Effects of safety pattern, cabin ergonomics, and sleep on work-related stress and burnout of city and transit bus drivers in Lahore, Pakistan.

The health and working environment of bus drivers is compromised in low-middle-income countries like Pakistan which leads to burnout and excessive Road Traffic Crashes. Hence, this study delves into factors affecting their safe operations from health and work environment perspectives and measures their associated stress and Burnout level. In a study of four hundred and ninety-nine (499), 86% city and 14% transit bus drivers are surveyed through a questionnaire. Stress is estimated for city and transit bus drivers, using the Effort/Reward Imbalance Model (ERI) of Siegrist, and burnout is calculated using the Copenhagen Burnout Inventory (CBI). For the determination of important determinants, descriptive and regression analyses are conducted. Findings show that stress has emerged as a negative factor for the physical and psychological health of city and transit bus drivers. Results based on bus drivers' responses suggest that organisational awareness and emphasis on health and safety levels can significantly reduce driver stress and burnout. Practitioner Summary: This study explores burnout and work-related stress of bus drivers in Lahore (Pakistan). City and transit bus drivers were interviewed through a questionnaire, containing three sections, using different subjective ratings based upon their past reliability. Results indicate that stress in bus drivers emerged as a physical and psychological health-damaging factor.

Uninterrupted versus interrupted direct oral anticoagulation for catheter ablation of atrial fibrillation: A systematic review and meta-analysis.

INTRODUCTION: To evaluate the safety of uninterrupted versus interrupted direct oral anticoagulation (DOAC) for patients undergoing catheter ablation (CA) of atrial fibrillation (AF). METHODS: We conducted a systematic search of MEDLINE and EMBASE for randomized controlled trials (RCT) and observational studies comparing uninterrupted versus interrupted DOAC for patients undergoing CA of AF. Primary outcome was major bleeding. Secondary outcomes included minor bleeding, stroke or transient ischemic attack (TIA) or thromboembolism (TE), silent cerebral ischemic events, and cardiac tamponade. Meta-analysis was stratified by study design. Risk ratios (RR) with 95% confidence intervals were calculated using random effects model and Mantel-Haenszel method was used to pool RR. RESULTS: A total of 13 studies (7 randomized, 6 observational) comprising 3595 patients were included. The RCT restricted analysis did not show any difference in terms of major bleeding (risk ratio [RR] = 0.79; [0.35-1.79]), minor bleeding (RR = 0.99 [0.68-1.43]), stroke or TIA or TE (RR = 0.80 [0.19-3.32]), silent cerebral ischemic events (RR = 0.64 [0.32-1.28]), and cardiac tamponade (RR = 0.61 [0.20-1.92]). Observational study restricted analysis showed a protective effect of uninterrupted DOAC on silent cerebral ischemic events (RR = 0.45 [0.31-0.67]) and no difference in other outcomes. CONCLUSIONS: There is no difference in bleeding and thromboembolic outcomes with uninterrupted versus interrupted DOAC for CA of AF and observational data suggests that uninterrupted DOACs are protective against silent cerebral ischemic lesions.

Survivorship Struggles: Navigating Etiologies and Clinical Parameters of Febrile Neutropenia During Induction Chemotherapy in Acute Leukemia Patients.

Background Acute leukemia, characterized by the uncontrolled proliferation of immature white blood cell precursors, poses significant challenges during induction chemotherapy, including the elevated risk of febrile neutropenia and its associated complications. Our study aims to explain the clinical and etiological parameters of these patients in a resource-limited setting. Methods This retrospective study focused on a total of 102 adult patients with acute leukemia who developed febrile neutropenia during the induction chemotherapy phase. Patients with disease relapse, prior bone marrow transplantation, and cases of acute promyelocytic leukemia were excluded from the study. Demographical characteristics, symptoms at presentation, diagnoses, infectious causes, and outcomes were systematically reported. Infectious etiologies and detailed culture reports were meticulously tabulated, and subsequent data were analyzed. Results Of the 102 patients, 43 (42.2%) were males, with a mean age of 31.9 ± 6.5 years. During the induction chemotherapy, a total of 31 patients died of complicated febrile neutropenia. Severe vomiting was the most common symptom present in 37 (36.2%), followed by cough in 35 (34.3%) and loose stools in 28 (27.5%). Community-acquired pneumonia, neutropenic sepsis, and neutropenic colitis were among the most common etiologies of febrile neutropenia. A total of 72 (70.6%) patients had culture-proven multidrug-resistant Gram-negative bacteremia that contributed to poor outcomes. Conclusions Acute leukemia patients undergoing induction chemotherapy face high infection-associated mortality due to their immunocompromised state. Inadequate infection control measures and antimicrobial resistance contribute to the emergence of multidrug-resistant organisms. Enhanced infection prevention strategies and evidence-based antibiotic prescription guidelines are need of time in resource-limited settings such as Pakistan to address febrile neutropenia complications and bridge the existing care gap in its management.

