Incremental Open Source Architecture Supporting a High Availability National PACS
Clinical information systems have evolved over the last years to accommodate the growing needs of digital medical imaging. Pixel data is stored alongside patient metadata on large data centers that can be remotely accessed by several users at the same time during diagnosis and screening. These systems need to be performant so as not to impair doctors during their work and provide a smooth and pleasant user experience. At the same time, these installations have grown in scale, containing both more imaging modalities and more users and institutions under the same installation, generating a lot of traffic on the information systems, with thousands of images arriving each second to the imaging archives. Achieving good performance under such large scales is a considerable challenge. This work presents a multi-node architecture tested in a production environment designed to support a nationwide imaging archive, with 150 connected clinics and 2000 daily active users. The work uses exclusively open-source components, describing in detail the challenges met during the implementation of the system and drawing comparisons with a single node installation.
Authors
- Eduardo Pinho (ORCID: https://orcid.org/0000-0002-4397-8792)
- Carlos Costa (ORCID: https://orcid.org/0000-0002-2707-5331)
- Rui Jesus (ORCID: https://orcid.org/0000-0002-9231-6744)
- Tibério Baptista
- Luís Bastião Silva
Institutions
- Universidade da Coruña (ES)
- Critical Software (Portugal) (PT)
- University of Aveiro (PT)
Publication Details
- Published
- 2025-06-18
- DOI
- https://doi.org/10.1109/ichi64645.2025.00058
- Primary Topic
- Distributed and Parallel Computing Systems
- Type
- article
- Field-Weighted Citation Impact
- 0.00