324139 research outputs found
Sort by
Clinical and cost-effectiveness of flexor digitorum profundus (FDP) versus FDP and flexor digitorum superficialis (FDS) repair for complete zone 2 flexor tendon injuries (FLARE): protocol for a randomised controlled trial
Get PDFBackground Flexor tendon injuries are common and lead to over 3200 admissions for specialist surgical repair annually in England and Wales. Surgery to repair complete division of both flexor tendons in zone 2 of the hand is technically challenging. There is variation in surgical repair techniques with no high-quality evidence to support decision-making. In particular, the decision to repair both tendons or just one is contested. Surgery is followed by specialist rehabilitation, which takes at least 12 weeks. The resulting hand function can impact the patient’s income, life satisfaction, well-being, self-worth, and mental health. The FLARE trial aims to determine the clinical and cost-effectiveness of repairing the flexor digitorum profundus (FDP) alone (intervention) versus the repair of both FDP and flexor digitorum superficialis (FDS) (control) for the treatment of complete zone 2, single-digit flexor tendon injuries in adults. Methods A multi-centre, two-arm, blinded, non-inferiority, parallel group, randomised controlled trial with an internal pilot, economic evaluation, and nested qualitative study. Participants will be randomised 1:1 to receive either repair of FDP alone or repair of both FDP and FDS. A total of 310 adults will be recruited from NHS Trusts within the UK, randomised at surgery, and followed up within 7 days, 6 weeks, 3 months, and 6 months post-randomisation. The primary outcome measure is the patient evaluation measure (PEM) administered 6 months post-randomisation. Secondary outcomes include the PEM at other timepoints, Patient Related Wrist/Hand Evaluation (PRWHE), EuroQol 5 Dimensions Score (EQ-5D-5L), complications, total range of motion, grip strength, adherence to splint and therapy regimens, work outcomes, treatment and outcome satisfaction, and healthcare resource use. Discussion FLARE is designed with sufficient power and rigour to provide evidence on the clinical and cost-effectiveness of two surgical repair methods for single-digit, complete zone 2 flexor tendon injuries in adults. If the repair of FDP alone is as beneficial to the patient as the repair of FDP and FDS, this could save the NHS £1.8 million annually through reduced time and material costs. Furthermore, the trial findings will facilitate better shared decision-making discussions between clinicians and patients. Trial registration ISRCTN 10918157. Prospectively registered: 12.01.202
Disingenuous ‘box‐ticking’: Undergraduate students' attitudes towards university mental health awareness efforts
Get PDFMental health problems are common among UK undergraduate students. In response, many universities have put considerable effort into raising awareness about student mental health problems and avenues of support (e.g., via workshops, posters, email newsletters and social media posts). Nonetheless, reported rates of mental health problems in students have continued to rise. Despite the ubiquity of awareness efforts, there has been limited research assessing students' attitudes towards and experience of these initiatives. To address this, in this study, N = 15 undergraduate students (aged 18 to 24) from 13 UK universities were interviewed in depth to explore their attitudes towards the mental health awareness efforts of their respective universities. N = 11 reported personal experience of mental health problems. Using reflexive thematic analysis, three themes were generated: (1) university life contradicts university mental health awareness efforts; (2) university mental health awareness efforts are perceived as disingenuous and inadequate; and (3) students don't want awareness—they want accessible help and supportive communities. These themes highlight the frustration students feel towards what they see as misguided efforts from their universities, and the structural problems at university that make living a mentally healthy life difficult. Simultaneously, the findings highlight the elements of mental health support that students value. The findings of this study have important implications for the design and implementation of universities' efforts to improve and support student mental health, while also adding constructively to the wider societal conversation critiquing the impact of mental health awareness efforts
The meaning of ca 'and': coordination and compounding in the Indian grammatical tradition
No full textThe tradition of linguistic analysis in ancient India, vyākaraṇa ‘grammar’, developed a unique and complex approach to the topic of coordination and the question of the meaning of ‘and’. This approach, moreover, was not static but evolved over the roughly 2,000-year history of the tradition. Although the basics of the ancient Indian theory of cārtha ‘the meaning of ca ‘and’’ are well-known, the finer details of the grammarians’ approach, and in particular the differences between the major grammarians, and the evolution of the theory over time, remain to be clearly described. In this paper we present a detailed account of the tradition’s approach to coordination, considering not only the mainstream Pāṇinian tradition, but also the competing traditions of the Cāndra and Kātantra schools
