arXiv Science⌕ Search
Home/Archive/arXiv

arXiv · 2610.02132

Diffusion broadening of the point spread function in steady-state MRI

Bibek Dhakal·John C. Gore
View original source ↗Read source document ↗Export metadata ↓

Abstract

Purpose: To demonstrate how diffusion blurs the longitudinal magnetization in steady-state gradient-echo imaging, quantify resolution dependence on flip angle and repetition time, and the effects on relaxometry at microscopic resolution. Methods: We incorporated diffusion in the Bloch-Torrey equation in k space to determine how the point spread function (PSF) of T1-weighted spoiled gradient-echo images varies with flip angle. Theoretical predictions were validated numerically and extended to b-SSFP. Results: Diffusion acts as a low-pass filter whose width depends on the diffusion rate (D), repetition time (TR), T1 and flip angle ($α$). The blurring length spans from $\sqrt{D \cdot TR/2}$ at large $α$ to $\sqrt{DT1}$ as the $α\rightarrow 0$, equaling $\sqrt{DT1/2}$ at the Ernst angle. Consequently, multi-angle acquisitions have mismatched resolution (PSF widths of 38 $μ$m at $α= 10^\circ$ versus 11 $μ$m at for TR = 50 ms). For the flip-angle pair optimizing variable-flip-angle T1 mapping, the effective resolutions differ by a factor of 2.4. In simulated DESPOT1 maps the of a 30 $μ$m feature is reduced to be 1.80 s$^{-1}$ (true value 2.00 s$^{-1}$). In b-SSFP blurring depends on by T1, T2, and $α$ but is independent of TR, with minimum of $\sqrt{DT2}$. Conclusion: The PSF of a steady-state sequence depends on the flip angle and T1. Combining images acquired at different flip angles implicitly assumes identical spatial resolution, but that is not valid at microscopic scales, affecting the accuracy of quantitative relaxometry. The theory identifies a new limitation on the ultimate resolution of MRI set by diffusion.

Explore related subjects

physics.med-ph

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Bibek Dhakal, John C. Gore. 2026-10-01. Diffusion broadening of the point spread function in steady-state MRI. https://arxiv.org/abs/2610.02132

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Validation of a Computational Respiratory System Model for Mechanical Ventilation

Computational modeling and simulation have emerged as powerful tools for assessing medical device performance and safety, particularly in silico clinical trials (ISCTs) for automated medical systems. In ventilation, where gas exchange, respiratory mechanics, and patient-ventilator interaction must be managed under evolving pathophysiology, clinical translation of automated control strategies remains slow and resource-intensive. These challenges are particularly relevant for AI-based therapy-control systems, whose data-driven decision-making must be evaluated across heterogeneous and safety-critical patient states that may be sparsely represented in clinical datasets. Mechanistic, physiology-based models provide a complementary and interpretable environment for testing such scenarios. This paper applies a standards-aligned framework for credibility assessment of a computational respiratory model, demonstrated using an automated weaning case study. The framework operationalizes ASME V&V 40 and FDA principles within a structured validation workflow. The model integrates respiratory mechanics, gas exchange, respiratory control, and a ventilator representation, with validation under a defined context of use and explicit questions of interest. Model credibility is assessed through calibration, physiological plausibility, population-based evaluation, and reproduction of emergent behavior. All model requirements derived from the intended context of use are addressed, and gaps are transparently reported. The resulting credibility argument supports applicability of the model for medium-low-risk preclinical ISCTs of automated weaning strategies. Residual limitations relate to the extent of in vivo evidence, population representativeness, and external validation. The validation procedure provides a blueprint for validation of this and similar models in mechanical ventilation and related use cases.

physics.med-ph↗
arXiv2026

Quiet, rapid 3D multiparametric mapping using magnetization-prepared zero echo time MRI

Purpose: To introduce and evaluate MuPa-ZTE, a quiet, rapid 3D framework combining native and magnetization-prepared zero echo time (ZTE) acquisitions for multiparametric mapping. Methods: MuPa-ZTE combines steady-state native ZTE with transient-state magnetization-prepared ZTE. Two implementations were evaluated: T2T1-ZTE for apparent proton density, T1, and T2 mapping, and T1-ZTE for apparent proton density and T1 mapping. Both were assessed in an ISMRM/NIST system phantom and two healthy volunteers; T2T1-ZTE was also demonstrated in a patient with brain metastases. The 10-minute phantom and 4.5-minute in vivo acquisitions were retrospectively truncated to 3, 2, and 1 minute and reconstructed with and without deep learning-based denoising. Evaluations included phantom agreement, precision, short-term repeatability, apparent SNR, edge sharpness, and consistency with full-duration in vivo maps. Results: T1 estimates remained close to nominal phantom values across implementations, durations, and reconstructions. T2 accuracy was maintained down to 2 minutes over the brain-relevant range, with limited sensitivity to longer T2 values. Denoising generally reduced variability and improved short-term repeatability. In vivo, 1.1-mm isotropic whole-brain maps were obtained in 4.5 minutes; denoising increased apparent SNR while preserving edge sharpness and yielded promising image quality after retrospective truncation to 2 minutes. Conclusion: MuPa-ZTE enables quiet, isotropic 3D multiparametric mapping within 4.5 minutes, supporting robust T1 mapping and T2 mapping over a brain-relevant range, with promising acceleration toward 2 minutes using deep learning-based denoising.

physics.med-ph↗
arXiv2026

AnatomIQ: An Open-Source Toolkit for Automated Background Detection in Medical Imaging

Manual background selection for contrast-to-noise ratio (CNR) calculations in CT image quality assessment is time-consuming, operator-dependent, and compromises reproducibility. Advanced metrics such as Noise Power Spectrum (NPS) and Task Transfer Function (TTF) traditionally require dedicated phantom acquisitions, adding cost, radiation exposure, and workflow disruption. We developed AnatomIQ, an open-source Python toolkit that automates background detection and performs phantom-free NPS and TTF analysis directly on patient anatomy. Automated background detection uses tissue-specific Hounsfield unit thresholding and morphological filtering, paired with a web interface for click-to-select lesion identification from PACS screenshots. Reported metrics include CNR, signal-to-noise ratio, detectability indices, and dose-normalized figures of merit with automated optimization recommendations. NPS quantifies noise texture from automatically detected uniform tissue regions, while TTF leverages natural circular anatomical structures (e.g., trachea, vessels) as edge phantoms, together characterizing the resolution-noise trade-off without dedicated phantom scans. Analysis takes 2-3 minutes per case, with CNR computed in under 10 seconds and NPS/TTF available on demand. Automated detection reliably identified valid regions across anatomical sites, including head-and-neck scans, a near-worst-case test of uniform-region availability. Both NPS and TTF pipelines were validated against phantom-derived references, confirming that in vivo measurements reflect genuine anatomical texture and resolution rather than pipeline artifacts. By eliminating operator variability and phantom requirements, AnatomIQ enables practical continuous quality monitoring and evidence-based reconstruction optimization. The toolkit is open-source, with optional commercial licensing for academic and clinical use.

physics.med-ph↗