Scalable deep learning to identify brick kilns and aid regulatory capacity.
Ontology highlight
ABSTRACT:
SUBMITTER: Lee J
PROVIDER: S-EPMC8092470 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Many applications in speech, robotics, finance, and biology deal with sequential data, where ordering matters and recurrent structures are common. However, this structure cannot be easily captured by standard kernel functions. To model such structure, we propose expressive closed-form kernel functions for Gaussian processes. The resulting model, GP-LSTM, fully encapsulates the inductive biases of long short-term memory (LSTM) recurrent networks, while retaining the non-parametric probabilistic advantages of Gaussian processes. We learn the properties of the proposed kernels by optimizing the Gaussian process marginal likelihood using a new provably convergent semi-stochastic gradient procedure, and exploit the structure of these kernels for scalable training and prediction. This approach provides a practical representation for Bayesian LSTMs. We demonstrate state-of-the-art performance on several benchmarks, and thoroughly investigate a consequential autonomous driving application, where the predictive uncertainties provided by GP-LSTM are uniquely valuable.
| S-EPMC6334642 | biostudies-other
Project description:BackgroundBrick kiln workers are often not aware of the health effects of their working environment and health-related respiratory problems. There have been few studies on the relationship between brick kiln pollution and its health impact on brick kiln workers.ObjectivesThe present study measured the association of brick kiln contamination with severe respiratory problems and lung function among brick kiln workers in the Kasur district, Pakistan.MethodsAir quality variables (PM2.5, PM10, sulfur dioxide (SO2), nitrogen dioxide (NO2) and volatile organic compounds (VOCs)) were monitored during operations in brick kiln modulation and kiln areas. Workers (n=60) were selected for participation if they were between the ages of 18 and 60 and had been working in brick kilns for at least one year and gave consent to participate. Their lung function was measured.ResultsThe average concentrations of measured air quality variables for all working sites were found to exceed the World Health Organization (WHO) and National Ambient Air Quality Standard (NAAQS) guidelines. These high values of brick kiln pollutants were associated with a significant decrease in spirometric values (forced vital capacity (FVC), forced expiratory volume in one second (FEV1), peak expiratory flow (PEF), and average flow between 25% and 75% of the FVC (FEF2575)) among workers and revealed that 78.33% of workers had abnormal lung function with 5% obstructive and 95% restrictive impairments. Occurrences of pulmonary problems like frequent cough (50%), chronic cough (11.67%), frequent phlegm (21.67%), chronic phlegm (11.67%), frequent wheezing (20%), chronic wheezing (15%), shortness of breath grade-I & grade-II (38.33%) and self-reported asthma (3.33%) were also found among the workers.ConclusionsPollution from brick kiln operations was significantly high and associated with respiratory problems as well as a decrease in lung function. There was a clear correlation between pulmonary function in workers with brick kiln contamination.Participant consentObtained.Ethics approvalThis study was approved by the Bioethics Committee of the Department of Zoology, University of the Punjab, Lahore, Pakistan (Ref.1443-UZ).Competing interestsThe authors declare no competing financial interests.
| S-EPMC8383793 | biostudies-literature
Project description:Non-governmental organizations (NGOs) work across the globe to improve the welfare of working equids. Despite decades of veterinary and other interventions, welfare issues persist with equids working in brick kilns. Engagement with all stakeholders is integral to creating abiding improvements to working equid welfare as interventions based purely on reactive measures fail to provide sustainable solutions. Equid owners, particularly those in low to middle-income countries (LMICs), may have issues such as opportunity, capacity, gender or socio-economic status, overriding their ability to care well for their own equids. These "blind spots" are frequently overlooked when organizations develop intervention programs to improve welfare. This study aims to highlight the lives of the poorest members of Indian society, and will focus on working donkeys specifically as they were the only species of working equids present in the kilns visited. We discuss culture, status, religion, and social influences, including insights into the complexities of cultural "blind spots" which complicate efforts by NGOs to improve working donkey welfare when the influence of different cultural and societal pressures are not recognized or acknowledged. Employing a mixed-methods approach, we used the Equid Assessment Research and Scoping (EARS) tool, a questionnaire based equid welfare assessment tool, to assess the welfare of working donkeys in brick kilns in Northern India. In addition, using livelihoods surveys and semi-structured interviews, we established owner demographics, socioeconomic status, ethnicity, religion and their personal accounts of their working lives and relationships to their donkeys. During transcript analysis six themes emerged: caste, ethnicity, inherited knowledge; social status, and impacts of ethnic group and caste; social status and gender; migration and shared suffering; shared suffering, compassion; religious belief, species hierarchy. The lives led by these, marginalized communities of low status are driven by poverty, exposing them to exploitation, lack of community cohesion, and community conflicts through migratory, transient employment. This vulnerability influences the care and welfare of their working donkeys, laying bare the inextricable link between human and animal welfare. Cultural and social perspectives, though sometimes overlooked, are crucial to programs to improve welfare, where community engagement and participation are integral to their success.