Overcoming barriers that limit the impact of ecological research
Get PDFEcology and conservation researchers have diverse goals that often include both personal career aspirations and desires to enhance the well‐being of the natural world and its inhabitants. Perception of ecological research by ecologists typically involves a triad—linking goals, research, and impact. Yet the realities of scientific practice are substantially more complicated due to numerous constraints that limit the ability of researchers to conduct ecological research and to have a genuine impact. Many of these barriers can be mitigated, leading to more effective contributions to society and biodiversity conservation. Here, we outline frequently encountered constraints in ecological research institutions and, by drawing upon many practices used internationally, we identify feasible mitigations and highlight examples of negative consequences that can occur in the absence of effective mitigation strategies. Finally, we propose changes to aspects of the culture and reward systems that would allow ecological research as a discipline to more effectively achieve societal, environmental, and personal goals
Select2Plan: training-free ICL-based planning through VQA and memory retrieval
Get PDFWe introduce Select2Plan (S2P), a novel training-free framework for high-level robot planning that leverages off-the-shelf VLMs for autonomous navigation. Unlike most learning-based approaches that require extensive task-specific training and large-scale data collection, S2P overcomes the need for fine-tuning by adapting inputs to align with the VLM's pretraining data. Our method achieves this through a combination of structured Visual Question Answering (VQA) to ground action selection on the image, and In-Context Learning (ICL) to exploit knowledge drawn from relevant examples from a memory bank of (visually) annotated data, which can include diverse, in-the-wild sources. We demonstrate S2P flexibility by evaluating it in both First-Person View (FPV) and Third-Person View (TPV) navigation. S2P improves the performance of a baseline VLM by 40% in TPV and surpasses end-to-end trained models by approximately 24% in FPV when tasked with navigating towards unseen objects in novel scenes. These results highlight the adaptability, simplicity, and effectiveness of our training-free approach, demonstrating that the use of pre-trained VLMs with structured memory retrieval enables robust high-level robot planning without costly task-specific training. Our experiments also show that retrieving samples from heterogeneous data sources, including online videos of different robots or humans walking, is highly beneficial for navigation. Notably, our method effectively generalizes to novel scenarios, requiring only a handful of demonstrations. Project Page: lambdavi.github.io/select2pla
Systems and methods for improving large language models with increased inference compute
Get PDFScaling the amount of compute used to train language models has led to massive increases in their capability. A relatively under-explored direction has been scaling the amount of compute used at inference time. This thesis explores the design space for test-time compute algorithms, focusing on the simple method of repeated sampling. Notably, this thesis demonstrates that coverage, the percent of problems solved by any sample, often scales log-linearly with the number of samples over four orders of magnitude. In domains where automatic verification tools are available, this allows Llama-8b to outperform the single-attempt performance of GPT4o across all tasks studied. These increases in coverage can often be modeled with an exponentiated power law, which enables forecasting coverage up to 10k samples (the largest sample size attempted) using an order of magnitude fewer samples with an average error of 3.84% across all models and datasets. In domains where tools for verification are available, these gains in coverage directly translate into increased capability. In other domains, methods for selecting a correct sample from a large sample collection are required. This thesis investigates this selection problem in the salient application of software engineering. Using a combination of unit-test based voting and model-based selection, half of the accuracy between the single-attempt lower bound and coverage upper bound can be recovered. Finally, this thesis explores the systems implications of test-time compute scaling and presents an exact implementation of attention that accelerates inference throughput in these settings by over 10x
Deterministic laser-writing of nitrogen-vacancy centres in high purity diamond
