Edinburgh Napier University

About: Edinburgh Napier University is a education organization based out in Edinburgh, United Kingdom. It is known for research contribution in the topics: Population & Context (language use). The organization has 2665 authors who have published 6859 publications receiving 175272 citations.

Finding coordinated atrial activity during ventricular fibrillation using wavelet decomposition

Abstract: Until recently, the ECG recorded during ventricular fibrillation was thought to represent disorganized and unstructured electrical activity of the heart. Using a new signal analysis technique based on wavelet decomposition, we have begun to reveal previously unreported structure within the ECG tracing. We report preliminary findings that provide the first evidence linking this structure to unexpected mechanical phenomena occurring in the heart.

Differentiating symptom profiles of ICD-11 PTSD, complex PTSD, and borderline personality disorder: A latent class analysis in a multiply traumatized sample.

Abstract: Complex posttraumatic stress disorder (CPTSD) has been included as a diagnostic category in the International Classification of Diseases, 11th Edition, consisting of six symptom clusters: the three PTSD criteria of reexperiencing, avoidance, and hypervigilance, in addition to three disturbances of self-organization (DSO) symptoms defined as emotional dysregulation, interpersonal difficulties, and negative self-concept. As borderline personality disorder (BPD) shares similar features to DSO presentations and is commonly associated with PTSD, there is debate as to whether and how CPTSD is distinct from PTSD comorbid with BPD. This article aimed to identify groups with distinct profiles of self-reported CPTSD and BPD symptoms and associated trauma history characteristics. A latent class analysis (LCA) using CPTSD and BPD symptom variables was conducted on a sample of 195 treatment-seeking adults at a specialist trauma service. The classes were then compared on demographic and clinical characteristics using a series of analysis of variance and χ2 tests. The latent class analysis determined three distinct classes: a CPTSD/High BPD class characterized by high symptom endorsement across both conditions, a CPTSD/Moderate BPD class characterized by high PTSD and DSO symptom endorsement and moderate BPD, and a PTSD/Low BPD class characterized by PTSD symptoms and low DSO and BPD symptom endorsement. The two CPTSD classes were associated with greater exposure to multiple, interpersonal traumas earlier in life and exhibited higher functional impairment. Findings support the construct of a CPTSD diagnosis as a separate entity although BPD features seem to overlap greatly with CPTSD symptoms in this highly traumatized clinical sample. (PsycINFO Database Record (c) 2019 APA, all rights reserved).

Measuring pain in dogs and cats using structured behavioural observation.

Abstract: The contemporary approach to pain measurement in people and animals seeks to measure the affective (emotional) component of the pain experience using structured questionnaires with formal scoring methodology. Chronic pain has wide-ranging impacts which affects the quality of life (QOL) of the individual, whether that is a person or an animal. Accordingly instruments to measure chronic pain are designed to measure its impact on QOL and are called health-related quality of life (HRQL) instruments. In veterinary science instruments to measure pain are based on behavioural observation by the veterinary surgeon/nurse in the case of acute pain and by the owner in the case of chronic pain. The development of HRQL instruments is an expanding field in veterinary science, not just for the measurement of pain, but for other chronic diseases, and it has a wide application in pharmaceutical research and clinical practice to improve patient care. This review highlights the challenges involved in creating such measures for dogs and cats, seeking to provide the reader with an understanding of their development process. It then provides an overview of the current status with regard to acute and chronic pain measurement.

Flux to acetate and lactate excretions in industrial fermentations: physiological and biochemical implications

Abstract: The efficiency of carbon conversion to biomass and desirable end products in industrial fermentations is diminished by the diversion of carbon to acetate and lactate excretions. In this study, the use of prototrophic and mutant strains of Escherichia coli, as well as enzyme active site directed inhibitors, revealed that flux to acetate excretion is physiologically advantageous to the organism as it facilitates a faster growth rate (mu) and permits growth to high cell densities. Moreover, the abolition of flux to acetate excretion was balanced by the excretion of lactate as well as 2-oxoglutarate, isocitrate and citrate, suggesting a 'bottle-neck' effect at the level of 2-oxoglutarate in the Krebs cycle. It is proposed that the acetate excreting enzymes, phosphotransacetylase and acetate kinase, constitute an anaplerotic loop or by-pass, the primary function of which is to replenish the Krebs cycle with reduced CoA, thus relieving the bottle-neck effect at the level of 2-oxoglutarate dehydrogenase. Furthermore, flux to lactate excretion plays a central role in regenerating proton gradient and maintaining the redox balance within the cell. The long-held view that flux to acetate and lactate excretions is merely a function of an 'over-flow' in central metabolism should, therefore, be re-evaluated.

Cyclical Annealing Schedule: A Simple Approach to Mitigating KL Vanishing

Abstract: Variational autoencoders (VAEs) with an auto-regressive decoder have been applied for many natural language processing (NLP) tasks. The VAE objective consists of two terms, (i) reconstruction and (ii) KL regularization, balanced by a weighting hyper-parameter \beta. One notorious training difficulty is that the KL term tends to vanish. In this paper we study scheduling schemes for \beta, and show that KL vanishing is caused by the lack of good latent codes in training the decoder at the beginning of optimization. To remedy this, we propose a cyclical annealing schedule, which repeats the process of increasing \beta multiple times. This new procedure allows the progressive learning of more meaningful latent codes, by leveraging the informative representations of previous cycles as warm re-starts. The effectiveness of cyclical annealing is validated on a broad range of NLP tasks, including language modeling, dialog response generation and unsupervised language pre-training.