In this paper, we propose MotionTTT a deep learning-based test-time-training (TTT) method for accurate motion estimation.

Magnetic Resonance Imaging (MRI) is a powerful medical imaging modality that offers diagnostic information without harmful ionizing radiation.

Quantitative imaging methods, such as magnetic resonance fingerprinting (MRF), aim to extract interpretable pathology biomarkers by estimating biophysical tissue parameters from signal evolutions. However, the pattern-matching algorithms or neural networks used in such inverse problems often lack principled uncertainty quantification, which limits the trustworthiness and transparency, required for clinical acceptance. Here, we describe a physics-structured variational autoencoder (PS-VAE) designed for rapid extraction of voxelwise multi-parameter posterior distributions. Our approach integrates a differentiable spin physics simulator with self-supervised learning, and provides a full covariance that captures the inter-parameter correlations of the latent biophysical space. The method was validated in a multi-proton pool chemical exchange saturation transfer (CEST) and semisolid magnetization transfer (MT) molecular MRF study, across in-vitro phantoms, tumor-bearing mice, healthy human volunteers, and a subject with glioblastoma. The resulting multi-parametric posteriors are in good agreement with those calculated using a brute-force Bayesian analysis, while providing an orders-of-magnitude acceleration in whole brain quantification. In addition, we demonstrate how monitoring the multi-parameter posterior dynamics across progressively acquired signals provides practical insights for protocol optimization and may facilitate real-time adaptive acquisition.

Purpose: The Unadjusted Langevin Algorithm (ULA) in combination with diffusion models can generate high quality MRI reconstructions with uncertainty estimation from highly undersampled k-space data. However, sampling methods such as diffusion posterior sampling or likelihood annealing suffer from long reconstruction times and the need for parameter tuning. The purpose of this work is to develop a robust sampling algorithm with fast convergence. Theory and Methods: In the reverse diffusion process used for sampling the posterior, the exact likelihood is multiplied with the diffused prior at all noise scales. To overcome the issue of slow convergence, preconditioning is used. The method is trained on fastMRI data and tested on retrospectively undersampled brain data of a healthy volunteer. Results: For posterior sampling in Cartesian and non-Cartesian accelerated MRI the new approach outperforms annealed sampling in terms of reconstruction speed and sample quality. Conclusion: The proposed exact likelihood with preconditioning enables rapid and reliable posterior sampling across various MRI reconstruction tasks without the need for parameter tuning.

To investigate the feasibility of zero-shot self-supervised learning reconstruction for reducing breath-hold times in magnetic resonance cholangiopancreatography (MRCP). Breath-hold MRCP was acquired from 11 healthy volunteers on 3T scanners using an incoherent k-space sampling pattern, leading to 14-second acquisition time and an acceleration factor of R=25. Zero-shot reconstruction was compared with parallel imaging of respiratory-triggered MRCP (338s, R=3) and compressed sensing reconstruction. For two volunteers, breath-hold scans (40s, R=6) were additionally acquired and retrospectively undersampled to R=25 to compute peak signal-to-noise ratio (PSNR). To address long zero-shot training time, the n+m full stages of the zero-shot learning were divided into two parts to reduce backpropagation depth during training: 1) n frozen stages initialized with n-stage pretrained network and 2) m trainable stages initialized either randomly or m-stage pretrained network. Efficiency of our approach was assessed by varying initialization strategies and the number of trainable stages using the retrospectively undersampled data. Zero-shot reconstruction significantly improved visual image quality over compressed sensing, particularly in SNR and ductal delineation, and achieved image quality comparable to that of successful respiratory-triggered acquisitions with regular breathing patterns. Improved initializations enhanced PSNR and reduced reconstruction time. Adjusting frozen/trainable configurations demonstrated that PSNR decreased only slightly from 38.25 dB (0/13) to 37.67 dB (12/1), while training time decreased up to 6.7-fold. Zero-shot learning delivers high-fidelity MRCP reconstructions with reduced breath-hold times, and the proposed partially trainable approach offers a practical solution for translation into time-constrained clinical workflows.

