Resource efficient aortic distensibility calculation by end to end spatiotemporal learning of aortic lumen from multicentre multivendor multidisease CMR images.

Aortic distensibility (AD) is important for the prognosis of multiple cardiovascular diseases. We propose a novel resource-efficient deep learning (DL) model, inspired by the bi-directional ConvLSTM U-Net with densely connected convolutions, to perform end-to-end hierarchical learning of the aorta from cine cardiovascular MRI towards streamlining AD quantification. Unlike current DL aortic segmentation approaches, our pipeline: (i) performs simultaneous spatio-temporal learning of the video input, (ii) combines the feature maps from the encoder and decoder using non-linear functions, and (iii) takes into account the high class imbalance. By using multi-centre multi-vendor data from a highly heterogeneous patient cohort, we demonstrate that the proposed method outperforms the state-of-the-art method in terms of accuracy and at the same time it consumes [Formula: see text] 3.9 times less fuel and generates [Formula: see text] 2.8 less carbon emissions. Our model could provide a valuable tool for exploring genome-wide associations of the AD with the cognitive performance in large-scale biomedical databases. By making energy usage and carbon emissions explicit, the presented work aligns with efforts to keep DL's energy requirements and carbon cost in check. The improved resource efficiency of our pipeline might open up the more systematic DL-powered evaluation of the MRI-derived aortic stiffness.

PHACTR1 modulates vascular compliance but not endothelial function: a translational study.

AIMS: The non-coding locus at 6p24 located in Intron 3 of PHACTR1 has consistently been implicated as a risk allele in myocardial infarction and multiple other vascular diseases. Recent murine studies have identified a role for Phactr1 in the development of atherosclerosis. However, the role of PHACTR1 in vascular tone and in vivo vascular remodelling has yet to be established. The aim of this study was to investigate the role of PHACTR1 in vascular function. METHODS AND RESULTS: Prospectively recruited coronary artery disease (CAD) patients undergoing bypass surgery and retrospectively recruited spontaneous coronary artery dissection (SCAD) patients and matched healthy volunteers were genotyped at the PHACTR1 rs9349379 locus. We observed a significant association between the PHACTR1 loci and changes in distensibility in both the ascending aorta (AA = 0.0053 ± 0.0004, AG = 0.0041 ± 0.003, GG = 0.0034 ± 0.0009, P < 0.05, n = 58, 54, and 7, respectively) and carotid artery (AA = 12.83 ± 0.51, AG = 11.14 ± 0.38, GG = 11.69 ± 0.66, P < 0.05, n = 70, 65, and 18, respectively). This association was not observed in the descending aorta or in SCAD patients. In contrast, the PHACTR1 locus was not associated with changes in endothelial cell function with no association between the rs9349379 locus and in vivo or ex vivo vascular function observed in CAD patients. This finding was confirmed in our murine model where the loss of Phactr1 on the pro-atherosclerosis ApoE-/- background did not alter ex vivo vascular function. CONCLUSION: In conclusion, we have shown a role for PHACTR1 in arterial compliance across multiple vascular beds. Our study suggests that PHACTR1 has a key structural role within the vasculature.

Variation in telangiectasia predisposing genes is associated with overall radiation toxicity.

PURPOSE: In patients receiving radiotherapy for breast cancer where the heart is within the radiation field, cutaneous telangiectasiae could be a marker of potential radiation-induced heart disease. We hypothesized that single nucleotide polymorphisms (SNPs) in genes known to cause heritable telangiectasia-associated disorders could predispose to such late, normal tissue vascular damage. METHODS AND MATERIALS: The relationship between cutaneous telangiectasia as a late normal tissue radiation injury phenotype in 633 breast cancer patients treated with radiotherapy was examined. Patients were clinically assessed for the presence of cutaneous telangiectasia and genotyped at nine SNPs in three candidate genes. Candidate SNPs were within the endoglin (ENG) and activin A receptor, type II-like 1 (ACVRL1) genes, mutations in which cause hereditary hemorrhagic telangiectasia and the ataxia-telangiectasia mutated (ATM) gene associated with ataxia-telangiectasia. RESULTS: A total of 121 (19.1%) patients exhibited a degree of cutaneous telangiectasiae on clinical examination. Regression was used to examine the associations between the presence of telangiectasiae in patients who underwent breast-conserving surgery, controlling for the effects of boost and known brassiere size (n=388), and individual geno- or haplotypes. Inheritance of ACVRL1 SNPs marginally contributed to the risk of cutaneous telangiectasiae. Haplotypic analysis revealed a stronger association between inheritance of a ATM haplotype and the presence of cutaneous telangiectasiae, fibrosis and overall toxicity. No significant association was observed between telangiectasiae and the coinheritance of the candidate ENG SNPs. CONCLUSIONS: Genetic variation in the ATM gene influences reaction to radiotherapy through both vascular damage and increased fibrosis. The predisposing variation in the ATM gene will need to be better defined to optimize it as a predictive marker for assessing radiotherapy late effects.