Ben-Gurion University of the Negev

Proprotein convertase subtilisin/kexin type 9 (PCSK9) monoclonal antibodies (mAbs) lower LDL cholesterol and may influence cancer through immunomodulatory pathways. However, their effect on human cancer incidence remains unknown. We conducted a retrospective, propensity score-matched study (Clalit Health Services, Israel, 2010-2023) comparing PCSK9 mAbs to ezetimibe. Adults prescribed PCSK9 mAbs for 6 months or more were matched 1:3 to ezetimibe-treated patients without prior cancer, applying a 1-year latency. The cohort included 9,876 patients (2,469 PCSK9 mAb; 7,407 ezetimibe; mean age 65). During a median 4.6-year follow-up, cancer occurred in 12% of PCSK9 mAb users and 11% of ezetimibe users (HR 1.09 [95% CI, 0.95-1.25]). In sex-stratified analysis, men on PCSK9 mAbs had a higher cancer incidence (12.5% vs. 10.3%, P = 0.03); no difference was observed in women. All-cause mortality was significantly lower in the PCSK9 mAb group (3% vs. 5%; HR 0.65 [95% CI, 0.54-0.80]). Post-cancer-diagnosis mortality did not differ. In this large cohort, PCSK9 mAb therapy appeared safe regarding overall cancer risk and was associated with a significant reduction in all-cause mortality; the slightly higher cancer incidence in men may likely be attributed to a higher prevalence of baseline risk factors.

BackgroundFlight attendants (FAs) are subject to unique occupational demands, increasing their risk of developing work-related musculoskeletal pain (WRMSP). WRMSP is influenced by physical, psychosocial, and demographic factors that may change over time due to evolving work environments and practices.ObjectivesCompare the prevalence of FAs' WRMSP in 2016 vs 2024; Identify key physical and psychosocial predictors of WRMSP in each period; explore implications for ergonomic interventions and health promotion strategies tailored to the aviation work environment.MethodsTwo cross-sectional cohorts of FAs, from 2016 (n = 94) and 2024 (n = 89), completed identical, validated questionnaires using the same sampling frame. Surveys covered demographics, occupational tasks, region-specific pain, and psychosocial factors.ResultsArm and hand pain were less prevalent in 2024. Logistic regressions showed higher odds of neck pain in 2024 associated with stress, attitudes, and satisfaction, but markedly lower odds of arm pain across psychosocial and task factors. Lower back pain was less predicted in 2024 when satisfaction was considered. Across both years, hand pain increased with workload and oven-loading but decreased with age and authority; finger pain rose with effort; upper back pain with BMI and time pressure; shoulder pain with dissatisfaction; elbow pain with standing.ConclusionsTechnological and organizational changes appear to have contributed to reduced arm and hand pain and to lower back pain being less predicted in 2024, while psychosocial stressors were linked to increased neck pain, underscoring the dual impact of ergonomic improvements and persistent organizational pressures on WRMSP.

To describe the clinical and laboratory findings, disease course, and treatment effectiveness of a large, genetically homogenous group of patients with deficiency of adenosine deaminase 2 (DADA2), and to identify factors associated with disease-related damage. We conducted a retrospective cohort study that included 45 patients with DADA2 vasculitis with long-term follow-up. Information was collected from patients' charts, including demographic information, clinical presentations, laboratory findings, and treatment. The extent of disease-related damage was determined based on the Vasculitis Damage Index. In this large homogenous cohort, we found that a young age at disease onset and periodic illness that included recurrent episodes of fever, myalgia, and abdominal pain were associated with disease-related damage. Patients with severe DADA2 disease responded favorably to tumor necrosis factor inhibitors (TNFi). Early age of onset and the periodic nature of the disease warrant prompt initiation of TNFi treatment to prevent later complications and severe disease progression.