To extract biologically interpretable information, a common approach is to fit a microstructural tissue model to a set of signals acquired with different dMRI acquisition settings (Alexander et al., 2019; Lampinen et al., 2023; Jelescu et al., 2020). In the absence of diffusion time dependence, these typically include different combinations of gradient strengths (commonly quantified by the b-value), directions (b-vector), and B-tensor shape (Westin et al., 2014). Microstructural parameters estimated by these models - including compartmental signal fractions and diffusivities - have shown to be sensitive to changes in brain structure due to diseases like multiple sclerosis (Alotaibi et al., 2021), Alzheimer's disease (Parker et al., 2018) and Parkinson's disease (Kim et al., 2016), and can provide a more fundamental understanding of tissue microstructure in both healthy and pathological tissues (Zhang et al., 2012). The Standard Model of white matter (SM) (Novikov et al., 2019) describes the signal arising from white matter by a kernel consisting of three compartments (intra-axonal, extra-axonal, and free water (occasionally omitted)) convolved with a fiber orientation distribution (FOD) (Tournier et al., 2007b). Compartmental signal fractions and diffusivities can be estimated, alongside the parameters that describe the FOD (usually in the form of a spherical harmonics (SH) series). Nevertheless, the high-dimensional parameter space of the SM complicates the estimation of its parameters, potentially leading to low accuracy, precision, and degeneracy of estimates (Jelescu et al., 2016).

A b stract: Clinical adoption of multi - shot diffusion - weighted magnetic resonance imaging ( multi - shot DWI) for body - wide tumor diagnostics is limited by severe motion - induced phase artifacts from respiration, peristalsis, and so on, compounded by multi - organ, multi - sl ice, multi - direction and multi - b - value complexities. Here, we introduce a reconstruction framework, LoSP - Prompt, that overcomes these challenges through physics - informed modeling and synthetic - data - driven prompt learning. We model int er - shot phase variations as a high - order Locally Smooth Phase (LoSP), integrated into a low - rank Hankel matrix reconstruction. Crucially, the algorithm's rank parameter is automat ically set via prompt learning trained exclusively on synthetic abdominal DWI data emulatin g physiological motion. The approach eliminates navigator signals and real istic data supervision, providing an interpretable, robust solution for high - resolution multi - organ multi - shot DWI. Ho wever, the ms - iEPI DWI is very sensitive to the inter - shot motion during the data acquisition of each shot ( Figure 1(a - d)) . E ven s light movement on the millimeter scale will cause the significant extra inter - shot phase (motion - induced phase in Fig . All t hese methods can successfully remove image artifacts in brain imaging ( F ig. 1 ( h)), g reatly promot ing applications of multi - shot high - resolution DWI . For the a bdominal tumor diagnosis, such as liver and kidney, ms - iEPI DWI has not been applied well ( F ig. 1 ( l)) . The se movement s bring organ - specific and high - order motion - induced phase s ( Figure 1 ( i, j)), which do not conform to the smooth phase prior assumption made in multi - shot DWI brain imaging .

DSC-MRI perfusion is a medical imaging technique for diagnosing and prognosing brain tumors and strokes. Its analysis relies on mathematical deconvolution, but noise or motion artifacts in a clinical environment can disrupt this process, leading to incorrect estimate of perfusion parameters. Although deep learning approaches have shown promising results, their calibration typically rely on third-party deconvolution algorithms to generate reference outputs and are bound to reproduce their limitations. To adress this problem, we propose a physics-informed autoencoder that leverages an analytical model to decode the perfusion parameters and guide the learning of the encoding network. This autoencoder is trained in a self-supervised fashion without any third-party software and its performance is evaluated on a database with glioma patients. Our method shows reliable results for glioma grading in accordance with other well-known deconvolution algorithms despite a lower computation time. It also achieved competitive performance even in the presence of high noise which is critical in a medical environment.

Magnetic Resonance Imaging (MRI) is a powerful medical imaging modality that offers diagnostic information without harmful ionizing radiation.