arXiv subjects

Joseph V. Hajnal

Publications and source records attributed to Joseph V. Hajnal.

At least 19 recordsLinked to original sources

arXiv2025

L-FUSION: Laplacian Fetal Ultrasound Segmentation & Uncertainty Estimation

Accurate analysis of prenatal ultrasound (US) is essential for early detection of developmental anomalies. However, operator dependency and technical limitations (e.g. intrinsic artefacts and effects, setting errors) can complicate image interpretation and the assessment of diagnostic uncertainty. We present L-FUSION (Laplacian Fetal US Segmentation with Integrated FoundatiON models), a framework that integrates uncertainty quantification through unsupervised, normative learning and large-scale foundation models for robust segmentation of fetal structures in normal and pathological scans. We propose to utilise the aleatoric logit distributions of Stochastic Segmentation Networks and Laplace approximations with fast Hessian estimations to estimate epistemic uncertainty only from the segmentation head. This enables us to achieve reliable abnormality quantification for instant diagnostic feedback. Combined with an integrated Dropout component, L-FUSION enables reliable differentiation of lesions from normal fetal anatomy with enhanced uncertainty maps and segmentation counterfactuals in US imaging. It improves epistemic and aleatoric uncertainty interpretation and removes the need for manual disease-labelling. Evaluations across multiple datasets show that L-FUSION achieves superior segmentation accuracy and consistent uncertainty quantification, supporting on-site decision-making and offering a scalable solution for advancing fetal ultrasound analysis in clinical settings.

eess.IV↗

arXiv2025

Meta-learning Slice-to-Volume Reconstruction in Fetal Brain MRI using Implicit Neural Representations

High-resolution slice-to-volume reconstruction (SVR) from multiple motion-corrupted low-resolution 2D slices constitutes a critical step in image-based diagnostics of moving subjects, such as fetal brain Magnetic Resonance Imaging (MRI). Existing solutions struggle with image artifacts and severe subject motion or require slice pre-alignment to achieve satisfying reconstruction performance. We propose a novel SVR method to enable fast and accurate MRI reconstruction even in cases of severe image and motion corruption. Our approach performs motion correction, outlier handling, and super-resolution reconstruction with all operations being entirely based on implicit neural representations. The model can be initialized with task-specific priors through fully self-supervised meta-learning on either simulated or real-world data. In extensive experiments including over 480 reconstructions of simulated and clinical MRI brain data from different centers, we prove the utility of our method in cases of severe subject motion and image artifacts. Our results demonstrate improvements in reconstruction quality, especially in the presence of severe motion, compared to state-of-the-art methods, and up to 50% reduction in reconstruction time.

eess.IV↗

arXiv2024

DCRA-Net: Attention-Enabled Reconstruction Model for Dynamic Fetal Cardiac MRI

Dynamic fetal heart magnetic resonance imaging (MRI) presents unique challenges due to the fast heart rate of the fetus compared to adult subjects and uncontrolled fetal motion. This requires high temporal and spatial resolutions over a large field of view, in order to encompass surrounding maternal anatomy. In this work, we introduce Dynamic Cardiac Reconstruction Attention Network (DCRA-Net) - a novel deep learning model that employs attention mechanisms in spatial and temporal domains and temporal frequency representation of data to reconstruct the dynamics of the fetal heart from highly accelerated free-running (non-gated) MRI acquisitions. DCRA-Net was trained on retrospectively undersampled complex-valued cardiac MRIs from 42 fetal subjects and separately from 153 adult subjects, and evaluated on data from 14 fetal and 39 adult subjects respectively. Its performance was compared to L+S and k-GIN methods in both fetal and adult cases for an undersampling factor of 8x. The proposed network performed better than the comparators for both fetal and adult data, for both regular lattice and centrally weighted random undersampling. Aliased signals due to the undersampling were comprehensively resolved, and both the spatial details of the heart and its temporal dynamics were recovered with high fidelity. The highest performance was achieved when using lattice undersampling, data consistency and temporal frequency representation, yielding PSNR of 38 for fetal and 35 for adult cases. Our method is publicly available at https://github.com/denproc/DCRA-Net.

eess.IV↗

arXiv2024

CINA: Conditional Implicit Neural Atlas for Spatio-Temporal Representation of Fetal Brains