Get PDFWe report the deterministic writing of single nitrogen-vacancy defects in commercially available high purity diamond using only laser processing. The positioning accuracy of the single defects is around 250 nm and about 20% of the written defects display near transform-limited optical transitions at a temperature of 5 K. The results represent a significant step towards defect-engineered devices using colour centres as coherent spin-photon interfaces for quantum technologies, and shed light on the physical mechanism for laser-induced vacancy diffusion in diamond
PAH Marks the Spot: Digging for Buried Clusters in Nearby Star-forming Galaxies
Get PDFThe joint capabilities of the Hubble Space Telescope (HST) and JWST allow for an unparalleled look at the early lives of star clusters at near- and mid-infrared wavelengths. We present here a multiband analysis of embedded young stellar clusters in 11 nearby, star-forming galaxies, using the PHANGS-JWST and PHANGS-HST data sets. We use the Zooniverse citizen science platform to conduct an initial by-eye search for embedded clusters in near-UV/optical/near-infrared images that trace stellar continuum emission, the Paschenα and Hα recombination lines, and the 3.3 μm polycyclic aromatic hydrocarbon feature and its underlying continuum. With this approach, we identify 292 embedded cluster candidates for which we characterize their ages, masses, and levels of line-of-sight extinction by comparing the photometric data to predictions from stellar population models. The embedded cluster candidates have a median age of 4.5 Myr and an average line-of-sight extinction 〈AV〉 = 6.0 mag. We determine lower limits on source stellar masses, resulting in a median stellar mass of 103 M⊙. We use this sample of embedded cluster candidates to train multiple convolutional neural network models to carry out deep transfer learning-based searches for embedded clusters. With the aim of optimizing models for future catalog production, we compare results for four variations of training data using two neural networks. Confusion matrices for all eight model configurations, as well as inter-model identification trends, are presented. With refinement of the training sample, we determine that optimized models could serve as a pathway for future embedded cluster identification beyond our 11 galaxy sample
Impact of real-time glucose monitoring using FreeStyle Libre 3 on glycaemia in type 2 diabetes managed with basal insulin plus SGLT2 inhibitor and/or GLP-1 agonist: the FreeDM2 randomised controlled trial protocol
Get PDFIntroduction: Effective management of type 2 diabetes mellitus (T2DM) consists of lifestyle modification and therapy optimisation. While glycaemic monitoring can be used as a tool to guide these changes, this can be challenging with self-monitoring of blood glucose (SMBG). The FreeStyle Libre 3 (FSL3) is a real-time continuous glucose monitoring (CGM) system designed to replace SMBG. The evidence for the benefit of CGM in people with T2DM on non-intensive insulin regimens is limited. This study aims primarily to assess the glycaemic impact of FSL3 in people with suboptimally controlled T2DM treated with basal-only insulin regimens plus sodium-glucose cotransporter-2 (SGLT-2) inhibitor and/or glucagon-like peptide (GLP)-1 agonist. Methods and analysis: This is an open-label, multicentre, parallel design, randomised (2:1) controlled trial. Recruitment has been offered across 24 clinical centres in the UK and nationally through self-referral. Adults with T2DM treated with basal-only insulin regimens plus SGLT-2 inhibitor and/or GLP-1 agonist and with screening HbA1c from ≥59 mmol/mol to ≤97 mmol/mol are included. Eligible participants will be randomised to either FSL3 (intervention) for 32 weeks or continuation of SMBG (control). The study is split into two phases, each of 16 weeks duration: phase 1 consisting of self-management with basal-insulin self-titration and phase 2 where additional therapies may be initiated. Control group participants may subsequently enter an optional extension phase to receive FSL3. The primary endpoint is the difference between treatment groups in mean change from baseline in HbA1c at 16 weeks. Secondary outcomes include HbA1c at 32 weeks, CGM-based metrics, therapy changes, physical activity levels and psychosocial measures. An economic evaluation for costs and patient outcomes will be undertaken. Ethics and dissemination: The study was approved by the Health Research Authority, Health and Care Research Wales and the West Midlands-Edgbaston Research Ethics Committee (reference: 23/WM/0092). Study results will be disseminated in peer-reviewed journals. Trial registration number: NCT05944432. Secondary identifying number: Identifier assigned by the sponsor: ADC-UK-PMS-22057. Protocol version: Revision D. Dated, 13 December 2024
A surgical approach to building impactful artificial intelligence
Get PDFTwo recent studies in surgical AI video analysis show divergent outcomes based on user expertise and AI explainability. While both demonstrate improved accuracy with AI-assistance, differences in trust and transparency critically shaped impact. This commentary explores human-computer interaction factors such as trust, usability, and explainability, and argues for structured evaluation frameworks to ensure effective, safe integration of AI into surgical practice