Training-Free Motion-Corrected 3D Brain MRI Reconstruction via Event-Guided References

Wei Jiang; Hongli Chen; Tianyi Ding; Feng Liu; Nan Ye; Hongfu Sun

cs.CVNov 1, 2026

Training-Free Motion-Corrected 3D Brain MRI Reconstruction via Event-Guided References

Authors: Wei Jiang, Hongli Chen, Tianyi Ding, Feng Liu, Nan Ye, Hongfu Sun

Organizations: The University of Queensland, Brisbane, Australia · RMIT University, Melbourne, Australia · The University of Newcastle, Newcastle, Australia

Abstract

Head motion causes artefacts in 3D brain MRI and can bias structural measurements. Model-based correction alternates between image and motion estimation, but artefacts in the reconstructed reference can reinforce motion errors. Learned priors can reduce these artefacts but require pretraining and may add test-time computation. We propose EventMC (event-guided motion correction), a training-free method for interleaved Cartesian acquisitions. It iteratively detects motion events and estimates motion using reference images reconstructed from stable intervals. It then refines the image and motion from low to high spatial frequencies. In a fair comparison in simulation (same reconstruction), EventMC's motion estimates improve mean PSNR by up to 4.0 dB over those of the learning-based MotionTTT. A single run of EventMC is up to 7.5 times faster than MotionTTT with less than half the GPU memory. On in vivo motion-corrupted scans, EventMC achieves similar mean image quality to MotionTTT without any training. Code is available at https://github.com/bayesbreeze/EventMC.

Figures & tables

Figure 1: EventMC overview (schematic). (a) Interleaved acquisition: consecutive shots (bars, one

k_{y}

each) form segments (boxes, one pose

\boldsymbol{\theta}_{s}

each) spread over

k_{y}

. (b) Stage A, fitting segment 7 (orange,

\mathcal{P}_{\{7\}}

) to a TV reference

\mathbf{x}_{7}

from the stable interval

\mathcal{A}

around the anchor

\alpha

(blue,

\mathcal{P}_{\mathcal{S}_{7}}

; hatched: excluded event-adjacent segments); the poses (orange line) jump at events

\mathcal{J}

. (c) Stage B, coarse-to-fine refinement: all segments, with the central band

\mathcal{P}_{f}

widening with

f

Figure 2: Example reconstructions under (a) simulated and (b) real motion; baselines in (b) are the released reconstructions. Top: central sagittal slices with volumetric PSNR (dB); bottom: absolute errors relative to the reference display range, within the head mask and valid FOV (a) or brain mask (b).

Panel of Figure 3 — Algorithm 1 EventMC-S: single-anchor estimation

Figure 3: True and estimated motion for one CC359 subject (moderate, R1; translations

t_{x},t_{y}

and rotations

r_{x},r_{y}

). Each method estimates one pose per segment (piecewise constant).

Appendix figures & tables22 assets

Supplementary material from the paper’s appendix.

Appendix

Candidate	Anchor	Stage A grid
(i) EventMC-S	$\alpha$	full (1 mm on CC359)
(ii)	$\alpha$	half (2 mm on CC359)
(iii), if $\alpha_{2}$ exists	$\alpha_{2}$	half (2 mm on CC359)

Appendix

Table A.1: Multi-anchor candidates. All candidates run Stage B and the final reconstruction at full resolution.

Parameter	Value
Segments $N_{s}$	13
Reference groups	$s\bmod 4$
Stage A rounds	5, after a one-round pre-run with temporary anchor $\lfloor N_{s}/2\rfloor=6$
Stage B bands $f$	0.25, 0.5, 1, 1
$\operatorname{ImgEst}$ iterations	15 in Stages A and B; 30 in the final reconstruction
$\operatorname{MotEst}$ iterations	30 per segment and start; 3 starts in Stage A, 1 in Stage B

Appendix

Table A.2: Parameters of EventMC, shared by all experiments unless stated.

$\lambda$	$10^{-4}$	$3{\times}10^{-4}$	$10^{-3}$	$3{\times}10^{-3}$	$10^{-2}$	$3{\times}10^{-2}$
Error	0.1140	0.1009	0.0869	0.0807	0.0816	0.0941

Appendix

Table B.1: k-space cross-validation of the TV weight on the PMoC3D motion-free scans: median held-out error over eight subjects. Lower is better; bold marks the minimum.