We introduce a conditional implicit neural atlas (CINA) for spatio-temporal atlas generation from Magnetic Resonance Images (MRI) of the neurotypical and pathological fetal brain, that is fully independent of affine or non-rigid registration. During training, CINA learns a general representation of the fetal brain and encodes subject specific information into latent code. After training, CINA can construct a faithful atlas with tissue probability maps of the fetal brain for any gestational age (GA) and anatomical variation covered within the training domain. Thus, CINA is competent to represent both, neurotypical and pathological brains. Furthermore, a trained CINA model can be fit to brain MRI of unseen subjects via test-time optimization of the latent code. CINA can then produce probabilistic tissue maps tailored to a particular subject. We evaluate our method on a total of 198 T2 weighted MRI of normal and abnormal fetal brains from the dHCP and FeTA datasets. We demonstrate CINA's capability to represent a fetal brain atlas that can be flexibly conditioned on GA and on anatomical variations like ventricular volume or degree of cortical folding, making it a suitable tool for modeling both neurotypical and pathological brains. We quantify the fidelity of our atlas by means of tissue segmentation and age prediction and compare it to an established baseline. CINA demonstrates superior accuracy for neurotypical brains and pathological brains with ventriculomegaly. Moreover, CINA scores a mean absolute error of 0.23 weeks in fetal brain age prediction, further confirming an accurate representation of fetal brain development.

cs.LG↗

arXiv2023

An automated pipeline for quantitative T2* fetal body MRI and segmentation at low field

Fetal Magnetic Resonance Imaging at low field strengths is emerging as an exciting direction in perinatal health. Clinical low field (0.55T) scanners are beneficial for fetal imaging due to their reduced susceptibility-induced artefacts, increased T2* values, and wider bore (widening access for the increasingly obese pregnant population). However, the lack of standard automated image processing tools such as segmentation and reconstruction hampers wider clinical use. In this study, we introduce a semi-automatic pipeline using quantitative MRI for the fetal body at low field strength resulting in fast and detailed quantitative T2* relaxometry analysis of all major fetal body organs. Multi-echo dynamic sequences of the fetal body were acquired and reconstructed into a single high-resolution volume using deformable slice-to-volume reconstruction, generating both structural and quantitative T2* 3D volumes. A neural network trained using a semi-supervised approach was created to automatically segment these fetal body 3D volumes into ten different organs (resulting in dice values > 0.74 for 8 out of 10 organs). The T2* values revealed a strong relationship with GA in the lungs, liver, and kidney parenchyma (R^2>0.5). This pipeline was used successfully for a wide range of GAs (17-40 weeks), and is robust to motion artefacts. Low field fetal MRI can be used to perform advanced MRI analysis, and is a viable option for clinical scanning.

eess.IV↗

arXiv2023

Generalizing Deep Learning MRI Reconstruction across Different Domains

We look into the robustness of deep learning based MRI reconstruction when tested on unseen contrasts and organs. We then propose to generalize the network by training with large publicly-available natural image datasets with synthesized phase information to achieve high cross-domain reconstruction performance which is competitive with domain-specific training. To explain its generalization mechanism, we have also analyzed patch sets for different training datasets.

cs.CV↗

arXiv2022

Placenta Segmentation in Ultrasound Imaging: Addressing Sources of Uncertainty and Limited Field-of-View

Automatic segmentation of the placenta in fetal ultrasound (US) is challenging due to the (i) high diversity of placenta appearance, (ii) the restricted quality in US resulting in highly variable reference annotations, and (iii) the limited field-of-view of US prohibiting whole placenta assessment at late gestation. In this work, we address these three challenges with a multi-task learning approach that combines the classification of placental location (e.g., anterior, posterior) and semantic placenta segmentation in a single convolutional neural network. Through the classification task the model can learn from larger and more diverse datasets while improving the accuracy of the segmentation task in particular in limited training set conditions. With this approach we investigate the variability in annotations from multiple raters and show that our automatic segmentations (Dice of 0.86 for anterior and 0.83 for posterior placentas) achieve human-level performance as compared to intra- and inter-observer variability. Lastly, our approach can deliver whole placenta segmentation using a multi-view US acquisition pipeline consisting of three stages: multi-probe image acquisition, image fusion and image segmentation. This results in high quality segmentation of larger structures such as the placenta in US with reduced image artifacts which are beyond the field-of-view of single probes.

eess.IV↗

arXiv2022

Parallel transmit PUlse design for Saturation Homogeneity (PUSH) for Magnetization Transfer imaging at 7T

