Search

Introduction: Many strategies have been implemented to address the shortage of medical practitioners in rural areas. One such strategy, the Rural Clinical School Program supporting 18 rural clinical schools (RCSs), represents a substantial financial investment by the Australian Government. This is the first collaborative RCS study summarising the rural work outcomes of multiple RCSs. The aim of this study was to combine data from all RCSs' 2011 graduating classes to determine the association between rural location of practice in 2017 and (i) extended rural clinical placement during medical school (at least 12 months training in a rural area) and (ii) having a rural background. Methods: All medical schools funded under the RCS Program were contacted by email about participation in this study. Deidentified data were supplied for domestic students about their gender, origin (rural background defined as having lived in an Australian Standard Geographic Classification-Remoteness Area (ASGC-RA) 2–5 area for at least 5 years since beginning primary school) and participation in extended rural clinical placement (attended an RCS for at least 1 year of their clinical training). The postcode of their practice location according to the publicly available Australian Health Practitioner Regulation Agency (AHPRA) register was collected (February to August 2017) and classified into rural and metropolitan areas using the ASGC 2006 and the more recent Modified Monash Model (MMM). The main outcome measure was whether graduates were working in a ‘rural’ area (ASGC categories RA2–5 or MMM categories 3–7) or ‘metropolitan’ area. Pearson’s χ test was used to detect differences in gender, rural background and extended placement at an RCS between rural and metropolitan practice locations. Binary logistic regression was used to determine odds of rural practice and 95% confidence intervals (CIs) were calculated. Results: Although data were received from 14 universities, two universities had not started collecting origin data at this point so were excluded from the analysis. The proportion of students with a rural background had a range of 12.3–76.6% and the proportion who had participated in extended RCS placement had a range of 13.7–74.6%. Almost 17% (16.6%) had a principal practice postcode in a rural area (according to ASGC), range 5.8–55.6%, and 8.3% had a principal practice postcode in rural areas (according to MMM 3–7), range 4.5–29.9%. After controlling for rural background, it was found that students who attended an RCS were 1.5 times more likely to be in rural practice (95%CI 1.2–2.1, p=0.004) using ASGC criteria. Using the MMM 3–7 criteria, students who participated in extended RCS placement were 2.6 times as likely to be practising in a rural location (95%CI 1.8–3.8, p<0.001) after controlling for rural background. Regardless of geographic classification system (ASGC, MMM) used for location of practice and of student background (metropolitan or rural), those students with an extended RCS had an increased chance of working rurally. Conclusion: Based on the combined data from three-quarters (12/16) of the Australian medical schools who had a graduating class in 2011, this suggests that the RCS initiative as a whole is having a significant positive effect on the regional medical workforce at 5 years post-graduation. [ABSTRACT FROM AUTHOR]

Ovarian cancer is a significant health issue with lasting impacts on the community. Despite recent advances in surgical, chemotherapeutic and radiotherapeutic interventions, they have had only marginal impacts due to an inability to identify biomarkers at an early stage. Biomarker discovery is challenging, yet essential for improving drug discovery and clinical care. Machine learning (ML) techniques are invaluable for recognising complex patterns in biomarkers compared to conventional methods, yet they can lack physical insights into diagnosis. eXplainable Artificial Intelligence (XAI) is capable of providing deeper insights into the decision-making of complex ML algorithms increasing their applicability. We aim to introduce best practice for combining ML and XAI techniques for biomarker validation tasks. We focused on classification tasks and a game theoretic approach based on Shapley values to build and evaluate models and visualise results. We described the workflow and apply the pipeline in a case study using the CDAS PLCO Ovarian Biomarkers dataset to demonstrate the potential for accuracy and utility. The case study results demonstrate the efficacy of the ML pipeline, its consistency, and advantages compared to conventional statistical approaches. The resulting guidelines provide a general framework for practical application of XAI in medical research that can inform clinicians and validate and explain cancer biomarkers. [Display omitted] • Machine learning and Shapley analysis are suitable for biomarker discovery tasks. • Pipeline achieves consistent result with statistical analysis on feature contribution. • Shapley analysis provides opportunities for accountability via model explainability. • Various challenges in this area persist and need further research. [ABSTRACT FROM AUTHOR]

The youth (15-24 years old) in South Africa remain at high risk of HIV infection despite varied efforts to control the disease. An understanding of the perspectives of relevant stakeholders of HIV-prevention interventions targeting the youth is important to guide research, policy and practice aimed at improving these interventions. This study explores youth and intervention implementers' perceptions of a resilience-based HIV-prevention intervention (You Only Live Once) aimed at reducing risky sexual behaviours among the youth in South Africa. Semi-structured interviews were conducted with 10 young people who participated in the intervention, and four intervention implementers at a not-for-profit organisation in Maluti-a-Phofung Local Municipality, South Africa. Data were analysed using thematic analysis. Three main themes emerged from the data: (1) Acceptability and impact of the intervention; (2) Factors influencing intervention implementation; and (3) Recommendations to improve intervention implementation. These findings provide insights into the acceptability, impact, barriers and facilitators of resiliencebased HIV-prevention interventions for the youth in South Africa and similar contexts, and how implementation of these interventions could be enhanced. The findings can help researchers, policy makers and health care practitioners in the field of HIV prevention to improve interventions targeting young people. [ABSTRACT FROM AUTHOR]

