Results
- Showing results for:
- Reset all filters
Search results
-
Journal articleJones B, Rane N, Finnegan M, et al., 2024,
Effect of evolocumab on carotid plaque composition in asymptomatic carotid artery stenosis (EVOCAR-1) using magnetic resonance imaging
, Journal of Clinical Lipidology, Vol: 18, Pages: e855-e866, ISSN: 1933-2874Background and AimsTo determine the effect of evolocumab treatment in patients with asymptomatic carotid artery stenosis ≥50% on carotid plaque morphology and composition, as determined by magnetic resonance imaging.MethodsWe conducted a double-blind randomized controlled trial in patients with asymptomatic carotid artery plaque with ≥50% stenosis and low-density lipoprotein-associated cholesterol (LDL-C) ≥1.8 mmol/L, despite standard lipid-lowering therapy, with 12 months of evolocumab or placebo injection every two weeks. The primary endpoint was the between group difference in the absolute change from baseline in carotid plaque lipid-rich necrotic core (LRNC), assessed by carotid magnetic resonance.ResultsDue to interrupted recruitment during the COVID-19 pandemic, 33 patients (36% female) were randomised, which was less than the target of 52. Mean age was 68.7 years (SD, 8.5) and baseline LDL-C 2.4 mmol/L (SD, 0.7). LDL-C was reduced with evolocumab to 0.8 mmol/L (SD, 0.5) vs 2.2 mmol/L (SD, 0.7) with placebo at 3 months (between group absolute difference -1.3 mmol/L [95% confidence interval [CI], -1.7 to -0.9], p < 0.001). Evolocumab treatment was associated with a favourable change in LRNC at 12 months of -16 mm3 (SD, 54) whereas the placebo group showed -4 mm3 (SD, 44). Between group differences did not show statistical significance with a placebo-adjusted LRNC change of -17 mm3 ([95% CI, -45 to 12], p = 0.25). Percentage carotid plaque LRNC also numerically reduced at 12 months, however this did not reach statistical significance (-2.4% vessel wall volume [95% CI, -5.7 to 0.9], p = 0.16).ConclusionIntensive LDL-C lowering with the addition of evolocumab to maximally tolerated lipid-lowering therapy did not lead to a statistically significant change in vulnerable plaque phenotype characteristics in patients with asymptomatic carotid artery stenosis, but the study was underpowered due to under-recruitment in the context of the COVID-19 pande
-
Conference paperFermoyle C, Walsh S, Mackintosh J, et al., 2024,
A deep learning algorithm for predicting progression in fibrosing interstitial lung disease
, European-Respiratory-Society Congress (ERS), Publisher: EUROPEAN RESPIRATORY SOC JOURNALS LTD, ISSN: 0903-1936 -
Journal articleMcDowell AR, Zambreanu L, Salhab HA, et al., 2024,
Initial findings using high-resolution magnetic resonance imaging for visualisation of the sural nerve and surrounding anatomy in healthy volunteers at 7 Tesla
, Journal of the Peripheral Nervous System, Vol: 29, Pages: 368-375, ISSN: 1085-9489Background and AimsHistopathological diagnosis is the gold standard in many acquired inflammatory, infiltrative and amyloid based peripheral nerve diseases and a sensory nerve biopsy of sural or superficial peroneal nerve is favoured where a biopsy is deemed necessary. The ability to determine nerve pathology by high-resolution imaging techniques resolving anatomy and imaging characteristics might improve diagnosis and obviate the need for biopsy in some. The sural nerve is anatomically variable and occasionally adjacent vessels can be sent for analysis in error. Knowing the exact position and relationships of the nerve prior to surgery could be clinically useful and thus reliably resolving nerve position has some utility.Methods7T images of eight healthy volunteers' (HV) right ankle were acquired in a pilot study using a double-echo in steady-state sequence for high-resolution anatomy images. Magnetic Transfer Ratio images were acquired of the same area. Systematic scoring of the sural, tibial and deep peroneal nerve around the surgical landmark 7 cm from the lateral malleolus was performed (number of fascicles, area in voxels and mm2, diameter and location relative to nearby vessels and muscles).ResultsThe sural and tibial nerves were visualised in the high-resolution double-echo in steady-state (DESS) image in all HV. The deep peroneal nerve was not always visualised at level of interest. The MTR values were tightly grouped except in the sural nerve where the nerve was not visualised in two HV. The sural nerve location was found to be variable (e.g., lateral or medial to, or crossing behind, or found positioned directly posterior to the saphenous vein).InterpretationHigh-resolution high-field images have excellent visualisation of the sural nerve and would give surgeons prior knowledge of the position before surgery. Basic imaging characteristics of the sural nerve can be acquired, but more detailed imaging characteristics are not easily evaluable in the v
-
Journal articleZhou X, Wang X, Ma H, et al., 2024,
Customized T-time inner sampling network with uncertainty-aware data augmentation strategy for multi-annotated lesion segmentation.
