Self-Expanding or Balloon-Expandable TAVR in Patients with a Small Aortic Annulus.

BACKGROUND: Patients with severe aortic stenosis and a small aortic annulus are at risk for impaired valvular hemodynamic performance and associated adverse cardiovascular clinical outcomes after transcatheter aortic-valve replacement (TAVR). METHODS: We randomly assigned patients with symptomatic severe aortic stenosis and an aortic-valve annulus area of 430 mm2 or less in a 1:1 ratio to undergo TAVR with either a self-expanding supraannular valve or a balloon-expandable valve. The coprimary end points, each assessed through 12 months, were a composite of death, disabling stroke, or rehospitalization for heart failure (tested for noninferiority) and a composite end point measuring bioprosthetic-valve dysfunction (tested for superiority). RESULTS: A total of 716 patients were treated at 83 sites in 13 countries (mean age, 80 years; 87% women; mean Society of Thoracic Surgeons Predicted Risk of Mortality, 3.3%). The Kaplan-Meier estimate of the percentage of patients who died, had a disabling stroke, or were rehospitalized for heart failure through 12 months was 9.4% with the self-expanding valve and 10.6% with the balloon-expandable valve (difference, -1.2 percentage points; 90% confidence interval [CI], -4.9 to 2.5; P<0.001 for noninferiority). The Kaplan-Meier estimate of the percentage of patients with bioprosthetic-valve dysfunction through 12 months was 9.4% with the self-expanding valve and 41.6% with the balloon-expandable valve (difference, -32.2 percentage points; 95% CI, -38.7 to -25.6; P<0.001 for superiority). The aortic-valve mean gradient at 12 months was 7.7 mm Hg with the self-expanding valve and 15.7 mm Hg with the balloon-expandable valve, and the corresponding values for additional secondary end points through 12 months were as follows: mean effective orifice area, 1.99 cm2 and 1.50 cm2; percentage of patients with hemodynamic structural valve dysfunction, 3.5% and 32.8%; and percentage of women with bioprosthetic-valve dysfunction, 10.2% and 43.3% (all P<0.001). Moderate or severe prosthesis-patient mismatch at 30 days was found in 11.2% of the patients in the self-expanding valve group and 35.3% of those in the balloon-expandable valve group (P<0.001). Major safety end points appeared to be similar in the two groups. CONCLUSIONS: Among patients with severe aortic stenosis and a small aortic annulus who underwent TAVR, a self-expanding supraannular valve was noninferior to a balloon-expandable valve with respect to clinical outcomes and was superior with respect to bioprosthetic-valve dysfunction through 12 months. (Funded by Medtronic; SMART ClinicalTrials.gov number, NCT04722250.).

Optimising brain age estimation through transfer learning: A suite of pre-trained foundation models for improved performance and generalisability in a clinical setting.

Estimated age from brain MRI data has emerged as a promising biomarker of neurological health. However, the absence of large, diverse, and clinically representative training datasets, along with the complexity of managing heterogeneous MRI data, presents significant barriers to the development of accurate and generalisable models appropriate for clinical use. Here, we present a deep learning framework trained on routine clinical data (N up to 18,890, age range 18-96 years). We trained five separate models for accurate brain age prediction (all with mean absolute error ≤4.0 years, R2 ≥ .86) across five different MRI sequences (T2 -weighted, T2 -FLAIR, T1 -weighted, diffusion-weighted, and gradient-recalled echo T2 *-weighted). Our trained models offer dual functionality. First, they have the potential to be directly employed on clinical data. Second, they can be used as foundation models for further refinement to accommodate a range of other MRI sequences (and therefore a range of clinical scenarios which employ such sequences). This adaptation process, enabled by transfer learning, proved effective in our study across a range of MRI sequences and scan orientations, including those which differed considerably from the original training datasets. Crucially, our findings suggest that this approach remains viable even with limited data availability (as low as N = 25 for fine-tuning), thus broadening the application of brain age estimation to more diverse clinical contexts and patient populations. By making these models publicly available, we aim to provide the scientific community with a versatile toolkit, promoting further research in brain age prediction and related areas.

Effects of complete revascularization according to age in patients with ST-segment elevation myocardial infarction and multivessel disease (COMPLETE-AGE).