Before the advent of ML in the 1950s, scientists needed to know the form of a model before attempting to solve it, whereas ML can devise the form of a model and solve it simultaneously. B Accelerating battery research b : This special collection is devoted to the field of Artificial Intelligence, including Machine Learning, applied to electrochemical energy storage systems. Since their invention, AI and ML have triggered fascination specially since demonstrated spectacular performances of AI/ML empowered computer programs. [Extracted from the article]

Context Longitudinal integrated clerkships ( LICs) represent a model of the structural redesign of clinical education that is growing in the USA, Canada, Australia and South Africa. By contrast with time-limited traditional block rotations, medical students in LICs provide comprehensive care of patients and populations in continuing learning relationships over time and across disciplines and venues. The evidence base for LICs reveals transformational professional and workforce outcomes derived from a number of small institution-specific studies. Objectives This study is the first from an international collaborative formed to study the processes and outcomes of LICs across multiple institutions in different countries. It aims to establish a baseline reference typology to inform further research in this field. Methods Data on all LIC and LIC-like programmes known to the members of the international Consortium of Longitudinal Integrated Clerkships were collected using a survey tool developed through a Delphi process and subsequently analysed. Data were collected from 54 programmes, 44 medical schools, seven countries and over 15 000 student-years of LIC-like curricula. Results Wide variation in programme length, student numbers, health care settings and principal supervision was found. Three distinct typological programme clusters were identified and named according to programme length and discipline coverage: Comprehensive LICs; Blended LICs, and LIC-like Amalgamative Clerkships. Two major approaches emerged in terms of the sizes of communities and types of clinical supervision. These referred to programmes based in smaller communities with mainly family physicians or general practitioners as clinical supervisors, and those in more urban settings in which subspecialists were more prevalent. Conclusions Three distinct LIC clusters are classified. These provide a foundational reference point for future studies on the processes and outcomes of LICs. The study also exemplifies a collaborative approach to medical education research that focuses on typology rather than on individual programme or context. [ABSTRACT FROM AUTHOR]

Background and objectives Rural clinical schools (RCSs) help address Australia's rural workforce shortfall, but they require an investment by rural clinicians and communities. Our objective was to determine the location of RCS graduates as one measure of the effectiveness of RCSs. Method This cross-sectional study obtained work location data for Australian National University Medical School (ANUMS) graduates and analysed both RCS and non-RCS data. Results The percentage of graduates working in rural areas after their fifth postgraduate year (PGY6-11: 34.7%) was significantly greater than that of graduates in PGY1-5 (15.2%, P <0.001). Discussion Many graduates who trained in rural sites spend time in cities before returning to work in rural areas. This is encouraging for rural clinicians and communities, but it can take time for graduates to return. [ABSTRACT FROM AUTHOR]

The economic and social future of nanotechnology depends on our ability and manufacture nanomaterials that avoid potential toxicity, by identifying them before they are made, used and released into the environment. Safety-by-design is a framework for including these issues at an early stage of the development process, but balancing multiple nanoparticle properties and selection criteria remains challenging. Based on a synthetic data set of over 19,000 possible sunscreen product specifications, we have used multi-target machine learning to predict the corresponding size, shape, concentration and polytype of titania nanoparticle additives. The study considers the optical properties responsible for the sun protection factor and product transparency, including the extinction coefficients for ultra violet and visible light, and the potential for toxicity due to the generation of reactive oxygen species from the photocatalytically active facets of both anatase and rutile nanoparticles, as a function of the size and shape. We predict a number of conventional forward structure/property and property/product relationships, but show that a direct structure/product relationship provides superior performance when predicting multiple properties or product specifications simultaneously. These models are then inverted, re-optimized and re-trained to provide focused, high performing inverse design models that do not require additional optimization, and are capable of identifying nanoparticle configurations outside of the training set. The ability to directly predict suitable nanoparticle structures that conform to prerequisite sun protection, transparently and potential toxicity thresholds represents a new approach to safety-by-design that can be applied to other products and materials where multiple design criteria must be met at the same time. [Display omitted] • Using multi-target random forest regression, the the performance of sunscreens can be directly predicted based on the type of TiO 2 nanoparticle additives. • Using inverse design, the type of TiO 2 nanoparticle additives is selected based on the desired SPF, aesthetic appearance and potential toxicity. • Difficult property/product relationships can be avoided by predicting structure/product relationships directly. [ABSTRACT FROM AUTHOR]