Evidence for transcatheter aortic valve implantation (TAVI) is scarce among patients with non-calcific aortic stenosis, and it is not known whether aortic valve calcification is associated with new cerebral ischemic lesions (CILs) that are detected by diffusion-weighted magnetic resonance imaging. So, our study enrolled 328 patients who underwent transfemoral TAVI using a self-expanding valve between December 2016 and June 2021 from the TORCH registry (NCT02803294). A total of 34 patients were finally confirmed as non-calcific AS and the remaining 294 patients were included in the calcific AS group. Incidence of new CILs (70.6% vs. 85.7%, p = 0.022), number of lesions (2.0 vs. 3.0, p = 0.010), and lesions volume (105.0 mm³ vs. 200.0 mm³, p = 0.047) was significantly lower in the non-calcific AS group. However, the maximum and average lesion volumes were comparable between two groups. Non-calcific AS was associated with lower risk for developing new CILs by univariate logistic regression analysis [Odds ratio (OR): 0.040, 95% confident interval (CI): 0.18–0.90, p = 0.026] and multivariate analysis (OR: 0.031, 95% CI: 0.13–0.76, p = 0.010). In summary, non-calcific AS patients had a lower risk of developing new cerebral ischemic infarction after TAVI compared to calcific AS patients. However, new ischemic lesions were still found in over 70% of patients.

Background: Comparative data of the Valve Academic Research Consortium (VARC-3)-defined technical success between bicuspid versus tricuspid aortic stenosis (AS) remain lacking. Aims: We sought to compare the technical success and other clinical outcomes between patients with bicuspid and tricuspid AS receiving transcatheter aortic valve replacement. Methods: A registration-based analysis was performed for 402 patients (211 and 191 cases of bicuspid and tricuspid AS, respectively). The primary outcome was VARC-3-defined technical success. Additional analysis was performed to assess outcomes for up to one year between the two groups. Results: Bicuspid AS patients tended to be younger (74 years vs. 77 years; p < 0.001) with a lower Society of Thoracic Surgeons score (4.4% vs. 5.4%; p = 0.003). Bicuspid AS patients showed a lower prevalence of hypertension and peripheral vascular diseases. Technical failure was encountered in 17.7% of these patients, driven primarily by the high incidence of second valve implantation. The technical success rates were comparable between the bicuspid and tricuspid AS groups (82.5% vs. 82.2%, p = 0.944). Chronic kidney disease (CKD) and larger sinotubular junctional diameter (STJ) were identified as predictors of technical failure, whereas CKD, impaired left ventricular ejection fraction (LVEF), along with larger STJ, were predictors of cardiac technical failure. Technical failure was associated with an increased risk of all-cause mortality at 30 days and 1 year, as evidenced by the Cox multivariable analysis. Conclusions: No significant differences were observed in the technical success rates and most clinical outcomes between the bicuspid and tricuspid AS groups. Technical failure conferred an increased risk for both 30-day and 1-year all-cause mortalities.

Discrete Ordinary Differential Equations (ODEs) have been employed to develop lightweight deep neural networks in recent years. In this letter, we introduce a novel lightweight UNet variant called the neural memory Self-Supervised State Space Model (nmS4M-UNet) for medical image segmentation, where discrete ODEs serve as the decoder. The proposed nmS4M block has learnable gates and performs multi-head computation to enhance memory updates. Additionally, the nmS4M-UNet incorporates a self-supervised learning branch to improve feature extraction capabilities. The intermediate features are reused as partial input to the decoder, helping to mitigate network overfitting. The nmS4M-UNet reduces the number of parameters by 29.70% compared to the standard UNet. Experimental results on the PH2, ISIC2018, and BU-COCO datasets demonstrate that the proposed nmS4M-UNet achieves performance comparable to state-of-the-art models.

Backgrounds: The risk of brain injury in bicuspid aortic valve (BAV) patients following transcatheter aortic valve replacement (TAVR) is currently unknown. Methods and Results: A total of 204 consecutive severe aortic stenosis patients who underwent TAVR were enrolled. 83 (40.7%) patients were BAV patients and the other 121 patients were tricuspid aortic valve (TAV) patients. All patients received diffusion-weighted magnetic resonance imaging (DW-MRI) at baseline, 2 to 7 days after TAVR. Mean ages (mean ± SD: 75.8 ± 6.7 years vs. 78.9 ± 6.6 years, p = 0.004) and STS scores (6.0 ± 3.7 vs. 7.1 ± 4.2, p = 0.044) of the BAV and TAV patients were significantly different, while the stroke rates (2.4% vs. 1.7%, p = 0.704) were comparable between two groups. BAV patients were associated with higher number of new lesions (5.69 ± 6.22 vs. 3.50 ± 4.16, p = 0.008), total lesion volume [median(interquartile range): 290(70-930) mm ³ vs. 140(35-480) mm ³ , p = 0.008], and the volume per lesion [70.0(45.0-115.0) mm ³ vs. 57.5(24.5-93.0) mm ³ , p = 0.037] in DW-MRI. Moreover, the proportion of patients with lesions larger than 1cm ³ (28.6% vs. 10.9%, p = 0.005) and the number of new lesions in the middle cerebral arteries zone (1.46 ± 2.07 vs. 0.98 ± 1.84, p = 0.039) and intermediate zone between the anterior cerebral and middle cerebral arteries (ACA/MCA) (1.07 ± 1.68 vs. 0.50 ± 1.05, p = 0.007), and between the vertebral artery and basilar artery (VA/BA) (1.01 ± 1.35 vs. 0.77 ± 1.44, p = 0.033) were higher in BAV patients than in TAV patients. Conclusions: BAV patients may encounter more severe brain injuries not only due to greater number of lesions but also due to larger lesion size, especially in the ACA/MCA, MCA and VA/BA lesions zone.