Skip to main content

Pre-Caching Images

Overview​

Caching is a vital feature in OmegaAI designed to improve the efficiency of image retrieval and enhance DICOM viewing performance. This functionality is implemented through two main methods: OmegaAI Link Caching and Auto-Precache using the browser. Each method optimizes data access differently to suit varying user needs and infrastructure setups.

Link Caching accelerates data retrieval by storing pixel data locally as it is received, which avoids repeated fetches from the server, a process that can be both costly and slow.

How it Works​

  • Identification of Local Link: The system uses the public IP address to identify local Links where OmegaAI is installed.

  • Storage of Pixel Data: To protect patient health information (PHI), pixel data is encrypted using the AESGCM algorithm before storage on the local disk.

  • Running the Link Server: A secure web server is initiated with the start of the link, and a domain certificate (HTTPS) is attached to ensure secure data transmission.

This advanced feature is designed to further speed up the application usage through predictive caching mechanisms.

Steps Involved​

  • Login Trigger: Upon user login, the system begins auto-precache operations.

  • Data Request and Caching: If a link is available, it requests specific pixel data. The Link checks if this data is already cached; if not, it retrieves and caches it from the server. Subsequently, this cached data is quickly available for the next user access.

Auto-Precache Using Browser​

This method targets a broader user base and is activated upon opening a study. It aims to provide predictive caching to facilitate immediate access to necessary images.

Caching Methodology:​

  • Selection of Studies: The system selects the top-listed studies in the worklist based on the set priority or default order.

  • Batch Caching: Initially, the first five studies are cached. Once complete, the next batch of studies (6-10) is processed, with a limitation to cache up to 15 studies.

  • Background Caching: When a user logs in, all assigned studies are cached in the browser in priority order using a web-worker. This includes downloading the studies and any relevant prior images in the background.

  • Purge Management: The browser automatically manages the purging of cached data.