Domain Adaptive Object Detection via Balancing Between Self-Training and Adversarial Learning.

Deep learning based object detectors struggle generalizing to a new target domain bearing significant variations in object and background. Most current methods align domains by using image or instance-level adversarial feature alignment. This often suffers due to unwanted background and lacks class-specific alignment. A straightforward approach to promote class-level alignment is to use high confidence predictions on unlabeled domain as pseudo-labels. These predictions are often noisy since model is poorly calibrated under domain shift. In this paper, we propose to leverage model's predictive uncertainty to strike the right balance between adversarial feature alignment and class-level alignment. We develop a technique to quantify predictive uncertainty on class assignments and bounding-box predictions. Model predictions with low uncertainty are used to generate pseudo-labels for self-training, whereas the ones with higher uncertainty are used to generate tiles for adversarial feature alignment. This synergy between tiling around uncertain object regions and generating pseudo-labels from highly certain object regions allows capturing both image and instance-level context during the model adaptation. We report thorough ablation study to reveal the impact of different components in our approach. Results on five diverse and challenging adaptation scenarios show that our approach outperforms existing state-of-the-art methods with noticeable margins.

Visual Object Tracking With Discriminative Filters and Siamese Networks: A Survey and Outlook.

Accurate and robust visual object tracking is one of the most challenging and fundamental computer vision problems. It entails estimating the trajectory of the target in an image sequence, given only its initial location, and segmentation, or its rough approximation in the form of a bounding box. Discriminative Correlation Filters (DCFs) and deep Siamese Networks (SNs) have emerged as dominating tracking paradigms, which have led to significant progress. Following the rapid evolution of visual object tracking in the last decade, this survey presents a systematic and thorough review of more than 90 DCFs and Siamese trackers, based on results in nine tracking benchmarks. First, we present the background theory of both the DCF and Siamese tracking core formulations. Then, we distinguish and comprehensively review the shared as well as specific open research challenges in both these tracking paradigms. Furthermore, we thoroughly analyze the performance of DCF and Siamese trackers on nine benchmarks, covering different experimental aspects of visual tracking: datasets, evaluation metrics, performance, and speed comparisons. We finish the survey by presenting recommendations and suggestions for distinguished open challenges based on our analysis.

Transformers in medical imaging: A survey.

Following unprecedented success on the natural language tasks, Transformers have been successfully applied to several computer vision problems, achieving state-of-the-art results and prompting researchers to reconsider the supremacy of convolutional neural networks (CNNs) as de facto operators. Capitalizing on these advances in computer vision, the medical imaging field has also witnessed growing interest for Transformers that can capture global context compared to CNNs with local receptive fields. Inspired from this transition, in this survey, we attempt to provide a comprehensive review of the applications of Transformers in medical imaging covering various aspects, ranging from recently proposed architectural designs to unsolved issues. Specifically, we survey the use of Transformers in medical image segmentation, detection, classification, restoration, synthesis, registration, clinical report generation, and other tasks. In particular, for each of these applications, we develop taxonomy, identify application-specific challenges as well as provide insights to solve them, and highlight recent trends. Further, we provide a critical discussion of the field's current state as a whole, including the identification of key challenges, open problems, and outlining promising future directions. We hope this survey will ignite further interest in the community and provide researchers with an up-to-date reference regarding applications of Transformer models in medical imaging. Finally, to cope with the rapid development in this field, we intend to regularly update the relevant latest papers and their open-source implementations at https://github.com/fahadshamshad/awesome-transformers-in-medical-imaging.

Lifecycle cost analysis of an insulated duct with an air gap.

