Technical Platform

Overview:

The EHPDCL technical platform (the DCL platform) offers a secure environment for researchers to explore healthcare data and conduct data analysis. Designed to support the full research lifecycle, it enables seamless data processing, application coding, model training, and data validation.

Built on a multi-tenancy architecture within an isolated network, our platform ensures strict data segregation and protection. Researchers interact with data exclusively through the DCL platform, with no ability to copy, download, or export data. File imports and exports of research outcomes are managed by the EHPDCL Office to maintain data security.

Hardware and Software:

For the EXPERT service, data is stored as CSV files within the Hadoop clusters. The DCL platform offers dedicated hardware resources, such as virtual machines and GPU resources, allowing researchers to access the data and perform analysis effectively.

To facilitate a wide range of research activities, the DCL platform also provides a suite of commonly used freeware and open-source software tools, including Python, R, DICOM viewers, and office productivity applications, among others.

Import Service:

Researchers may request to import their own materials, such as datasets, programme scripts and software tools into DCL platform to facilitate their research. All imports are subject to review by EHPDCL Office for compliance with data security, privacy and other requirements. In general, the materials for import should

• be legally owned, properly licensed or clearly identified as free to use and have no copyright or licensing issues;
• contain no sensitive information or information that can lead to re-identification of individuals, households, or organisations;
• be compatible with DCL platform’s infrastructure and contains no computer viruses or harmful programmes.