans were performed to assess the AREX metric accuracy. The CEST signal was simulated under different relaxation delays, RF saturation amplitudes, and durations. The AREX was evaluated as a function of the bulk water T1 and labile proton concentration using the multiple linear regression model. AREX MRI was also assessed in brain tumor rodent models, with both apparent CEST scans and QUASS reconstruction.ResultsSimulation showed that the AREX calculation from apparent CEST scans, under non‐equilibrium conditions, had significant dependence on labile proton fraction ratio, RF saturation time, and T1. In comparison, QUASS‐boosted AREX depended on the labile proton fraction ratio without significant dependence on T1 and RF saturation time. Whereas the apparent (2.7 ± 0.8%) and QUASS MTR asymmetry (2.8 ± 0.8%) contrast between normal and tumor regions of interest (ROIs) were significant, the difference was small. In comparison, AREX contrast between normal and tumor ROIs calculated from the apparent CEST scan and QUASS reconstruction was 3.8 ± 1.1%/s and 4.4 ± 1.2%/s, respectively, statistically different from each other.ConclusionsAREX analysis benefits from the QUASS‐reconstructed equilibrium CEST effect for improved T1 correction and quantitative CEST analysis.