P8 Big data platform for cardiovascular healthcare data storage and data viewer. (May 2019)

• Record Type: Journal Article
• Authors: Navarro-Paredes, Cesar; Jing, Min; Finlay, Dewar; McLaughlin, James
• Is Part Of: Heart. Volume 105(2019)Supplement 4
• Page Start: A8
• Page End: A8
• Publication Date: 2019-05
• Subjects: Heart -- Diseases -- Treatment -- Periodicals; Cardiology -- Periodicals; 616.12
• DOI: 10.1136/heartjnl-2019-SCF.16
• Languages: English
• ISSNs: 1355-6037

Abstract

Abstract : With increasingly generated big data in the healthcare industry, it is vital to develop the platforms for more effective capture, storage and manipulation of such large volume, high velocity and variable data. A central database has been established for cardiovascular healthcare, which enables the data storage and data transformation between the hospital and data analysis centre with a PHP data viewer. The data are collected from sensors and devices in general applied to patients, which are organised and can be searched according to four fields: projects, events, patients and recordings. Data organised in events englobe a session where the patient was assessed. Different events (or sessions) can form a project and the same event could belong to different projects. Each event contains recordings and each recording can be a time series, an image, a video or a punctual datum, such as, video taken for patient's eye conjunctival, Electrocardiogram (ECG), temperature, blood pressure or respiration rate. The recordings can have associated annotations for classifications, diagnoses, and additional files such as images, reports or notes. Data are stored within the university servers following the approved guidelines on availability and security of the data. Only authorized users can access the database. The data can be accessed by a PHP viewer application which resides in the university servers, in which login and password are required. The user can navigate and search data by four fields. It is also possible to upload retrospective data into the database through the PHP viewer (to create events). Data can be also uploaded live by Android apps which capture data from a sensor or directly from electronic boards (controlling the sensors) connected via Wi-Fi. Figure 1 (a) shows the relational model of the database, (b) lists different recordings for an event and (c) shows a time series data (ECG). Further extension of the platform will enable the algorithms, trend analysis and analytical methodologies to be developed and operated, which will form the basis of a static and cloud-based platform solution for cardiovascular healthcare research and applications. This project has been funded by the EU's INTERREG VA programme, managed by the Special EU Programmes Body (SEUPB).