A Data Sharing and Integration Technology for Heterogeneous Databases

For a long time, there are a large number of heterogeneous databases on the network, and their heterogeneity is manifested in many aspects. With the development of enterprise informatization and e-government, the system database of each department constitutes a real heterogeneous database framework with its independence and autonomy in the network system of many different functional departments. This paper will design information sharing between heterogeneous databases of network database system of many similar functional departments by using XML data model. The solution of data sharing between heterogeneous databases can accelerate the integration of information systems with departments and businesses as the core among enterprises, form a broader and more efficient organic whole, improve the speed of business processing, broaden business coverage, and strengthen cooperation and exchange among enterprises. In addition, heterogeneous database sharing can avoid the waste of data resources caused by the heterogeneity of database, and promote the availability rate of data resources. Due to the advantages of XML data model, the system has good scalability.

Research Trend Analysis of Artificial Intelligence Rainfall Prediction Algorithms Based on Knowledge Networks

This research uses CiteSpace software as a tool to sort out the research results of artificial intelligence-based rainfall prediction models and algorithms in the China Knowledge Network Database (CNKI) and the “Web of Science” database, summarize relevant research hotspots and topics, and identify the latest research Trends, provide a reference for further advancement of rainfall prediction models and algorithms. Through knowledge network analysis, the following conclusions are drawn: (1) The literature based on rainfall prediction models and algorithms has shown an increasing trend over time. (2) It is scientific research institutions and colleges and universities of various countries that publish a large number of relevant documents. (3) The current research trend is deep learning and meteorological satellites. Neural networks tend to study a variety of data assimilation and hybrid models. (4) Global artificial intelligence-based rainfall prediction models and algorithms research results show that more emphasis is placed on deep learning algorithms Application trends.

Faecal calprotectin testing in UK general practice: a retrospective cohort study using The Health Improvement Network database

BackgroundFaecal calprotectin (FC) testing to detect inflammatory bowel disease (IBD) was recommended for use in UK general practice in 2013. The actual use of FC testing following the national recommendations is unknown.AimTo characterise the use of FC testing for IBD in UK general practice.Design and settingA retrospective cohort study of routine electronic patient records from The Health Improvement Network database from UK general practice.MethodThe study included 6 965 853 adult patients (aged ≥18 years), between 2006 and 2016. FC test uptake, the patients tested, and patient management following testing were characterised.ResultsA total of 17 027 patients had 19 840 FC tests recorded. The mean age of tested patients was 44.2 years. The first FC tests were documented in 2009. FC test use was still increasing in 2016. By 2016, 66.8% (n = 493/738) of practices had started FC testing. About one-fifth (20.7%, n = 1253/6051) of tests were carried out in patients aged ≥60 years. Only 7.8% (n = 473/6051) of the FC test records were preceded by symptoms eligible for FC testing. Only 3.1% (n = 1720/55 477) of patients with eligible symptoms have received FC testing since the national recommendations were published. There was only a small number of patients with symptoms, FC test, and a IBD diagnosis. In total, 71.3% (n = 1416/1987) of patients with a positive and 47.7% (n = 1337/2805) with a negative FC test were referred or further investigated.ConclusionUptake of FC testing in clinical practice has been slow and inconsistent. The indication of non-compliance with national recommendations may suggest that these recommendations lack applicability to the general practice context.

TP6.2.12Prognostic significance of symptomatic presentation in patients with Gastro-Intestinal Stromal Tumours

Background Gastro-Intestinal Stromal tumours (GISTs) are rare neoplasms of mesenchymal origin that arise in the GI tract and account for approximately 0.1-3% of all GI tumours. The aim of this study was to determine whether the mode of clinical presentation of GISTs has an effect on outcome. Methods Consecutive 143 patients diagnosed with GISTs were identified from the regional network database (median age 67 (21-88) years, 85 male). Data were collected on presenting symptoms, tumour, treatment undertaken, and the primary outcome measure was survival. Results The commonest GIST anatomical sites were stomach (69.9%), small bowel (13.3%), and colon (4.9%). Commonest symptoms were abdominal pain (30.0%), dyspepsia (11.8%), and anaemia (11.1%). Symptomatic presentations were associated with older age (p = 0.056) and higher mitotic index (>5 per 50HPF, p = 0.045). On univariable analysis the factors associated with cumulative 10-year survival were age (p = 0.076), the presence of symptoms (78.1% vs. 96.4%, p = 0.028), mitotic index (p = 0.011), and modified National Institutes of Health (NIH) prognostic index (p = 0.010), but not deprivation index, anatomical site or GIST diameter. Symptomatic patients aged over 60 years with mitotic index >5 per 50 HPF had the poorest 10-year survival at 63.6% (HR 2.577, 95%CI 1.132-5.867). On multivariable analysis, NIH index (HR 4.283, 95%CI 1.395-13.149, p = 0.011) and age (HR1.061, 95%CI 1.006-1.118, p = 0.029) were independently associated with 10-year survival. Conclusions Age, symptoms, and NIH criteria represent the most important prognostic biomarkers in patients diagnosed with GISTs.

The risk of COVID-19 in survivors of domestic violence and abuse

A ‘shadow pandemic’ of domestic violence and abuse (DVA) has emerged secondary to strict public health measures containing the spread of SARS-CoV-2. Many countries have implemented policies to allow the free movement of DVA survivors in attempts to minimise their exposure to abusive environments. Although these policies are well received, as a result there is a possibility of increased COVID-19 transmission within this vulnerable group who are not currently prioritised for vaccination. Therefore, we aimed to compare the risk of developing suspected or confirmed COVID-19 in women (aged over 16 years) exposed to DVA against age-sex-matched unexposed controls, following adjustment for known COVID-19 risk factors. A population-based retrospective open cohort study was undertaken between the 31 January 2020 and 28 February 2021 using ‘The Health Improvement Network’ database. We identified 10,462 eligible women exposed to DVA who were matched to 41,467 similarly aged unexposed women. Following adjustment for key covariates, women exposed to DVA were at an increased risk (aHR 1.57; 95% CI 1.29–1.90) of suspected/confirmed COVID-19 compared to unexposed women. These findings support previous calls for positive policy action improving DVA surveillance and prioritising survivors for COVID-19 vaccination.