BACKGROUND: In ST-segment elevation myocardial infarction (STEMI), complete revascularization with percutaneous coronary intervention (PCI) reduces major cardiovascular events compared with culprit-lesion-only PCI. Whether age influences these results remains unknown. METHODS: COMPLETE was a multinational, randomized trial evaluating a strategy of staged complete revascularization, consisting of angiography-guided PCI of all suitable nonculprit lesions, versus a strategy of culprit-lesion-only PCI. In this prespecified subgroup analysis, treatment effect according to age (≥65 years vs <65 years) was determined for the first coprimary outcome of cardiovascular (CV) death or new myocardial infarction (MI) and the second coprimary outcome of CV death, new MI, or ischemia-driven revascularization (IDR). Median follow-up was 35.8 months (interquartile range [IQR]: 27.6-44.3 months). RESULTS: Of 4,041 patients randomized in COMPLETE, 1,613 were aged ≥ 65 years (39.9%). Higher event rates were observed for both coprimary outcomes in patients aged ≥ 65 years comparted with those aged < 65 years (11.2% vs 7.9%, HR 1.49, 95% CI 1.22-1.83; 14.4% vs 11.8%, HR 1.28, 95% CI 1.07-1.52, respectively). Complete revascularization reduced the first coprimary outcome in patients ≥ 65 years (9.7% vs 12.5%, HR 0.77; 95% CI, 0.58-1.04) and < 65 years (6.7% vs 9.1%, HR 0.72; 95% CI, 0.54-0.96)(interaction P = .74). The second coprimary outcome was reduced in those ≥ 65 years (HR 0.56, 95% CI, 0.43-0.74) and < 65 years (HR 0.48, 95% CI, 0.37-0.61 (interaction P = .37). A sensitivity analysis was performed with consistent results demonstrated using a 75-year threshold (albeit attenuated). CONCLUSIONS: In patients with STEMI and multivessel CAD, complete revascularization compared with culprit-lesion-only PCI reduced major cardiovascular events regardless of patient age and could be considered as a revascularization strategy in older adults.

Trend-attribute forecasting of hourly PM2.5 trends in fifteen cities of Central England applying optimized machine learning feature selection.

Recorded particulate matter (PM2.5) hourly trends are compared for fifteen urban recording sites distributed across central England for the period 2018 to 2022. They include 10 urban-background and five urban-traffic (roadside) sites with some located within the same urban area. The sites all show consistent background and peak distributions with mean annual values and standard deviations higher for 2018 and 2019 than for 2020 to 2022. The objective of this study is to demonstrate that trend attributes extracted from hourly recorded univariate PM2.5 trends at these sites can be used to provide reliable short-term hourly predictions and provide valuable insight into the regional variations in the recorded trends. Fifteen trend attributes extracted from the prior 12 h (t-1 to t-12) of recorded PM2.5 data were compiled and used as input to four supervised machine learning models (SML) to forecast PM2.5 concentrations up to 13 h ahead (t0 to t+12). All recording sites delivered forecasts with similar ranges of error levels for specific hours ahead which are consistent with their PM2.5 recorded ranges. Forecasting results for four representative sites are presented in detail using models trained and cross-validated with 2020 and 2021 hourly data to forecast 2021 and 2022 hourly data, respectively. A novel optimized feature selection procedure using a suite of five optimizers is used to improve the efficiency of the forecasting models. The LASSO and support vector regression models generate the best and most generalizable hourly PM2.5 forecasts from trained and validated SML models with mean average error (MAE) of between ∼1 and ∼3 µg/m3 for t0 to t+3 h ahead. A novel overfitting indicator, exploiting the cross-validation mean values, demonstrates that these two models are not affected by overfitting. Forecasts for t+6 to t+12 h forward generate higher MAE values between ∼3 and ∼4 µg/m3 due to their tendency to underestimate some of the extreme PM2.5 peaks. These findings indicate that further model refinements are required to generate more reliable short-term predictions for the t+6 to t+24 h ahead.

Stochastic lithofacies and petrophysical property modeling for fast history matching in heterogeneous clastic reservoir applications.

For complex and multi-layered clastic oil reservoir formations, modeling lithofacies and petrophysical parameters is essential for reservoir characterization, history matching, and uncertainty quantification. This study introduces a real oilfield case study that conducted high-resolution geostatistical modeling of 3D lithofacies and petrophysical properties for rapid and reliable history matching of the Luhais oil reservoir in southern Iraq. For capturing the reservoir's tidal depositional setting using data collected from 47 wells, the lithofacies distribution (sand, shaly sand, and shale) of a 3D geomodel was constructed using sequential indicator simulation (SISIM). Based on the lithofacies modeling results, 50 sets of porosity and permeability distributions were generated using sequential Gaussian simulation (SGSIM) to provide insight into the spatial geological uncertainty and stochastic history matching. For each rock type, distinct variograms were created in the 0° azimuth direction, representing the shoreface line. The standard deviation between every pair of spatial realizations justified the number of variograms employed. An upscaled version of the geomodel, incorporating the lithofacies, permeability, and porosity, was used to construct a reservoir-flow model capable of providing rapid, accurate, and reliable production history matching, including well and field production rates.

Hydrodynamic Assessment of Explanted Degenerated Transcatheter Aortic Valves: Novel Insights Into Noncalcific and Calcific Mechanisms.

