West Coast University

ART adherence, essential to preventing disease progression for people living with HIV (PLHIV), remains challenged by factors like poor mental health condition, insufficient social support, and substance use. This study investigates the roles of social support and substance use as mediators in the relationship between mental health and antiretroviral therapy (ART) adherence among PLHIV in a multisite, international context. Cross-sectional data from N = 1598 PLHIV across eight countries were analyzed using parallel mediation analyses in R. Results indicated that both social support and substance use independently mediated the relationship between mental health and ART adherence. Hallucinogens mediated the mental health-adherence link. Cocaine, Amphetamine-type stimulants, inhalants, and opioids directly influenced adherence but did not mediate this relationship. In contrast, no significant effects were observed for tobacco, alcohol, cannabis or sedatives. This large-scale, nurse-led, international study offers critical insights into the roles of social support and substance use as psychosocial influencers in the cascade of care for PLHIV. Clinical implications include prioritizing educative, preventative, and harm-reductive approaches to specific substances that impact ART adherence. Findings may also inform interventions that emphasize strengthening social support to improve adherence in ART and care while addressing substance use as a barrier to care and well-being.

IntroductionRunning is a popular recreational activity worldwide, with women’s participation growing rapidly over the past decade. Compared with men, women runners are more likely to sustain a running-related injury, such as bone stress and knee injuries. Following a serious knee injury and subsequent surgery, women also experience worse knee and health-related outcomes than men. However, little is known about the intersection of female-specific health, physical, psychological and social-gendered factors with knee health in women runners with and without a history of knee surgery.Methods and analysisBuilding on the established ‘TRAjectory of knee heaLth in runners’ (TRAIL) prospective cohort study and designed with patient and content-expert partners, the nested TRAIL-W study will comprehensively explore the associations of multiple factors (ie, female-specific health, physical, psychological and social gendered) with knee health (symptoms and structural features), device-measured running load and running-related pain in women runners. Where appropriate, we will explore sex and/or gender differences. Alongside their scheduled TRAIL 6-monthly data collection, all active TRAIL female and male participants will be invited to complete a once-off ‘TRAIL-W survey’ and attend an additional laboratory-based assessment. The survey will include questionnaires measuring psychological, social gendered and, for women only, female-specific health factors. The laboratory assessment will measure body composition, bone mineral density and blood biomarkers. A subset of women will be invited to participate in qualitative interviews to understand women runners’ experiences of female-specific health factors and their association with running behaviour.Ethics and disseminationFindings from TRAIL-W, approved by the La Trobe University Human Ethics Committee, will address critical research gaps by describing and exploring the diverse factors that may influence women runners’ knee health.

With the high rate of Internet of Things (IoT) technologies, a new era of smart cities has been ushered in where the cities have been made more connected, efficient and innovative. However, this development poses great problems in ensuring security and stability of IoT networks. The proposed paper suggests a scalable machine-learning architecture that will enhance the security of IoT networks in smarter cities, and in particular, cyber-resilience in the modern AI age. The proposed architecture merges machine-learned models with digital governance approaches to provide real-time anomaly detection capabilities, intrusion prevention, and privacy-preserving solutions; and outlines the ways AI models can be used in IoT security and discusses the essential concerns, e.g., data privacy, compliance with regulations, and resource utilization. The suggested solution is tested with the help of simulations that prove its efficiency in resisting the process of cyber-attacks and in providing the means of scalability and flexibility in dynamic smart-city settings, which, in its turn, preconditions the establishment of the conceptual background of the formation of cyber-resilient smart cities in which the IoT networks will be able to operate safely and effectively even under the influence of the surfacing cybersecurity threats.

Gesture-based control systems are gaining popularity for their intuitive interfaces. This project demonstrates a hardware system using hand postures to control media players. It employs an Arduino Uno microcontroller and infrared sensors to detect gestures like play/pause and volume adjustment. The setup includes five IR sensors placed strategically to recognize hand positions, connected to the Arduino Uno board. Software components consist of Arduino code for sensor data processing and a Python script for interpreting signals and triggering actions using the pyautogui module. Through systematic testing, the system accurately identifies gestures, translating them into commands for improved media playback control. This innovation advances gesture-based interactions, with potential applications in smart homes, IoT, and beyond.

Artificial Intelligence (AI) has significantly reshaped the landscape of language education, particularly in Teaching English to Speakers of Other Languages (TESOL). Traditional assessment strategies often fail to capture learner-specific progress, provide timely feedback, or adapt to diverse linguistic backgrounds. This paper proposes an AI-driven adaptive assessment model that leverages Natural Language Processing (NLP), machine learning, and deep learning to provide real-time, personalized feedback to learners. The approach focuses on adaptive diagnostic assessments, formative evaluation, and continuous feedback loops to enhance language fluency and learner autonomy. Through a system architecture comprising speech recognition, grammar error detection, and semantic analysis, the model dynamically adjusts assessment difficulty and provides scaffolded guidance tailored to each learner's proficiency. Experimental evaluation with ESL learners demonstrated improved engagement, increased accuracy in language use, and higher satisfaction compared to conventional assessment methods. The results suggest that AI-driven adaptive feedback can foster learner motivation, improve performance, and align with modern personalized learning paradigms. This research contributes to the integration of AI technologies in TESOL by bridging the gap between automated assessment and individualized learning needs, offering scalable solutions for global ESL classrooms.

