Search

Noisy situations cause huge problems for suffers of hearing loss as hearing aids often make the signal more audible but do not always restore the intelligibility. In noisy settings, humans routinely exploit the audio-visual (AV) nature of the speech to selectively suppress the background noise and to focus on the target speaker. In this paper, we present a causal, language, noise and speaker independent AV deep neural network (DNN) architecture for speech enhancement (SE). The model exploits the noisy acoustic cues and noise robust visual cues to focus on the desired speaker and improve the speech intelligibility. To evaluate the proposed SE framework a first of its kind AV binaural speech corpus, called ASPIRE, is recorded in real noisy environments including cafeteria and restaurant. We demonstrate superior performance of our approach in terms of objective measures and subjective listening tests over the state-of-the-art SE approaches as well as recent DNN based SE models. In addition, our work challenges a popular belief that a scarcity of multi-language large vocabulary AV corpus and wide variety of noises is a major bottleneck to build a robust language, speaker and noise independent SE systems. We show that a model trained on synthetic mixture of Grid corpus (with 33 speakers and a small English vocabulary) and ChiME 3 Noises (consisting of only bus, pedestrian, cafeteria, and street noises) generalise well not only on large vocabulary corpora but also on completely unrelated languages (such as Mandarin), wide variety of speakers and noises., 34 pages, 11 figures, Submitted to Information Fusion

Background Approximately 20% of all orthotropic heart transplant (OHT) recipients in the US receive anti-thymocyte globulin (ATG) post-transplant to prevent acute rejection, however use varies between centers. There is lack of prospective multi center clinical trials assessing long-term survival benefit of early ATG treatment including readmission for infections compared with no induction in OHT patients. We aim to review the role of ATG use during the index hospitalization for OHT to prevent acute cellular rejection (ACR) and evaluate the impact of ATG in readmission rates for infection at our center. Hypothesis We hypothesize that therapy with ATG improves freedom from ACR but may result in higher readmission rates for infection. Methods We performed a retrospective cohort study of 98 adults who underwent OHT at a single center from January 2014 to December 2019. Patients without one year follow up and those who underwent heart-kidney transplant were excluded. We compared baseline characteristics and analyzed outcomes between patients with and without ATG administration using standard statistical methods. Unadjusted Kaplan-Meier curves were used to study differences in ACR and hospitalization for infection between patients with and without ATG administration. Results 53 OHT patients received ATG and 45 patients did not. There was no significant difference in death at one year (p=1.0), ACR (p=0.17) or readmission for infection (p=0.35) between the two groups. Patients receiving ATG had delayed date to first ACR (p=0.038), with a median time of first ACR of 100 days versus 26 days post-OHT in the non-ATG group. Conclusion Treatment with ATG during the index hospitalization for OHT delayed the time of first ACR but not the overall burden of rejection in the first year. The rate of freedom from infection was similar in both groups along with overall mortality. As this is a retrospective observation we will analyze the use of ATG in selected patients at high risk for rejection or for renal sparing.

Introduction Acute kidney injury (AKI) complicates up to 25% of patients undergoing orthotopic heart transplant (OHT) and can be exacerbated by the introduction of calcineurin inhibition (CNI). Induction with rabbit anti-thymocyte globulin (ATG) can provide immunosuppression to prevent early rejection until CNI therapy is introduced. However, current data regarding ATG use to allow delayed CNI initiation is limited. Hypothesis ATG induction with delayed initiation of tacrolimus as a renal sparing strategy will result in less renal failure without a significant increase in infection and rejection. Methods From January 2014 to December 2019, 143 patients underwent OHT. Dual organ transplantation, those requiring ATG for non-renal sparing reasons, and patients without 1 -year follow-up were excluded. Using standard statistical analysis, we compared in-hospital and post discharge outcomes at 1, 6, and 12 months between patients with and without ATG administration. Results ATG was given to 26/71 (36.6%) patients as a renal sparing strategy. Compared to the no ATG group, the ATG group had higher pre-transplant creatinine (1.5 vs 1.2 mg/dL, p=0.001) and lower glomerular filtration rate (GFR) (58.6 vs 76.4 ml/min, p=0.002), but no other significant differences in baseline characteristics. Tacrolimus was started later in the ATG group (3.6 vs 0.9 days, p=0.0001). The ATG group had lower GFR at 1 month (64.1 vs 82.5 ml/min, p = 0.009) and 6 months (44.2 vs 58.1 ml/min, p= 0.028) with a non-significant difference at 1 year (50.1 vs 62.4 ml/min, p=0.08). The ATG group was more likely to require renal replacement therapy (RRT) during the index admission (34.6 vs 4.4%, p=0.01). This difference persisted at discharge (19.2% vs 2.2%, p= 0.022) and 1 year (15.4% vs 0%, p= 0.015). No significant difference between the two groups in the rate of worsening chronic kidney disease at 1, 6 and 12 months. Freedom from readmission for infection was lower in the ATG group (53.8 % vs 80%, p= 0.03). There were no significant differences in in-hospital (p=1.00), 1 year mortality (p=0.19), and 1 year freedom from rejection (p=0.80) between groups. Conclusions ATG use as a renal sparing strategy in OHT recipients with renal dysfunction at baseline was associated with a high rate of early RRT but overall change in renal function over one year was similar. Freedom from readmission for infection was lower in the ATG group and freedom from rejection was equal in both groups. Thus, it appears in this retrospective analysis, that a strategy of induction therapy for renal sparing was not beneficial and increased the risk of infection.

Acute pulmonary embolism (PE) is a common and highly lethal diagnosis. Surgical management has been marginalized despite recent promising outcomes. Survival and right ventricular (RV) recovery from the largest single center experience of surgically managed acute PE is reported. Surgical