BACKGROUND: The etiology of transcatheter aortic valve (TAV) degeneration is poorly understood, particularly noncalcific mechanisms. OBJECTIVES: The authors sought to investigate noncalcific and calcific mechanisms of TAV degeneration and evaluate their impact on leaflet function by bench testing, imaging, and histology. METHODS: TAV explants were obtained from the EXPLANT THV registry and clinical institutions. Hydrodynamic assessment was performed using a heart valve pulse duplicator system under physiological conditions. Micro-computed tomography, high-resolution photography, high speed video, and hematoxylin and eosin staining were used to evaluate the morphological appearance, leaflet kinematics, and calcium burden of TAVs. RESULTS: A total of 14 explants were evaluated: 10 self-expanding CoreValve/Evolut TAVs (Medtronic), 3 balloon-expandable SAPIEN 3 TAVs (Edwards Lifesciences), and 1 mechanically expandable Lotus TAV (Boston Scientific). The median patient age at explantation was 73.0 years (Q1-Q3: 64.5-80.0 years), with a time to explantation of 4 years 1 month (1 year 5 months to 4 years 11 months). Six TAV explants were found to have leaflet calcification (162.4 mm3; 58.8-603.0 mm3), and 8 had no calcification detectable by micro-computed tomography and histology. All samples had impaired leaflet kinematics. There was no significant difference in the hydrodynamic mean gradient between calcified (47.2 mm Hg; 26.6-74.1 mm Hg) and noncalcified (27.6 mm Hg; 15.2-36.7 mm Hg; P = 0.28) TAVs. Leaflet calcification had a weak but nonsignificant association with the hydrodynamic mean gradient (r = 0.42; P = 0.14). CONCLUSIONS: TAV function can be severely impacted by noncalcific and calcific mechanisms of tissue degeneration. Importantly, functional stenosis can occur in TAVs in the absence of obvious and significant calcification.

Redo-TAVI with the ACURATE neo2 and Prime XL for balloon-expandable transcatheter heart valve failure.

BACKGROUND: There are limited data regarding treatment for failed balloon-expandable transcatheter heart valves (THVs) in redo-transcatheter aortic valve implantation (TAVI). AIMS: We aimed to assess THV performance, neoskirt height and expansion when performing redo-TAVI with the ACURATE platform inside a SAPIEN 3 (S3) compared to redo-TAVI with an S3 in an S3. METHODS: Redo-TAVI was performed on the bench using each available size of the S3, the ACURATE neo2 (ACn2) and the next-generation ACURATE Prime XL (AC XL) implanted at 2 different depths within 20 mm/23 mm/26 mm/29 mm S3s serving as the "failed" index THV. Hydrodynamic testing was performed to assess THV function. Multimodality assessment was performed using photography, X-ray, microcomputed tomography (micro-CT), and high-speed videos. RESULTS: The ACURATE in S3 combinations had favourable hydrodynamic performance compared to the S3 in S3 for all size combinations. In the 20 mm S3, redo-TAVI with the ACn2 had lower gradients compared to the S3 (mean gradient 16.3 mmHg for the ACn2 vs 24.7 mmHg for the 20 mm S3 in 20 mm S3). Pinwheeling was less marked for the ACURATE THVs than for the S3s. On micro-CT, the S3s used for redo-TAVI were underexpanded across all sizes. This was also observed for the ACURATE platform, but to a lesser extent. CONCLUSIONS: Redo-TAVI with an ACn2/AC XL within an S3 has favourable hydrodynamic performance and less pinwheeling compared to an S3 in S3. This comes at the price of a taller neoskirt.

Glioblastoma and radiotherapy: A multicenter AI study for Survival Predictions from MRI (GRASP study).

BACKGROUND: The aim was to predict survival of glioblastoma at 8 months after radiotherapy (a period allowing for completing a typical course of adjuvant temozolomide), by applying deep learning to the first brain MRI after radiotherapy completion. METHODS: Retrospective and prospective data were collected from 206 consecutive glioblastoma, isocitrate dehydrogenase -wildtype patients diagnosed between March 2014 and February 2022 across 11 UK centers. Models were trained on 158 retrospective patients from 3 centers. Holdout test sets were retrospective (n = 19; internal validation), and prospective (n = 29; external validation from 8 distinct centers). Neural network branches for T2-weighted and contrast-enhanced T1-weighted inputs were concatenated to predict survival. A nonimaging branch (demographics/MGMT/treatment data) was also combined with the imaging model. We investigated the influence of individual MR sequences; nonimaging features; and weighted dense blocks pretrained for abnormality detection. RESULTS: The imaging model outperformed the nonimaging model in all test sets (area under the receiver-operating characteristic curve, AUC P = .038) and performed similarly to a combined imaging/nonimaging model (P > .05). Imaging, nonimaging, and combined models applied to amalgamated test sets gave AUCs of 0.93, 0.79, and 0.91. Initializing the imaging model with pretrained weights from 10 000s of brain MRIs improved performance considerably (amalgamated test sets without pretraining 0.64; P = .003). CONCLUSIONS: A deep learning model using MRI images after radiotherapy reliably and accurately determined survival of glioblastoma. The model serves as a prognostic biomarker identifying patients who will not survive beyond a typical course of adjuvant temozolomide, thereby stratifying patients into those who might require early second-line or clinical trial treatment.