Purpose: This work proposes a novel RF pulse design for parallel transmit (pTx) systems to obtain uniform saturation of semisolid magnetization for Magnetization Transfer (MT) contrast in the presence of transmit field ($B_1^+$) inhomogeneities. The semisolid magnetization is usually modeled as being purely longitudinal, with the applied $B_1^+$ field saturating but not rotating its magnetization, thus standard pTx pulse design methods do not apply. Theory and Methods: Pulse design for Saturation Homogeneity (PUSH) optimizes pTx RF pulses by considering uniformity of root-mean squared $B_1^+$, $B_1^{rms}$, which relates to the rate of semisolid saturation. Here we considered designs consisting of a small number of spatially non-selective sub-pulses optimized over either a single 2D plane or 3D. Simulations and in vivo experiments on a 7T Terra system with an 8-TX Nova head coil in 5 subjects were carried out to study the homogenization of $B_1^{rms}$ and of the MT contrast by acquiring MT ratio maps. Results: Simulations and in vivo experiments showed up to 6 and 2 times more uniform $B_1^{rms}$ compared to circular polarized (CP) mode for 2D and 3D optimizations, respectively. This translated into 4 and 1.25 times more uniform MT contrast, consistently for all subjects, where 2 sub-pulses were enough for the implementation and coil used. Conclusion: The proposed PUSH method obtains more uniform and higher MT contrast than CP mode within the same SAR budget.

physics.med-ph↗

arXiv2021

Fetal MRI by robust deep generative prior reconstruction and diffeomorphic registration: application to gestational age prediction

Magnetic resonance imaging of whole fetal body and placenta is limited by different sources of motion affecting the womb. Usual scanning techniques employ single-shot multi-slice sequences where anatomical information in different slices may be subject to different deformations, contrast variations or artifacts. Volumetric reconstruction formulations have been proposed to correct for these factors, but they must accommodate a non-homogeneous and non-isotropic sampling, so regularization becomes necessary. Thus, in this paper we propose a deep generative prior for robust volumetric reconstructions integrated with a diffeomorphic volume to slice registration method. Experiments are performed to validate our contributions and compare with a state of the art method in a cohort of $72$ fetal datasets in the range of $20-36$ weeks gestational age. Results suggest improved image resolution and more accurate prediction of gestational age at scan when comparing to a state of the art reconstruction method. In addition, gestational age prediction results from our volumetric reconstructions compare favourably with existing brain-based approaches, with boosted accuracy when integrating information of organs other than the brain. Namely, a mean absolute error of $0.618$ weeks ($R^2=0.958$) is achieved when combining fetal brain and trunk information.

eess.IV↗

arXiv2021

Magnetization Transfer-Mediated MR Fingerprinting

Purpose: Magnetization transfer (MT) and inhomogeneous MT (ihMT) contrasts are used in MRI to provide information about macromolecular tissue content. In particular, MT is sensitive to macromolecules and ihMT appears to be specific to myelinated tissue. This study proposes a technique to characterize MT and ihMT properties from a single acquisition, producing both semiquantitative contrast ratios, and quantitative parameter maps. Theory and Methods: Building upon previous work that uses multiband radiofrequency (RF) pulses to efficiently generate ihMT contrast, we propose a cyclic-steady-state approach that cycles between multiband and single-band pulses to boost the achieved contrast. Resultant time-variable signals are reminiscent of a magnetic resonance fingerprinting (MRF) acquisition, except that the signal fluctuations are entirely mediated by magnetization transfer effects. A dictionary-based low-rank inversion method is used to reconstruct the resulting images and to produce both semiquantitative MT ratio (MTR) and ihMT ratio (ihMTR) maps, as well as quantitative parameter estimates corresponding to an ihMT tissue model. Results: Phantom and in vivo brain data acquired at 1.5T demonstrate the expected contrast trends, with ihMTR maps showing contrast more specific to white matter (WM), as has been reported by others. Quantitative estimation of semisolid fraction and dipolar T1 was also possible and yielded measurements consistent with literature values in the brain. Conclusions: By cycling between multiband and single-band pulses, an entirely magnetization transfer mediated 'fingerprinting' method was demonstrated. This proof-of-concept approach can be used to generate semiquantitative maps and quantitatively estimate some macromolecular specific tissue parameters.

physics.med-ph↗

arXiv2021

Complementary Time-Frequency Domain Networks for Dynamic Parallel MR Image Reconstruction

