Jennifer Anne Steeden

The value of exercise cardiovascular magnetic resonance (CMR) has been shown in many clinical scenarios. We have developed a novel MR‐compatible exercise device and aim to validate it against the reference standard MR‐conventional ergometer. The novel device consisted of two half‐pipes fixed to a wooden base, with participants wearing knee‐length socks with a 0.5 kg weight in each sock. Increased workload was achieved by increasing the rate of alternating leg flexion and extension in time with a bleep sound of increasing frequency. Twenty subjects (10 healthy volunteers, 10 patients with pulmonary arterial hypertension) performed two CMR‐augmented cardiopulmonary exercise tests (CMR/CPET) using the novel exercise device and a conventional ergometer in a randomised order. Comparing peak metrics elicited on both exercise devices, there was a moderate correlation in peak oxygen consumption (VO ₂ , r = 0.86, p < 0.001), cardiac output (CO, r = 0.73, p < 0.001), stroke volume (SV, r = 0.90, p < 0.001), peak heart rate (HR, r = 0.65, p = 0.002) and peak arteriovenous oxygen content gradient (ΔavO ₂ , r = 0.71, p < 0.001). However, all metrics were significantly lower with the novel device. Both devices were able to elicit statistically significant differences in VO ₂ , HR and right ventricular ejection fraction (RVEF) between patients and healthy subjects ( p ≤ 0.04). We have created a simple, easy to use and affordable exercise device for the CMR environment. This may encourage greater dissemination of exercise CMR in clinical and research practice.

Purpose: To investigate whether synthetically generated fractal data can be used to train deep learning (DL) models for dynamic MRI reconstruction, thereby avoiding the privacy, licensing, and availability limitations associated with cardiac MR training datasets. Methods: A training dataset was generated using quaternion Julia fractals to produce 2D + time images. Multi‐coil MRI acquisition was simulated to generate paired fully sampled and radially undersampled k‐space data. A 3D UNet deep artifact suppression model was trained using these fractal data (F‐DL) and compared with identical models trained on natural videos (NV‐DL) and cardiac MRI data (CMR‐DL). All models were evaluated on prospectively acquired radial real‐time cardiac MRI from 10 patients. Reconstructions were compared against compressed sensing (CS) and low‐rank deep image prior (LR‐DIP). All reconstructions were ranked for image quality, while ventricular volumes and ejection fraction were compared with reference breath‐hold cine MRI. Results: There was no significant difference in qualitative ranking between F‐DL, NV‐DL, and CMR‐DL ( p > 0.75), while both outperformed CS and LR‐DIP ( p < 0.05). Ventricular volumes and function derived from F‐DL were similar to CMR‐DL, showing no significant bias and acceptable limits of agreement compared to reference cine imaging. Conclusion: A DL model trained using synthetic fractal data reconstructed real‐time cardiac MRI with image quality and clinical measurements comparable to a model trained on true cardiac MRI data. Fractal training data provide an open, scalable alternative to clinical datasets and may enable development of more generalisable DL reconstruction models for dynamic MRI.

Purpose: The clinical feasibility and translation of many advanced quantitative MRI (qMRI) techniques are inhibited by their restriction to ‘research mode’, due to resource‐intensive, offline parameter estimation. This work aimed to achieve ‘clinical mode’ qMRI, by real‐time, inline parameter estimation with a trained neural network (NN) fully integrated into a vendor's image reconstruction environment, therefore facilitating and encouraging clinical adoption of advanced qMRI techniques. Methods: The Siemens Image Calculation Environment (ICE) pipeline was customized to deploy trained NNs for advanced diffusion MRI parameter estimation with Open Neural Network Exchange (ONNX) Runtime. Two fully‐connected NNs were trained offline with data synthesized with the neurite orientation dispersion and density imaging (NODDI) model, using either conventionally estimated (NN _MLE ) or ground truth (NN _GT ) parameters as training labels. The strategy was demonstrated online in two healthy volunteers (one rescanned) and evaluated offline with synthetic data, testing two diffusion protocols. Results: NNs were successfully integrated and deployed natively in ICE, performing inline, whole‐brain, in vivo NODDI parameter estimation in < 10 s. The proposed workflow was reproducible across protocols, volunteers and rescans. DICOM parametric maps were exported from the scanner for further analyses. Comparisons between NN _MLE and NN _GT suggested NN _MLE parameter estimates to be more consistent with conventional fitting, a finding supported by offline evaluations. Conclusion: Real‐time, inline parameter estimation with the proposed generalizable framework resolves a key practical barrier to the potential clinical uptake of advanced qMRI methods, enabling their efficient integration into clinical workflows. Next steps include incorporation of pre‐processing methods and evaluation in pathology.

