Truncated singular value decomposition (TSVD) is an effective method for the deconvolution of dynamic contrast enhanced (DCE) MRI. Two robust methods for the selection of the truncation threshold on a pixel-by-pixel basis--generalized cross validation (GCV) and the L-curve criterion (LCC)--were optimized and compared to paradigms in the literature. GCV and LCC were found to perform optimally when applied with a smooth version of TSVD, known as standard form Tikhonov regularization (SFTR). The methods lead to improvements in the estimate of the residue function and of its maximum, and converge properly with SNR. The oscillations typically observed in the solution vanish entirely, and perfusion is more accurately estimated at small mean transit times. This results in improved image contrast and increased sensitivity to perfusion abnormalities, at the cost of 1-2 min in calculation time and hyperintense clusters in the image. Preliminary experience with clinical data suggests that the latter problem can be resolved using spatial continuity and/or hybrid thresholding methods. In the simulations GCV and LCC are equivalent in terms of performance, but GCV thresholding is faster.
Receptor tyrosine kinase signaling causes profound neo-angiogenesis in high-grade gliomas (HGG). The KIT, PDGFR-α, and VEGFR2 genes are frequently amplified and expressed in HGG and are molecular targets for therapeutic inhibition by the small-molecule kinase inhibitor sunitinib malate. Twenty-one patients with progressive HGG after prior radiotherapy and chemotherapy received a daily dose of 37.5 mg sunitinib until progression or unacceptable toxicity. Magnetic resonance imaging (MRI) and dynamic susceptibility contrast (DSC)-enhanced perfusion measurements were performed before and during therapy. Cerebral blood volume (CBV) and cerebral blood flow (CBF) lesion-to-normal-white matter ratios were measured to evaluate the antiangiogenic effects of sunitinib. The most frequent grade ≥3 adverse events were skin toxicity, neutropenia, thrombocytopenia, and lymphocytopenia. None of the patients achieved an objective response, whereas a decrease in CBV and CBF within the lesion compared with the normal brain was documented in four out of 14 (29%) patients evaluable for DSC-enhanced perfusion measurements. All patients experienced progression of their disease before or after eight weeks of therapy. Median time-to-progression and overall survival were 1.6 (95%CI 0.8-2.5) and 3.8 (95% CI 2.2-5.3) months, respectively. No correlation could be established between VEGFR2, PDGFR-α, and KIT gene copy numbers or protein expression and the effects of sunitinib. Single-agent sunitinib at 37.5 mg/day had insufficient activity to warrant further investigation of this monotherapy regimen in recurrent HGG.
The feasibility of a voxel-by-voxel deconvolution analysis of T 1 -weighted DCE data in the human kidney and its potential for obtaining quantification of perfusion and filtration was investigated. Measurements were performed on 14 normal humans and 1 transplant at 1.5 T using a Turboflash sequence. Signal time-courses were converted to tracer concentrations and deconvolved with an aorta AIF. Parametric maps of relative renal blood flow (rRBF), relative renal volume of distribution (rRVD), relative mean transit time (rMTT), and whole cortex extraction fraction (E) were obtained from the impulse response functions. For the normals average cortical rRBF, rRVD, rMTT, and E were 1.6 mL/min/mL (SD 0.8), 0.4 mL/mL (SD 0.1), 17s (SD 7), and 22.6% (SD 6.1), respectively. A gradual voxelwise rRBF increase is found from the center of two infarction zones toward the edges. Voxel IRFs showed more detail on the nefron substructure than ROI IRFs. In conclusion, quantitative voxelwise perfusion mapping based on deconvolved T 1 -DCE renal data is feasible, but absolute quantification requires inflow correction. rRBF maps and quantitative values are sufficiently sensitive to detect perfusion abnormality in pathologic areas, but further research is necessary to separate perfusion from extraction and to characterize the different compartments of the nephron on the ( Key words: renal; T-weighted; perfusion; deconvolution; quantificationDynamic contrast enhanced MRI is a promising noninvasive method for imaging renal perfusion and function (1-5), since it avoids the use of ionizing radiation and combines high temporal and spatial resolution. Moreover, Gd-DTPA is very well tolerated, not nephrotoxic, and has properties comparable to the radioisotopic 99mTc-DTPA (6,7). Tissue perfusion imaging with dynamic Gd-DTPA enhanced MRI therefore offers the potential for obtaining important information about organ viability, anatomy, and function in the normal as well as in the compromised kidney.Experiments with a rabbit model (1) have shown that Gd-DTPA enhanced bolus tracking and a deconvolution analysis of ROI signals permit quantitative measures of perfusion and filtration, independent of the effect of the shape and size of the bolus. On the other hand, results obtained with an intravascular contrast agent have shown that a pixel-by-pixel approach to perfusion quantification can be a useful tool for distinguishing normality from renal artery stenosis in dogs as well as humans (2,3).In this study we investigate the combination of both approaches: a pixel-by-pixel deconvolution analysis in the human kidney, using Gd-DTPA enhanced bolus tracking. We investigate the quality of the resulting parametric maps, evaluate to what extent they represent perfusion and extraction, and assess the potential of the technique for obtaining reliable quantification of renal blood flow and a measure of filtration. We also investigate the possibility of obtaining other parameters like the tracers' volume of distribution and mean transit time. Our first results...
Quantification of haemodynamic parameters with a deconvolution analysis of bolus-tracking data is an ill-posed problem which requires regularization. In a previous study, simulated data without structural errors were used to validate two methods for a pixel-by-pixel analysis: standard-form Tikhonov regularization with either the L-curve criterion (LCC) or generalized cross validation (GCV) for selecting the regularization parameter. However, problems of image artefacts were reported when the methods were applied to patient data. The aim of this study was to investigate the nature of these problems in more detail and evaluate strategies of optimization for routine application in the clinic. In addition we investigated to which extent the calculation time of the algorithm can be minimized. In order to ensure that the conclusions are relevant for a larger range of clinical applications, we relied on patient data for evaluation of the algorithms. Simulated data were used to validate the conclusions in a more quantitative manner. We conclude that the reported problems with image quality can be removed by appropriate optimization of either LCC or GCV. In all examples this could be achieved with LCC without significant perturbation of the values in pixels where the regularization parameter was originally selected accurately. GCV could not be optimized for the renal data, and in the CT data only at the cost of image resolution. Using the implementations given, calculation times were sufficiently short for routine application in the clinic.
Pixelwise deconvolution analysis of DCE MR data in patients with breast cancer can provide preoperative information regarding TBF. These results also support the hypothesis that there is increased TBF in HER2-positive tumors.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.