Ziad Al-Haj Hemidi

Mr. Ziad Al-Haj Hemidi

Medical Computer Science, University of Lübeck, Lübeck, Schleswig-Holstein, Germany
orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications


5 October 2024
Verified
Book Chapter
Closed access
PDF file
CGChristoph Großbröhmer
Christoph Großbröhmer
ZHZiad Al-Haj Hemidi
Ziad Al-Haj Hemidi
FFFenja Falta
Fenja Falta
MHMattias P. Heinrich
Mattias P. Heinrich

4 April 2024 in Sensors
Verified
Journal Article
Open access
PDF file
CWChristian Weihsbach
Christian Weihsbach
NVNora Vogt
Nora Vogt
ZHZiad Al-Haj Hemidi
Ziad Al-Haj Hemidi
ABAlexander Bigalke
Alexander Bigalke
LHLasse Hansen
Lasse Hansen
MHMattias P. Heinrich
Mattias P. Heinrich
In cardiac cine imaging, acquiring high-quality data is challenging and time-consuming due to the artifacts generated by the heart’s continuous movement. Volumetric, fully isotropic data acquisition with high temporal resolution is, to date, intractable due to MR physics constraints. To assess whole-heart movement under minimal acquisition time, we propose a deep learning model that reconstructs the volumetric shape of multiple cardiac chambers from a limited number of input slices while simultaneously optimizing the slice acquisition orientation for this task. We mimic the current clinical protocols for cardiac imaging and compare the shape reconstruction quality of standard clinical views and optimized views. In our experiments, we show that the jointly trained model achieves accurate high-resolution multi-chamber shape reconstruction with errors of 80%, indicating its effectiveness in both simulated cardiac cine MRI and clinical cardiac MRI with a wide range of pathological shape variations.

12 March 2026
Verified
Book Chapter
Open access
PDF file
ZHZiad Al-Haj Hemidi
Ziad Al-Haj Hemidi
EKEytan Kats
Eytan Kats
MHMattias P. Heinrich
Mattias P. Heinrich
Magnetic resonance imaging (MRI) provides excellent diagnostic detail but suffers from long scan times. Accelerated imaging through undersampling reduces acquisition but introduces aliasing, while parallel imaging and compressed sensing degrade quality at higher acceleration. Deep learning methods trained on large datasets suppress artifacts effectively, yet often over-smooth or hallucinate features. Implicit neural representations (INRs), which optimize continuous image functions per-instance, preserve fine details but remain under-constrained, leaving residual artifacts. We propose PrIINeR (prior-informed implicit neural representation) [1], a framework that unites population-trained priors with INR optimization. PrIINeR employs a hash-grid encoded implicit network jointly optimized with coil sensitivity maps and guided by a dual data consistency objective, enforcing fidelity to both undersampled k-space and prior-informed reconstructions. Total variation regularization further suppresses aliasing while maintaining sharp edges. On the NYU fastMRI [2] knee dataset (4–10× undersampling), PrIINeR consistently improves reconstruction quality over both INR-only methods and population-trained priors of varying complexity. Gains in SSIM and PSNR are statistically significant (p < 0.05) across nearly all settings, while qualitative results confirm better artifact suppression and preservation of anatomical detail. By combining global priors with instance-specific fidelity, PrIINeR offers a robust, flexible approach to accelerated MRI. Code is publicly available at: https://github.com/multimodallearning/PrIINeR.

11 August 2025 in arXiv
Verified
Preprint
Open access
PDF file
ZHZiad Al-Haj Hemidi
Ziad Al-Haj Hemidi
EKEytan Kats
Eytan Kats
MHMattias P. Heinrich
Mattias P. Heinrich
Accelerating Magnetic Resonance Imaging (MRI) reduces scan time but often degrades image quality. While Implicit Neural Representations (INRs) show promise for MRI reconstruction, they struggle at high acceleration factors due to weak prior constraints, leading to structural loss and aliasing artefacts. To address this, we propose PrIINeR, an INR-based MRI reconstruction method that integrates prior knowledge from pre-trained deep learning models into the INR framework. By combining population-level knowledge with instance-based optimization and enforcing dual data consistency, PrIINeR aligns both with the acquired k-space data and the prior-informed reconstruction. Evaluated on the NYU fastMRI dataset, our method not only outperforms state-of-the-art INR-based approaches but also improves upon several learning-based state-of-the-art methods, significantly improving structural preservation and fidelity while effectively removing aliasing artefacts.PrIINeR bridges deep learning and INR-based techniques, offering a more reliable solution for high-quality, accelerated MRI reconstruction. The code is publicly available on https://github.com/multimodallearning/PrIINeR.