, Comput Biol Med, Vol: 180Segmentation in medical images is inherently ambiguous. It is crucial to capture the uncertainty in lesion segmentations to assist cancer diagnosis and further interventions. Recent works have made great progress in generating multiple plausible segmentation results as diversified references to account for the uncertainty in lesion segmentations. However, the efficiency of existing models is limited, and the uncertainty information lying in multi-annotated datasets remains to be fully utilized. In this study, we propose a series of methods to corporately deal with the above limitation and leverage the abundant information in multi-annotated datasets: (1) Customized T-time Inner Sampling Network to promote the modeling flexibility and efficiently generate samples matching the ground-truth distribution of a number of annotators; (2) Uncertainty Degree defined for quantitatively measuring the uncertainty of each sample and the imbalance of the whole multi-annotated dataset from a brand new perspective; (3) Uncertainty-aware Data Augmentation Strategy to help probabilistic models adaptively fit samples with different ranges of uncertainty. We have evaluated each of them on both the publicly available lung nodule dataset and our in-house Liver Tumor dataset. Results show that our proposed methods achieves the overall best performance on both accuracy and efficiency, demonstrating its great potential in lesion segmentations and more downstream tasks in real clinical scenarios.
-
Journal articleO'Connor SAJ, Watson EJR, Grech-Sollars M, et al., 2024,
Perioperative research into memory (PRiMe), part 2: Adult burns intensive care patients show altered structure and function of the default mode network
, BURNS, Vol: 50, Pages: 1908-1915, ISSN: 0305-4179 -
Journal articleBoyalla V, Haldar S, Khan H, et al., 2024,
Long-term clinical outcomes and cost-effectiveness of catheter vs thoracoscopic surgical ablation in long-standing persistent atrial fibrillation using continuous cardiac monitoring: CASA-AF randomized controlled trial
, HEART RHYTHM, Vol: 21, Pages: 1562-1569, ISSN: 1547-5271- Cite
- Citations: 5
-
Journal articleYang Y, Zhou Z, Zhang N, et al., 2024,
Performance of artificial intelligence in detecting the chronic total occlusive lesions of coronary artery based on coronary computed tomographic angiography
, CARDIOVASCULAR DIAGNOSIS AND THERAPY, Vol: 14, ISSN: 2223-3652 -
Journal articleLi C, Tan J, Li H, et al., 2024,
The value of multiparametric MRI-based habitat imaging for differentiating uterine sarcomas from atypical leiomyomas: a multicentre study
, ABDOMINAL RADIOLOGY, ISSN: 2366-004X -
Journal articleHu J, Wu P, Wang S, et al., 2024,
A human feedback strategy for photoresponsive molecules in drug delivery: utilizing GPT-2 and time-dependent density functional theory calculations
, Pharmaceutics, Vol: 16, ISSN: 1999-4923Photoresponsive drug delivery stands as a pivotal frontier in smart drug administration, leveraging the non-invasive, stable, and finely tunable nature of light-triggered methodologies. The generative pre-trained transformer (GPT) has been employed to generate molecular structures. In our study, we harnessed GPT-2 on the QM7b dataset to refine a UV-GPT model with adapters, enabling the generation of molecules responsive to UV light excitation. Utilizing the Coulomb matrix as a molecular descriptor, we predicted the excitation wavelengths of these molecules. Furthermore, we validated the excited state properties through quantum chemical simulations. Based on the results of these calculations, we summarized some tips for chemical structures and integrated them into the alignment of large-scale language models within the reinforcement learning from human feedback (RLHF) framework. The synergy of these findings underscores the successful application of GPT technology in this critical domain.
-
Journal articleWang S, Nan Y, Zhang S, et al., 2024,
Probing perfection: The relentless art of meddling for pulmonary airway segmentation from HRCT via a human-AI collaboration based active learning method
, ARTIFICIAL INTELLIGENCE IN MEDICINE, Vol: 154, ISSN: 0933-3657
This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.
Contact
For enquiries about the MRI Physics Collective, please contact:
Mary Finnegan
Senior MR Physicist at the Imperial College Healthcare NHS Trust
Pete Lally
Assistant Professor in Magnetic Resonance (MR) Physics at Imperial College
Jan Sedlacik
MR Physicist at the Robert Steiner MR Unit, Hammersmith Hospital Campus