17628 research outputs found
Sort by
Advancing the Understanding of Clinical Sepsis Using Gene Expression-Driven Machine Learning to Improve Patient Outcomes
Get PDFSepsis remains a major challenge that necessitates improved approaches to enhance patient outcomes. This study explored the potential of Machine Learning (ML) techniques to bridge the gap between clinical data and gene expression information to better predict and understand sepsis. We discuss the application of ML algorithms, including neural networks, deep learning, and ensemble methods, to address key evidence gaps and overcome the challenges in sepsis research. The lack of a clear definition of sepsis is highlighted as a major hurdle, but ML models offer a workaround by focusing on endpoint prediction. We emphasize the significance of gene transcript information and its use in ML models to provide insights into sepsis pathophysiology and biomarker identification. Temporal analysis and integration of gene expression data further enhance the accuracy and predictive capabilities of ML models for sepsis. Although challenges such as interpretability and bias exist, ML research offers exciting prospects for addressing critical clinical problems, improving sepsis management, and advancing precision medicine approaches. Collaborative efforts between clinicians and data scientists are essential for the successful implementation and translation of ML models into clinical practice. ML has the potential to revolutionize our understanding of sepsis and significantly improve patient outcomes. Further research and collaboration between clinicians and data scientists are needed to fully understand the potential of ML in sepsis management
Library catalogue’s search interface: Making the most of subject metadata
No full textThis article addresses the underutilization of knowledge organization systems (KOS) elements in online library catalogues, hindering effective subject-based search and discovery. It highlights the International Society for Knowledge Organization's initiative to develop metadata guidelines for library catalog procurement, focusing on maximizing the value of subject metadata from classification systems and controlled vocabularies. The paper discusses the rationale for quality subject access, proposes desirable search functionalities based on research, explores implementation challenges, and outlines future developments. The conclusion emphasizes the importance of providing quality subject access in digital services and calls for further research on interface design, guideline adoption, KOS evolution, and the impact of language models on subject metadata use. The work underscores the need for applying controlled vocabularies in search interfaces across libraries, archives, and museums while acknowledging the complementary role of alternative approaches like social tagging and automatic indexing. Extensive future research is suggested to implement search functionalities, promote guidelines adoption, enhance KOS evolution, and assess the influence of language models on subject metadata utilization
CoDeS: A Deep Learning Framework for Identifying COVID-Caused Depression Symptoms
No full textDepression is a serious mental health condition that affects a person’s ability to feel happy and engaged in activities. The COVID-19 pandemic has led to an increase in depression due to factors such as isolation, financial stress, and uncertainty about the future. Additionally, restrictions on travel and socializing have contributed to feelings of loneliness and isolation. In this research, we present a deep learning framework named CoDeS (COVID-caused depression symptoms) for detecting prodromes of depression in online users caused due to COVID pandemic. This framework uses a combination of CNN, LSTM, and integrated CNN-LSTM techniques, with three different feature representation methods, viz. Word2Vec, TF-IDF, and BERT. Nine experiments were conducted on individual and integrated models, and the results were evaluated based on the accuracy, precision, recall, F1-score, and Matthews correlation coefficient (MCC) performance metric. The highest accuracy value of 98.95% was recorded for the TF-IDF-based integrated CNN+LSTM model. When the same integrated model was trained using Word2Vec and BERT-based features, it still performed well with an accuracy of 97.32% and 98.51% respectively. The results demonstrate that the TF-IDF-based feature representation performed better than the Word2Vec and BERT-based feature representations for the CNN and LSTM models in identifying COVID-caused depression symptoms. The proposed approaches showcased substantial advancements over the existing ones, with significant improvements in accuracy. TF-IDF-CNN+LSTM achieved an accuracy approximately 37.28% higher, while BERT-CNN, BERT-LSTM, and BERT-CNN+LSTM achieved accuracy enhancements of approximately 29.78%, 34.44%, and 27.14% respectively. These accuracy improvements demonstrate the superior classification capabilities of the proposed approaches, leading to more precise depression analysis outcomes. In terms of F1 measure, the proposed approaches consistently demonstrated superior performance, with F1 measure values ranging from 0.965 to 0.987. BERT-CNN+LSTM achieved the highest F1 measure, highlighting its balanced precision and recall. Overall, the proposed approaches outperformed existing ones in terms of recall, precision, accuracy, and F1 measure, with improvements ranging from 27.14 to 44.85%. By incorporating advanced techniques such as TF-IDF, CNN, LSTM, and BERT, more accurate and reliable sentiment analysis outcomes can be achieved, offering the potential for enhanced applications in this field
