Publications by authors named "R Neji"
Purpose: To demonstrate the feasibility of hepatic 3D MR elastography (MRE) at 0.55 T in healthy volunteers using Hadamard encoding and to study the effects of concomitant fields in the domain of MRE in general.
Methods: Concomitant field effects in MRE are assessed using a Taylor series expansion and an encoding scheme is proposed to study the corresponding effects on 3D MRE at 0.
Purpose: To develop a 3D distortion-free reduced-FOV diffusion-prepared gradient-echo sequence and demonstrate its application in vivo for diffusion imaging of the spinal cord in healthy volunteers.
Methods: A 3D multi-shot reduced-FOV diffusion-prepared gradient-echo acquisition is achieved using a slice-selective tip-down pulse in the phase-encoding direction in the diffusion preparation, combined with magnitude stabilizers, centric k-space encoding, and 2D phase navigators to correct for intershot phase errors. The accuracy of the ADC values obtained using the proposed approach was evaluated in a diffusion phantom and compared to the tabulated reference ADC values and to the ADC values obtained using a standard spin echo diffusion-weighted single-shot EPI sequence (DW-SS-EPI).
Article Synopsis
- The study aimed to create and assess a new method for conducting free-breathing 3D whole-heart Cardiac Magnetic Resonance Angiography (CMRA) without contrast agents at a lower magnetic field strength of 0.55T.
- To achieve this, researchers optimized pulse sequences and imaging techniques, incorporating advanced methods like low-rank denoising and respiratory motion correction, and tested their approach on 11 healthy volunteers.
- Results showed that the new method produced high-quality images with minimal artifacts in just 6 minutes, matching the performance of higher field strength systems, paving the way for future testing in patients with heart conditions.
Deep Convolutional Backbone Comparison for Automated PET Image Quality Assessment.
IEEE Trans Radiat Plasma Med Sci
November 2024
Article Synopsis
- Pretraining deep convolutional networks using natural images can enhance medical imaging analysis, particularly given the scarcity of annotated medical data.
- The study compared 18 different pretrained backbone networks for evaluating the quality of PET brain scans, using data from multiple patients and testing against clinical quality metrics.
- Results indicated that using residual units in backbone architectures improved performance, with many networks achieving a mean-absolute-error below 0.5, highlighting the potential of over-parameterization for automated image quality assessments.
Background: Coronary computed tomography angiography (CCTA) is recommended as the first-line diagnostic imaging modality in low-to-intermediate-risk individuals suspected of stable coronary artery disease (CAD). However, CCTA exposes patients to ionizing radiation and potentially nephrotoxic contrast agents. Invasive coronary angiography is the gold-standard investigation to guide coronary revascularisation strategy; however, invasive procedures incur an inherent risk to the patient.
View Article and Find Full Text PDF