Search

This study aimed to examine the accuracy with which multiple natural language processing artificial intelligence models could predict discharge and readmissions after general surgery. Natural language processing models were derived and validated to predict discharge within the next 48 hours and 7 days and readmission within 30 days (based on daily ward round notes and discharge summaries, respectively) for general surgery inpatients at 2 South Australian hospitals. Natural language processing models included logistic regression, artificial neural networks, and Bidirectional Encoder Representations from Transformers. For discharge prediction analyses, 14,690 admissions were included. For readmission prediction analyses, 12,457 patients were included. For prediction of discharge within 48 hours, derivation and validation data set area under the receiver operator characteristic curves were, respectively: 0.86 and 0.86 for Bidirectional Encoder Representations from Transformers, 0.82 and 0.81 for logistic regression, and 0.82 and 0.81 for artificial neural networks. For prediction of discharge within 7 days, derivation and validation data set area under the receiver operator characteristic curves were, respectively: 0.82 and 0.81 for Bidirectional Encoder Representations from Transformers, 0.75 and 0.72 for logistic regression, and 0.68 and 0.67 for artificial neural networks. For readmission prediction within 30 days, derivation and validation data set area under the receiver operator characteristic curves were, respectively: 0.55 and 0.59 for Bidirectional Encoder Representations from Transformers and 0.77 and 0.62 for logistic regression. Modern natural language processing models, particularly Bidirectional Encoder Representations from Transformers, can effectively and accurately identify general surgery patients who will be discharged in the next 48 hours. However, these approaches are less capable of identifying general surgery patients who will be discharged within the next 7 days or who will experience readmission within 30 days of discharge. [ABSTRACT FROM AUTHOR]

Despite a long history of Improved Cookstove (ICS) interventions by Non-Governmental Organizations, International Development partners and the Government of Nepal, the majority of rural Nepalese people cook on a traditional open fire for their large-scale cooking needs due to a significant lack of approved institutional-scale cooking solutions. Whilst 65.8% of rural Nepalese households cook with biomass as their primary fuel source to satisfy their personal energy needs, there is no information collected on institutional cooking use by the Government of Nepal. In this paper our main objective was to design, implement and evaluate a novel Institutional Improved Cookstove (IICS) to satisfy this gap and following its manufacturing and testing in a Government of Nepal approved test center, to identify the complex contextual factors that often override the technical capabilities of IICS. Our three-phase method combined qualitative and quantitative research approaches, as well as north-south collaborations involving a transdisciplinary research team to create an integrated systems approach taking into account the voices of all key energy stakeholders. Phase 1 included UK based co-design and testing at the University of Nottingham in 2017 to develop a novel IICS that could be used in rural Nepal. Phase 2 involved adapting the design to accommodate contextual factors highlighted by Nepalese partners and to meet testing requirements at a Government of Nepal approved testing center in late 2017. Phase 3 was conducted between December 2017 and April 2020 and focused on piloting the novel IICS in a range of locations, altitudes, socio-economic and cultural settings, monitoring sustained use and obtaining user feedback. We present our results through three case studies that highlight the highly contextualized nature of IICS adoption and sustained use, the importance of stacking, usability and cost savings, and a number of pathways to scale in an institutional setting. • Focus on institutional scale improved cookstoves. • Co-creation of cooking technology with UK and Nepali Engineers. • Present a novel Institutional scale Natural Draft Top-Lit Up-Draft (TLUD) Gasifier for use in off-grid rural settings. • 36 month long Small-Scale Study identifying complex contextual factors that act as barriers to adoption and sustained use. [ABSTRACT FROM AUTHOR]

Purpose: Prolonged time to diagnosis of primary intracerebral haemorrhage (ICH) can result in delays in obtaining appropriate blood pressure control, reversal of coagulopathy or surgical intervention in select cases. We sought to characterise the time to diagnosis in a cohort of patients with ICH and identify factors associated with delayed diagnosis.Methodology: The stroke database of our hospital was retrospectively reviewed to identify patients presenting to our hospitals emergency department with ICH over two years (January 2017-December 2018.) Data collected included demographics (age and sex), comorbidities, anticoagulation status, clinical scores (NIHSS, GCS, ICH score), and imaging (anatomical site, haematoma size). Time from symptom onset to diagnosis and hospital presentation were recorded. Factors associated with diagnosis >8 h post ictus were assessed using a univariate and then multivariable analysis.Results: 235 patients were identified with 125 males (53%) and a median age of 76 (range 40-98). For the 200 patients that initially presented to our hospital, median time to presentation was 179 min (IQR 77-584 min), and median time from ictus to imaging diagnosis was 268 min (IQR 114-717 min). 139 (70%) presented within 8 h of symptom onset, and 129 (65%) patients had imaging of the brain performed within 8 h of symptom onset. Factors associated with presentation >8 h post symptom onset included wake up stroke (OR 5.31, 95% confidence interval (CI) 2.36-11.96, p < 0.0001) and age (OR 1.04, 95% CI 1.01-1.08, p = 0.01). Patients with hemiplegia were less likely to present >8 h following ictus (OR 0.41, 95% CI 0.21-0.84, p = 0.01).Conclusions: The majority of patients with ICH presented within 8 h of ictus. Cases of delayed diagnosis involved patients who had not incurred hemiplegia. [ABSTRACT FROM AUTHOR]