University of Melbourne

To evaluate the efficacy and safety of conservative (oxygen saturation [Sp o2 ] 88-94% or Pa o2 < 80 mm Hg) vs. liberal oxygen targets (Sp o2 ≥ 94% or Pa o2 ≥ 90 mm Hg) in mechanically ventilated critically ill adults. PubMed, Cochrane CENTRAL, Embase, and ClinicalTrials.gov . We conducted the OXY-BREATHES, a systematic review and meta-analysis of randomized controlled trials (RCTs) comparing conservative vs. liberal oxygen targets in mechanically ventilated ICU patients. Primary outcomes were 90-day mortality and ICU length of stay. Secondary outcomes included ventilator- and vasopressor-free days, renal replacement therapy, nosocomial pneumonia, and cardiac or cerebral ischemic events. Subgroup analyses included patients with sepsis/septic shock and post-cardiac arrest. Data were collected according to study selection criteria. Certainty of evidence was appraised with Grading of Recommendations, Assessment, Development, and Evaluation, and risk of bias with the Cochrane tool. Data were analyzed using a random-effects model. Nine RCTs enrolling 20,447 patients were included. Conservative and liberal targets showed no substantial differences in 90-day (risk ratio [RR], 1.01; 95% CI, 0.94-1.09) or ICU length of stay (mean difference [MD], -0.17; 95% CI, -0.41 to 0.06). Secondary outcomes, including organ support-free days and the incidence of adverse events, were comparable between groups. In subgroup analyses, conservative targets yielded more vasopressor-free days in septic patients (MD, 2.0; p = 0.008) and a potential survival benefit in post-cardiac arrest patients (RR, 0.89; p = 0.05). Certainty of evidence was rated moderate for 90-day mortality, ICU length of stay, vasopressor-free days, and ventilator-free days; low for renal replacement therapy and nosocomial pneumonia; and very low for cerebral and cardiac ischemia due to imprecision and open-label trial designs. Conservative oxygenation is comparable to liberal oxygen targets in mechanically ventilated critically ill patients, with possible advantages in sepsis and post-cardiac arrest. Future condition-specific RCTs are warranted to define optimal ICU oxygen strategies.

Efficient job scheduling in high-performance computing (HPC) systems necessitates the simultaneous consideration of system-centric objectives, such as maximizing resource utilization, and user-centric objectives, such as minimizing job waiting times. In practice, the relative importance of these objectives is not static, but shifts dynamically in response to fluctuations in workload characteristics and system state. However, existing scheduling frameworks — including both traditional workload managers and reinforcement learning (RL)-based methods — typically rely on fixed policies or fixed reward functions that encode a predetermined combination of objectives. As a result, they lack the flexibility to adjust their scheduling priorities as workload intensity or system conditions change. In this work, we propose MetaPilot, a deep reinforcement learning-based controller designed to enable dynamic adaptation to shifting scheduling objectives. MetaPilot functions as an adaptive decision layer, continuously monitoring real-time system state and workload characteristics in order to select the most appropriate scheduling objective prior to each scheduling event. The selected objective is subsequently optimized by established scheduling frameworks, ensuring seamless integration with existing systems. Comprehensive experiments conducted on real-world HPC workload traces from both CPU-based (Physical) and GPU-enabled (Deeplearn) clusters demonstrate that MetaPilot achieves substantial improvements in both maximum and average job waiting times, as well as in resource utilization across CPU, GPU, and memory dimensions, relative to state-of-the-art baseline schedulers. Notably, MetaPilot reduces maximum waiting time by up to 19% on the Physical partition and achieves best-in-class utilization and balanced waiting times on the Deeplearn partition, thereby demonstrating robust adaptability and effectiveness across heterogeneous HPC workloads.

School-based programs represent a potential avenue for conducting population-wide paediatric blood pressure (BP) screening. The aim of this review was to systematically scope peer-reviewed literature reporting school-based BP screening, with respect to measurement protocols, diagnostic process coverage, and implementation considerations. Only peer-reviewed articles in English across PubMed, OVID Medline and OVID Embase were included. Two authors independently screened the article titles and abstracts prior to undertaking a full-text review. All disagreements were resolved through discussion and agreement. From each study, four categories of information were extracted: general information, BP measurement methodology, diagnostic process coverage, and implementation strategies. Each article was then assigned to one of three categories regarding the stated or implied study objectives: general school-based research incorporating BP measurement, hypertension prevalence studies, or hypertension screening studies. Of the 112 articles meeting the inclusion criteria, only 17 were categorised as hypertension screening studies. Within these, there was substantial variability in BP measurement techniques and adherence to the diagnostic process recommended by the American Academy of Pediatrics. Additionally, there was minimal reporting on implementation strategies. A pragmatic, standardised protocol for school-based BP screening is needed that includes recommended measurement methods, considers the trade-offs (in terms of feasibility and economics) of covering more or less of the diagnostic process in schools vs health care settings, and covers approaches to optimise implementability.