Optical ultrasound (OpUS) imaging is an ultrasound modality that utilizes fiber-optic ultrasound sources and detectors to perform pulse-echo ultrasound imaging. These probes can be constructed entirely from glass optical fibers and plastic components, and as such, these devices have been predicted to be compatible with computed tomography (CT) and magnetic resonance imaging (MRI), modalities that use intense electromagnetic fields for imaging. However, to date, this compatibility has not been demonstrated. In this work, a free-hand OpUS imaging system was developed specifically to investigate the compatibility of OpUS systems with CT and MRI imaging systems. The OpUS imaging platform discussed in this work was used to perform real-time OpUS imaging under (separately) concurrent CT and MRI. CT and MRI imaging of the OpUS probe was used to determine if the probe itself would induce artifacts in the CT and MRI imaging, and ultrasound resolution targets and background measurements were used to assess any impact of CT and MRI on the OpUS signal fidelity. These measurements demonstrate that there was negligible interaction between the OpUS system and both the CT and MRI systems, and to further demonstrate this capability, concurrent OpUS-CT and OpUS-MRI imaging was conducted of a tissue-mimicking phantom and a dynamic motion phantom. This work presents a comprehensive demonstration of an OpUS imaging system operating alongside CT and MRI, which opens up new applications of ultrasound imaging in electromagnetically challenging settings.

Computational fluid dynamics (CFD) can be used for non-invasive evaluation of hemodynamics. However, its routine use is limited by labor-intensive manual segmentation, CFD mesh creation, and time-consuming simulation. This study aims to train a deep learning model to both generate patient-specific volume-meshes of the pulmonary artery from 3D cardiac MRI data and directly estimate CFD flow fields. This proof-of-concept study used 135 3D cardiac MRIs from both a public and private dataset. The pulmonary arteries in the MRIs were manually segmented and converted into volume-meshes. CFD simulations were performed on ground truth meshes and interpolated onto point-point correspondent meshes to create the ground truth dataset. The dataset was split 110/10/15 for training, validation, and testing. Image2Flow, a hybrid image and graph convolutional neural network, was trained to transform a pulmonary artery template to patient-specific anatomy and CFD values, taking a specific inlet velocity as an additional input. Image2Flow was evaluated in terms of segmentation, and the accuracy of predicted CFD was assessed using node-wise comparisons. In addition, the ability of Image2Flow to respond to increasing inlet velocities was also evaluated. Image2Flow achieved excellent segmentation accuracy with a median Dice score of 0.91 (IQR: 0.86–0.92). The median node-wise normalized absolute error for pressure and velocity magnitude was 11.75% (IQR: 9.60–15.30%) and 9.90% (IQR: 8.47–11.90), respectively. Image2Flow also showed an expected response to increased inlet velocities with increasing pressure and velocity values. This proof-of-concept study has shown that it is possible to simultaneously perform patient-specific volume-mesh based segmentation and pressure and flow field estimation using Image2Flow. Image2Flow completes segmentation and CFD in ~330ms, which is ~5000 times faster than manual methods, making it more feasible in a clinical environment.