	R1		R2		R4
Method	PSNR $\uparrow$	SSIM $\uparrow$	PSNR $\uparrow$	SSIM $\uparrow$	PSNR $\uparrow$	SSIM $\uparrow$
Uncorrected	26.4	0.731	26.0	0.708	25.8	0.691
Stacked U-Net	26.5	0.771	26.3	0.753	26.3	0.750
Joint-AltOpt	31.6	0.893	30.7	0.870	30.3	0.858
MotionTTT	27.4	0.751	27.2	0.705	31.6	0.897
MotionTTT+TV	30.8	0.840	29.9	0.835	32.6	0.913

Appendix

Table C.1: Simulated moderate motion (mean over nine subjects; PSNR in dB). Rows other than Stacked U-Net and True segment poses repeat Table 1 (a). Bold: best method values, including ties at the displayed precision; the true-pose reference is excluded.

	R1		R2		R4
Method	PSNR $\uparrow$	SSIM $\uparrow$	PSNR $\uparrow$	SSIM $\uparrow$	PSNR $\uparrow$	SSIM $\uparrow$
Uncorrected	23.9	0.608	23.7	0.589	23.6	0.577
Stacked U-Net	23.7	0.645	23.9	0.641	24.1	0.653
Joint-AltOpt	29.9	0.865	29.4	0.845	29.2	0.838
MotionTTT	31.0	0.846	31.6	0.878	31.9	0.911
MotionTTT+TV	32.3	0.858	32.7	0.891	33.3	0.934

Appendix

Table C.2: Simulated sustained mid-scan motion, M (mean over nine subjects; PSNR in dB). Bold: best method values, including ties at the displayed precision; the true-pose reference is excluded.

EventMC versus	Metric	Subj.	$p$	Cases	Subj.	$p$	Cases	Subj.	$p$	Cases
		R1			R2			R4
MotionTTT	PSNR	9	0.0039	18	9	0.0039	17	8	0.020	16
	SSIM	9	0.0039	17	8	0.0078	17	8	0.13	16
MotionTTT+TV	PSNR	8	0.0078	17	8	0.012	15	4	0.57	10
	SSIM	8	0.0078	17	8	0.012	15	4	0.43	9
Joint-AltOpt	PSNR	9	0.0039	18	9	0.0039	18	9	0.0039	17

Appendix

Table C.3: Subject-level comparisons in simulation, moderate and M together. Subj.: subjects in which EventMC is better (of 9);

p

: two-sided Wilcoxon; Cases: cases in which EventMC is better (of 18).

Figure C.1: Per-case PSNR in simulation, EventMC against MotionTTT+TV: (a) moderate and (b) M motion, 27 cases each (nine subjects at R1, R2 and R4). Points above the diagonal: EventMC is better.

Sampling	Error $\downarrow$	Joint-AltOpt	MotionTTT	EventMC-S	EventMC	$p$
R1	Translation (mm)	0.57	0.52	0.30	0.29	0.0078
	Rotation (deg)	0.54	0.50	0.31	0.26	0.0078
R2	Translation (mm)	0.57	0.47	0.27	0.25	0.0039
	Rotation (deg)	0.55	0.56	0.26	0.26	0.0078
R4	Translation (mm)	0.56	0.30	0.31	0.19	0.43
	Rotation (deg)	0.51	0.30	0.33	0.24	0.50

Appendix

Table C.4: Pose error: median per-case RMSE over 18 cases per sampling level (moderate and M). Lower is better; bold marks the minimum at the displayed precision. The last column gives exploratory, unadjusted subject-level Wilcoxon p-values for EventMC versus MotionTTT (

n=9

Figure C.2: All six pose components for the case of Fig. 3 (CC359, moderate motion, R1). The true motion is shown per shot; each method estimates one pose per segment. Estimates are shown after removing the global rigid offset (Appendix B.4 ).

	Median	Smallest	Cases with	Cases with
Sampling	$\Delta$ PSNR	$\Delta$ PSNR	lower PSNR	lower SSIM
R1	$+0.44$	$+0.22$	0	0
R2	$+0.28$	$+0.15$	0	0
R4	$+0.17$	$+0.12$	0	0

Appendix

Table C.5: Drift-only control: EventMC-S minus the uncorrected reconstruction on the 27 subthreshold cases (

\Delta

PSNR in dB; counts out of nine cases).

Variant	What changes
$-$ Event refs	All five Stage A rounds use the Round 1 references (all segments except the target’s group). Events are still detected and still split the smoothing.
$-$ Disconnection	Smoothing also spans events (all second differences are kept). Events still select the references.
$-$ Stage A	Stage B starts from $\Theta=\mathbf{0}$ .
$-$ Coarse-to-fine	Stage B is replaced by four rounds at $f=1$ , starting from the Stage A poses.
$+$ Multi-anchor	EventMC (Appendix A.5 ).

Appendix

Table C.6: Ablation variants of EventMC-S.