The insulation materials are used to reduce heat loss to/from the ducts with additional investment. This study introduces an air gap between insulation and duct surface to reduce the quantity of insulation. It uses lifecycle cost (LCC) analysis to determine the economic benefits of the air gap, considering four insulation materials for insulating the duct and natural gas as an energy source for chiller operation. The preliminary data regarding design and operating parameters were obtained from a renowned pharmaceutical company. The duct's annual energy loss was estimated for given operation hours in a year using the preliminary data and ambient conditions. The estimated energy loss through the duct was fed in LCC analysis to determine the impact of the air gap on optimum insulation thickness (OIT) corresponding to the minimum LCC and payback period. Results revealed that OIT thickness for a duct with an air gap was lower than insulated duct without an air gap, resulting in maximum cost savings within a shorter payback period. Among different insulation materials, insulated duct with expanded polystyrene was investigated as cost-effective insulation material with maximum cost savings of USD (508.8-766.8)/m/year and a payback period of 1.15-1.17 years. On the contrary, the air gap was the most effective in terms of cost savings for the ducts insulated with rock wool. In conclusion, an air gap is a cost-effective design approach for duct applications.

Safety and feasibility of same-day discharge for catheter ablation of atrial fibrillation: a systematic review and meta-analysis.

PURPOSE: Most centers performing catheter ablation (CA) of atrial fibrillation (AF) admit the patients for an overnight hospital stay to monitor for post-procedure complications, but the clinical benefits of this overnight hospital admission policy have not been carefully investigated. We hypothesized that same-day discharge strategy is safe and feasible in patients with AF undergoing CA. METHODS: A systematic review of studies comparing the safety of same-day discharge vs hospital admission for AF patients undergoing CA was conducted in PubMed/MEDLINE, Embase, Scopus, and Web of Science. No randomized controlled trials met the inclusion criteria; therefore, observational cohort studies were included. Mantel-Haenszel risk ratios were calculated and I2 statistics were reported for heterogeneity assessment. RESULTS: A total of 8 observational studies with 10,102 patients were included. There were no statistically significant differences between same-day discharge vs hospital admission in all studied outcomes including post-discharge 30-day hospital visits (RR: 0.90; 95% CI: 0.40-2.02; p = 0.81), post-discharge vascular/bleeding complications (RR: 0.93; 95% CI: 0.46-1.88; p = 0.85), post-discharge stroke/transient ischemic attack/thromboembolism (RR: 0.70; 95% CI: 0.23-2.20; p = 0.55), and post-discharge recurrent arrhythmias (RR: 0.81; 95% CI: 0.60-1.09; p = 0.1). CONCLUSION: In carefully selected AF patients undergoing CA, same-day discharge strategy is feasible and safe. There are no significant differences in post-discharge 30-day hospital visits, post-discharge vascular complications, and other safety outcomes. Randomized trials are needed to validate these hypothesis-generating findings.

Mask-Guided Attention Network and Occlusion-Sensitive Hard Example Mining for Occluded Pedestrian Detection.

Pedestrian detection relying on deep convolution neural networks has made significant progress. Though promising results have been achieved on standard pedestrians, the performance on heavily occluded pedestrians remains far from satisfactory. The main culprits are intra-class occlusions involving other pedestrians and inter-class occlusions caused by other objects, such as cars and bicycles. These result in a multitude of occlusion patterns. We propose an approach for occluded pedestrian detection with the following contributions. First, we introduce a novel mask-guided attention network that fits naturally into popular pedestrian detection pipelines. Our attention network emphasizes on visible pedestrian regions while suppressing the occluded ones by modulating full body features. Second, we propose the occlusion-sensitive hard example mining method and occlusion-sensitive loss that mines hard samples according to the occlusion level and assigns higher weights to the detection errors occurring at highly occluded pedestrians. Third, we empirically demonstrate that weak box-based segmentation annotations provide reasonable approximation to their dense pixel-wise counterparts. Experiments are performed on CityPersons, Caltech and ETH datasets. Our approach sets a new state-of-the-art on all three datasets. Our approach obtains an absolute gain of 10.3% in log-average miss rate, compared with the best reported results on the heavily occluded HO pedestrian set of the CityPersons test set. Code and models are available at: https://github.com/Leotju/MGAN.