Johannes Kepler University of Linz

About: Johannes Kepler University of Linz is a education organization based out in Linz, Oberösterreich, Austria. It is known for research contribution in the topics: Computer science & Thin film. The organization has 6605 authors who have published 19243 publications receiving 385667 citations.

Characteristics and Outcomes in Patients With COVID-19 and Acute Ischemic Stroke: The Global COVID-19 Stroke Registry.

Abstract: Recent case-series of small size implied a pathophysiological association between coronavirus disease 2019 (COVID-19) and severe large-vessel acute ischemic stroke. Given that severe strokes are typically associated with poor prognosis and can be very efficiently treated with recanalization techniques, confirmation of this putative association is urgently warranted in a large representative patient cohort to alert stroke clinicians, and inform pre- and in-hospital acute stroke patient pathways. We pooled all consecutive patients hospitalized with laboratory-confirmed COVID-19 and acute ischemic stroke in 28 sites from 16 countries. To assess whether stroke severity and outcomes (assessed at discharge or at the latest assessment for those patients still hospitalized) in patients with acute ischemic stroke are different between patients with COVID-19 and non-COVID-19, we performed 1:1 propensity score matching analyses of our COVID-19 patients with non-COVID-19 patients registered in the Acute Stroke Registry and Analysis of Lausanne Registry between 2003 and 2019. Between January 27, 2020, and May 19, 2020, 174 patients (median age 71.2 years; 37.9% females) with COVID-19 and acute ischemic stroke were hospitalized (median of 12 patients per site). The median National Institutes of Health Stroke Scale was 10 (interquartile range [IQR], 4-18). In the 1:1 matched sample of 336 patients with COVID-19 and non-COVID-19, the median National Institutes of Health Stroke Scale was higher in patients with COVID-19 (10 [IQR, 4-18] versus 6 [IQR, 3-14]), P=0.03; (odds ratio, 1.69 [95% CI, 1.08-2.65] for higher National Institutes of Health Stroke Scale score). There were 48 (27.6%) deaths, of which 22 were attributed to COVID-19 and 26 to stroke. Among 96 survivors with available information about disability status, 49 (51%) had severe disability at discharge. In the propensity score-matched population (n=330), patients with COVID-19 had higher risk for severe disability (median mRS 4 [IQR, 2-6] versus 2 [IQR, 1-4], P<0.001) and death (odds ratio, 4.3 [95% CI, 2.22-8.30]) compared with patients without COVID-19. Our findings suggest that COVID-19 associated ischemic strokes are more severe with worse functional outcome and higher mortality than non-COVID-19 ischemic strokes.

Orai1 contributes to the establishment of an apoptosis-resistant phenotype in prostate cancer cells

Abstract: The molecular nature of calcium (Ca(2+))-dependent mechanisms and the ion channels having a major role in the apoptosis of cancer cells remain a subject of debate. Here, we show that the recently identified Orai1 protein represents the major molecular component of endogenous store-operated Ca(2+) entry (SOCE) in human prostate cancer (PCa) cells, and constitutes the principal source of Ca(2+) influx used by the cell to trigger apoptosis. The downregulation of Orai1, and consequently SOCE, protects the cells from diverse apoptosis-inducing pathways, such as those induced by thapsigargin (Tg), tumor necrosis factor α, and cisplatin/oxaliplatin. The transfection of functional Orai1 mutants, such as R91W, a selectivity mutant, and L273S, a coiled-coil mutant, into the cells significantly decreased both SOCE and the rate of Tg-induced apoptosis. This suggests that the functional coupling of STIM1 to Orai1, as well as Orai1 Ca(2+)-selectivity as a channel, is required for its pro-apoptotic effects. We have also shown that the apoptosis resistance of androgen-independent PCa cells is associated with the downregulation of Orai1 expression as well as SOCE. Orai1 rescue, following Orai1 transfection of steroid-deprived cells, re-established the store-operated channel current and restored the normal rate of apoptosis. Thus, Orai1 has a pivotal role in the triggering of apoptosis, irrespective of apoptosis-inducing stimuli, and in the establishment of an apoptosis-resistant phenotype in PCa cells.

Central Moment Discrepancy (CMD) for Domain-Invariant Representation Learning

Abstract: The learning of domain-invariant representations in the context of domain adaptation with neural networks is considered. We propose a new regularization method that minimizes the discrepancy between domain-specific latent feature representations directly in the hidden activation space. Although some standard distribution matching approaches exist that can be interpreted as the matching of weighted sums of moments, e.g. Maximum Mean Discrepancy (MMD), an explicit order-wise matching of higher order moments has not been considered before. We propose to match the higher order central moments of probability distributions by means of order-wise moment differences. Our model does not require computationally expensive distance and kernel matrix computations. We utilize the equivalent representation of probability distributions by moment sequences to define a new distance function, called Central Moment Discrepancy (CMD). We prove that CMD is a metric on the set of probability distributions on a compact interval. We further prove that convergence of probability distributions on compact intervals w.r.t. the new metric implies convergence in distribution of the respective random variables. We test our approach on two different benchmark data sets for object recognition (Office) and sentiment analysis of product reviews (Amazon reviews). CMD achieves a new state-of-the-art performance on most domain adaptation tasks of Office and outperforms networks trained with MMD, Variational Fair Autoencoders and Domain Adversarial Neural Networks on Amazon reviews. In addition, a post-hoc parameter sensitivity analysis shows that the new approach is stable w.r.t. parameter changes in a certain interval. The source code of the experiments is publicly available.