Variant	$\Delta$ PSNR	moderate / M	below $-1$ dB	R1	R2
	Mean	Median $\Delta$ PSNR	Cases	Subjects better / worse ( $p$ )
$-$ Event refs	$-2.1$	$-0.04$ / $-4.11$	16	2 / 7 (0.039)	0 / 9 (0.0039)
$-$ Disconnection	$-0.7$	$-1.07$ / $-0.48$	13	3 / 6 (0.43)	0 / 9 (0.0039)
$-$ Stage A	$-6.6$	$-4.88$ / $-8.56$	35	0 / 9 (0.0039)	0 / 9 (0.0039)
$-$ Coarse-to-fine	$-0.3$	$0.00$ / $0.00$	4	3 / 6 (0.30)	6 / 3 (0.91)
$+$ Multi-anchor	$+0.5$	$0.00$ / $0.00$	0	4 / 1 (0.125)	2 / 1 (1.0)

Appendix

Table C.7: Ablation on the 36 R1 and R2 cases, relative to EventMC-S on the same platform (

\Delta

PSNR in dB). Subject-level tests are exploratory and unadjusted; ties are omitted from better/worse counts.

Figure C.3: Multi-anchor selection on the 54 moderate and M cases: PSNR of the selected candidate (EventMC) against candidate (i) (EventMC-S). Points on the diagonal: the rule kept candidate (i) or changed little. Hollow markers: the three subjects inspected when the rule was designed. Here PSNR is computed within the intersection of the candidates’ valid FOVs, so values can differ slightly from Table C.1 .

	R1		R2
Cohort	Mean PSNR	Subjects better ( $p$ , Holm)	Mean PSNR	Subjects better ( $p$ , Holm)
All nine subjects	35.4 vs 31.6	8/9 (0.0078, 0.016)	34.2 vs 31.3	8/9 (0.012, 0.016)
Without the three design subjects	36.2 vs 31.9	6/6 (0.031, 0.063)	34.9 vs 31.7	6/6 (0.031, 0.063)

Appendix

Table C.8: Sensitivity to the three design subjects, moderate and M together: mean PSNR (dB) of EventMC versus MotionTTT+TV, and subjects in which EventMC is better. Bold marks the higher mean within each pair;

p

is raw, followed by its Holm-adjusted value.

Method	R1	R2	R4	Peak GPU memory (GiB)
MotionTTT	52.3 (47.8–54.5)	32.0 (29.3–34.2)	18.3 (18.0–18.9)	27.9–29.1
Joint-AltOpt	2.5 (2.4–2.8)	2.5 (2.3–2.7)	2.5 (2.3–2.7)	8.3–12.3
EventMC-S	6.9 (6.5–7.9)	6.8 (6.3–7.8)	6.6 (6.0–7.7)	8.8–12.5
EventMC (2 cases per level)	11.6 (11.4–11.8)	11.2 (11.1–11.3)	10.9 (10.9–11.0)	8.7–9.3

Appendix

Table C.9: Run time per case in minutes (median, with range) and peak GPU memory range on CC359, one H100 PCIe. Bold marks the lowest median run time. Memory ranges are not ranked across the different drivers.

Figure C.4: (a) Per-scan PSNR on PMoC3D, EventMC against MotionTTT; the band marks

\pm 1

dB, and the orange points are the two scans with the largest losses (Appendix C.6 ). (b, c) Run time and peak GPU memory per case on CC359 (one H100 PCIe). Bars: medians; dots: cases; labels: ratio of the median run times of MotionTTT and EventMC-S.

EventMC versus	Subjects better ( $p$ )	Scans better	$\Delta$
MotionTTT	5/8 (0.64)	17/26	$+0.31$
AltOpt	5/8 (0.46)	17/26	$+0.41$
Stacked U-Net	8/8 (0.0078)	26/26	$+1.54$
Uncorrected (L1)	8/8 (0.0078)	26/26	$+1.92$

Appendix

Table C.10: Real motion: EventMC versus each method, PSNR, 26 scans of eight subjects.

\Delta

: median per-scan PSNR difference (dB).

sub-01	sub-02	sub-03	sub-04	sub-05	sub-06	sub-07	sub-08
$+0.68$	$-0.79$	$+0.19$	$+0.42$	$+0.52$	$+0.89$	$-0.25$	$-0.61$

Appendix

Table C.11: Real motion: per-subject mean PSNR difference, EventMC minus MotionTTT (dB).

Figure C.5: Central sagittal slices of the case of Fig. 2 a (CC359, moderate motion) at (a) R2 and (b) R4, with volumetric PSNR. Bottom rows: absolute error relative to the reference display range, within the head mask and each method’s valid FOV.