Purpose: To introduce a novel deep learning based approach for fast and high-quality dynamic multi-coil MR reconstruction by learning a complementary time-frequency domain network that exploits spatio-temporal correlations simultaneously from complementary domains. Theory and Methods: Dynamic parallel MR image reconstruction is formulated as a multi-variable minimisation problem, where the data is regularised in combined temporal Fourier and spatial (x-f) domain as well as in spatio-temporal image (x-t) domain. An iterative algorithm based on variable splitting technique is derived, which alternates among signal de-aliasing steps in x-f and x-t spaces, a closed-form point-wise data consistency step and a weighted coupling step. The iterative model is embedded into a deep recurrent neural network which learns to recover the image via exploiting spatio-temporal redundancies in complementary domains. Results: Experiments were performed on two datasets of highly undersampled multi-coil short-axis cardiac cine MRI scans. Results demonstrate that our proposed method outperforms the current state-of-the-art approaches both quantitatively and qualitatively. The proposed model can also generalise well to data acquired from a different scanner and data with pathologies that were not seen in the training set. Conclusion: The work shows the benefit of reconstructing dynamic parallel MRI in complementary time-frequency domains with deep neural networks. The method can effectively and robustly reconstruct high-quality images from highly undersampled dynamic multi-coil data ($16 \times$ and $24 \times$ yielding 15s and 10s scan times respectively) with fast reconstruction speed (2.8s). This could potentially facilitate achieving fast single-breath-hold clinical 2D cardiac cine imaging.

eess.IV↗

arXiv2021

Efficiency analysis for quantitative MRI of T1 and T2 relaxometry methods

This study presents a comparison of quantitative MRI methods based on an efficiency metric that quantifies their intrinsic ability to extract information about tissue parameters. Under a regime of unbiased parameter estimates, an intrinsic efficiency metric $η$ was derived for fully-sampled experiments which can be used to both optimize and compare sequences. Here we optimize and compare several steady-state and transient gradient-echo based qMRI methods, such as magnetic resonance fingerprinting (MRF), for joint T1 and T2 mapping. The impact of undersampling was also evaluated, assuming incoherent aliasing that is treated as noise by parameter estimation. In-vivo validation of the efficiency metric was also performed. Transient methods such as MRF can be up to 3.5 times more efficient than steady-state methods, when spatial undersampling is ignored. If incoherent aliasing is treated as noise during least-squares parameter estimation, the efficiency is reduced in proportion to the SNR of the data, with reduction factors of 5 often seen for practical SNR levels. In-vivo validation showed a very good agreement between the theoretical and experimentally predicted efficiency. This work presents and validates an efficiency metric to optimize and compare the performance of qMRI methods. Transient methods were found to be intrinsically more efficient than steady-state methods, however the effect of spatial undersampling can significantly erode this advantage.

physics.med-ph↗

arXiv2020

Impact of aperture, depth, and acoustic clutter on performance of Coherent Multi-Transducer Ultrasound imaging

Transducers with larger aperture size are desirable in ultrasound imaging to improve resolution and image quality. A coherent multi-transducer ultrasound imaging system (CoMTUS) enables an extended effective aperture through coherent combination of multiple transducers. In this study, the discontinuous extended aperture created by CoMTUS and its performance for deep imaging and through layered-media are investigated by both simulations and experiments. Typical image quality metrics - resolution, contrast and contrast-to-noise ratio - are evaluated and compared with a standard single probe imaging system. Results suggest that, the image performance of CoMTUS depends on the relative spatial location of the arrays. The resulting effective aperture significantly improves resolution, while the separation between the arrays may degrade contrast. For a limited gap in the effective aperture (less than few centimetres), CoMTUS provides benefits to image quality compared to standard single probe imaging system. Overall, CoMTUS shows higher sensitivity and reduced loss of resolution with imaging depth. In general, CoMTUS imaging performance was unaffected when imaging through a layered-medium with different speed of sound values and resolution improved up to 80% at large imaging depths.

physics.med-ph↗

arXiv2020

Diffusion tensor driven image registration: a deep learning approach

Tracking microsctructural changes in the developing brain relies on accurate inter-subject image registration. However, most methods rely on either structural or diffusion data to learn the spatial correspondences between two or more images, without taking into account the complementary information provided by using both. Here we propose a deep learning registration framework which combines the structural information provided by T2-weighted (T2w) images with the rich microstructural information offered by diffusion tensor imaging (DTI) scans. We perform a leave-one-out cross-validation study where we compare the performance of our multi-modality registration model with a baseline model trained on structural data only, in terms of Dice scores and differences in fractional anisotropy (FA) maps. Our results show that in terms of average Dice scores our model performs better in subcortical regions when compared to using structural data only. Moreover, average sum-of-squared differences between warped and fixed FA maps show that our proposed model performs better at aligning the diffusion data.