| S-EPMC7201042 | biostudies-literature
Project description:Predictive modeling with electronic health record (EHR) data is anticipated to drive personalized medicine and improve healthcare quality. Constructing predictive statistical models typically requires extraction of curated predictor variables from normalized EHR data, a labor-intensive process that discards the vast majority of information in each patient's record. We propose a representation of patients' entire raw EHR records based on the Fast Healthcare Interoperability Resources (FHIR) format. We demonstrate that deep learning methods using this representation are capable of accurately predicting multiple medical events from multiple centers without site-specific data harmonization. We validated our approach using de-identified EHR data from two US academic medical centers with 216,221 adult patients hospitalized for at least 24?h. In the sequential format we propose, this volume of EHR data unrolled into a total of 46,864,534,945 data points, including clinical notes. Deep learning models achieved high accuracy for tasks such as predicting: in-hospital mortality (area under the receiver operator curve [AUROC] across sites 0.93-0.94), 30-day unplanned readmission (AUROC 0.75-0.76), prolonged length of stay (AUROC 0.85-0.86), and all of a patient's final discharge diagnoses (frequency-weighted AUROC 0.90). These models outperformed traditional, clinically-used predictive models in all cases. We believe that this approach can be used to create accurate and scalable predictions for a variety of clinical scenarios. In a case study of a particular prediction, we demonstrate that neural networks can be used to identify relevant information from the patient's chart.
| S-EPMC6550175 | biostudies-literature
Project description:As deep neural network (DNN) models grow ever-larger, they can achieve higher accuracy and solve more complex problems. This trend has been enabled by an increase in available compute power; however, efforts to continue to scale electronic processors are impeded by the costs of communication, thermal management, power delivery and clocking. To improve scalability, we propose a digital optical neural network (DONN) with intralayer optical interconnects and reconfigurable input values. The path-length-independence of optical energy consumption enables information locality between a transmitter and a large number of arbitrarily arranged receivers, which allows greater flexibility in architecture design to circumvent scaling limitations. In a proof-of-concept experiment, we demonstrate optical multicast in the classification of 500 MNIST images with a 3-layer, fully-connected network. We also analyze the energy consumption of the DONN and find that digital optical data transfer is beneficial over electronics when the spacing of computational units is on the order of [Formula: see text]m.
| S-EPMC7862662 | biostudies-literature
Project description:Household and ambient air pollution remain public health problems in much of the world. Brick kiln employees in Nepal may be particularly at risk of high air pollution exposures and resulting health effects due to high levels of outdoor air pollution, substandard housing, and indoor biomass cooking. We conducted a cross-sectional study of indoor and outdoor air pollution concentrations at workers' homes at four fixed chimney Bull's trench brick kilns in Bhaktapur, Kathmandu Valley, Nepal. We measured air concentrations of carbon monoxide (CO), carbon dioxide (CO2), nitrogen dioxide (NO2), sulfur dioxide (SO2), and respirable crystalline silica (SiO2; cristobalite, quartz, tridymite) using established methods and conducted a survey about characteristics of homes or samples that may be associated with air pollution concentrations. Geometric mean concentrations of CO, CO2, and SiO2 (quartz) were 0.84 ppm, 1447.34 ppm, and 6.22 µg/m3, respectively, whereas concentrations of all other air pollutants measured below lower detection limits. Most characteristics of homes or samples were not associated with air pollution concentrations. We found a positive association between the variable how long lived in house and SiO2 (quartz) concentrations, which may reflect sustained take-home exposure to SiO2 (quartz) over time. Interventions should focus on administrative controls to reduce take-home exposure to SiO2 (quartz) in this population.
| S-EPMC9566599 | biostudies-literature
Project description:The present investigation focuses on assessing the water quality of groundwater surrounding brick kilns in the Jammu district of Jammu and Kashmir (J&K). At 43 different brick kiln sites in both north and south regions of Jammu, concentrations of heavy metals were measured using established techniques. The elements zinc, copper, iron, lead, cadmium, nickel, and manganese were analyzed utilizing an Atomic Absorption Spectrophotometer (AAS). The pollution load index value was consistently below unity across all sites, suggesting an absence of pollution and making the water suitable for consumption. The average concentrations, listed in ascending order, were found to be 0.38 mg/L for copper, 0.31 mg/L for zinc, 0.01 mg/L for iron, and 0.09 mg/L for manganese. Notably, concentrations of lead, cadmium, and nickel were found below the detectable levels. Evaluation of contamination factors revealed the sequence Cu > Fe > Zn > Mn, while the geo accumulation index followed the sequence Cu > Fe > Mn > Zn. Comparison of these findings with the established standards of World Health Organization and Bureau of Indian Standards indicated that the recorded ranges were within permissible limits. The study's outcomes suggest that heavy metal emissions from brick kilns may not significantly impact the quality of groundwater. Elevated copper levels found near brick kilns were likely to result from plumbing materials in the study area. Iron and manganese in groundwater seems to have geo-genic origin and not emission-related. This research represents a foundational step in examining groundwater contamination by heavy metals specifically in the neighborhood of brick kilns in Jammu district. It contributes to the establishment of a comprehensive database and serves as a reference point for future studies. Additionally, the study recommends regular monitoring of groundwater to ensure the maintenance of drinking water quality.