Large Language Models (LLMs) demonstrate increasing capabilities in medical knowledge assessment, yet limitations remain in cross-population validation, direct human-AI comparisons, and evaluation of newer models in anesthesiology contexts. This study addresses these gaps by conducting a head-to-head comparison between newer LLMs and human examinees on official Israeli multiple-choice board examinations. We evaluated two LLMs (Claude 3.7 Sonnet and ChatGPT-4) against anonymized aggregate data from 381 examinees on three consecutive official Israeli anesthesiology board examinations (2023–2024), comprising 450 multiple-choice questions stratified by difficulty, discrimination ability, and topic. Each model was tested twice per exam. Claude 3.7 Sonnet achieved 73.67% accuracy, significantly outperforming both human examinees (62.77%, P < 0.001) and ChatGPT-4 (64.44%, P < 0.001). However, both LLMs performed below the upper quartile of human performance (78.05%). While LLMs excelled on easy questions and theoretical domains like cardiac physiology (Claude: 96.88%, ChatGPT-4: 81.25%), they showed lower performance in areas such as ambulatory (Claude: 30.00%, ChatGPT-4: 10.00%) and regional anesthesia (Claude: 44.44%, ChatGPT-4: 38.89%). Human examinees demonstrated consistent performance across all domains, whereas LLMs showed extreme variability. Self-consistency was substantial for both LLMs (κ = 0.66–0.68), but agreement with human responses was moderate (κ = 0.34–0.39). While advanced LLMs currently exceed average examinee performance on anesthesiology board examinations, they fall short of top-quartile examinees at present and demonstrate significant performance variability across different topic areas.

Autonomous systems are expanding rapidly, driving a demand for robust perception technologies capable of navigating challenging, unstructured environments. While urban autonomy has made significant progress, off-road environments pose unique challenges, including dynamic terrain and limited communication infrastructure. This research addresses these challenges by introducing a novel three-part solution for off-road autonomous vehicles. First, we present a large-scale off-road dataset curated to capture the visual complexity and variability of unstructured environments, providing a realistic training ground that supports improved model generalization. Second, we propose a Confusion-Aware Loss (CAL) that dynamically penalizes systematic misclassifications based on class-level confusion statistics. When combined with cross-entropy, CAL improves segmentation mean Intersection over Union (mIoU) on the off-road test set from 68.66% to 70.06% and achieves cross-domain gains of up to ~0.49% mIoU on the Cityscapes dataset. Third, leveraging semantic segmentation as an intermediate representation, we introduce a spatial overlay video encoding scheme that preserves high-fidelity RGB information in semantically critical regions while compressing non-essential background regions. Experimental results demonstrate Peak Signal-to-Noise Ratio (PSNR) improvements of up to +5 dB and Video Multi-Method Assessment Fusion (VMAF) gains of up to +40 points under lossy compression, enabling efficient and reliable off-road autonomous operation. This integrated approach provides a robust framework for real-time remote operation in bandwidth-constrained environments.

The possibility of realizing Higgs inflation in a model with a small non-minimal coupling constant, which was demonstrated recently, provides grounds for further development of the model. Incorporating the electroweak SM into the Two-Measure theory (TMT) in a way that fully accounts for the TMT structure leads to a theory we call the Two-Measure Standard Model (TMSM). The TMSM is realized in the context of cosmology as a set of cosmologically modified copies of the Glashow–Weinberg–Salam (GWS) theory, such that each of the copies exists as a local quantum field theory defined on the classical cosmological background at the appropriate stage of its evolution. This basic idea is studied in detail for two stages of the cosmological background evolution: for slow-roll inflation and for the stage of approaching the vacuum. Mainly due to the presence of the ratio of two volume measures in all equations of motion, all TMSM coupling constants turn into a kind of “running” (classical) TMT-effective parameters. During the evolution of the cosmological background, changing these parameters yields new results: (1) the classical “running” TMT-effective Higgs self-coupling parameter increases from λ∼10−11 (which provides Higgs inflation consistent with the Planck CMB data at ξ=16) to λ∼0.1 at the stage close to the vacuum; (2) the mass term in the TMT-effective Higgs potential changes sign from positive to negative, which provides SSB in the standard way of GWS theory; (3) the classical “running” parameters of the gauge and Yukawa couplings change by several orders of magnitude; (4) the GWS theory is reproduced when the Yukawa constant in the original action is chosen to be universal for three generations of fermions. We show that, due to these classical-level results, taking into account quantum corrections in the one-loop approximation preserves the slow-roll inflation regime and does not violate the vacuum stability during inflation.