Figure C.6: Sustained mid-scan motion (M): central sagittal slices of the case with the median PSNR gain of EventMC over MotionTTT+TV among the nine subjects, (a) at R1 and (b) at R2. Display as in Fig. C.5 .

Figure C.7: Real motion (PMoC3D), central sagittal slices of the two median scans by per-scan PSNR difference between EventMC and MotionTTT. PSNR as in Table 1 (b). Bottom rows: absolute error within the brain mask, relative to the normalised intensity range (Appendix B.4 ).

Figure C.8: Real motion (PMoC3D), central sagittal slices: (a) the scan with the largest gain over MotionTTT, also shown in Fig. 2 b, and (b) the scan with the largest loss, with multiple detected events. PSNR as in Table 1 (b). Display as in Fig. C.7 .

Magnetic Resonance Imaging

Magnetic resonance imaging uses magnetic fields and radio waves to create images of anatomy and tissue function for research and clinical care.

See all 258 papers

MotionDPS: Motion-Compensated 3D Brain MRI Reconstruction
May 21, 2026Antonio Ortiz-Gonzalez, Erich Kobler, Lukas Schletter +1Accelerated MRI Reconstruction Magnetic Resonance Imaging
Life and Medical Sciences Institute, University of Bonn, Bonn, Germany · Institute for Machine Learning, LIT AI Lab, Department of Virtual Morphology, Clinical Research Institute Medical AI, Johannes Kepler University Linz, Linz, Austria · German Center for Neurodegenerative Diseases (DZNE), Bonn, Germany +1
Motion Artifact-Aware Self-Supervised Representation Learning for 3D Brain MRI Motion Artifact Reduction
Aug 10, 2026Mojtaba Safari, Shansong Wang, Zach Eidex +4Magnetic Resonance Imaging Medical Image Representation Learning
Department of Radiation and Cellular Oncology, The University of Chicago, Chicago, IL, USA · Radiation Oncology and Winship Cancer Institute, Emory University, Atlanta, GA, USA · Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, NY, USA +1
MoSAIK: Motion Search with Artificial Intelligence in K-space
Oct 19, 2026Oscar Dabrowski, Abijith Jagannath, Frédéric Grouiller +1Magnetic Resonance Imaging Brain MRI
CIBM Center for Biomedical Imaging, Geneva, Switzerland · Department of Radiology and Medical Informatics, Faculty of Medicine, University of Geneva, Geneva, Switzerland · Biomedical Imaging Group, École polytechnique fédérale de Lausanne (EPFL), Lausanne, Switzerland +1

May 21, 2026J/K move · Enter open · S save

MotionDPS: Motion-Compensated 3D Brain MRI Reconstruction

Antonio Ortiz-Gonzalez, Erich Kobler, Lukas Schletter, Alexander Effland

Life and Medical Sciences Institute, University of Bonn, Bonn, Germany · Institute for Machine Learning, LIT AI Lab, Department of Virtual Morphology, Clinical Research Institute Medical AI, Johannes Kepler University Linz, Linz, Austria · German Center for Neurodegenerative Diseases (DZNE), Bonn, Germany+1

Accelerated MRI Reconstruction Magnetic Resonance Imaging Brain MRI

Open paper AskPDF arXiv

Magnetic resonance imaging (MRI) is highly susceptible to patient motion due to its relatively long acquisition times and the fact that data are acquired sequentially in k-space. Even small patient movements introduce phase inconsistencies across measurements, leading to severe artifacts such as blurring, ghosting, and geometric distortions that can compromise diagnostic quality. Retrospective motion compensation remains challenging, particularly in accelerated acquisitions, due to the ill-posed nature of the joint reconstruction and motion estimation problem. In this work, we propose a unified Bayesian framework for motion-compensated 3D MRI that jointly estimates the anatomical image, rigid-body motion parameters, and coil sensitivity maps directly from motion-corrupted k-space data. Our approach integrates pretrained 3D complex-valued score-based diffusion models as expressive anatomical image priors within a physics-based forward model. Inference is performed by alternating diffusion posterior image updates with efficient proximal optimization steps for motion and coil sensitivity estimation, enabling fully unsupervised reconstruction without the need for paired motion-free training data. Experiments on simulated and real-motion brain MRI datasets demonstrate that the proposed method achieves improved image quality and motion robustness compared to state-of-the-art classical and learning-based motion correction techniques, particularly in the presence of severe motion and high acceleration.

Abstract

Magnetic Resonance Imaging

Explore similar work

MotionDPS: Motion-Compensated 3D Brain MRI Reconstruction

Motion Artifact-Aware Self-Supervised Representation Learning for 3D Brain MRI Motion Artifact Reduction

MoSAIK: Motion Search with Artificial Intelligence in K-space