Excessive cell death in the epithelium due to prolonged immune activation is associated with intestinal diseases such as Crohn’s disease. Mice with mutations in the Crohn’s disease susceptibility gene Atg16l1 are susceptible to inflammation associated with intestinal epithelial cell (IEC) death, including loss of antimicrobial Paneth cells triggered by infection with murine norovirus (MNV). Here, we show that intestinal disease downstream of MNV depends on IFN-α/β and IFN-λ signaling in IECs. In mouse organoids, IFNs synergize with TNF to induce RIPK1-dependent cell death, amplified by ATG16L1 deficiency. We further show that human intestinal organoids harboring the ATG16L1 risk allele exhibit heightened sensitivity to TNF and IFN co-stimulation and to serum from severe COVID-19 patients. Our findings reveal that virally triggered cytokines including IFNs exacerbate epithelial vulnerability in genetically predisposed hosts.

People evaluate their singing ability to varying degrees of accuracy, but why these individual differences occur is largely unknown. Objectively evaluated, everyday singing ability is influenced by both genetic and environmental factors. Since singing self-evaluation (SSE) strongly correlates with everyday singing ability, it is possible that the same factors might influence both. Using a classical twin design, we estimated genetic and environmental influences on SSE and its phenotypic relationship with everyday singing ability in Australian twins (N = 996). Two validated measures of SSE were used: a single self-report item assessing overall singing ability (SSE-Single) and a four-item composite measure (SSE-Factor). Both measures showed significant genetic (SSE-Single A = 42%; SSE-Factor A = 53%) and unshared environmental influences (both E = 29%). Phenotypic correlations between the SSE measures and everyday singing ability ranged from rp = 0.67 - 0.68 and were significantly influenced by genetic, shared and unshared environmental factors (SSE-Single A = 41%, C = 51%, E = 8%; SSE-Factor A = 37%, C = 52%, E = 11%). Strong genetic correlations (rg = 0.63 - 0.64) were observed between SSE measures and everyday singing ability, as well as between both SSE measures (rg = 0.82), suggesting shared underlying genes. Taken together, our findings indicate that both genes and environments affect how individuals evaluate their own singing ability and how accurate these evaluations are compared to their actual ability.

Awe is theorized to serve a social function, enabling individuals to integrate into collaborative groups and engage in collective action. Across five studies (N = 1,124), we examined awe's role in promoting identity fusion-an apex form of group connection. Two cross-sectional studies (1a and 1b) revealed that dispositional awe predicted stronger identity fusion. Three subsequent experiments (Studies 2-5) demonstrated that awe experiences strengthened identity fusion, with analyses revealing that the small-self sense of "vastness vis-à-vis the self" provided a significant indirect pathway linking awe (vs. controls) to increased fusion. These effects replicated across varying awe manipulations (emotion recall and virtual reality), target groups (country, university, local community, and nature), and cultural contexts (Australian and American samples). Our findings suggest that awe primes a readiness to fuse with groups, creating an openness to deeper collective bonds. Importantly, rather than diminishing personal agency, awe appears to foster an interdependent alignment where personal and collective goals converge, motivating individuals to direct their capabilities toward shared goals through mutual strengthening between self and group. (PsycInfo Database Record (c) 2026 APA, all rights reserved).

The current study used ecological momentary assessment (EMA) to examine whether body mass index (BMI) and weight suppression (highest minus current weight) predicted momentary body dissatisfaction and disordered eating urges, including dietary restriction, excessive exercise, binge eating, and unhealthy eating, and whether trait eating disorder (ED) symptoms moderated these associations. Data were collected from 686 adults (75% female), comprising community participants and undergraduate students, through six daily EMA surveys over seven days (42 possible assessments). Multilevel models showed that lower BMI (p=0.005) and greater weight suppression (p=0.004) predicted higher average state body dissatisfaction, while higher BMI (p<0.001) and greater weight suppression (p=0.039) predicted stronger urges for unhealthy eating. ED symptomatology moderated the relationship between BMI and dietary restraint, such that BMI positively predicted restraint urges at low levels of ED symptoms but negatively predicted them at high levels. No other moderating effects of ED symptomatology were observed for BMI or weight suppression on the remaining state-based outcomes. Overall, both weight-based severity indicators (BMI and weight suppression) demonstrated limited utility for indexing ED-related state-based variables in a female non-clinical sample. Future studies should examine additional weight-related severity indicators across both non-clinical and clinical ED samples.