eess.IV↗

arXiv2020

Scattered slice SHARD reconstruction for motion correction in multi-shell diffusion MRI

Diffusion MRI offers a unique probe into neural microstructure and connectivity in the developing brain. However, analysis of neonatal brain imaging data is complicated by inevitable subject motion, leading to a series of scattered slices that need to be aligned within and across diffusion-weighted contrasts. Here, we develop a reconstruction method for scattered slice multi-shell high angular resolution diffusion imaging (HARDI) data, jointly estimating an uncorrupted data representation and motion parameters at the slice or multiband excitation level. The reconstruction relies on data-driven representation of multi-shell HARDI data using a bespoke spherical harmonics and radial decomposition (SHARD), which avoids imposing model assumptions, thus facilitating to compare various microstructure imaging methods in the reconstructed output. Furthermore, the proposed framework integrates slice-level outlier rejection, distortion correction, and slice profile correction. We evaluate the method in the neonatal cohort of the developing Human Connectome Project (650 scans). Validation experiments demonstrate accurate slice-level motion correction across the age range and across the range of motion in the population. Results in the neonatal data show successful reconstruction even in severely motion-corrupted subjects. In addition, we illustrate how local tissue modelling can extract advanced microstructure features such as orientation distribution functions from the motion-corrected reconstructions.

physics.med-ph↗

arXiv2020

Deformable Slice-to-Volume Registration for Motion Correction in Fetal Body MRI

In in-utero MRI, motion correction for fetal body and placenta poses a particular challenge due to the presence of local non-rigid transformations of organs caused by bending and stretching. The existing slice-to-volume registration (SVR) reconstruction methods are widely employed for motion correction of fetal brain that undergoes only rigid transformation. However, for reconstruction of fetal body and placenta, rigid registration cannot resolve the issue of misregistrations due to deformable motion, resulting in degradation of features in the reconstructed volume. We propose a Deformable SVR (DSVR), a novel approach for non-rigid motion correction of fetal MRI based on a hierarchical deformable SVR scheme to allow high resolution reconstruction of the fetal body and placenta. Additionally, a robust scheme for structure-based rejection of outliers minimises the impact of registration errors. The improved performance of DSVR in comparison to SVR and patch-to-volume registration (PVR) methods is quantitatively demonstrated in simulated experiments and 20 fetal MRI datasets from 28-31 weeks gestational age (GA) range with varying degree of motion corruption. In addition, we present qualitative evaluation of 100 fetal body cases from 20-34 weeks GA range.

physics.med-ph↗

arXiv2020

Motion corrected MRI with DISORDER: Distributed and Incoherent Sample Orders for Reconstruction Deblurring using Encoding Redundancy

Purpose: To enable rigid-body motion tolerant parallel volumetric magnetic resonance imaging by retrospective head motion correction on a variety of spatio-temporal scales and imaging sequences. Theory and methods: Tolerance against rigid-body motion is based on distributed and incoherent sampling orders for boosting a joint retrospective motion estimation and reconstruction framework. Motion resilience stems from the encoding redundancy in the data, as generally provided by the coil array. Hence, it does not require external sensors, navigators or training data, so the methodology is readily applicable to sequences using 3D encodings. Results: Simulations are performed showing full inter-shot corrections for usual levels of in-vivo motion, large number of shots, standard levels of noise and moderate acceleration factors. Feasibility of inter- and intra-shot corrections is shown under controlled motion in-vivo. Practical efficacy is illustrated by high quality results in most corrupted of 208 volumes from a series of 26 clinical pediatric examinations collected using standard protocols. Conclusion: The proposed framework addresses the rigid motion problem in volumetric anatomical brain scans with sufficient encoding redundancy which has enabled reliable pediatric examinations without sedation.

physics.med-ph↗

arXiv2019

dAUTOMAP: decomposing AUTOMAP to achieve scalability and enhance performance

AUTOMAP is a promising generalized reconstruction approach, however, it is not scalable and hence the practicality is limited. We present dAUTOMAP, a novel way for decomposing the domain transformation of AUTOMAP, making the model scale linearly. We show dAUTOMAP outperforms AUTOMAP with significantly fewer parameters.

cs.LG↗