This paper presents a far-field text-dependent speaker verification database named HI-MIA. We aim to meet the data requirement for far-field microphone array based speaker verification since most of the publicly available databases are single channel close-talking and text-independent. The database contains recordings of 340 people in rooms designed for the far-field scenario. Recordings are captured by multiple microphone arrays located in different directions and distance to the speaker and a high-fidelity close-talking microphone. Besides, we propose a set of end-to-end neural network based baseline systems that adopt single-channel data for training. Moreover, we propose a testing background aware enrollment augmentation strategy to further enhance the performance. Results show that the fusion systems could achieve 3.29% EER in the far-field enrollment far field testing task and 4.02% EER in the close-talking enrollment and far-field testing task.

The INTERSPEECH 2020 Far-Field Speaker Verification Challenge (FFSVC 2020) addresses three different research problems under well-defined conditions: far-field text-dependent speaker verification from single microphone array, far-field textindependent speaker verification from single microphone array, and far-field text-dependent speaker verification from distributed microphone arrays. All three tasks pose a cross-channel challenge to the participants. To simulate the real-life scenario, the enrollment utterances are recorded from close-talk cellphone, while the test utterances are recorded from the far-field microphone arrays. In this paper, we describe the database, the challenge, and the baseline system, which is based on a ResNetbased deep speaker network with cosine similarity scoring. For a given utterance, the speaker embeddings of different channels are equally averaged as the final embedding. The baseline system achieves minDCFs of 0.62, 0.66, and 0.64 and EERs of 6.27%, 6.55%, and 7.18% for task 1, task 2, and task 3, respectively.

Background: Corona Virus Disease 2019 (COVID-19) has become a global pandemic. This study established prognostic scoring models based on comorbidities and other clinical information for severe and critical patients with COVID-19. Material and Methods: We retrospectively collected data from 51 patients diagnosed as severe or critical COVID-19 who were admitted between January 29, 2020, and February 18, 2020. The Charlson (CCI), Elixhauser (ECI), and age-and smoking-adjusted Charlson (ASCCI) and Elixhauser (ASECI) comorbidity indices were used to evaluate the patient outcomes. Results: The mean hospital length of stay (LOS) of the COVID-19 patients was 22.82 ± 12.32 days; 19 patients (37.3%) were hospitalized for more than 24 days. Multivariate analysis identified older age (OR 1.064, P = 0.018, 95%CI 1.011-1.121) and smoking (OR 3.696, P = 0.080, 95%CI 0.856-15.955) as positive predictors of a long LOS. There were significant trends for increasing hospital LOS with increasing CCI, ASCCI, and ASECI scores (OR 57.500,

Aim:To investigate the possible effects of telmisartan and losartan on cardiac function in adriamycin (ADR)-induced heart failure in rats, and to explore the changes in plasma level of angiotensin-(1-7)[Ang-(1-7)] and myocardial expression of angiotensin II type 1/2 receptors (AT 1 R / AT 2 R) and Mas receptor caused by the two drugs. Methods: Male Sprague-Dawley rats were randomly divided into 4 groups: the control group, ADR-treated heart failure group (ADR-HF), telmisartan plus ADR-treated group (Tel+ADR) and losartan plus ADR-treated group (Los+ADR). ADR was administrated (2.5 mg/kg, ip, 6 times in 2 weeks). The rats in the Tel+ADR and Los+ADR groups were treated orally with telmisartan (10 mg/kg daily po) and losartan (30 mg/kg daily), respectively, for 6 weeks. The plasma level of Ang-(1-7) was determined using ELISA. The mRNA and protein expression of myocardial Mas receptor, AT 1 R and AT 2 R were measured using RT-PCR and Western blotting, respectively. Results: ADR significantly reduced the plasma level of Ang-(1-7) and the expression of myocardial Mas receptor and myocardial AT 2 R, while significantly increased the expression of myocardial AT 1 R. Treatment with telmisartan and losartan effectively increased the plasma level of Ang-(1-7) and suppressed myocardial AT 1 R expression, but did not influence the expression of Mas receptor and AT 2 R. Conclusion: The protective effects of telmisartan and losartan in ADR-induced heart failure may be partially due to regulation of circulating Ang-(1-7) and myocardial AT 1 R expression.Keywords: angiotensin-(1-7); Mas; angiotensin I receptor; angiotensin II receptor; adriamycin; heart failure; telmisartan; losartan Acta Pharmacologica Sinica (2011Sinica ( ) 32: 1345Sinica ( -1350 doi: 10.1038/aps.2011 published online 3 Oct 2011 Original Article # The first two authors contributed equally to this work. * To whom correspondence should be addressed. npg an oncogene as well as a receptor for Ang-(1-7) [13] . In a previous work, Mas receptor-deficient mice showed higher blood pressure values, impaired endothelial function, decreased NO production and lower endothelial NO synthetase expression [14] , indicating that the Ang-(1-7)/Mas receptor axis plays an important role in cardioprotective and antihypertensive effects. Importantly, Ang-(1-7) can prevent heart failure after myocardial ischemia [15] . Some recent studies have reported that angiotensin converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs), two classes of drugs that target RAS, may prevent ADR-induced cardiotoxicity [16][17][18][19] . However, the underlying mechanisms are largely unclear. In the present study, we investigated the regulation of two ARBs (telmisartan and losartan) on plasma Ang-(1-7) levels and the mRNA and protein expression of the myocardial AT 1 R, AT 2 R and Mas receptors in ADR-induced heart failure. Materials and methods Animals Experimental protocolRats were randomly divided into four groups: (1) the control group (n=10, intraperitoneally injected an...