BBox-Free SAR Ship Instance Segmentation Method Based on Gaussian Heatmap
Get PDFRecently, deep learning methods have been widely adopted for ship detection in synthetic aperture radar (SAR) images. However, many of the existing methods miss adjacent ship instances when detecting densely arranged ship targets in inshore scenes. Besides, they suffer from the lack of precision in the instance indication information and the confusion of multiple instances by a single mask head. In this paper, we propose a novel center point prediction algorithm, which detects the center points by finding a long distance variation relationship between two points. The whole prediction process is anchor-free and does not require additional bounding box (BBox) predictions for non-maximum suppression (NMS). Therefore, our algorithm is BBox-free and NMS-free, solving the problem of low recall rates when conducting NMS for densely arranged targets. Furthermore, to tackle the deficiency of position indication information in localization tasks, we introduce a feature fusion module with feature decoupling (FD). This module uses classification branch to provide guidance information for localization branch, while suppressing the influence of the gradient flow mixing, effectively improving the algorithm’s segmentation performance of ship contours. Finally, through principal component analysis (PCA) of the Gaussian distribution covariance matrix, we propose a loss function based on the distance between centroids and the difference of angle, called centroid and angle constraint (CAC). CAC guides the network in learning the criterion that a single dynamic mask head is only valid for a single instance. Experiments conducted on PSeg-SSDD and HRSID demonstrate the effectiveness and robustness of our method
Crop and landscape heterogeneity increase biodiversity in agricultural landscapes: A global review and meta‐analysis
Get PDFAgricultural intensification not only increases food production but also drives widespread biodiversity decline. Increasing landscape heterogeneity has been suggested to increase biodiversity across habitats, while increasing crop heterogeneity may support biodiversity within agroecosystems. These spatial heterogeneity effects can be partitioned into compositional (land-cover type diversity) and configurational heterogeneity (land-cover type arrangement), measured either for the crop mosaic or across the landscape for both crops and semi-natural habitats. However, studies have reported mixed responses of biodiversity to increases in these heterogeneity components across taxa and contexts. Our meta-analysis covering 6397 fields across 122 studies conducted in Asia, Europe, North and South America reveals consistently positive effects of crop and landscape heterogeneity, as well as compositional and configurational heterogeneity for plant, invertebrate, vertebrate, pollinator and predator biodiversity. Vertebrates and plants benefit more from landscape heterogeneity, while invertebrates derive similar benefits from both crop and landscape heterogeneity. Pollinators benefit more from configurational heterogeneity, but predators favour compositional heterogeneity. These positive effects are consistent for invertebrates and vertebrates in both tropical/subtropical and temperate agroecosystems, and in annual and perennial cropping systems, and at small to large spatial scales. Our results suggest that promoting increased landscape heterogeneity by diversifying crops and semi-natural habitats, as suggested in the current UN Decade on Ecosystem Restoration, is key for restoring biodiversity in agricultural landscapes
Childhood adversity and mental health admission patterns prior to young person suicide (CHASE): a case-control 36 year linked hospital data study, Scotland UK 1981–2017
Get PDFBackgroundChildhood adversity (CA) is associated with increased later mental health (MH) problems and suicidal behaviour. Opportunities for earlier healthcare identification and intervention are needed.AimTo determine associations between hospital admissions for CA and MH in children who later die by suicide.MethodPopulation-based longitudinal case-control study. Scottish inpatient general and psychiatric records were summarised for individuals born 1981 or later who died by suicide between 1991-2017 (cases), and matched controls (1:10), for CA and MH (broadly defined as psychiatric diagnoses and general hospital admissions for self-harm and substance use).ResultsRecords were extracted for 2,477 ‘cases’ and 24,777 ‘controls’; 2,106 cases (85%) and 13,589 controls (55%) had lifespan hospitalisations. Mean age at death was 23.7; 75.9% were male. Maltreatment or violence-related CA codes were recorded for 7.6% cases aged 10-17 (160/2,106) versus 2.7% controls (371/13,589), OR=2.9 (95%CI,2.4-3.6); MH-related admissions were recorded for 21.7% cases (458/2,106), versus 4.1% controls (560/13,589), OR=6.5 (95%CI, 5.7-7.4), 80% of MH admissions were in general hospital. Using conditional logistic models we found a dose-response effect of MH admissions <18y, with highest aOR for three or more MH admissions: aORmale=8.17 (95%CI,5.02-13.29), aORfemale=15.08 (95%CI,8.07-28.17). We estimated each type of CA multiplied odds of suicide by aORmale=1.90 (95%CI,1.64-2.21), aORfemale=2.65 (95%CI,1.94-3.62), and each MH admission by aORmale=2.06 (95%CI,1.81-2.34), aORfemale=1.78 (95%CI,1.50-2.10).ConclusionsOur lifespan study found that experiencing CA (primarily maltreatment or violence-related admissions) or MH admissions increased odds of young person suicide, with highest odds for those experiencing both. Healthcare practitioners should identify and flag potential ‘at-risk’ adolescents to prevent future suicidal acts, especially those in general hospitals
Making a Case for the Adoption of Industry 4.0 Technologies for Sustainable Housing Delivery in Saudi Arabia
Get PDFThe Kingdom of Saudi Arabia (KSA) grapples with the challenge of achieving sustainable housing delivery amidst rapid urbanization and rising construction costs. Current housing strategies have failed to offer a lasting solution to the crisis. To address these issues, this study advocates the adoption of fourth industrial revolution (4IR) technologies for sustainable housing. The previous literature highlights the versatility of 4IR technologies, prompting an examination of their suitability and benefits for housing delivery. Thus, this study was aimed at evaluating suitable 4IR technologies for housing delivery and the benefits of adopting the technologies for sustainable housing delivery. The data used were collected via random sampling from stakeholders in the housing sector and analyzed using SPSS V 24, including mean scores, frequencies, and principal component analysis (PCA). The KMO and Bartlett’s test of sphericity confirmed that the data were appropriate for PCA and identified three key components of 4IR technology: Immersive technologies, smart connectivity, and automated construction sites suitable for sustainable housing delivery. These components enhance decision-making, operational efficiency, and project management throughout the housing delivery process. The study emphasizes the potential of 4IR technologies to transform the housing sector in the KSA sustainably, offering insights for both practice and research
Reflective evaluations of perinatal bereavement care provision in the US and UK: an exploratory qualitative comparative study
Get PDFThere is increased recognition of the need to improve standards of perinatal bereavement care, due to its frequency and potential sequelae. As part of a Fulbright Scholar award, United States (US) and United Kingdom (UK) researchers collaborated to explore similarities and differences in perinatal bereavement care between two nations. Using an explorative qualitative comparative method, key categories within perinatal bereavement provision were identified and analysed. Themed findings include: (1) Differences in definitions of miscarriage and stillbirth impact care pathways; (2) For the experiencer grief is the same regardless of legal lines drawn; (3) The meaning of loss is personal and ‘fetal personhood’ needs to be acknowledged during care; (4) Appropriate psychological care is required whether miscarriage or stillbirth is experienced. We conclude that perinatal bereavement care should include screening for Postnatal Depression (PND) and Post-Traumatic-Stress-Disorder (PTSD), and support should be equally available to all women who experience perinatal bereavement, irrespective of type of loss. Acknowledging that cultures react to loss in different ways, we recommend that strategies are developed to build human resilience. For example, Compassionate-Mind-Training (CMT), which helps people cope with trauma through cultivating compassion and teaching self-care strategies to build resilience, reduce self-criticism, and decrease threat-based emotions
The British Rail Total Operations Processing System And the Birth of Telematics
Get PDFContemporary mobility is almost universally supported by telecommunications networks and computing facilities known collectively as telematics. This has allowed closer integration of freight logistics into supply chains, and supported the growth of the on-demand economy. From a passenger perspective it has allowed real-time journey tracking, planning and re-planning in response to disruption. We examine the design and implementation of the British Rail (BR) Total Operations Processing System and make the case that it pioneered the field of Telematics. When introduced in 1971 TOPS was the first computer system that created a digital model of a complex transportation network, updated in real-time, supporting operation and management activities. Although based on an earlier IBM system, BR expanded TOPS and implemented it into a scenario that was significantly more complex than previous usage in the USA. We argue that the fundamental principles that underpin contemporary telematics systems were established through TOPS
Data sources:
Access Repository Dashboard
Do you manage Repository@Napier? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!