In this paper, we provide necessary and sufficient conditions for the existence and multiplicity of periodic smooth solutions to a hybrid system of differential equations that describes animal running at constant angular velocity during stance. We show that solutions to this system exist if and only if the angular velocity during stance is lower than the angular velocity of the corresponding linear harmonic oscillator and satisfies a non-resonance condition. The proof of our main theorem relies on a combination of tools from oscillation theory for ODEs, intersection theory for real plane conics and the analysis of Poincaré maps. Furthermore, we demonstrate the validity of our result with numerical simulations for combinations of parameters corresponding to periodic, resonant, and unbounded solutions. Finally, we show that our model fits accurately publicly available data on human running on different conditions, including comfortable and fast running on a treadmill and overground.

Endocrine therapies for breast cancer (BC) (i.e., selective estrogen receptor modulators (SERMs) or aromatase inhibitors (AI)) lower the risk for cancer recurrence but are linked to an increased risk of type 2 diabetes (T2D). This study examined associations between cardiorespiratory fitness (CRF), physical activity (PA), and sedentary behavior with T2D risk markers, and whether the magnitude of these associations varied between SERM and AI. This study was a cross-sectional study. Participants were BC survivors receiving either SERM or AI for ≥1 yr. A graded exercise test determined CRF (peak oxygen consumption). Participants wore an accelerometer for 14 d to assess time in sedentary behavior, light PA, moderate to vigorous PA, and number of sit-to-stand transitions. T2D risk was measured by an oral glucose tolerance test to determine fasting glucose, glucose area under the curve, insulin sensitivity (Matsuda Index), and insulin resistance (homeostatic model assessment of insulin resistance). Regression models estimated associations between CRF/activity behaviors and markers of T2D by endocrine therapy type adjusting for age, fat mass (measured using dual-energy x-ray absorptiometry scan, time receiving therapy, and amount of moderate or vigorous PA. Participants (SERM ( n = 19) or AI ( n = 20)) were M = 54 ± 12 yr old and had received therapy for M = 3.2 ± 2.8 yr. Sit-to-stand transitions were associated with lower glucose tolerance (-221.52; 95% confidence interval (CI), -442.44 to -0.59; P = 0.049), higher insulin sensitivity (0.45; 95% CI, 0.25 to 0.66; P < 0.001), and lower insulin resistance (-0.06; 95% CI, -0.13 to 0; P = 0.047) but only for those on SERMs. Breaking up sedentary time may be a promising intervention target to lowering T2D risk among BC survivors treated with SERMs. Further studies are needed to better understand how SERMS and AI are differentially influencing glucoregulatory pathways.

Abstract Date Presented 04/04/2025 Falls are leading cause of injury-related death among older adults, with more than 14 million incidents annually. This mixed-methods research aims to understand the impact of interprofessional education (IPE) on fall prevention for OT, PT, and pharmacy students. Primary Author and Speaker: Manisha Sheth

This study sought to understand the experience of women veterans transitioning from military to civilian life. A community-based research design and Ricoeur's hermeneutic phenomenology guided the philosophical framework. Seventeen women veterans were recruited from a non-profit organisation. Data was collected in six virtual focus groups using Zoom in the fall of 2022 using dialogical engagement. Analysis of the data was done using hermeneutic phenomenology. Five participants in the research validated the data and recommended changes to their peer-to-peer support group. The veterans described their environment as service to confusion and from pride to embodied shame. The phenomenology of this experience was screaming in silence, harsh reality, creating a village, and taking the high road. Lists of deployment and military sexual trauma triggers were created. There are many obstacles for women transitioning from military to civilian life. Some veterans suffer more than others. For those who are challenged by living in civilian life, this study strongly emphasised that veterans are often their own best source of support. Peer-to-peer support groups for women veterans were highly encouraged as a vital part of their well-being. The data obtained in this study provided us with rich information for nursing practice, including an understanding of the way of being of women veterans, suicide, military sexual trauma, and homelessness assessment, significant community resources, the significance of having peer-to-peer support groups, and the policies and research needed. Consolidated criteria for reporting qualitative research. This study did not include patient or public involvement in its design, conduct, or reporting.

We propose a dynamic fractional volatility model that incorporates a time-varying Hurst exponent estimated via Daubechies-4 wavelet analysis on 252-day rolling windows to capture evolving market memory effects in equity markets. This approach overcomes the limitations of traditional GARCH-type and static fractional volatility models, which assume a constant memory parameter and struggle during regime shifts and market stress. The model is applied to daily closing prices of the S&amp;amp;P 500 Index over 1,258 trading days from January 1, 2015 to December 31, 2019, yielding statistically significant improvements in forecasting performance. Empirical results indicate a 12.3 % reduction in RMSE and a 9.8 % improvement in MAPE, with an out-of-sample R-squared exceeding 0.72 compared to benchmark models. Maximum likelihood estimation with Fisher scoring is used for daily parameter updates, ensuring the model remains responsive to rapidly changing market conditions. Additionally, the model achieves an average absolute option pricing error of 1.8 %, markedly lower than that of traditional specifications. These enhancements are further corroborated by pairwise Diebold–Mariano tests, which confirm the statistical significance of the forecast improvements. Overall, this framework offers a rigorous and computationally efficient method for real-time volatility forecasting that delivers substantial benefits for risk management, derivative pricing, and automated trading strategies, grounded in robust statistical methodology.