Genetic deletion of the A2AR promotes development of relative myopia with increased axial length and altered scleral collagen fiber structure during postnatal development in mice. Thus, the A2AR may be important in normal refractive development.

Background/aimsTo describe the clinicopathological and immunohistochemical characteristics of 10 patients representing a new entity of benign conjunctival myxoid stromal tumours.MethodsRetrospective review of clinical findings, histopathological and immunohistochemical studies identified 10 cases of low-grade conjunctival myxoid stromal tumours. Specimens were routinely processed and stained with H&E. Immunohistochemical stains for CD34, CD68, vimentin, S100, smooth muscle actin (SMA), myosin, desmin, actin, Bcl-2 and Ki-67 were performed. Specific stains for Alcian-blue periodic acid-Schiff (AB-PAS) and aldehyde fuchsin stains were also performed.ResultsTen patients with an average age of 45.6±11.1 years had a tender white or faint yellow to red mass on the bulbar conjunctiva. All the lesions were completely removed, and none of the patients relapsed. Histologically, all neoplasms consisted of spindle-shaped cells that showed signs of pseudonuclear inclusions, multinuclear cells and had no atypia. The stroma consisted of a large amount of mucus and was infiltrated with delicate to ropey collagens, a few mast cells and new vessels. Immunohistochemical stains were positive for CD34, vimentin and Bcl-2; partial positive for CD68; very low for Ki-67; and negative for S100, SMA, myosin, desmin and actin. AB-PAS suggested that the stroma was mucinous.ConclusionsThese rare benign mesenchymal conjunctival tumours are mostly unilateral and occur in the bulbar conjunctiva. Complete resection is the radical treatment. These lesions are characterised by multiple spindle cells, a large amount of mucus, and sharing similar basic histopathological features with conjunctival myxoma and conjunctival stromal tumour. We suggest naming these lesions ‘conjunctival myxoid stromal tumours’.

Supplemental Digital Content is available in the text

In this paper, we focus on the far-field end-to-end textdependent speaker verification task with a small-scale far-field text dependent dataset and a large scale close-talking text independent database for training. First, we show that simulating far-field text independent data from the existing large-scale clean database for data augmentation can reduce the mismatch. Second, using a small far-field text dependent data set to finetune the deep speaker embedding model pre-trained from the simulated far-field as well as original clean text independent data can significantly improve the system performance. Third, in special applications when using the close-talking clean utterances for enrollment and employing the real far-field noisy utterances for testing, adding reverberant noises on the clean enrollment data can further enhance the system performance. We evaluate our methods on AISHELL ASR0009 and AISHELL 2019B-eval databases and achieve an equal error rate (EER) of 5.75% for far-field text-dependent speaker verification under noisy environments.