China
Africa
Explore chapters and articles related to this topic
Software and Technology Standards as Tools
Published in Jim Goodell, Janet Kolodner, Learning Engineering Toolkit, 2023
Jim Goodell, Andrew J. Hampton, Richard Tong, Sae Schatz
Data architectures address the structure of data and the capabilities of associated data platforms. Data platforms are the various components required to acquire, store, prepare, deliver, and manage your data (along with the requisite security).10 Most data platforms include one or more databases, including relational databases (for example, built using SQL) and / or non-relational or NoSQL databases (such as object databases, graph databases, document stores, key / value stores, triple / quad stores, and hybrid platforms). The different database types each have strengths and weaknesses. For instance, SQL needs predefined schema and structured data, while NoSQL can handle dynamic schema and unstructured data. The different database types also scale and perform differently, depending on how they’re accessed and how data are structured within them.11 For example, object databases are a convenient choice for applications built using object query languages, and they can handle complex relationships between objects. Meanwhile, graph databases excel at managing highly connected data and complex queries where the relationships between data elements are as important to the data elements themselves.
Role of Knowledge Graphs in Analyzing Epidemics and Health Disasters
Published in Adarsh Garg, D. P. Goyal, Global Healthcare Disasters, 2023
A graph database is a type of NoSQL database (a NoSQL database does not store data in the form of tables as in traditional relational databases). It stores data in the form of nodes and edges, nodes store the data items and edges store the relationships between these nodes. It preserves the structure of a connected or interlinked dataset in the form of entities, their properties, and the relationship among them. Graph database is recommended for a connection-based dataset, when one requires to track the links between the data items.
The Value of Health Data for Patients
Published in Disa Lee Choun, Anca Petre, Digital Health and Patient Data, 2023
But privacy-preserving technologies can take us step even further by reducing the need to share data with researchers altogether. Today, data is duplicated time and time again in multiple databases depending on the needs. When data is duplicated, errors can occur, and quality can be deteriorated. The more data changes hands, the more risks there are to alter it. So, instead of duplicating and sending data to third parties, why couldn’t we allow these third parties to send their processing tools where the data is. These tools are called federated technologies. They allow researchers to analyze data (federated analytics) or train their machine learning models (federated learning) across many devices and databases without having to duplicate and centralize data. For instance, researchers can apply data science methods on datasets that are locally stored in a hospital or a patient’s device without requesting a copy of the dataset.
Machine learning approaches to improve disease management of patients with rheumatoid arthritis: review and future directions
Published in Expert Review of Clinical Immunology, 2021
Joanna Kedra, Thomas Davergne, Ben Braithwaite, Hervé Servy, Laure Gossec
Technical issues are related to the fact that huge amounts of validated data are required to properly train ‘supervised’ ML models – which represent most ML methods used nowadays [100]. Indeed, depending on the physician or research team, disease parameters are not assessed in the same way (e.g. disease activity can be assessed in 28 or 44 joints, or using erythrocyte sedimentation rate or C reactive protein), which results in heterogeneous data. A crucial and time-consuming data management process is therefore required to homogenize the data, and check if the data are adequately computed in the database. A potential solution would be to standardize the way data are computed by the means of recommendations for good research practice [103,105], or to develop validated ‘equivalence’ scales to homogenize data collected in different ways for a same disease parameter.
Use of Routinely Collected Registry Data for Undergraduate and Postgraduate Medical Education in Denmark
Published in Journal of European CME, 2021
Kasper Bonnesen, Cecilia Hvitfeldt Fuglsang, Søren Korsgaard, Katrine Hjuler Lund, Natascha Gaster, Vera Ehrenstein, Morten Schmidt
Administrative databases register individuals from a certain geographic area or attending a certain health service (e.g. hospital department or out-patient clinic). The CRS is an administrative database. Other types of information in Danish administrative databases include hospital encounters [13], prescription redemptions [14], and laboratory results [15]. Figure 2 displays examples of such databases. Health databases include, e.g. disease registries containing information on the time of diagnosis or treatment for a specific disease (e.g. the Danish Cancer Registry) [16], procedure registries containing information on time and type of procedure and other procedure-specific data (e.g. the Western Denmark Heart Registry) [17], and biological biobanks containing blood and tissue samples. Clinical quality databases aim to use clinical care data to improve treatment of specific diseases or clinical procedures, to improve management of specific departments, and for research [18,19]. Currently, the Danish Clinical Quality Program – National Clinical Registries (RKK) has listed 84 clinical quality databases [20] categorised into (1) heart/vascular, surgery, and emergency (e.g. the Danish in-hospital cardiac arrest registry) [21], (2) cancer and cancer screening (e.g. the Danish Colorectal Cancer Group Database) [22], and (3) psychiatry, gynecology/obstetrics, and chronic diseases (e.g. the Danish Depression Database) [23].
Radiation databases and archives – examples and comparisons
Published in International Journal of Radiation Biology, 2019
Alia Zander, Tatjana Paunesku, Gayle Woloschak
Descriptive work observing the effects of ionizing radiation dates back to the discovery of X-rays in 1895 (Hall and Giaccia 2012). Initially, observational notes were created from patient samples collected after outwardly adverse radiation events. As interest in the field grew, more accurate and detailed information describing the impact of ionizing radiation was necessary. Eventually, researchers and doctors recognized that there is a significant delay in expression of radiation pathologies following ionizing radiation exposures. This resulted in the collection of data and materials from radiation-exposed subjects who did not exhibit any noticeable and/or immediate complications. Data collection and storage methods gradually improved over time, and ultimately led to valuable archives that we still use today. In current terms, a dataset is the actual data that has been collected, while a database is an organized way to store datasets, typically controlled by a management system. Once all of the data have been collected and there will be no new alterations, it can be stored as a searchable archive for future reference and analysis. This review provides some examples of open and closed databases and archives available for studying the impact ionizing radiation has on health (Figure 1).