Major Depressive Disorder (MDD) is a severe illness that affects millions of people, and it is critical to diagnose this disorder as early as possible. Detecting depression from voice signals can be of great help to physicians and can be done without any invasive procedure. Since relevant labelled data are scarce, we propose a modified Instance Discriminative Learning (IDL) method, an unsupervised pre-training technique, to extract augment-invariant and instance-spread-out embeddings. In terms of learning augment-invariant embeddings, various data augmentation methods for speech are investigated, and time-masking yields the best performance. To learn instance-spreadout embeddings, we explore methods for sampling instances for a training batch (distinct speaker-based and random sampling). It is found that the distinct speaker-based sampling provides better performance than the random one, and we hypothesize that this result is because relevant speaker information is preserved in the embedding. Additionally, we propose a novel sampling strategy, Pseudo Instance-based Sampling (PIS), based on clustering algorithms, to enhance spread-out characteristics of the embeddings. Experiments are conducted with DepAudioNet on DAIC-WOZ (English) and CONVERGE (Mandarin) datasets, and statistically significant improvements, with -value 0.0015 and 0.05, respectively, are observed using PIS in the detection of MDD relative to the baseline without pre-training.
Background: Angiopoietin-like protein 2 (Angptl2) is a cytokine that is released to stimulate inflammation and accelerate atherogenesis. Our study sought to assess the predictive significance of serum Angptl2 in individuals diagnosed with acute coronary syndrome (ACS) and determine whether it can enhance prognostic performance beyond the GRACE risk score.
Methods: We recruited a total of 1060 patients with ACS in a consecutive manner.
Purpose: To assess and compare, through a retrospective cohort study, the relationships between frailty, comorbidity, multimorbidity, and levels of adherence to lipid-lowering drugs (LLDs), antihypertensives and antidepressants.
Methods: In a primary care database, we selected a cohort of patients aged 60 or older on December 31, 2022. The date of the first prescription of the aforementioned medications was the study index date.
Background: Immature granulocytes can be measured easily in a complete blood count by new automated hemolytic analyzers and have recently been studied as bio-markers in many infectious/inflammatory diseases. This study aims to investigate whether immature granulocyte percentage (IG%) would enable greater discrimination than conventionally utilized laboratory values in terms of early clinical prediction in instances with respiratory syncytial virus (RSV) bronchiolitis.
Methods: A prospective observational cohort study involved 149 individuals with RSV bronchiolitis.
There are few psychometrically valid measures of exposure to social maltreatment that simultaneously assess sexism, racism, and anti-LGBTQ+ (lesbian, gay, bisexual, transgender, queer, and other nonheteronormative) behavior, despite the commonness of these phenomena. The (SDMS) meets this requirement but is, as a result, somewhat lengthy (36 items). This article introduces a short form of the SDMS containing only half the number of items but generally retaining the psychometric qualities of the original measure.
Background: There has been an increasing interest in the development and prevention of sports injuries from a complex dynamic systems perspective. From this perspective, injuries may occur following critical fluctuations in the psychophysiological state of an athlete. Our objective was to quantify these so-called Early Warning Signals (EWS) as a proof of concept to determine their explanatory performance for injuries.
