Related Experiment Video
Updated: Apr 11, 2026
07:13
Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities
Published on: October 27, 2023
1.8K
Accelerating PREFUL MRI: Comparison of Registration Methods and Impact of Time Series Reduction
A Voskrebenzev1,2,3, G H Pöhler4, F Klimeš1,2,3
1Institute of Diagnostic and Interventional Radiology, Hannover Medical School, Hannover, Germany.
NMR in Biomedicine
|April 10, 2026
Summary
Accelerated phase-resolved functional lung (PREFUL) MRI using Forsberg registration and 30-second acquisition times maintains robustness for chronic obstructive pulmonary disease (COPD) patients and healthy volunteers. This enhances clinical feasibility and speeds up results without compromising key imaging parameters.
Area of Science:
- Medical Imaging
- Pulmonary Medicine
- Radiology
Background:
- Phase-resolved functional lung (PREFUL) MRI is crucial for assessing lung function in conditions like COPD.
- Current PREFUL protocols can be time-consuming, limiting clinical integration.
- Optimizing registration and acquisition times is essential for improving PREFUL MRI's efficiency.
Purpose of the Study:
- To evaluate if accelerated registration and reduced acquisition times maintain the robustness of PREFUL MRI.
- To determine if these optimizations improve clinical feasibility and reduce result turnaround times.
- To compare the efficacy of Advanced Normalization Tools (ANTs) and Forsberg registration methods.
Main Methods:
- Retrospective analysis of PREFUL MRI data from 28 COPD patients and 56 healthy volunteers at 1.5T.
- Image registration using ANTs and Forsberg, with time series (TS) truncated to 45, 30, and 15 seconds.
- Comparison of PREFUL parameters (regional ventilation, quantified perfusion (QQ), flow volume loop correlation metric (FVL-CM), defect percentages) across methods and TS lengths.
- Statistical analysis included Wilcoxon tests, Pearson correlations, intraclass correlation coefficients (ICC), and regional overlap analysis.
Main Results:
- Forsberg registration reduced processing time ~20-fold compared to ANTs, maintaining trends and repeatability.
- Correlations between registration methods exceeded 90% with regional overlap ≥97% (healthy) and ≥80% (COPD).
- Reducing TS to 30 seconds preserved overlap (≥79%) and strong correlations (≥77%) compared to 55 seconds, except for QQ.
- Repeatability (ICC ≥68%) was maintained down to 30 seconds for most parameters, with FVL-CM and defect percentages being most stable.
Conclusions:
- Forsberg registration offers comparable results to ANTs at significantly reduced processing time, enabling practical PREFUL analysis.
- Acquisition time per slice can be reduced to 30 seconds without compromising PREFUL MRI robustness.
- Optimized PREFUL protocols enhance clinical feasibility, though caution is advised for QQ analysis with reduced acquisition times.