| S-EPMC10963316 | biostudies-literature
Project description:BackgroundIndia holds the world's largest burden of chronic and acute child undernutrition. Poverty and systemic inequities are basic causes of undernutrition that also drive households to engage in circular migration for livelihood. Short-term, temporary movement of the whole family, including young children, is common; yet, the nutritional implications of recurrent movements beginning in early life has not been studied. We sought to estimate the association of repeat and early life migration with stunting and wasting outcomes among circular migrant children under three.MethodsUsing a stratified cluster design, we conducted two waves of primary data collection among 2564 randomly selected circular migrant children under three years of age temporarily residing across 1156 brick kilns in Bihar, India. We conducted multilevel modeling to estimate the association of the number of migration episodes and age at first migration with stunting (<-2 standard deviations (SD) height-for-age z scores (HAZ)) and wasting (<-2 SD weight-for-height z-scores (WHZ)) and examined the determinants of nutrition status among migrant children, including diet, illness, food security and the health environment.ResultsThe overall prevalence of stunting was 51.6%. Among children who were either born during migration or first migrated before six months of age, those who experienced multiple episodes of migration were more likely to be stunted compared to those who migrated once (adjusted odds ratio (aOR) = 2.10; 95% confidence interval (CI) = 1.30-3.41). Children were over three times as likely to be wasted in the summer compared to the winter (aOR = 3.28; 95% CI = 2.68-4.01); in the summer, the overall prevalence of wasting was 38.8%. Public health access indicators such as interaction with frontline health workers at the destination was low (5.3%), whereas feeding indicators such as exclusive breastfeeding among 0-5 months was high (81.1%).ConclusionsPolicy efforts should ensure continuity of social protection and welfare entitlements between home and destinations for circular migrant families, with an explicit focus on rural-to-rural movement.
| S-EPMC8818295 | biostudies-literature
Project description:Slavery, in the form of 'debt-bondage', is rife in Indian brick kilns, where the enforcement of labour laws is poor. Working equids support brick-kiln workers by transporting raw bricks into the kilns, but the situation of equids and their owners within the brick kilns is relatively unknown. We describe the welfare of donkeys (Equus asinus) owned under conditions of debt-bondage, examine the links between owner and donkey behaviour, and outline the living conditions of both donkeys and humans working in the brick kilns of Gujarat, India. We then explore the unique experience of debt-bondage by donkey owners, compare migration trends to those of non-donkey-owning workers and assess impacts on their children's education. The physical and behavioural conditions of donkeys reflected that of their owners, creating negative feedback loops and potentially reducing productivity. All donkey owners experienced debt-bondage and were particularly vulnerable to unexpected financial loss. Donkey owners, unlike non-owners, migrated within their home state, enabling their children to attend school. Our work highlights the need for policy reform within the brick-kiln industry to acknowledge the pivotal role of working donkeys in supporting human livelihoods.
| S-EPMC10936255 | biostudies-literature
Project description:Target enrichment sequencing techniques are gaining widespread use in the field of genomics, prized for their economic efficiency and swift processing times. However, their success depends on the performance of probes and the evenness of sequencing depth among each probe. To accurately predict probe coverage depth, a model called Deqformer is proposed in this study. Deqformer utilizes the oligonucleotides sequence of each probe, drawing inspiration from Watson-Crick base pairing and incorporating two BERT encoders to capture the underlying information from the forward and reverse probe strands, respectively. The encoded data are combined with a feed-forward network to make precise predictions of sequencing depth. The performance of Deqformer is evaluated on four different datasets: SNP panel with 38 200 probes, lncRNA panel with 2000 probes, synthetic panel with 5899 probes and HD-Marker panel for Yesso scallop with 11 000 probes. The SNP and synthetic panels achieve impressive factor 3 of accuracy (F3acc) of 96.24% and 99.66% in 5-fold cross-validation. F3acc rates of over 87.33% and 72.56% are obtained when training on the SNP panel and evaluating performance on the lncRNA and HD-Marker datasets, respectively. Our analysis reveals that Deqformer effectively captures hybridization patterns, making it robust for accurate predictions in various scenarios. Deqformer leads to a novel perspective for probe design pipeline, aiming to enhance efficiency and effectiveness in probe design tasks.
| S-EPMC10835675 | biostudies-literature