You might also read
Articles linked to this work by shared authors, journal, and citation graph.
Updated: Jun 27, 2026
Magnetic Resonance Imaging Quantification of Pulmonary Perfusion using Calibrated Arterial Spin Labeling
Published on: May 31, 2011
1Deptartment of Biomedicine, University of Bergen, Bergen, Norway. renate@fmri.no
This article introduces a new computational method to improve the accuracy of brain blood flow measurements. By using a mathematical technique called iterative blind deconvolution, the researchers can better account for how contrast agents travel through the brain, leading to more precise hemodynamic maps.
Area of Science:
Background:
Standard perfusion imaging often struggles with the complex movement of contrast agents through the brain. Researchers frequently rely on global arterial input functions that may not accurately reflect local blood delivery. This limitation creates significant errors when calculating hemodynamic parameters across different brain regions. No prior work had fully resolved the challenges posed by contrast delay and dispersion in individual voxels. That uncertainty drove the development of more sophisticated mathematical models for signal processing. Prior research has shown that ignoring these local variations leads to biased flow estimates. This gap motivated the exploration of advanced algorithms to separate tissue signals from input functions. The current study addresses these persistent technical hurdles in magnetic resonance imaging.
Purpose Of The Study:
The aim of this study is to develop a method for simultaneous estimation of voxel-specific arterial input functions and tissue residue functions. Researchers seek to overcome the limitations of current perfusion imaging techniques. The primary problem involves the delay and dispersion of contrast agents during the first pass. This gap motivated the creation of a more precise mathematical framework for hemodynamic analysis. The authors intend to eliminate signal errors caused by global input function assumptions. They propose using an iterative blind deconvolution approach to enhance data quality. This work addresses the need for accurate blood flow quantification in individual brain voxels. The study evaluates the performance of an extended contrast concentration model through simulations and patient data.
Main Methods:
The review approach utilizes a computational framework based on the Richardson-Lucy algorithm for signal estimation. Investigators designed an extended concentration model to isolate first-pass bolus signals from noise. This strategy incorporates both simulated datasets and clinical in vivo measurements for validation. The team applied the algorithm to separate voxel-specific arterial input functions from tissue residue functions. Researchers performed computer simulations to assess the mathematical feasibility of the proposed technique. They subsequently evaluated the model using patient data from a case of fibromuscular dysplasia. The analysis focused on comparing these results against standard conventional processing methods. This comprehensive approach ensures the robustness of the hemodynamic parameter calculations.
Main Results:
Key findings from the literature indicate that the proposed method yields higher flow values and shorter mean transit times than conventional techniques. These improvements suggest that the negative impacts of contrast dispersion are effectively reduced. Preliminary in vivo data from a patient with fibromuscular dysplasia revealed specific territories with delayed or dispersed input functions. These identified areas coincided precisely with regions supplied by collateral circulation as confirmed by complete radiologic examinations. The estimated arterial input functions successfully visualized blood supply patterns as a function of time. Simulations supported the feasibility of the blind deconvolution approach for perfusion imaging applications. The extended model successfully separated the first-pass bolus from recirculation and leakage signals. This evidence demonstrates the capability of the algorithm to refine hemodynamic mapping in clinical settings.
Conclusions:
The researchers propose that their mathematical framework effectively minimizes errors caused by contrast agent dispersion. This approach allows for the simultaneous estimation of voxel-specific input and tissue residue functions. Synthesis and implications suggest that the method provides a more accurate representation of blood supply patterns over time. The findings indicate that higher flow values and shorter mean transit times are achievable compared to standard techniques. This suggests that the model successfully accounts for local hemodynamic variations in patients. The authors claim that the visualized blood supply patterns align with clinical observations of collateral circulation. These results demonstrate the potential for improved diagnostic precision in perfusion imaging. The study confirms that the extended concentration model helps separate bolus signals from recirculation effects.
The researchers propose an iterative blind deconvolution approach using the Richardson-Lucy algorithm. This method simultaneously estimates voxel-specific arterial input functions and tissue residue functions, effectively separating the first-pass bolus from recirculation and leakage signals to improve hemodynamic parameter accuracy.
The authors utilize an extended contrast concentration model. This framework is necessary to isolate the initial bolus signal from secondary influences like contrast recirculation and leakage, which otherwise confound the calculation of quantitative blood flow metrics.
Voxel-specific measurements are necessary because they eliminate errors caused by the delay and dispersion of contrast agents as they travel from the injection site to different brain regions, which global functions fail to capture.
The researchers employ both computer simulations and in vivo patient data. Simulations test the feasibility of the algorithm, while the patient data validates the model's ability to map blood supply patterns in clinical scenarios involving collateral circulation.
The study measures hemodynamic parameters, specifically flow values and mean transit times. These metrics are compared against conventional methods to demonstrate that the new approach minimizes the negative effects of contrast dispersion.
The authors propose that this method provides a more accurate visualization of blood supply patterns. They claim that this approach is particularly useful for identifying territories supplied by collateral circulation in patients with vascular conditions like fibromuscular dysplasia.
